Molecular dynamics simulation analysis method for polycrystalline structure cascade collision

Abstract

The invention relates to a molecular dynamics simulation analysis method for cascade collision of a polycrystalline structure, and belongs to the technical field of metal materials and molecular simulation. The molecular dynamics simulation analysis method for cascade collision of the polycrystalline structure comprises the following steps: S1, constructing a polycrystalline structure model by adopting a Voronoi algorithm; s2, performing cascade collision tests of the simulation polycrystalline structure model under different incident energies and different incident directions, and observing irradiation damage behaviors and microstructure evolution characteristics of the polycrystalline structure; and S3, carrying out intragranular and grain boundary region separation on the microscopic structure of the polycrystalline structure model subjected to the cascade collision simulation test, and analyzing by adopting OVITO software so as to distinguish the body-centered cubic structure of the block region in the crystal, the crystal boundary region and the irradiation defect in the polycrystalline structure after cascade collision. A molecular dynamics method is adopted, and series cascade collision simulation is carried out on polycrystalline tungsten systems with different grain sizes at the temperature of 363K.

Description

Technical Field

[0001] The present application relates to the technical field of metal materials and molecular simulation, and in particular to a molecular dynamics simulation and analysis method for cascade collision of polycrystalline structures. Background Art

[0002] Metallic tungsten (W) is considered to be a promising plasma-facing material in high-temperature, high-load fusion environments due to its high melting point, excellent thermal conductivity, good mechanical properties, and chemical stability. However, plasma-facing materials in fusion reactors must withstand the violent impact of high-energy particles, and the degradation of their service performance and shortened life will seriously affect the safety and stability of fusion reactor operation. Therefore, under the current conditions where fusion neutron source experiments are still relatively scarce, cascade collision simulations of tungsten materials will help us further understand the impact of irradiation on the evolution characteristics of the microscopic defect structure of the material, which has important theoretical and practical significance.

[0003] There are many studies on the simulation of irradiation damage of tungsten materials in fusion environments. Setyawan et al. used the Large-scale Atomic Molecular Massively Parallel Simulator (LAMMPS) code to simulate cascade collisions of metallic tungsten with incident particle energies below 100 keV in a temperature gradient range of 300K to 2050K, and found that the number of Frankel defect pairs is highly dependent on the incident energy, but has no obvious dependence on the irradiation temperature. Liu et al. used molecular dynamics methods to build a more systematic single crystal tungsten cascade collision simulation database, and considered the effects of different incident energies, directions and temperatures. They found that at lower incident energies, statistically speaking, the impact of the incident direction and temperature on the distribution of defect numbers and sizes is not significant, but it does affect the type of dislocation loops. Fu et al. simulated cascade collisions of tungsten and tungsten-rhenium alloys with energies up to 300 keV. The results showed that the effect of rhenium atoms on the number of defects was small, but compared with the pure tungsten system, the movement of defect clusters and dislocation loops in the tungsten-rhenium alloy was suppressed after the cascade, indicating that rhenium has a pinning effect on dislocation loops. Zhang et al. used molecular dynamics methods to simulate cascade collisions in tungsten at different temperatures near the grain boundary. The results showed that the number of vacancy defects is independent of temperature. However, the above studies focus more on issues such as neutron irradiation damage of single-crystal tungsten materials in fusion and the influence of typical irradiation defects on the evolution of the material's microstructure.

[0004] In general, the properties of polycrystalline materials are usually better than their respective single crystal materials. Changes in grain size in polycrystalline materials can greatly affect the performance of the material. When the grain size reaches the nanoscale, the mechanical properties of the material can be greatly improved, which makes polycrystalline materials attractive for engineering applications in many fields. In the field of nuclear fusion, researchers have also conducted some studies on irradiation damage of polycrystalline materials. Gao et al. studied the irradiation damage behavior of polycrystalline silicon carbide with grains of different sizes. The results showed that with the increase of grain size, the number of defects generated inside the grains increased, and the efficiency of defect generation increased with the increase of grain size. Samaras et al. conducted a cascade collision simulation of polycrystalline nickel (Ni) materials. They found that there are multiple possible movement mechanisms of Ni self-interstitial in the process of moving to the grain boundary area, and conducted detailed research. Kong et al. studied the hardening effect caused by helium in polycrystalline. They found that injecting helium (He) into tungsten will cause a significant hardening effect. Its hardness first increases rapidly with the increase of He injection. When the injection is greater than 5×10 16 cm -2 When the hardness increases slowly. Du et al. used the classical molecular dynamics method to simulate the cascade collision in zirconia. The results showed that compared with the single crystal structure, the polycrystalline structure produced more dislocated atoms during the cascade collision process. Park et al. compared the irradiation cascade simulation in single crystal tungsten and polycrystalline. The results showed that the decrease rate of the number of self-interstitials (SIA) in single crystal tungsten was much faster than that in nano-polycrystalline tungsten, which means that the SIA and vacancy recombination rate in single crystal tungsten was faster. Levo et al. studied the radiation resistance of nano-polycrystalline Ni and nano-multinary Ni alloys. They found that compared with nano-polycrystalline Ni, nano-multinary Ni alloys can withstand higher irradiation doses. These studies show that compared with single crystals, the influence of polycrystalline structure on material properties and defect evolution behavior under irradiation conditions is very significant. Therefore, the simulation of irradiation damage of polycrystalline tungsten materials and the qualitative exploration of the influence of grain size, incident direction and energy on defect distribution and evolution during irradiation cascade are of great significance for understanding the performance changes of related structural materials under irradiation environment. Summary of the invention

[0005] In view of this, the present application provides a molecular dynamics simulation analysis method for cascade collision of polycrystalline structures. The present application adopts the molecular dynamics method to first establish a model of polycrystalline tungsten with different grain sizes, and further studies the irradiation damage behavior and microstructure evolution characteristics under the incident energy of 10 to 150 keV and different incident directions, and analyzes, counts and discusses the information such as defect formation, cluster distribution and structural evolution during the irradiation process in detail. The research results can not only further enrich and improve the understanding of the microstructure behavior characteristics and damage mechanism of polycrystalline tungsten under irradiation environment, but also provide an important theoretical basis for the optimization and application of tungsten-based plasma materials in fusion environment, which is of great significance and can effectively overcome the defects of the above-mentioned prior art.

[0006] The present application provides a molecular dynamics simulation analysis method for cascade collision of polycrystalline structures, comprising the following steps:

[0007] S1, using Voronoi algorithm to build a polycrystalline structure model;

[0008] S2. Conduct cascade collision tests on simulated polycrystalline structure models under different incident energies and different incident directions to observe the radiation damage behavior and microstructure evolution characteristics of the polycrystalline structure;

[0009] S3. The microstructure of the polycrystalline structure model after the simulated cascade collision test is separated into the intracrystalline and grain boundary regions, and then analyzed using OVITO software to distinguish the body-centered cubic structure of the bulk region inside the crystal, the crystal interface region, and the irradiation defects in the polycrystalline structure after the cascade collision.

[0010] Preferably, the S1 specifically comprises the following steps:

[0011] S11. Determine key parameters: Calculate cell side length (l x / l y / l z ) and the number of grains N included;

[0012] S12, determining the center positions of N grains and the corresponding rotation matrices;

[0013] S13, establishing a single crystal, rotating it according to the rotation matrix and translating it to the corresponding center position;

[0014] S14, judging the grain to which each atom belongs based on the distance from the atom to the center of the grain;

[0015] S15. For the constructed grain boundary, determine the distance between atoms at the grain boundary and delete the overlapping atoms;

[0016] S16. Complete the construction of polycrystalline grains and output the corresponding position file.

[0017] Preferably, in step S12, the determining of the center positions of the N grains is achieved by a random number method.

[0018] Preferably, the specific process of the random number method is:

[0019] Given the lengths l of the three sides of the computational cell x , l y , l z After that, the average size of each grain is first approximated by using the cubic structure or spherical structure method, thereby obtaining the average side length of each grain under the cubic structure approximation condition, or the radius length of the spherical grain under the spherical structure approximation condition; after determining the relevant key size parameters of the grain, the center position of each grain is obtained by random numbers.

[0020] Preferably, the calculation formula for the average side length of each grain under the cubic structure approximation condition is as follows:

[0021]

[0022] Preferably, the radius length calculation formula of the spherical grains under the spherical structure approximation condition is as follows:

[0023]

[0024] Preferably, the specific process of obtaining the center position of each grain by random numbers is:

[0025] For grains with cubic structure, it is assumed that the distance between grains is not less than 2D; for grains with spherical structure, it can be assumed that the distance between grains is not less than the diameter 2R; the assumptions are all first-order approximations. Based on this assumption, a polycrystalline structure with relatively uniformly distributed grain positions can be constructed. After determining the relevant key size parameters of the grains (such as the diameter, radius, shape, etc. of the grains), set the spatial range of the grain distribution area, and then select a random number generator (such as Python's random module) to generate the center coordinates of the grains using an appropriate random number generation function; according to the required situation, randomly generate coordinates (x, y, z) within the specified range to ensure that these coordinates conform to the size and shape of the grains.

[0026] Preferably, the specific process of the rotation matrix in step S12 is:

[0027] According to the definition of grain boundary freedom, the grains are rotated at any angle around the three main axes to achieve free rotation in three-dimensional space and construct the corresponding grain boundaries; according to the commonly used rotation matrix, the given angle θ is rotated around the Z axis to obtain the corresponding rotation matrix as follows:

[0028]

[0029] Therefore, when the entire grain rotates around three given directions in three-dimensional space by α, β and γ, the rotation matrix of the three-dimensional space can be obtained as follows:

[0030]

[0031] Preferably, the specific process of step S14 is:

[0032] After determining the corresponding grain center position pi, the Voronoi method can be used to give the position of the mid-vertical plane between the centers of two grains, confirming all p i The area enclosed by the following formula satisfies the range r i :

[0033] r i ∈{x,y,z|d(r i ,P i ) <d(r i ,P j ),j={1,2,...n},j≠i}

[0034] Among them, r i is the position vector, d(r i ,p i ) is the distance from a given position to the center of the grain. Therefore, according to the above formula, the grain at a given position can be confirmed to which it belongs, thereby realizing the initial establishment of a polycrystalline periodic structure.

[0035] Preferably, step S1 further comprises: after the polycrystalline structure is constructed, firstly, a molecular statics method is used for relaxation, and then a molecular dynamics method is used for dynamic equilibrium relaxation to obtain a corresponding stable structure.

[0036] Compared with the prior art, this application has the following beneficial effects:

[0037] This application uses molecular dynamics methods to simulate a series of cascade collisions at 363K for polycrystalline tungsten systems with different grain sizes, including different primary collision atom (PKA) directions and PKA energies of 10 to 150 keV. The results show that grain size, PKA direction and energy all have an impact on the formation and evolution of defects during cascade collisions. Smaller grain sizes will lead to more defects forming at grain boundaries. Under high energy, systems with small grain sizes have a greater probability of distortion in the crystal interface region, and this phenomenon becomes more obvious with the increase of PKA energy. The PKA direction affects the formation and diffusion path of defects. When PKA is incident perpendicular to the crystal interface, defects will preferentially form in the area near the crystal interface, otherwise they will tend to form inside the grain. The research results are of great significance for understanding the performance changes of polycrystalline tungsten in a high-energy fusion neutron environment, and provide theoretical guidance for the further optimization and application of tungsten-based plasma-facing materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the description of the present application or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 To construct a polycrystalline computing cell algorithm flow chart based on the Voronoi algorithm;

[0040] Figure 2 is a schematic diagram of the polycrystalline tungsten structure, where Figure 2 (a) indicates that the number of grains is 4, Figure 2 (b) indicates that the number of grains is 12;

[0041] Figure 3 The defect configuration after polycrystalline cascade in different PKA directions, where the red enlarged atoms are defects and the white atoms are crystal interfaces;

[0042] Figure 4 The defect configuration distribution after cascading of different grain size systems (50keV, PKA vertical incidence as an example);

[0043] Figure 5 is the number of defects that survive under different PKA energies and incident directions, where Figure 5 (a) indicates 4Grian, Figure 5 (b) indicates 12Grian;

[0044] Figure 6 The number and size distribution of vacancy clusters under different incident directions for a polycrystalline model containing four grains;

[0045] Figure 7 The number and size distribution of vacancy clusters under different incident directions for a polycrystalline model containing 12 grains;

[0046] Figure 8 is a snapshot of the defect distribution of the 4Grian system when PKA is incident parallel to the crystal interface at 100 keV energy, where Figure 8 (a) is the initial model, Figure 8 (b) is the collision stage, Figure 8 (c) is the heat peak stage, Figure 8 (d) is the quenching stage, Figure 8 (e) is the final stable stage, Figure 8 (f) shows the grain boundary distortion in the stable phase, where atoms are colored by their potential energy. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0048] Unless otherwise specified, the experimental methods used in the examples of this application are all conventional methods.

[0049] In the following examples, unless otherwise specified, all raw materials can be obtained by commercial purchase or conventional methods.

[0050] Example

[0051] 1. Model and Method

[0052] The establishment of polycrystalline structure is mainly realized by Voronoi algorithm, the algorithm diagram is as follows Figure 1 According to this algorithm, the single crystal modeling is completed first, and then the single crystal is rotated according to the given rotation matrix. The translation process uses the Voronoi method to solve the key problem of determining the atomic belonging to the grain between two single crystal pairs, and the periodic method is used to achieve the establishment of the polycrystalline structure.

[0053] Among them, the position of the Voronoi grain center is mainly achieved through the random number method, and the corresponding judgment criterion is closely related to the number of grains (N) in the calculation cell.

[0054] For example, given the lengths l of the three sides of the computational cell x , l y , l zAfter that, the average size of each grain can be approximated by using the cubic structure or spherical structure method to obtain the average side length of each grain under the cubic structure condition, or the radius length of the spherical grain under the spherical structure approximation condition:

[0055] Cubic structure approximation:

[0056] Spherical structure approximation:

[0057] Therefore, after determining the relevant key size parameters of the grains, the center position of each grain can be obtained by random numbers. For cubic grains, it can be assumed that the distance between grains is not less than 2D, and for spherical grains, it can be assumed that the distance between grains is not less than the diameter 2R. It should be noted that this assumption belongs to the first-order approximation, based on which a polycrystalline structure with relatively uniformly distributed grain positions can be constructed.

[0058] In determining the corresponding grain center position p i Finally, the Voronoi method can be used to determine the position of the mid-vertical plane between the centers of two grains and confirm that all p i The area enclosed by the following formula satisfies the range r i :

[0059] r i ∈{x,y,z|d(r i , P i )<d(r i ,P j )={1,2,...n},j≠i} (3)

[0060] Among them, r i is the position vector, d(r i ,p i ) is the distance from a given position to the center of the grain. Therefore, according to the above formula, it is relatively easy to confirm to which the grain at a given position belongs, thereby realizing the initial establishment of a polycrystalline periodic structure.

[0061] like Figure 1 The figure shows the algorithm flow chart for constructing polycrystalline computational cells based on the Voronoi algorithm. In this algorithm, the operation on the grains is mainly realized by the method of rotation matrix. According to the definition of grain boundary freedom, the grains can be rotated at any angle around the three main axes, so as to achieve free rotation in three-dimensional space and construct the corresponding grain boundaries. According to the commonly used rotation matrix, for example, rotating around the Z axis by a given angle θ, the corresponding rotation matrix obtained is:

[0062]

[0063] Therefore, when the entire grain can be rotated around three given directions α, β and γ in three-dimensional space, the rotation matrix in three-dimensional space can be obtained:

[0064]

[0065] According to the above algorithm, we first established Figure 2 The polycrystalline tungsten structure shown in Figure 2 (a) indicates that the number of grains is 4; Figure 2 (b) indicates that the number of grains is 12.

[0066] In order to better understand the effect of different grain sizes on the irradiation displacement cascade, this application established two polycrystalline systems with the same side length simulation box containing different numbers of grains (i.e., calculation cells with different grain sizes), such as Figure 2 As shown, Figure 2 (a) The calculated cell contains 4 grains (4Grain), with an average grain size of 20.0nm and a total of 1,993,041 atoms. Figure 2 (b) contains 12 grains (12Grain), with an average grain size of 13.9nm and a total of 1,989,660 atoms. Periodic boundary conditions were used in the construction process. After the polycrystalline model was established, the molecular statics method was first used for relaxation, and then the molecular dynamics method was used for dynamic equilibrium relaxation to obtain the corresponding stable structure, which provided the basis for the subsequent displacement cascade simulation.

[0067] The temperature of the cascade collision simulation is 363K. In order to make the cascade collision occur in the center area of ​​the box as much as possible, the primary knock-on atom (PKA) is randomly selected at a distance of about 5 nanometers from the crystal interface inside the most central grain of the system, and its direction considers two directions, vertical and parallel to the crystal interface. In addition, the PKA energy first considers low-energy 10keV, 20keV, 35keV and 50keV, and then further considers medium and high energy 100keV and 150keV. Because the cascade collision process has a large randomness, this application simulates the cascade collision process more than 10 times for each case to reduce errors and contingency. In this application, based on the particularity of the crystal interface structure, the statistical definition of the number of defects is to remove the distance of about 7 angstroms near the crystal interface and count the defects in the bulk area inside the grain. The cascade collision model and condition settings are shown in Table 1.

[0068] Table 1 Cascade collision model and condition settings

[0069]

[0070] The interatomic potential function is the FS type potential function independently developed by this application. In order to better simulate the short-range interaction between atoms in the cascade process, this application also performs Ziegler-Bier sack-Littmark (ZBL) correction on it. The detailed connection function form and corresponding parameters can be found in the literature (Liu LX, Li XC, Chen YC, et al. Evaluation of tungsten interatomic potentials for radiation damage simulations [J]. Tungsten, 2020, 2: 3-14). The cascade process uses a variable time step and is controlled at 10 -3 and 10 - 7ps, and the total simulation time is selected to be 40 to 120 ps according to the different PKA energy values. The specific average grain size, simulation box size, PKA energy and other parameter settings of different systems are shown in Table 1. In addition, this application also conducted a small amount of high-energy 200 keV cascade simulation tests. The type and number of point defects generated in the irradiation cascade process are analyzed and identified using the Winger-Seitz (WS) primitive cell method (Nordlund K, Averback R S. Point defect movement and annealing incollision cascades [J]. Physical Review B, 1997, 56 (5): 2421). The common neighbor analysis (CNA) method (Honeycutt JD, Andersen H C. Molecular dynamics study of melting and freezing of small Lennard-Jones clusters [J]. Journal of Physical Chemistry, 1987, 91 (19): 4950-4963) can be used to analyze the local crystal structure around atoms, so as to distinguish the body-centered cubic (BCC) structure of the bulk region inside the grain, the crystal interface region (CNA judged to be other types) and the irradiation defects in the crystal after the cascade. Both methods can be implemented through the visualization and data analysis software OVITO (Open Visualization Tool) (Stukowski A. Visualization and analysis of atomic simulation data with OVITO-the Open Visualization Tool[J]. Modelling and simulation in materials science and engineering, 2009, 18(1):015012).

[0071] 2. Results and Discussion

[0072] 2.1 Origin of Defects

[0073] The distribution configuration of defects after cascade in different PKA directions is as follows Figure 3 As shown, this application takes 12 grains and a PKA energy of 10 keV as an example. When the PKA direction is parallel to the grains (P) and perpendicular to the grains (V), respectively, Figure 3 (a)-(c) and Figure 3 (d)-(e) As shown. When the PKA direction is parallel to the grain, at the thermal peak stage, Figure 3 As shown in (a), each atom is colored by its potential energy. For convenience, only the crystal interface and defect atoms are shown in the figure. It can be seen that at 10keV energy, defects are basically only generated inside the grains, and most defects will not pass through the crystal interface. The edge of the cascade region may overlap with the crystal interface region; the structural morphology of the final stable stage is as follows Figure 3 (b) shows that the red magnified area is the defect atom and the white area is the crystal interface; Figure 3 (c) The defective atomic part is magnified and each atom is colored by its potential energy. The number of vacancies inside the grains in the stable stage is generally greater than the number of interstitials, which means that some interstitials are captured by the grain boundary region during the cascade process, making it impossible for them to annihilate and recombine with the vacancies inside the grains.

[0074] When the PKA direction is perpendicular to the grain, the cascade thermal peak stage is as follows Figure 3 As shown in (d), at this time, the possibility of defects passing through the crystal interface is greater, so that more defects can be absorbed by the grain boundary absorption traps in the polycrystal. The defect morphology in the final stable stage is as follows: Figure 3 As shown in (e) and (f), the results show that the defects remaining in the inner area of ​​the grain are relatively few in general, especially the interstitial defects. Figure 3 (a)-(c) PKA parallel incidence and Figure 3 (d)-(e) It can be seen from the vertical incidence of PKA that when PKA is incident perpendicular to the crystal interface, the final number of defects inside the grain is generally slightly less than the number of defects inside the grain when PKA is incident parallel to the crystal interface. This may be related to the large overlap between the cascade region and the crystal interface region when PKA is incident perpendicular to the crystal interface. The slightly larger overlap between the cascade region and the crystal interface region makes it easier for more defects to be captured by the grain boundary absorption trap.

[0075] At the same time, this application further compares the defect distribution configurations of the polycrystalline cascade thermal peak stage and the final stable stage of two different grain sizes, 4Grian and 12Grian, taking the PKA energy of 50keV and the PKA direction perpendicular to the crystal interface as an example. Figure 4As shown. At 50keV energy, defects in the thermal peak stage will basically pass through the adjacent crystal interfaces. In the 12Grian polycrystalline system, because there are a large number of crystal interfaces inside the system, the defects in the thermal peak stage of the cascade process pass through more crystal interfaces, that is, the overlap between the cascade region and the crystal interface region is large, and a large number of defects will be quickly absorbed by the crystal interface region, so that the number of defects remaining in the bulk region inside the grain in the final stable stage of the 12Grian system is less than that in the 4Grian system, and interstitial defects are preferentially captured by the crystal interface, and most of the defects remaining inside the grain are vacancy defects. In addition, compared Figure 3 and Figure 4 It can be found that with the increase of PKA energy, the possibility of defects passing through the crystal interface during the thermal peak stage increases.

[0076] 2.2 Distribution of defects and defect clusters

[0077] Figure 5 (a) and Figure 5 (b) The number of defects surviving the post-cascade stable stage varies with PKA energy at different PKA energies of 10 to 150 keV for 4-grain and 12-grain systems, when the incident directions are parallel (P) and perpendicular (V) to the crystal interface, respectively. It can be seen that in both the 4-grain and 12-grain systems, the number of vacancies (vac) in the bulk of the grains is much greater than the number of interstitials (int), and the number of interstitials in the bulk region is almost negligible when the PKA energy is less than 35 keV. This indicates that interstitials are more easily absorbed by the grain boundary region, which is consistent with the results of many previous model studies containing symmetrical tilted grain boundaries (Samaras M, Derlet PM, Van Swygenhoven H, et al. Computer simulation of displacement cascades in nanocrystalline Ni [J]. Physical review letters, 2002, 88 (12): 125505). The strong absorption of interstitials by grain boundaries makes most of the interstitials captured by the grain boundary region, so that the number of vacancies remaining in the bulk region is much greater than the number of interstitials. At the same time, with the increase of PKA energy, the number of defects generally shows an increasing trend, which is consistent with the phenomenon in single crystal tungsten. And with the increase of PKA energy, the number of crystal interfaces passed through by the cascade region in the thermal peak stage and the number of defect atoms captured by the grain boundary region also increase significantly.

[0078] In addition, when the PKA direction is parallel to the crystal interface, the number of defects is greater than that when the PKA direction is perpendicular to the crystal interface. This may be because when the PKA direction is parallel to the crystal interface, most of the defects are generated inside the grain and do not pass through the crystal interface. Therefore, the defects do not fully interact with the crystal interface, resulting in a larger number of defects remaining inside the grain in this case. Figure 5 (a) and Figure 5 (b) It can be seen that in the 12Grian system, the number of defects that survive the cascade is less than that in the 4Grian system; that is, in the simulation box with the same side length, the number of defects that survive the cascade inside the grains decreases as the number of grains increases (that is, the size of the grains decreases). This may be because the system with a large number of grains has more crystal interfaces in the material, so its ability to absorb defects is stronger and faster. This phenomenon is consistent with the previous research results of Gao et al. (Gao F, Chen D, Hu W, et al. Energy dissipation and defect generation in nanocrystalline silicon carbide [J]. Physical Review B, 2010, 81 (18): 184101).

[0079] Will Figure 5 The number of defects in the polycrystalline W system after cascading shown in the table is compared with the previous research results in the single crystal W system (Liu L, Qiu R, Chen Y, et al. Displacement cascades database from molecular dynamics simulations in tungsten [J]. Journal of Nuclear Materials, 2023, 580: 154415) as shown in Table 2. It can be found that the number of defects that survived in the bulk region inside the grain in the polycrystalline W system, whether gaps or vacancies, is less than that in the single crystal W system. This shows that during the cascade process, the crystal interface of the polycrystalline material will not only absorb a large number of gaps, but also absorb vacancies; however, due to the strong interaction between the crystal interface and the gap, the gap will be preferentially absorbed by the grain boundary region, so that the number of gap defects in the bulk region (inside the grain) in the polycrystalline model is much smaller than the number of gaps in the single crystal. At the same time, in the polycrystalline system, since the number of interstitial defects absorbed by the grain boundary region is much greater than the number of vacancies, many vacancies cannot be annihilated and recombined with the interstitial defects and exist in the bulk region inside the grain in the form of point defects or defect clusters. The number of defects surviving at different PKA energies in single crystal W is shown in Table 2.

[0080] Table 2 Number of defects surviving at different PKA energies in single crystal W

[0081] <![CDATA[E PKA ]]> 10keV 20keV 35keV 50keV 80keV 100keV 150keV FPs 17(5) 25(8) 41(12) 58(15) 101(31) 124(39) 235(113)

[0082] (Note: The values ​​in brackets are standard deviations)

[0083] Based on the above results, the distribution of defect clusters in this application focuses on the statistics of vacancy clusters. According to previous studies (J. Fu, Y. Chen, J. Fang, N. Gao, W. Hu, C. Jiang, H.-B. Zhou, G.-H. Lu, F. Gao, H. Deng, Molecular dynamics simulations of high-energy radiation damage in W and W–Realloys, J. Nucl. Mater. 524 (2019) 9–20), the vacancy defect cluster cutoff distance criterion in this application is selected as the second nearest neighbor. Figure 6 and Figure 7 The number and size distribution of vacancy clusters in the 4Grian and 12Grian polycrystalline models under different incident directions are shown respectively. It can be seen that the number of vacancy clusters in both systems is greater when the PKA direction is parallel to the crystal plane than when the PKA direction is perpendicular to the crystal plane, and the number of defect clusters generally increases with the increase of PKA energy. Figure 6 and Figure 7 It can be found that, in general, the number of larger vacancy clusters in the 12Grian system is greater than that in the 4Grian system. The largest vacancy cluster observed in the 12Grian system contains 46 vacancies, and the largest vacancy cluster observed in the 4Grain system contains 34 vacancies. In addition, although the total number of vacancies in the polycrystalline system with 12Grian grains is less than that in the 4Grain system, the number of vacancy clusters in the 12Grian system is greater than that in the 4Grain system, that is, the number of vacancy clusters increases with the increase in the number of grains in the system.

[0084] 2.3 Defect Evolution

[0085] Snapshot of defect evolution in polycrystalline tungsten during high-energy PKA cascade. Figure 8 As shown, this application takes the 4Grian system, PKA with energy of 100 keV and incident parallel to the crystal interface as an example. Figure 8 (a) is the initial model, at which the atoms on the crystal interface are arranged in an orderly manner. Figure 8 (b) is the initial stage of the cascade collision. The system reaches the state of Figure 8 (c) shows the thermal peak state, and then as time evolves, the cascade process enters the quenching stage as shown in Figure 8As shown in (d), at this time, some interstitials and vacancies begin to annihilate and recombine, and the defects near the grain boundary area will be quickly absorbed by the crystal interface, especially the interstitial defects. The configuration distribution snapshot of the final stable stage is shown in Figure 8 As shown in (e), most of the interstitial defects are eventually absorbed into the crystal interface region, resulting in more vacancy defects remaining in the bulk region. At the same time, the grain boundary region absorbs a large number of interstitial atoms and violent cascade collision impacts, causing the originally regularly arranged crystal interface to be distorted to a certain extent, as shown in Figure 8 In addition, as for the crystal interface near the cascade center, by comparing the width of the crystal interface in the final stable stage and the initial stage, it can be found that the width of the crystal interface increases after the cascade, which can further explain that a large number of interstitial atoms are absorbed by the adjacent crystal interface during the cascade process, causing a certain degree of migration and growth of the crystal interface, which may affect the service performance of the material.

[0086] This application further tests the cascade collision process of polycrystalline systems with different grain sizes at a higher energy of 200keV, and finds that the defect distribution morphology in the cascade process at 200keV energy is similar to that at 100keV energy. At the same time, with the increase of PKA energy, the number of defects generated in the thermal peak stage increases significantly, and the possibility of distortion or even partial migration of the crystal interface is greater.

[0087] In addition, the present application found that the grain size has an effect on the formation and evolution of defects during the cascade collision process. In the 12-grain system with a smaller grain size, more defects will be formed in the crystal interface area after the cascade, resulting in the overall distortion of the grain boundary in the stable stage being slightly greater than that of the 4-grain system. This may be due to the smaller barrier effect of the grain boundary on the diffusion of defects. By comparing the effects of different PKA incident directions on the cascade collision process, the present application found that the incident direction has a great influence on the way defects are formed and evolved. Specifically, the relationship between the incident direction and the crystal orientation of the crystal determines the formation and diffusion path of the defects. The present application observed that when PKA is incident perpendicular to the crystal interface, defects will preferentially form near the grain boundary, while when PKA is incident parallel to the crystal interface, they will tend to form inside the grain. The incident energy affects the rate of defect formation and evolution. Higher incident energy leads to more defect formation and faster defect diffusion rate, which may be due to the fact that the incident energy is sufficient to overcome the binding force inside the crystal.

[0088] In summary, in this application, through the cascade collision simulation study of polycrystalline tungsten, the influence of grain size, PKA direction and PKA energy on the cascade collision process is systematically explored. The generation and distribution of defects, the number of defects, the distribution information of defect clusters and the evolution of defect microstructure in the polycrystalline tungsten system during the neutron irradiation cascade collision are systematically analyzed and discussed. The research results are of great significance to the performance changes of polycrystalline tungsten in high-temperature and high-load fusion environments, and can provide a basis for understanding the long-term evolution of defect microstructures during irradiation damage, and provide a reference for the application and optimization of W-based plasma.

[0089] The main conclusions of this application are as follows: (1) interstitial atoms are preferentially absorbed by the crystal interface region, and the defects remaining in the bulk region inside the grain are mostly vacancy type, and their number tends to increase with the increase of PKA energy; (2) with the increase of the number of grains, that is, the decrease of grain size, the number of defects in the bulk region inside the grain in the stable stage tends to decrease, while the defects formed in the crystal interface region tend to increase; (3) the incident direction of PKA affects the formation and diffusion path of defects. When PKA is incident perpendicular to the crystal interface, the number of defects in the bulk region inside the grain is less than when PKA is incident parallel to the crystal interface. At this time, the defects will preferentially form near the crystal interface, otherwise they will tend to form inside the grain; (4) The impact of high-energy PKA will cause a certain degree of distortion or even migration of the grain boundary.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A molecular dynamics simulation analysis method for cascade collision of polycrystalline structures, characterized in that: The following steps are involved: S1, using Voronoi algorithm to build a polycrystalline structure model; S2. Conduct cascade collision tests on simulated polycrystalline structure models under different incident energies and different incident directions to observe the radiation damage behavior and microstructure evolution characteristics of the polycrystalline structure; S3. The microstructure of the polycrystalline structure model after the simulated cascade collision test is separated into the intracrystalline and grain boundary regions, and then analyzed using OVITO software to distinguish the body-centered cubic structure of the bulk region inside the crystal, the crystal interface region, and the irradiation defects in the polycrystalline structure after the cascade collision.

2. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 1, characterized in that: The S1 specifically includes the following steps: S11. Determine key parameters: Calculate cell side length (l x / l y / l z ) and the number of grains N included; S12, determining the center positions of N grains and the corresponding rotation matrices; S13, establishing a single crystal, rotating it according to the rotation matrix and translating it to the corresponding center position; S14, judging the grain to which each atom belongs based on the distance from the atom to the center of the grain; S15. For the constructed grain boundary, determine the distance between atoms at the grain boundary and delete the overlapping atoms; S16. Complete the polycrystalline grain construction and output the corresponding position file.

3. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 2, characterized in that: In step S12, the center positions of the N grains are determined by a random number method.

4. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 3, characterized in that: The specific process of the random number method is as follows: Given the lengths l of the three sides of the computational cell x , l y , l z After that, the average size of each grain is first approximated by using the cubic structure or spherical structure method, thereby obtaining the average side length of each grain under the cubic structure approximation condition, or the radius length of the spherical grain under the spherical structure approximation condition; after determining the relevant key size parameters of the grain, the center position of each grain is obtained by random numbers.

5. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 4, characterized in that: The calculation formula for the average side length of each grain under the cubic structure approximation condition is as follows:

6. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 5, characterized in that: The calculation formula for the radius length of spherical grains under the spherical structure approximation condition is as follows:

7. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 6, characterized in that: The specific process of obtaining the center position of each grain by random numbers is as follows: For grains with cubic structure, it is assumed that the distance between grains is not less than 2D; for grains with spherical structure, it can be assumed that the distance between grains is not less than the diameter 2R; the assumptions here are all first-order approximations. Based on this assumption, a polycrystalline structure with relatively uniformly distributed grain positions can be constructed. After determining the relevant key size parameters of the grains, the spatial range of the grain distribution area is set, and then a random number generator is selected to generate the center coordinates of the grains using an appropriate random number generation function; according to the required situation, coordinates (x, y, z) are randomly generated within the specified range to ensure that these coordinates conform to the size and shape of the grains.

8. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 7, characterized in that: The specific process of the rotation matrix in step S12 is: According to the definition of grain boundary freedom, the grains are rotated at any angle around the three main axes to achieve free rotation in three-dimensional space and construct the corresponding grain boundaries; according to the commonly used rotation matrix, the given angle θ is rotated around the Z axis to obtain the corresponding rotation matrix as follows: Therefore, when the entire grain rotates around three given directions in three-dimensional space by α, β and γ, the rotation matrix of the three-dimensional space can be obtained as follows:

9. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 7, characterized in that: The specific process of step S14 is: After determining the corresponding grain center position pi, the Voronoi method can be used to give the position of the mid-vertical plane between the centers of two grains, and confirm that the area enclosed by all pi satisfies the range r determined by the following formula: i : r i ∈{x,y,z|d(r i ,P i )<d(r i ,P j ),j={1,2,...n},j≠i} Among them, ri is the position vector, and d(ri,pi) is the distance from the given position to the center of the grain. Therefore, according to the above formula, the grain at the given position can be confirmed, thereby realizing the initial establishment of the polycrystalline periodic structure.

10. The molecular dynamics simulation analysis method for cascade collision of polycrystalline structures according to claim 7, characterized in that: Step S1 also includes: after the polycrystalline structure is constructed, firstly, a molecular statics method is used for relaxation, and then a molecular dynamics method is used for dynamic equilibrium relaxation to obtain a corresponding stable structure.

Images