Search

Influence of hydrogen concentration and defects introduced by neutron irradiation on hydrogen diffusion in tungsten has been investigated by molecular dynamics simulation at elevated temperatures. Hydrogen diffusion is shown to be significantly restrained at high concentrations due to spontaneous formation of platelet-like hydrogen clusters. For neutron irradiation defects, self-interstitials, mono-vacancies and vacancy clusters are considered. By clustering and acting as dislocation loops, self-interstitials show considerable trapping effects on hydrogen, leading to the suppression of hydrogen effective diffusion and the change of diffusion model in which hydrogen mainly diffuses along dislocation lines instead of hopping between tetrahedral interstitial sites. Moreover, an equation connecting hydrogen diffusion parameters and the total length of dislocation loops is empirically established. Different influences of mono-vacancies and vacancy clusters on hydrogen diffusion have been carefully identified. With the same vacancy concentration, hydrogen diffusivity is lower with mono-vacancies than that with vacancy clusters because more isolated trapping sites are provided by mono-vacancies. This work is not only helpful for understanding the synergistic effects of neutron irradiation and plasma interaction, but also potentially applicable for larger scale simulations as input data.

Atomistic simulations were employed to investigate the stress field spatial distributions, strain energy of the 1/2 〈1 1 1〉 {1 1 0} edge dislocation, and the binding energies between the edge dislocation and the point defects in tungsten, based on two different embedded-atom method (EAM) potentials. The basic static properties are consistent with the elastic theory of dislocations for these two potentials. Comparatively speaking, the map of the binding energy between the edge dislocation and the monovacancy illustrates that the results based on the Marinica’s potential fits better with the elastic theory. In addition, we obtained the radii of absorption between the point defects and the edge dislocation in the slip plane (the maximum value is 19 A for the monovacancy, while 34 A for the self-interstitial atom (SIA)) at 0 K. By calculating the binding energy and the interaction radii, we found the intensity of the interaction between the SIA and the edge dislocation is stronger than that of the vacancy and the edge dislocation.

Atomistic simulations have been used to investigate the core structures, static properties of isolated 1/2 screw dislocations, and their interaction with vacancies in bcc tungsten (W) based on three empirical interatomic potentials. Differential displacement maps show that only one embedded atom method potential is able to reproduce the compact non-degenerate core as evidenced by ab initio calculations. The obtained strain energy and stress distribution from atomistic simulations are, in general, consistent with elasticity theory predictions. In particular, one component of the calculated shear stress, which is not present according to elasticity theory, is non-negligible in the core region of our dislocation model. The differences between the results calculated from three interatomic potentials are in details, such as the specific value and the symmetry, but the trend of spatial distributions of static properties in the long range are close to each other. By calculating the binding energies between the dislocations and vacancies, we demonstrate that the dislocations act as vacancy sinks, which may be important for the nucleation and growth of hydrogen bubbles in W under irradiation.

Hydrogen (H) defect interactions have been investigated by molecular statics simulations in tungsten (W), including H-H interactions and interactions between H and W self-interstitial atoms. The interactions between H and small H-vacancy clusters are also demonstrated; the binding energies of an H, a vacancy and a self-interstitial W to an H-vacancy cluster depend on the H-to-vacancy ratio. We conclude that H bubble formation needs a high concentration of H in W for the H bubble nucleation and growth, which are also governed by the H-to-vacancy ratio of the cluster. The vacancy first combines with H atoms and a cluster forms, then the H-vacancy cluster goes through the whole process of vacancy capture, H capture, and vacancy capture again, and as a result the H-vacancy cluster grows larger and larger. Finally, the H bubble forms.

We have clarified the performance of two tungsten–helium analytical interatomic potentials, one of which, developed by Li et al., is a bond-order potential, and another, developed by Juslin et al., is a combination of embedded atom method potential and pair potential. Using these two potentials, we have simulated and made a full comparison of formation energy and migration energy of different defects including helium and vacancy, binding energies of helium and vacancy with helium-vacancy cluster, surface energy, as well as melting point, with reference to the corresponding results from the first-principles and experiments.

Tungsten (W) is considered to be a promising candidate for plasma facing materials in fusion reactors because of its high melting point, high thermal diffusivity and low sputtering yield. To investigate the multiple hydrogen (H) diffusion behaviours at the W(001) surface, molecular dynamics simulations were used to study the H distributions after diffusion of 1 ns under different conditions. The results of using different H concentrations and temperatures showed the preference of bulk to surface diffusion, and the H distribution patterns of the H/W ratio of 0·1 are analysed in detail with an ‘empty’ region near the surface. The effect of temperature on H diffusion is also investigated, which gives the highest adsorption peak at a mediate temperature of 900 K. The trapping effect of vacancies on H atoms is demonstrated at different vacancy concentrations. The effect of H embedded depth on H distribution is also investigated together with the effect of temperature, which gives an increasing trend of...

The diffusion process of multiple He atoms in W is simulated by a molecular dynamics (MD) method with the W-H-He analytic bond-order potential. The diffusivities of different number of helium (He) atoms in W are determined by the mean squared displacement (MSD) method at different temperatures. The diffusivity-temperature (D-T) relationship is fitted to the Arrhenius equation to obtain the pre-factor and the diffusion barrier. Under the temperature of 1200K He atoms diffuse together, and above 1200K they separate from each other. When the number of He atoms is greater than three, all He atoms oscillate at the tetrahedral interstitial site (TIS) instead of diffusing under 400K. In the temperature range of 400-1200K, the diffusion barriers of He atoms, the number of which is from two to five, are 0.098, 0.170, 0.125 and 0.112eV, respectively. Contrasting with one He atom (0.058eV), the higher diffusion barriers reflect a greater difficulty in diffusion of multiple He atoms in W. In addition, when the number of He atoms is over five, vacancies are formed in W, and He atoms occupy the vacancies.

The interstitial helium (He) atom diffusion in tungsten (W) was studied by the Molecular Dynamics Simulation with the drag method, the Nudged Elastic Band method (NEB) and the mean square displacement (MSD) method. The diffusion barriers and the possible microscopic diffusion path were calculated by the drag method. It has the characteristics of simple, intuitive, and occupies less computer resources, but can't get the diffusion equation. The NEB method is more reasonable than the drag method to calculate the diffusion barriers, and determine the diffusion path which, but the former spends more computer resources than the latter, and it also can't get the diffusion equation. The diffusion equation is obtained by MSD method, including the diffusion per-factor and diffusion barriers. It is suggested that the mechanism of He diffusion changes with difference temperature, which spends the most computer resources among the three methods.

The size and the electronic factor are used to analyze the occupation of doped Co atom in L10 FePt alloy as well as its influence on crystal structure and magnetic property. The Co atom located in Fe sublattice site is determined by the variance of the calculated structure and the doping energy results investigated by the first-principles calculations. The charge-density difference contour plot and Bader charge analysis can quantitatively distinguish the effects of the size and the electronic factor on physical properties of the doping alloy. The effect of the electronic factor on the doping energy is more dominant than that of the size factor. The size factor almost has no effect on saturation magnetization.

Shear-coupled grain boundary (GB) migration can be an efficacious mechanism to accommodate plastic deformation when the grain size of polycrystalline materials goes small. Nevertheless, how this kind of GB motion comes into play at the atomic level has not been fully revealed. Here, we have investigated the shear-coupled migration (SCM) of typical [100] group symmetrical tilt GBs in bcc W using atomistic simulations. Depending on GB character, the SCM is found to proceed via dislocation slipping in the 〈100〉 or 〈110〉 mode with striking shear strength difference between them. We demonstrate that there exists an unusual atomic shuffling along the tilt axis, which greatly assists SCM to operate in the easier 〈110〉 mode instead of the 〈100〉 one. The present results highlight the significant role of GB character in the atomistic SCM process and contribute to the future design and fabrication of high-performance materials in GB engineering.

The point defects and their related physical properties in L10 FePt are investigated by molecular dynamics simulations based on an analytic bond-order potential. The calculated results agree well with the experimental value, indicating that the analytic bond-order potential is suitable to describe the structural properties and surface energies of the FePt alloy in the L10 phase. However, the calculated vacancy formation energy of an Fe atom is higher than that of a Pt atom, which disagrees with some other previously calculated results. This result indicates that the analytic bond-order potential is unable to describe the related point defect properties. The analytic bond-order potential needs to be modified in order to study these defect properties of an FePt alloy. analytic bond-order potential, FePt, point defects, molecular dynamics simulation

The preference site and magnetic property of doped Cr atom in L10 FePt alloy are investigated by the first-principles calculations. Due to the energy of Cr located in Fe and Pt sublattice are -0.15 and 0.48 eV, respectively, the Cr atom occupying Fe site is more energetic favorable than in Pt site in L10 FePt alloy. The calculated saturation magnetizations for Cr additive replacing Fe or Pt atom in L10 FePt are 959 or 989 emu/cm3, respectively. The results are lower than that of L10 FePt (1085emu/cm3). The additive of Cr in L10 FePt will reduce the saturation magnetization.

The dislocation 1/2[111](1-10) in iron is constructed. The motion behaviors of dislocation 1/2[111](1-10) in iron, as well as the dislocation with the Carbon located in dislocation core, are simulated by the molecular dynamics method with a modified analytic embedded atomic method (MAEAM). The carbon atoms are easy captured by the dislocation core and they can block the dislocation moving. After the carbon atoms are separated from the dislocation, the moving velocities of the dislocation are similar as the case of single dislocation.

We have performed a molecular dynamics study on the growth of helium (He) clusters in the subsurface of tungsten (W) (1 0 0) at 300 K, focusing on the role of cluster depth. Irregular 'stick–slip' behavior exhibited during the evolution of the He cluster growth is identified, which is due to the combined effects of the continuous cluster growth and the loop punching induced pressure relief. We demonstrate that the He cluster grows via trap-mutation and loop punching mechanisms. Initially, the self-interstitial atom SIA clusters are almost always attached to the He cluster; while they are instantly emitted to the surface once a critical cluster pressure is reached. The repetition of this process results in the He cluster approaching the surface via a 'stop-and-go' manner and the formation of surface adatom islands (surface roughening), ultimately leading to cluster bursting and He escape. We reveal that, for the Nth loop punching event, the critical size of the He cluster to trigger loop punching and the size of the emitted SIA clusters are correspondingly increased with the increasing initial cluster depth. We tentatively attribute the observed depth effects to the lower formation energies of Frenkel pairs and the greatly reduced barriers for loop punching in the stress field of the W subsurface. In addition, some intriguing features emerge, such as the morphological transformation of the He cluster from 'platelet-like' to spherical, to ellipsoidal with a 'bullet-like' tip, and finally to a 'bottle-like' shape after cluster rupture.

The effects of different amounts of segregated Ga (substitutional) on an Al grain boundary have been investigated by using a first-principles pseudopotential method. The segregated Ga is found to draw charge from the surrounding Al due to the electronegativity difference between Ga and Al, leading to a charge density reduction between Ga and Al as well as along the Al grain boundary. Such an effect can be enhanced by increasing the Ga segregation amount. With further Ga segregated, in addition to the charge-drawing effect that occurs in the Al–Ga interface, a heterogeneous α-Ga-like phase can form in the grain boundary, which greatly alters the boundary structure. These effects are suggested to be responsible for Ga-induced Al intergranular embrittlement.

The point defect properties of L12 Ni3Al are determined by MD with a simple modified analytic EAM model with the basic physical properties of pure constituents Ni and Al. On the basis of the calculated lattice constants, the formation energies of vacancy and antisite are calculated and the defect types are also discussed. The energies of alloying element Re replacing Ni site and Al site in Ni3Al are estimated. The occupational site of Re is found in Al site. The present calculations are in agreement with the experimental value and the theoretical results obtained from other authors.

The diffusion properties of iron and nickel in iron are investigated by the molecular dynamics method with a modified analytic embedded-atom method. In the monovacancy diffusion, the energy curve of self-diffusion and diffusion of nickel in iron is symmetric. However, the energy curve of self-diffusion and diffusion of nickel in iron is not symmetric in the divacancy diffusion mechanism. The self-diffusion mechanism is the monovacancy diffusion of nearest neighbor (NN) jump at low temperature. The activation energy, vacancy formation energy and migration energy of self-diffusion are 2.425, 1.657 and 0.768 eV, respectively. In high temperature, the divacancy diffusion must be considered. The solute nickel diffusion mechanism is that the nickel migrates to the NN vacancy. The activation energy and migration energy of diffusion of nickel in iron are 2.260 and 0.582 eV, respectively. The calculated results are in good agreement with the experimental values and other calculated results.

In the present paper, the modified analytic embedded-atom method (MAEAM) many-body potentials are provided for fcc and hcp elements. The segregation energies of Mg in Al host are calculated with the MAEAM potentials. The composition depth profiles of Al–Mg alloys are obtained with Monte Carlo simulation method based on grand canonical ensemble. It is found that Mg segregates strongly to the surface layers at 600 K and the composition near the surface is monotonic to the bulk one. The simulation results are reasonable agreement with other experimental and theoretical data.

Modified analytic embedded atom method (MAEAM) type potentials have been constructed for seven binary immiscible alloy systems: Al–Pb, Ag–Ni, Fe–Cu, Ag–Cu, Cu–Ta, Cu–W and Cu–Co. The potentials are fitted to the lattice constant, cohesive energy, unrelaxed monovacancy formation energy and elastic constants for only pure metals which consist the immiscible alloy systems. In order to test the reliability of the constructed MAEAM potentials, formation enthalpies of disordered alloys for those seven binary immiscible alloy systems have been calculated. The calculated results are in general agreement with the experimental data available and those theoretical results calculated by other authors. As only very limited experimental information is available for alloy properties in immiscible alloy systems, the MAEAM is demonstrated to be a reasonable method to construct the interatomic potentials for immiscible alloy systems because only the properties of pure elements are needed in calculation.

Using the analytic modified EAM potentials, the surface concentrations and concentration depth profiles of (111) and (10 0) faces of Pt-Pd and Pt-Ir binary alloys are studied with Monte Carlo simulation based on the grand canonical ensemble statistical rule. In Pt-Pd alloy, the topmost surface is enriched with Pd, and the subsurface layer is depleted of Pd. The simulation results show a damped oscillation of Pd concentration in the whole composition range. A strong Pd segregation is observed for both the orientation faces, while the amount of Pd segregation in the (I 11) face is significantly less than that in the more open (10 0) face. However, Pt segregates to the surface strongly in the Pt-Ir alloy, and the amount of Pt segregation in the (10 0) and (I 11) face is almost the same. The simulation results are in good agreement with the available experiment data and other theoretical values. (C) 2002 Elsevier Science B.V. All rights reserved.

Analytic embedded-atom method (EAM) interatomic potentials for body-centered cubic (bcc) transition metals have been constructed. The total energy is regarded as consisting of a pair potential part, an embedding interaction part and a modification term. All these parts are analytic functions of the atomic separations only and are fitted to various bulk properties, such as cohesive energy, vacancy formation energy, lattice constant and elastic constants. The present potentials are shown to predict a more realistic pressure–volume relationship so that interactions at separations smaller than that of the first-nearest neighbors can be treated. Interstitial formation energies are calculated for various configurations, using quenched molecular dynamics. Vacancy, surface and phonon spectra have been also studied and satisfactory agreement with available experimental data has been found.

An embedded-atom method potential has been developed for modeling hydrogen in body-centered-cubic (bcc) tungsten by fitting to an extensive database of density functional theory (DFT) calculations. Comprehensive evaluations of the new potential are conducted by comparing various hydrogen properties with DFT calculations and available experimental data, as well as all the other tungsten-hydrogen potentials. The new potential accurately reproduces the point defect properties of hydrogen, the interaction among hydrogen atoms, the interplay between hydrogen and a monovacancy, and the thermal diffusion of hydrogen in tungsten. The successful validation of the new potential confirms its good reliability and transferability, which enables large-scale atomistic simulations of tungsten-hydrogen system. The new potential is afterward employed to investigate the interplay between hydrogen and other defects, including [1 1 1] self-interstitial atoms (SIAs) and vacancy clusters in tungsten. It is found that both the [1 1 1] SIAs and the vacancy clusters exhibit considerable attraction for hydrogen. Hydrogen solution and diffusion in strained tungsten are also studied using the present potential, which demonstrates that tensile (compressive) stress facilitates (impedes) hydrogen solution, and isotropic tensile (compressive) stress impedes (facilitates) hydrogen diffusion while anisotropic tensile (compressive) stress facilitates (impedes) hydrogen diffusion.

Using the analytic modified embedded atom method (MEAM), the embedding, potential and modifying functions for the Mg and rare earth (RE) metals (RE = Sc, Y, Pr, Nd, Gd, Tb, Dy, Ho and Er) are presented. The thermodynamic properties, such as the dilute-limit heats of solution, enthalpies of formation of disordered solid solutions and intermetallic compounds, for the binary Mg-RE alloys are calculated. The obtained results are in good agreement with the available experimental data and with the results calculated using Miedema theory. The calculations of the enthalpies of formation of Mg3 RE, Mg2 RE, MgRE, MgRE2 and MgRE3 with various ordered structures (DO3 , DO19 , L12 , C15, MoPt2 , B2, W1 and L10 ) indicate that the trends in the structural stability can be interpreted directly in terms of the formation energy. Moreover, the lattice constants and volume contractions of alloys with various compositions are determined based on the relation between the formation energy and the interatomic distance. The correlation between the enthalpy of formation and volume contraction for intermetallic compounds is discussed.

Ion irradiation has been used to promote ordering processes and to modify the magnetic properties of magnetic thin films. The major reason for ion irradiation reducing the ordering temperature is the introduction of a number of vacancies. The vacancy and its influence on the ordering temperature and magnetic properties in L1 0 ordered FePt are investigated by first-principle simulation. The vacancy formation energy for Fe and Pt in FePt alloy are 1.45 and 2.25 eV respectively. The calculated order–disorder transition temperature of Fe 50 Pt 50 is 1680 K. The order–disorder transition temperatures for Fe vacancy and Pt vacancy models are about 50 K and 200 K lower than that of the stoichiometric Fe 50 Pt 50 alloy respectively. The results suggested that the vacancy in FePt alloy favors the ordering process. The saturation magnetization of stoichiometric L1 0 FePt is 1070 emu/cc and these of Fe and Pt vacancy are 1027 and 1075 emu/cc, respectively.

Tungsten and vanadium (W–V) alloys (with 1, 5 and 10 wt.% V) are fabricated by hot pressing (HP) at 1800[Formula: see text]C under 20 MPa for 2 h. The effects of V content on the microstructures and mechanical properties of W–V alloy are investigated. The results indicate that with increasing V content, (i) the formation of W–V alloying phase is enhanced and the grain size of W-matrix is significantly refined; (ii) the relative density gradually increases from 92.16% to 97.72% in the case of pure W to W-10 wt.% V; (iii) the hardness rises linearly while the bending strength decreases, which is related to the enhanced alloy phase formation.

We perform a first-principles calculation to investigate the effects of He in an Fe � 5(310)/[001] grain boundary (GB) with the SIESTA code, for which the reliability of the pseudopotential and the basis set are systematically tested. We calculate the formation and segregation energies for different substitutional and interstitial cases in order to determine the site preference and the segregation properties of He in the Fe GB. It is demonstrated that the He segregation either breaks (substitution) or weakens (interstitial) the surrounding interfacial Fe–Fe bonds, leading to the GB tensile strength reduction. (Some figures in this article are in colour only in the electronic version)

The energetics and kinetics regarding helium (He) cluster growth in bcc tungsten (W) are unveiled using combined techniques of molecular statics and molecular dynamics. The principal mechanisms accounting for the decrease of system potential energy are identified to be trap mutation, → cluster transformation, loop punching, coalescence between 1/2[1 1-1] and 1/2[1-1-1] loops, and loop capturing. The kinetic barriers associated with these key atomistic events are estimated. This work provides new insights into the complex yet intriguing atomistic evolution sequence of the He cluster and interstitial loop in W-based nuclear fusion materials under irradiation.

We have used molecular statics in conjunction with an embedded atom method to explore the interplay between native point defects (vacancies and self-interstitials (SIAs)) and a low-angle grain boundary (GB) in bcc tungsten. The low-angle GB has biased absorption of SIAs over vacancies. We emphasize the significance of phenomena such as vacancy delocalization and SIA instant absorption around the GB dislocation cores in stabilizing the defect structures. Interstitial loading into the GB can dramatically enhance the interaction strength between the point defects and the GB due to SIA clustering (SIA cloud formation) or SIA vacancy recombination. We propose that the 'maximum atom displacement' can complement the 'vacancy formation energy' in evaluating unstable vacancy sites. Calculations of point defect migration barriers in the vicinity of GB dislocation cores show that vacancies and SIAs preferentially migrate along the pathways in the planes immediately above and below the core, respectively.

L10 FePt nanocomposite with high magnetocrystalline anisotropy energy has been extensively investigated in the fields of ultra-high density magnetic recording media. However, the order—disorder transition temperature of the nanocomposite is higher than 600 °C, which is a disadvantage for the use of the material due to the sustained growth of FePt grain under the temperature. To address the problem, addition of Ag atoms has been proposed, but the magnetic properties of the doped system are still unclear so far. Here in this paper, we use first-principles method to study the lattice parameters, formation energy, electronic structure, atomic magnetic moment and order—disorder transition temperature of L10 FePt with Ag atom doping. The results show that the formation energy of a Ag atom substituting for a Pt site is 1.309 eV, which is lower than that of substituting for an Fe site 1.346 eV. The formation energy of substituting for the two nearest Pt sites is 2.560 eV lower than that of substituting for the further sites 2.621 eV, which indicates that Ag dopants tend to segregate L10 FePt. The special quasirandom structures (SQSs) for the pure FePt and the FePt doped with two Ag atoms at the stable Pt sites show that the order—disorder transition temperatures are 1377 °C and 600 °C, respectively, suggesting that the transition temperature can be reduced with Ag atom, and therefore the FePt grain growth is suppressed. The saturation magnetizations of the pure FePt and the two Ag atoms doped FePt are 1083 emu/cc and 1062 emu/cc, respectively, indicating that the magnetic property of the doped system is almost unchanged.

We demonstrate a quantitative stress indicator as well as a qualitative hard-sphere lattice model to characterize the effective volume and stability of interstitial helium in fcc and bcc metals, based on extensive first-principles total-energy and lattice stress calculations in combination with continuum elastic theory analyses. The concept of stress indicator is believed to be generally applicable to quantify the relative stability of other inert gas elements in metals.

We perform a first-principles calculation to investigate the effect of a vacancy on the sliding of an iron (Fe) Σ5 (310) [001] tilt grain boundary (GB). We calculate vacancy formation energies in order to determine the site preference of the vacancy at the Fe GB. We show that vacancies can easily segregate to the GB region, which is attributed to special atomic configurations of the GB. We demonstrate that the maximal sliding energy barrier of the GB with a monovacancy is 2.78 J/m2, 15% lower than that of the vacancy-free GB, suggesting that the presence of vacancy at the GB increases the GB mobility.