Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDevelopment of interatomic potential appropriate for simulation of dislocation migration in fcc Fe
62
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Molecular dynamics (MD) simulation of dislocation migration requires semi-empirical potentials of the interatomic interaction. While there are many reliable semi-empirical potentials for the bcc Fe, the number of the available potentials for the fcc is very limited. In the present study we tested three EAM potentials for the fcc Fe (ABCH97 [Phil. Mag. A, 75, 713-732 (1997)], BCT13 [MSMSE 21, 085004 (2013)] and ZFS18 [J. Comp. Chem. 39, 2420-2431 (2018)]) from literature. It was found that the ABCH97 potential does not provide that the fcc phase is the most stable at any temperature. On the other hand, the fcc phase is always more stable than the bcc phase for the BCT13, ZFS18 potentials. The hcp phase is the most stable phase for the BCT13 potential at any temperature. In order to fix these problems we developed two new EAM potentials (MB1 and MB2). The fcc phase is still more stable than the bcc phase for the MB1 potential but the MB2 potential provides that the bcc phase is the most stable phase from the upper fcc-bcc transformation temperature, T_gamma-delta, to the melting temperature, Tm, and the fcc phase is the most stable phase below T_gamma-delta. This potential also leads to an excellent agreement with the experimental data on the fcc elastic constants and reasonable stacking fault energy which makes it the best potential for the simulation of the dislocation migration in the fcc Fe among all semi-empirical potentials considered in the present study. The MD simulation demonstrated that only the ZFS18, MB1 and MB2 potentials are actually suitable for the simulation of the dislocation migration in the fcc Fe. They lead to the same orders of magnitude for the dislocation velocities and all of them show that the edge dislocation is faster than the screw dislocation. However, the actual values of the dislocation velocities do depend on the employed semi-empirical potential.
rate research
Read More
New interatomic potentials describing defects, plasticity and high temperature phase transitions for Ti are presented. Fitting the martensitic hcp-bcc phase transformation temperature requires an efficient and accurate method to determine it. We apply a molecular dynamics (MD) method based on determination of the melting temperature of competing solid phases, and Gibbs-Helmholtz integration, and a lattice-switch Monte Carlo method (LSMC): these agree on the hcp-bcc transformation temperatures to within 2 K. We were able to develop embedded atom potentials which give a good fit to either low or high temperature data, but not both. The first developed potential (Ti1) reproduces the hcp-bcc transformation and melting temperatures and is suitable for the simulation of phase transitions and bcc Ti. Two other potentials (Ti2 and Ti3) correctly describe defect properties, and can be used to simulate plasticity or radiation damage in hcp Ti. The fact that a single EAM potential cannot describe both low and high temperature phases may be attributed to neglect of electronic degrees of freedom, notably bcc has a much higher electronic entropy. A temperature-dependent potential obtained from the combination of potentials Ti1 and Ti2 may be used to simulate Ti properties at any temperature.
An interatomic potential for Al-Tb alloy around the composition of Al90Tb10 was developed using the deep neural network (DNN) learning method. The atomic configurations and the corresponding total potential energies and forces on each atom obtained from ab initio molecular dynamics (AIMD) simulations are collected to train a DNN model to construct the interatomic potential for Al-Tb alloy. We show the obtained DNN model can well reproduce the energies and forces calculated by AIMD. Molecular dynamics (MD) simulations using the DNN interatomic potential also accurately describe the structural properties of Al90Tb10 liquid, such as the partial pair correlation functions (PPCFs) and the bond angle distributions, in comparison with the results from AIMD. Furthermore, the developed DNN interatomic potential predicts the formation energies of crystalline phases of Al-Tb system with the accuracy comparable to ab initio calculations. The structure factor of Al90Tb10 metallic glass obtained by MD simulation using the developed DNN interatomic potential is also in good agreement with the experimental X-ray diffraction data.
We develop an Fe-C-H interatomic potential based on the modified embedded-atom method (MEAM) formalism based on density functional theory to enable large-scale modular dynamics simulations of carbon steel and hydrogen.
Dipolar dislocation loops, prevalent in fcc metals, are widely recognized as controlling many physical aspects of plastic deformation. We present results of 3D dislocation dynamics simulations that shed light on the mechanisms of their formation, motion, interactions, and large-scale patterning. We identify two main formation mechanisms, enabled by cross-slip, and show that arrays of dipoles can be easily formed as a result of the interaction between glide screw dislocations. We present a systematic analysis of the spectrum of possible junctions that can form as a result of mutual interaction between dipoles, and between dipoles and glide dislocations. We show that fully immobile dislocation segments arise in particular cases of these interactions, leading to hardening and Frank-Read type sources. We reveal that the collective motion of dipolar loop arrays can be induced by glide dislocations in the channels of Persistent Slip Bands (PSB), and result in their clustering within PSB channel walls. An efficient tripolar drag mechanism is found to contribute to the clustering of dipolar loops near channel walls.
Accuracy of molecular dynamics simulations depends crucially on the interatomic potential used to generate forces. The gold standard would be first-principles quantum mechanics (QM) calculations, but these become prohibitively expensive at large simulation scales. Machine learning (ML) based potentials aim for faithful emulation of QM at drastically reduced computational cost. The accuracy and robustness of an ML potential is primarily limited by the quality and diversity of the training dataset. Using the principles of active learning (AL), we present a highly automated approach to dataset construction. The strategy is to use the ML potential under development to sample new atomic configurations and, whenever a configuration is reached for which the ML uncertainty is sufficiently large, collect new QM data. Here, we seek to push the limits of automation, removing as much expert knowledge from the AL process as possible. All sampling is performed using MD simulations starting from an initially disordered configuration, and undergoing non-equilibrium dynamics as driven by time-varying applied temperatures. We demonstrate this approach by building an ML potential for aluminum (ANI-Al). After many AL iterations, ANI-Al teaches itself to predict properties like the radial distribution function in melt, liquid-solid coexistence curve, and crystal properties such as defect energies and barriers. To demonstrate transferability, we perform a 1.3M atom shock simulation, and show that ANI-Al predictions agree very well with DFT calculations on local atomic environments sampled from the nonequilibrium dynamics. Interestingly, the configurations appearing in shock appear to have been well sampled in the AL training dataset, in a way that we illustrate visually.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
