Structure of the glass-forming metallic liquids by ab-initio and classical molecular dynamics, a case study: Quenching the Cu60Ti20Zr20 alloy

Abstract

We consider the question of the amorphization of metallic alloys by melt quenching, as predicted by molecular dynamics simulations with semi-empirical potentials. The parametrization of the potentials is discussed on the example of the ternary Cu-Ti-Zr transition metals alloy, using as reference the ab-initio simulation. The pair structure in the amorphous state is computed from a potential of the Stillinger Weber form. The transferability of the parameters during the quench is investigated using two parametrizations: from solid state data, as usual, and from a new parametrization on the liquid structure. When the adjustment is made on the pair structure of the liquid, a satisfactory transferability is found between the pure components and their alloys. The liquid structure predicted in this way agrees well with experiment, in contrast with the one obtained using the adjustment on the solid. The final structure, after quenches down to the amorphous state, determined with the new set of parameters is shown to be very close to the ab-initio one, the latter being in excellent agreement with recent X-rays diffraction experiments. The corresponding critical temperature of the glass transition is estimated from the behavior of the heat capacity. Discussion of the consistency between the structures predicted using semi-empirical potentials and ab-initio simulation, and comparison of different experimental data underlines the question of the dependence of the final structure on the thermodynamic path followed to reach the amorphous state.

The critical model size for simulating the structure-dynamics correlation in bulk metallic glasses

“…The values of structural parameter W at different equilibrium temperatures are roughly close to the earlier study [35]. All these results indicate that the system size has only a negligible effect on the short-range features of atomic structure and the averaged structure features in simulated MGs which is consistent with previous studies [27]. It demonstrates that the AIMD could be an effective method for investigating the local structure information of MGs.…”

“…It is of interest to investigate the size effect in a three-dimensional system with the periodic boundary condition representing the bulk MG with ignorable surface effects. Amokrane et al [27] indicated that the system size has only a negligible effect on the atomic structure and the metallic bond characteristics in the bulk Cu-Ti-Zr system. A recent study [28] reported that the structural relaxation behaviors show an evident system size-dependence in MG-forming liquids.…”

Liquid metals: early contributions and some recent developments

“…Finally, in connection with the work of JPB and coll. for copper, we show in subsection 4.2 the recent AIMD results on Cu based alloys by Amokrane et al [22] in a study that illustrates the efficiency of some modern methods combining AIMD and CMD on the non-trivial case of the glass forming ternary alloy Cu 60 Ti 20 Zr 20 , a task hardly conceivable with the methods used 40 years ago.…”

“…Starting in the early nineties, they have been implemented by different academic groups. Among the various implementations, the self-consistent plane wave (PWscf) fortran code in the open source Quantum-Espresso package [49] has been used in this work and in [22].…”

Modelling and Simulation of Solidification Phenomena during Additive Manufacturing of Bulk Metallic Glass Matrix Composites (BMGMC)—A Brief Review and Introduction of Technique

“…GFA has been increasingly studied and considerable progress has been made in its improvement [37] [38] [39] [40] by alterations and improvement in both composition and processing condition's window [15] [41] [42]. Now, alloys having multicomponent composition can be cast in glassy state even at slow cooling rates owing to their superior glass forming ability [38] [43]- [48] which in turn is governed by various theories [40] [43] [49]- [60] and analytical models [61] [62].…”