How a single aluminum atom makes a difference to gallium: First-principles simulations of bimetallic cluster melting

Abstract

First principles molecular dynamics simulations of Ga19Al(+) have been performed in the microcanonical ensemble using parallel tempering. We perform a thorough investigation of the changes induced by the presence of an Al atom in the Ga dominated cluster. Dynamic analysis indicates that the Al atom prefers to occupy the internal sites of the cluster structure, at all temperatures, and above 450 K, the Al atom is less mobile than the central Ga atom throughout the simulation. Using the multiple histogram method, canonical specific heat curves are obtained that compare well with previous experimental measurements of the specific heat and equivalent simulations for the Ga20 (+) cluster. The first-principles melting temperature agrees well with the experimental value for Ga19Al(+). Analysis of the root mean squared fluctuation in bond length, velocity auto-correlation function, and the corresponding power spectrum, confirms the solid-liquid-like phase transition in Ga19Al(+), as for Ga20 (+).

Impurity effects on solid–solid transitions in atomic clusters

“…Also, note that it is not necessary to transform the velocities ẋ in unison with the coordinates x, but doing so is useful for separating out the kinetic energy contributions associated with particular normal modes. Now, substituting Ũ m (x) for U(x) in (14) and transforming the coordinates according to (15) yields…”

“…In the present study we focus on solid-solid transitions, where one solidlike phase is supplanted by a distinctly different one prior to complete melting, and we explore how such transitions are affected when the atomic identity of a single substituent is varied. We consider various models and set three main objectives: (i) to qualitatively compare with previous experimental 10,11 and computational [12][13][14][15][16][17][18][19] studies examining impurity/dopant effects on cluster melting; (ii) to discern generic behaviour that could be exploited for tuning various thermal instabilities in nanoalloys; and (iii) to demonstrate how the harmonic superposition approach (HSA) 20,21 can be used to identify and explain impurity effects in atomic clusters at temperatures below melting.…”

“…11 From previous computational studies we also know that impurity effects on cluster melting depend on the impurity 15 as well as the host cluster, 18 suggesting that selective substitutional doping may be a feasible strategy for tuning the finite-temperature behaviour. Indeed, a single impurity can cause the melting temperature of a cluster to decrease 13 in some cases and increase 14,16,17 in others. In particular, molecular dynamics simulations of Mottet et al 16 show the melting temperature of geometrically closed-shell Ag N Mackay icosahedra 22 increasing by up to 70 K when the central silver atom is substituted by nickel or copper, and this trend was correlated with impurity-dependent strain relaxation, which evidently favours the solidlike icosahedron over the liquidlike state.…”

Characterizing the Greater-Than-Bulk Melting Behavior of Ga–Al Nanoalloys

“…Models typically used for bulk phases [12,30] are unsuitable for these systems and hence, sensitive first-principles treatment of the electronic structure is necessary. In our previous study [37], we showed that density functional theory (DFT) was able to capture the solid-liquid-like phase transition temperature of Ga 19 Al + in excellent agreement with experiments [35]. In this work, we explore two more cluster sizes in the Ga (20-x) Al x + cluster series: Ga 11 Al 9 + and Ga 3 Al 17 + , and attempt to create a phase diagram of Ga-Al nanoalloy system using the phase transition temperature data of monometallic phases, Ga 20 + and Al 20 + respectively [50,38].…”

Cluster melting: new, limiting, and liminal phenomena

“…However BOMD simulations on such clusters will be required to provide further insight. Recent studies on hybrid gallium-aluminium clusters have attempted to map out the phase diagram as a function of composition for a single specific size [93,94]; there is however no doubt that size-dependent effects on the interactions within two-component systems will provide fertile ground for new phenomena, within the liminal space of the multi-component system. A further question -and limitation of the Born-Oppenheimer approximation in Molecular Dynamics calculations -is the role of quantum behaviour in the dynamics of lighter nuclei.…”