Second-moment interatomic potential for Cu-Au alloys based on total-energy calculations and its application to molecular-dynamics simulations

Abstract

We have evaluated interatomic potentials of Cu, Au and Cu-Au ordered alloys in the framework of the second-moment approximation to the tight-binding theory by fitting to the volume dependence of the total energy of these materials computed by first-principles augmented-plane-wave calculations. We have applied this scheme to calculate the bulk modulus and elastic constants of the pure elements and alloys and we have obtained a good agreement with experiment. We also have performed molecular-dynamics simulations at various temperatures, deducing the temperature dependence of the lattice constants and the atomic mean square displacements, as well as the phonon density of states and the phonon-dispersion curves of the ordered alloys. A satisfactory accuracy was obtained, comparable to previous works based on the same approximation, but resulting from fitting to various experimental quantities.

The Slater Koster tight-binding method: a computationally efficient and accurate approach

“…We will only mention here the second-moment approximation (SMA) because of our own work in this area. In our implementation of the SMA [36][37][38] we determine the potential by fitting to the volume dependence of the total energy computed by carrying out first-principles APW calculations instead of fitting to experimental quantities, as other workers do.…”

“…This will serve to illustrate the care needed to construct good parameters for binary compounds. Note that we originally picked this system because it seemed rather simple, and because it is well described by atomistic potentials such as the SMA [38].…”

Force-based many-body interatomic potential for ZrC

“…To empirically extend the second-moment potential to AB alloys, it is common to write, 27,[31][32][33] by neglecting the on-site energy differences…”

Size-, Shape-, and Composition-Dependent Model for Metal Nanoparticle Stability Prediction

“…Noncrystalline structures and chemical ordering in nanoalloys and MNPs can be accurately modeled using ab initio methods (i.e., density functional theory (DFT) on metal clusters). , DFT, however, becomes computationally intractable at even moderate MNP sizes (∼1–3 nm diameter MNPs) and is largely prohibitively expensive in studying nanoalloys due to their near infinite homotops. , For example, a single 25 atom nanoalloy structure with no identical positions (i.e., amorphous) composed of 15 Au and 10 Ag atoms has more than 3 268 760 distinct homotops. To accelerate nanoalloy analysis, several less-expensive empirical and semiempirical methods such as tight-binding models, − embedded atom models, , effective medium theory, and pair-wise potentials (e.g., Finnis–Sinclair and Sutton–Chen potentials) have been developed. However, such methods require parameter tuning against large ab initio (DFT) data for accurate nanoalloy energetics, − limiting their broad applicability (i.e., diverse compositions) and time acceleration in analyzing nanoalloy systems.…”