Interlayer surface relaxations and energies of fcc metal surfaces by a tight-binding method

Abstract

The authors examine the interlayer surface relaxations and surface energies for the low-index faces of fcc Ni, Pd, Rh, Pt, Au, and Ir using the Naval Research Laboratory (NRL) tight-binding (TB) method. We compare the TB calculations, utilizing self-consistent charge transfer, with experimental measurements, density functional theory (DFT) calculations, and semiempirical methods. We find that for these metals the NRL-TB method largely reproduces the trends with respect to the exposed face and periodic table position obtained in DFT calculations and experimental measurements. We find that the inclusion of self-consistency in the TB surface calculations is essential for obtaining this agreement, as the TB calculations without it predict large first interlayer expansions for many of these surfaces. We also examine the energetics and relaxations of the 2 ϫ 1 (011) missing row reconstruction for these metals. The TB method predicts that, in agreement with experiment, Au and Pt undergo this reconstruction, while Ni, Pd, and Rh do not, but predicts the Ir ground state structure to be unreconstructed 1 ϫ 1, opposite to experiment. The interatomic relaxations of the (011) missing row structure for Pt, Au, and Ir are in good agreement with DFT calculations and experiment. Finally, we analyze the bonding characteristics of these metals using a decomposition of the TB total energy over neighboring atoms and angular momentum character.

Density functional theory investigation of surface-stress-induced phase transformations in fcc metal nanowires

“…We employ the Naval Research Laboratory tight-binding ͑NRL TB͒ 24 parametrizations for the fcc metals, along with charge self-consistency for surfaces and nanowires. The NRL TB method has previously been shown to give reasonable agreement with DFT and experiment for fcc metal surfaces 25 and Au nanowires. 26 In our TB calculations we employ the same geometries and k-point sampling as the DFT calculations.…”

Compositional structures and thermodynamic properties of Pd-Cu, Rh-Pd, and Rh-Pd-Cu nanoclusters computed by a combined free-energy concentration expansion method and tight-binding approach

“…23 The basic idea is to incorporate elemental bond energies and their surface-induced variations, obtained by means of the Naval Research Laboratory ͑NRL͒ tightbinding method ͑NRL-TB͒. 24 The results of this FCEM/ NRL-TB approach, applied to surface segregation at flat surfaces ͑Rh 75 Pt 25 ͑111͒͒ as a test case, exhibit good agreement with experiment. 23 The TB method, developed from the original work by Slater and Koster, 25 can cope with the electronic structure of systems of hundreds of atoms that would otherwise be computationally intractable with ab initio techniques such as DFT.…”

“…27,28 In its original form, the NRL-TB method has been tested in many ways including elastic constants, vacancy formation and stacking fault energies, ductility and thermal expansion, with gener-ally good agreement with experiment. 27,28 It also performs well for predicting properties of low index faces in d-shell fcc metals 24 when charge self-consistency is included in the calculations. In particular, calculated nearest neighbor bond energies in bulk and surface environments, used in the present work, largely account for trends in surface energies and interlayer relaxations.…”

Spin-polarized field emission from Ni(001) and Ni(111) surfaces

“…[34][35][36][37][38][39][40][41] The change in the first interlayer spacing, ⌬ 12 , was also investigated both experimentally 42,43 and computationally, 39 and was found to be a few percent. In our calculations, the surface relaxations change work functions and magnetic moments only of the order of 0.1% and the spin-polarization of tunneling conductance by less than a few percent.…”