Transition-pathway models of atomic diffusion on fcc metal surfaces. II. Stepped surfaces

Abstract

Action-derived molecular dynamics was demonstrated in the companion paper ͑Paper I͒ to be effective for the analysis of atomic surface diffusion. The method is here applied to the search of minimum-energy paths and the calculation of activation energy barriers in more complex single-adatom diffusion processes on fcc metal surfaces containing steps. Diverse diffusion routes are investigated along and across one-or two-layer steps on different surface orientations. Fundamental diffusion mechanisms near the step corners are also studied. Results are analyzed in relation to the island growth mechanism, which is of importance to surface nanoengineering.

Formation and morphological evolution of self-similar 3D nanostructures on weakly interacting substrates

“…Moreover, 3D island formation is also known to occur in the absence of deposition flux due to surface restructuring via dewetting [29]. These observations suggest an atomistic pathway which facilitates material transport from the base of atomic islands to their upper layer, a highly unlikely process in standard homoepitaxial thin-film growth theory due to large activation barriers associated with upward interlayer transport across step edges [30].…”

“…Equations (2) through (4) include a total of eight parameters which are uniquely determined (see Supplemental Material [57] for values of these parameters) to reproduce published theoretical barriers for Ag/Ag(111) [30,70]. Moreover, BCS values are also compared to experimental results [16,54,69,[71][72][73][74][75][76][77][78], as well as barriers obtained from our nudged-elasticband (NEB) calculations using an embedded atom method interatomic potential [79].…”

Diffusion barriers for Ag and Cu adatoms on the terraces and step edges on Cu(100) and Ag(100): Anab initiostudy

“…21͒ and 0.35 eV. 22 For the exchange mechanism, barriers are found to be somewhat higher ͑between 0.62 and 0.78 eV͒ in theoretical calculations, 15,16,[18][19][20] while experiments report it to be 0.46 eV. 22 Interestingly, theoretical results for the energy barriers for diffusion via hopping of Cu adatoms on Cu͑100͒ also lie in the range of 0.48-0.69 eV, 7,16,18,[23][24][25][26] while the range of experimental values is between 0.28 and 0.40 eV.…”

“…22 For the exchange mechanism, barriers are found to be somewhat higher ͑between 0.62 and 0.78 eV͒ in theoretical calculations, 15,16,[18][19][20] while experiments report it to be 0.46 eV. 22 Interestingly, theoretical results for the energy barriers for diffusion via hopping of Cu adatoms on Cu͑100͒ also lie in the range of 0.48-0.69 eV, 7,16,18,[23][24][25][26] while the range of experimental values is between 0.28 and 0.40 eV. [27][28][29] The energy barrier for diffusion of Cu adatoms on Cu͑100͒ via exchange is, on the other hand, found to be much larger, close to 1 eV, in first-principles calculations, 7,16,23 and between 0.69 and 0.80 eV in results based on model interaction potentials.…”

“…For Ag adatoms at the step edge of Ag͑100͒ first-principles calculations find the ES barrier for exchange to be 0 meV, and that for hopping to be 100 meV, 15 signifying the importance of the former. Semiempirical methods find the barrier via hopping to lie in the range 110-220 meV, 20,31,32 while the barrier via exchange is between 30 and 160 meV. 20,31,32 Experimental studies report homoepitaxial growth on Ag͑100͒ to proceed layer by layer and the ES barrier to lie in the range 30-70 meV.…”