Surfactant-Induced Layer-by-Layer Growth of Ag on Ag(111): Origins and Side Effects

Abstract

It has been shown that submonolayer deposits of Sb change the growth mode of Ag(111) from multilayer to layer-by-layer.Using scanning tunneling microscopy, we identify the two main origins of this behavior. (i) Sb lowers the mobility of Ag adatoms on terraces and growing islands. As a direct consequence, the additional edge barrier (barrier to descend a step minus surface diffusion barrier) is reduced. (ii) Sb lowers the mobility along step edges, inducing dendritic island shapes. Both effects favor smoother growth. Neither a lower Ag surface mobility nor heterogeneous nucleation are sufficient for a smooth growth. PACS numbers: 68.55. -a, 61.16.ChThe mechanisms underlying epitaxial growth have been extensively studied over the past few years. While first work has concentrated on clean surfaces, the effect of adsorbates ("surfactants") on the morphology has been included recently [1,2]. These studies have revealed that the film morphology obtained is strongly dependent on the kinetics of intra-and interlayer mass transport, more specifically, on whether or not the material deposited on top of a growing two-dimensional island can reach its edge and fall down, so that the growing layer is completed before the next one starts to grow [3 -5].In a recent x-ray diffraction study, a striking change in growth behavior of Ag was observed for going from clean to Sb-precovered Ag(111) [6]. No layer-by-layer growth behavior is observed for Ag on clean Ag(111) over the whole temperature range studied, between 575 and 175 K. Rather the growth gradually changes from step flow to multilayer growth with several layers simultaneously exposed: The reflected x-ray intensity is constant at 575 K and shows a continuous decay at lower temperatures. In contrast, on a surface precovered with 0.2 monolayer (ML) of Sb, a layer-by-layer growth behavior is observed at temperatures between 225 and 375 K. This is evident from periodic oscillations in the reflected intensity which die out only after -25 ML of Ag have been deposited. Hence, Sb acts as a "surfactant. " These findings were tentatively explained by proposing a repulsive barrier for Ag adatoms to move over descending step edges on the clean surface, which would impede interlayer Ag mass transport. It was suggested that this barrier is reduced due to the Sb being adsorbed at the island edges and that thereby interlayer Ag transport becomes more efficient. From the steep initial decay in the reflected intensity it was concluded that Sb adatoms also act as nucleation centers for Ag islands.This peculiar growth behavior has motivated further work. Oppo, Fiorentini, and Scheffler [7] found in ab initio calculations that Ag is more strongly bound on Ag than on Sb. From this they suggested that the mobility of Ag adatoms should be reduced in the vicinity of Sb and that the resulting increased Ag island density would cause the transition to layer-by-layer growth. That proposal contradicts other available data. Not layer-bylayer, but instead multilayer growth along with increased island dens...

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

“…Additional step-edge growth simulations have shown that A edges grow considerably rougher than B edges, which further validates the effective implementation of the corner-crossing asymmetry in our KMC model [83]. Moreover, the smoothening of island B edges, leading to compact islands with increasing growth temperature, is consistent with Ag/Ag (111) homoepitaxial data [53,54] and shows that our model correctly treats adatom diffusion at the island periphery.…”

“…A physical model based upon a bond-counting scheme is first developed using Ag/Ag(111) homoepitaxy as a starting point, chosen due to the immense experimental and theoretical literature on this film/substrate system [41,43,[53][54][55][56]. The model is validated by showing that simulations: (i) reproduce experimentally observed trends of island morphological evolution during Ag/Ag(111) homoepitaxy as a function of film-growth temperature [54], and (ii) yield equilibrium island shapes upon annealing, at elevated temperatures and in the absence of deposition, of hemispherical islands on weakly interacting substrates [57].…”

Influence of strain in Ag on Al(111) and Al on Ag(100) thin film growth

“…This is in contrast to Ag on Ag͑111͒ homoepitaxy, where limited interlayer transport caused by the step edge barrier results in 3D growth. 7,8 It was proposed that the large density of kink sites originating from the boundaries of stacking-fault islands could offer channels with reduced step edge barriers to promote interlayer diffusion.…”

Initial stages of metal encapsulation during epitaxial growth studied by STM: Rh/Ag(100)

“…A similar reduction of the overall mobility had been observed experimentally, and deduced from ab initio calculations for Ag͑111͒ homoepitaxy in the presence of Sb surfactant atoms. [23][24][25] At the same time that Rh-Ag islands form, the substrate undergoes complementary morphological changes. Specifically, the steps roughen, starting at 0.02 ML, until by 0.30 ML, fjordlike vacancy structures extend from the step edges.…”