Search

Using a modification of the Clavilier crystal preparation method, a series of palladium–platinum (1 0 0) and (1 1 0) single crystal alloy electrodes have been prepared covering the bulk composition range 3% [1] . Voltammetric data displayed systematic changes as a function of bulk palladium composition. These changes could be related to a gradual increase in the population of surface terrace sites by randomly distributed palladium atoms leading to attenuation of voltammetric peaks associated with platinum terraces. It was found also that at certain alloy compositions, the addition of palladium was sufficient to suppress the reconstruction characteristics of the pure platinum (1 0 0) and (1 1 0) surfaces. It was also established that palladium strongly surface segregates. Furthermore, the segregation was surface structure dependent with the more open (1 1 0) surfaces exhibiting greater segregation than the closer-packed (1 0 0) and (1 1 1) surfaces for a fixed bulk single crystal alloy composition. On the basis of these results, it is proposed that the surface segregation characteristics of platinum–palladium alloys are controlled by thermodynamics and that the composition of the selvedge region plays a crucial role in disfavouring surface reconstruction.

Using cyclic voltammetry as a probe of surface morphology combined with AES and LEED, a detailed study is presented of the growth of well-defined rhodium films on Pt(111). Deviations from strictly layer-by-layer growth, as predicted by theoretical plots of the rhodium 302 eV AES peak-to-peak intensity versus time, were elucidated by voltammetry. It was shown that the structure of the rhodium film was temperature dependent. For T ⩽ 400K, “pseudo”-Frank-van der Merwe behaviour is observed whereby the second rhodium monolayer begins forming prior to the completion of the first with both layers giving rise to pseudomorphic (1 × 1) islands. At higher substrate temperatures, the ratio of the amount of second layer to first layer rhodium (for a fixed exposure) increases and at the highest temperatures investigated, evidence for the existence of bulk rhodium microcrystallites was deduced from LEED and voltammetry. The behaviour of rhodium contrasts significantly with palladium on Pt(111) which previously was shown to display good layer-by-layer growth irrespective of temperature or coverage, in agreement with expectations based on lattice mis-match and surface energy criteria. The differences between rhodium and palladium may be understood in terms of the increased strain associated with rhodium films and more especially, the larger surface energy of rhodium as compared to platinum.

The growth of palladium on a reconstructed and deconstructed Pt(100) electrode has been characterised using LEED, AES and cyclic voltammetry. Two quite distinct types of behaviour were observed. Palladium on Pt(100)-(1 × 1) grew essentially layer-by-layer as expected theoretically on the basis of the small lattice mis-match between both metals and the significantly larger surface energy displayed by platinum as compared to palladium. This result is in agreement with previous purely electrochemical studies of the same system. However, for palladium deposition onto the reconstructed Pt(100)-hex-R0.7° surface, the second palladium monolayer forms well before completion of the first. Possible reasons for the departure from layer-by-layer growth are suggested. Finally, it is demonstrated that contact of the electrode with electrolytes containing specifically adsorbing anions switches the Pt(100)-hex-R0.7° phase to unreconstructed (1 × 1) with small (1 × 1) islands of platinum sitting proud of the surface. These “excess” platinum atoms derive from the change in surface density upon deconstruction at ambient temperatures.

Low energy ion scattering spectroscopy (ISS), with both 20 Ne + and 40 Ar + ions at various scattering geometries, has been used to investigate surface structural transformations when silver and gold are deposited on W{100} at coverages up to one monolayer. If half a monolayer of either silver or gold is annealed at 800 K an intense, sharp, c(2×2) LEED pattern is produced. AES demonstrates that there is no significant bulk alloying of the metal overlayer and substrate upon heating. By contrast, ISS shows that a surface rearrangement occurs whereby shadowing of tungsten by both silver and gold is greatly reduced. This behaviour is rationalised in terms of the formation of two-dimensional alloys analogous to those formed by copper on W{100} at submonolayer coverages.

Structural transformations when up to two monolayers of Cu are adsorbed on W(100) and then annealed have been studied by ion scattering spectroscopy, low energy electron diffraction (LEED), Auger electron spectroscopy (AES), work function changes and thermal desorption spectroscopy (TDS). Adsorption at 300 K produces epitaxial growth up to two monolayers, but annealing the surface at ≈700 K induces coverage-dependent structural transformations. In the coverage range 0.2 4 He + collected at various scattering geometries show that this structure consists of a coplanar two-dimensional double-spaced Cu−W surface alloy. AES demonstrates that there is no significant bulk alloying. Annealing at θ=0.75 produces a p(2×2) structure which disorders at 900 K, while with higher Cu coverages the LEED patterns after annealing show disordering and domain formation. These results provide a structural basis for a coherent reinterpretation of the data produced by Bauer and coworkers for the same system.

At normal photoelectron emission from the surface of W{100} an intense surface state has been well characterised. Adsorption of Cu, Ag and Au causes an attenuation of this state, which is completely absent for coverages in excess of one monolayer. This result does not confirm earlier field emission energy distribution studies which appeared to demonstrate the reappearance of the \ gG state at exactly one, two and three monolayers. Further, we find that annealing to 800 K causes attenuation of the \ gG state at even lower coverages, and attribute this to surface alloy formation. Even after annealing there is no return of the \ gG state at higher coverages.

The reactivity of copper, silver and gold films (up to $ 2 monolayers in thickness) on W{100} towards hydrogen chemisorption has been examined using thermal desorption spectroscopy (TDS). For metal overlayers produced during deposition at 300 K, a linear decrease in the amount of hydrogen adsorbed at 300 K is obtained for increasing metal loadings up to one monolayer, by which time all hydrogen chemisorption ceases. It is believed that a site-blocking mechanism is responsible in which four-fold hollow sites are occupied by the metal adsorbate. In contrast, adsorption of hydrogen by metal films annealed at 800 K is more strongly inhibited than films produced at 300K. It is proposed that the irreversible formation of a two-dimensional surface alloy between the tungsten substrate and metal overlayer is the cause of this behaviour. We conclude that dissociative adsorption of room-temperature H 2 does not occur onto a surface containing only mixed W-Cu (Ag or Au) bridge sites. This behaviour is similar to that reported recently by Lambert and co-workers for N 2 adsorbed on annealed Pt/W{100} surface alloys.