Average mesoscopic properties

A structural surface characterization with high resolution is always restricted to a small fraction of the total surface area. The question of how representative the characterized fraction is for the whole sample is thus crucial. The utilization of well-defined substrate electrodes facilitates the interpretation of local structural information to a large extent. Starting from well-defined substrate surfaces, composite model electrodes with defined mesoscopic structures can be prepared. Out of the various approaches two examples will be discussed, one describing the nanometer-scale modification of Pt(lll) surfaces by ruthenium, the other the modification of conductive surfaces by nanometer-sized particles.

The first example addresses ruthenium-modified platinum electrodes, which show an enhanced electrochemical activity for the oxidation of H2/CO gas mixtures as compared with pure platinum [4, 5]. This makes them interesting electrocatalysts for low-temperature fuel cell applications. Here we discuss the mesoscopic structure of

Ru/Pt(lll) electrodes only; the catalytic implications are described elsewhere [6], The current efficiency of Ru deposition from RuCb solutions is low. This allows an easy tuning of the amount of deposition in the submonolayer regime. The amount of electrodeposited Ru cannot be directly derived from integration of the charge during the deposition process (low current efficiency), but a phenomenological approach was developed for its determination by potentiodynamic electrodissolution of the previously deposited Ru [4]. From this approach, which can only be applied to polycrystalline platinum substrates, a relation between coverage and deposition potential was derived. Deposition for 5 min on polycrystalline platinum from a 5 mM solution of RuCU in 0.1 M H2SO4 resulted in coverages of 6 » 0.3 and 6 « 0.7 of a monolayer at 0.6 V and 0.3 V (RHE), respectively.

STM results related to these conditions are shown in Fig. 1. The electrode surfaces for this sequence were prepared in a standard electrochemical glass cell then, protected with a drop of water, transferred to the STM cell. Figure 1(a) shows in a blank experiment the surface of the Pt( 111) electrode without any Ru deposit. The terrace structure of the (111) surface is clearly recognized. The islands on the terraces can be attributed either to insufficient flame annealing of the surface or to contamination during the transfer of the electrode. Fig. 1(b) and (c) show two examples of Ru-modified Pt(lll) electrodes. In Fig. 1(b), a considerable density of islands is observed on the surface. The islands are between 2 and 5 nm in diameter and their height is close to the height of a monatomic step of the substrate. From the evaluation of several STM images it was derived that the coverage of the surface with islands varies between 0.25 and 0.38. This is consistent with a Ru coverage of 0.3, which was determined by electrochemical techniques for a polycrystalline platinum electrode subjected to the same deposition conditions. These results confirm the approach of Watanabe and Motoo [4], and they allow the conclusion that the observed islands represent the Ru deposit, which was recently also supported by X-ray surface diffraction data. In addition, it indicates that the deposition of Ru is comparable for polycrystalline platinum and Pt(lll). In order to cross-check the results, a Ru-modified Pt(lll) electrode was prepared under conditions, where a coverage of 0.7 is expected. An image of this surface is shown in Fig. 1(c). The surface is densely covered with structures similar to those in Fig. 1(b). The coverage is, however, too high for a detailed evaluation, but the image is in reasonable agreement with a coverage of 6 » 0.7.

The second example concerns an important class of technical electrodes which are based on dispersed catalyst particles. For those, the relation between structure and reactivity is important [7], but its elucidation is hampered by the problem that reactivity is usually referred to a macroscopic sample while the structural characterization gives local information. For fundamental investigations it would be

Fig. 1. Ru-modified Pt(l 11) electrodes in 0.1 M HC104 at E = 0.5 RHE, Et = 0.8 V, 7t = 0.4 nA (A) without Ru deposit, (B) after Ru deposition at 0.6 V, (C) after Ru deposition at 0.3 V; deposition time 5 min for (B) and (C).

desirable to control the average mesoscopic structure of such electrodes, i.e., to control the size and distribution of the catalyst particles on the substrate surface.

Recently a technique for the preparation of catalyst particles with a narrow size distribution was developed [8], yielding colloidal metal clusters stabilized by a shell of surfactants. By adsorbing these clusters on substrate surfaces, model electrodes for dispersed electrocatalysts can be prepared [9]! Figure 2 compares two samples prepared from different colloidal solutions of such clusters adsorbed on a gold surface. It is evident that both samples differ significantly with respect to their mesoscopic structure.

0 0

Post a comment