Recent Developments in the Electrocatalysis of the O Reduction Reaction - Fuel Cell Catalysis: A Surface Science Approach - page 289

detailed theoretical calculations of free energy changes in all possible ORR elemen-
tary steps, including potential bias, entropy, and zero-point energy corrections accord-
ing to the methodology developed by Nørskov and co-workers [Nørskov et al., 2004],
suggest that OH removal from Pt and from supported Pt ML surfaces and O - O bond
scission are the limiting factors to the ORR performance of these catalysts [Nilekar
and Maurikakis, 2008]. Clearly, regardless the level of detail used in the analysis,
and despite the additional complexity that the electrochemical environment presents,
the intrinsic reactivity of the metal surface remains a crucial factor in determining its
catalytic properties.
An additional factor favoring Pt ML /Pd(111) is the reduced binding energy of OH
on the Pt ML surface compared with Pt(111) and Pd(111), since a high OH coverage
may inhibit the ORR. Voltammetry and in situ XANES [Zhang et al., 2004] have con-
firmed that Pt - OH adsorption occurs at more positive potentials on carbon-supported
Pt ML /Pd nanoparticles (Pt ML /Pd/C) than on Pt/C, which in turn occurs at higher
potential than Pd - OH on Pd/C, in line with the calculations. Increase in the potential
of OH formation on Pt skin on Pt 3 Co(111) versus Pt(111) has also been reported
[Roques et al., 2005]. It stands to reason that catalyst performance may be further
improved if ways to destabilize OH can be devised.
One possibility would be to replace some of the Pt atoms on the surface with atoms
of other metals that oxidize more easily than Pt. At low potentials, these metal atoms
should attract oxygen-containing species (e.g., O and OH) to themselves and, through
the electronic modification of the surface plus the enhanced lateral repulsion among
OH groups, destabilize OH on adjacent Pt sites and decrease the lifetime and coverage
of OH on those sites. As a test, Zhang and co-workers deposited a mixed Pt-Ru and
Pt-Ir monolayer of varying compositions on Pd(111) and tested the ORR reactivity
of these surfaces by performing rotating disk experiments [Zhang et al., 2005b].
The kinetic current density was observed to increase substantially with the Ru mole
fraction up to a maximum at a Pt : Ru ratio of 4 : 1, after which it decreased as the
Pt content of the surface diminished. Similar results were obtained for (Pt - Ir) ML /
Pd(111) (Fig. 9.16). In fact, the kinetic current density for the Pt : M ¼ 4 : 1 compo-
sition is significantly enhanced compared with that on the Pt ML /Pd(111) surface,
which already has a higher ORR activity than Pt(111).
To put these observations into a broader perspective, additional DFT calculations
on (Pt 3 M) ML /Pd(111) and rotating disk experiments on (Pt 0.8 M 0.2 ) ML /Pd(111) with
M ¼ Au, Pd, Rh, Re, or Os were carried out [Zhang et al., 2005b]. The calculated
interaction energies between OH groups and the geometries of the most stable OH
states at

1

2 ML coverage on all the mixed Pt-M monolayer surfaces are shown in
Fig. 9.17a. Negative (positive) interaction energy implies attractive (repulsive) inter-
action between two neighboring OH groups. For metals with oxidation potential com-
parable to or higher than that of Pt (M ¼ Au, Pt, and Pd), the OH groups have an
attractive interaction and prefer to adsorb alternating on bridge and top Pt sites. For
surfaces where M is Rh, Ru, and Ir, all of which have lower oxidation potential
than Pt, the OH groups experience an increased destabilization, with half of
the OH groups being adsorbed atop M atoms and half atop Pt atoms. The repul-
sive interaction is the largest among all the Pt-M monolayers for M ¼ Re and Os.