Size Effects in Electrocatalysis of Fuel Cell Reactions on Supported Metal Nanoparticles - Fuel Cell Catalysis: A Surface Science Approach - page 521

Considering the above discussion, the currents in Fig. 15.5 are normalized to the
surface areas estimated from CO stripping. Similar to massive electrodes, CVs for
Pt nanoparticles show three characteristic potential regions (all potentials vs. RHE):
(i) an H UPD region 0.05 V , E , 0.40 V, followed by (ii) the so-called “double-
layer region” 0.40 V , E , 0.60 V and (iii) for E . 0.7 V, the oxygen adsorption/
desorption region.
Two adsorption peaks and three desorption peaks can be distinguished in the H UPD
region. Based on the data obtained for basal [Clavilier et al., 1980] and stepped [Attard
et al., 2004; Climent et al., 2006] Pt single-crystal electrodes, the so-called “strongly
bound” (about 0.25 - 0.28 V vs. RHE) hydrogen may be attributed to processes on
Pt(111) and Pt(100) facets [Clavilier et al., 1980, 1981a, b; Markovic and Ross
2002], while “weakly bound” hydrogen (about 0.13 - 0.15 V vs. RHE) may be tenta-
tively assigned to Pt(110) sites [Clavilier et al., 1980; Inaba et al., 2006; Komanicky
et al., 2005]. We have observed that decreasing the size of the crystallites results in
suppression of H UPD charge, which is especially remarkable in the potential interval
of “strongly bound” hydrogen. Note that suppression of hydrogen adsorption has
also been claimed by Zoval and co-workers for Pt nanoparticles with d 4nm
supported on HOPG [Zoval et al., 1998].
There is no consensus in the literature regarding H UPD adsorption sites on Pt metals.
It is usually believed that H UPD occupies multicoordinated sites (three-fold sites on
(111) and four-fold sites on (100) surfaces) (see Jerkiewicz [1998] and references
therein). However, somewhat conflicting evidence comes from infrared - visible
sum frequency generation (IR-VIS SFG) investigations by Tadjeddine and
co-workers, which favor monocoordinated H UPD [Peremans and Tadjeddine, 1994;
Tadjeddine and Peremans, 1996]. According to DFT calculations by Watson and
co-workers, on Pt(111), the sites with maximum adsorption energy are the atop and
three-fold hollow sites [Watson et al., 2002]. The study by Teliska and co-workers,
performed for highly dispersed 1.5 - 2.0 nm Pt particles supported on carbon in an
HClO 4 electrolyte with Pt L 2,3 XAS combined with real-space full-multiple-scattering
calculations on model clusters [Teliska et al., 2004], lead the authors to the following
conclusions: (i) at low coverage, a chemisorbed hydrogen atom is highly mobile and
possibly delocalized on the surface; (ii) at higher coverage, it localizes into fcc sites;
and (iii) at very high coverage, H is also found in atop sites, presumably at or near
edges. Note that a study of bimetallic Pd x Au y electrodes of different compositions
has shown that hydrogen adsorption coverage correlates with the fraction of Pd 3
sites, while CO adsorption correlates with Pd 1 sites [Maroun et al., 2001].
Given the above discussion, the observed decrease in the fraction of “strongly
adsorbed” hydrogen may be tentatively ascribed to the “ensemble effect” related to
the decrease of the fraction of the most energetically favourable three-fold hollow
sites with decreasing particle size. On the other hand, one cannot exclude a possible
“electronic effect” resulting in a change in the adsorption energy with decreasing
size of Pt crystallites. Yet another factor that may have an impact on H UPD on Pt nano-
particles is the negative shift of the potential of total zero charge with decreasing par-
ticle size documented by Mayrhofer and co-workers [Mayrhofer et al., 2005b]. In any
case, the observation of reduced H UPD coverage for nanometer-size Pt particles may