34 Experimental setups and methods
nanoparticles in Chapter 5. It should be noticed that all these methods rely on
approximations; therefore they don’t necessarily give equivalent results and they
are not necessarily the best choice in the case of extended surfaces. Hence a few
comments are advisable.
The three methods rely on the adsorption of different species on the active
sites of an electrode and on the total desorption of these species from the surface
while raising the applied potential 75, 116. In the case of hydrogen the number
of active sites is calculated by integrating the adsorption-desorption charge of an
adsorbed layer in the so-called hydrogen underpotential deposition (Hupd) region.
The CO-stripping involves the adsorption of a monolayer of CO and the measurement
of its potentiodynamic (or chronoamperometric) oxidation charge. The third
method relies on the under-potential deposition of a layer of Cu (Cuupd) and its
oxidation to Cu2+. In all cases the measured adsorption-desorption charges are expected
to be proportional to the number of active sites and therefore to the ECSA.
All these methods can give satisfactory results for establishing the active surface
area of Pt catalysts. The Hupd is probably the most common choice as it can be
directly derived from the stable CVs. When the CO oxidation reaction is studied,
CO-stripping is also often performed as a further estimate of the ECSA. However,
it is well known that the saturation coverages of H and CO might vary depending
on the chemical properties of a surface; for instance they are highly dependent on
the exposed Pt facets. For the same reason the saturation coverages might strongly
differ in the case of Pt alloys whose surface chemistry is different from pure Pt
2.5.2 Deposition of Pt-Y nanoparticles
Although a good part of this thesis presents fundamental studies on new materials
for the ORR, mainly performed on extended surfaces, Chapters 5 and 6 will focus
on Pt-Y alloys in the more technologically relevant form of nanoparticles. These
were fabricated in a cluster source under UHV conditions as described in this section.
The preparation relies on a magnetron sputter gas aggregation source (Birmingham
Instruments Inc.), combined with time-of-flight mass filtering. The gas aggregation
method consists in the Ar+ sputtering of an Pt-Y alloy target to produce
an atomic vapor that is subsequently condensed into nanoparticles through collissions
with a liquid nitrogen cooled Ar and He gas. For the particles of this work the
alloy target was purchased from Kurt J. Lesker Inc. with nominal composition of
Pt9Y. Compared to chemical or electrochemical methods this technique has the advantage
of forming the particles under UHV conditions where the oxygen pressure
is negligible. Indeed, the high reactivity and oxophilicity of Y (or any other rare
earth) and the very negative standard reduction potentials are the main challenges
for a chemical-electrochemical synthesis. At the same time the particles produced
through this gas aggregation technique are generally ionized, allowing the mass
selection through a time-of-flight filter selecting the desired mass to charge ratio.