Methanol, Formaldehyde, and Formic Acid Adsorption/Oxidation on a Carbon-Supported Pt Nanoparticle Fuel Cell Catalyst: A Comparative Quantitative DEMS Study - Fuel Cell Catalysis: A Surface Science Approach - page 413

The CO ad coverages resulting from 10-minute interaction with formaldehyde,
formic acid and methanol containing solution at 0.11 V were determined by compar-
ing with the mass spectrometric charges for CO 2 obtained for saturated CO adlayer
stripping. Under present adsorption conditions, the CO ad coverages correspond to
about 87% (formaldehyde), 36% (formic acid), and 10% (methanol), respectively,
of that of a saturated CO adlayer (see Table 13.1), i.e., they are significantly lower
than after CO adsorption. It is important to note that the shift of the adsorbate stripping
peak to lower potentials compared with CO ad stripping cannot be explained by the
reduced adsorbate coverage, since in a previous study on similar Pt/Ccatalysts,the
coverage dependence of the CO ad stripping peak after exposure to different amounts of
CO was found to be marginal (,40 mV at 15% of the CO ad saturation coverage) [Behm
and Jusys, 2006]. Other effects, originating, for example, from a different structure
and/or lateral distribution of the adlayer, must contribute as well, and are dominant.
Similar trends to much lower CO ad coverages upon adsorption of methanol,
compared with formaldehyde and formic acid adsorption, at potentials in the H upd
region were found in recent in situ IR spectroscopy [Park et al., 2002] and DEMS
[Lanova et al., 2006] measurements on Pt/Vulcan (adsorption potentials 0.06 V
[Park et al., 2002] and 0.3 V [Lanova et al., 2006], respectively) and on Pt films
[Chen et al., unpublished results] (adsorption potential 0.06 V). Other studies, in con-
trast, reported significantly higher adsorbate coverages, up to 60% of CO ad saturation,
for methanol adsorption on a Pt/Vulcan catalyst in the H upd region [Rice et al., 2000].
The discrepancy between the latter study on the one hand and our as well as other
results on the other may be due to the much longer adsorption time in the experiments
in [Rice et al., 2000]. While slow adsorption of additional methanol cannot be ruled
out, such long-term adsorption experiments can also be affected by adsorption of
trace impurities present in the reactant. Similarly, even very slow oxidation of methanol
to formaldehyde at low potentials [Korzeniewski and Childers, 1998], and subsequent
adsorption of the resulting formaldehyde, could contribute to long-term adsorption
experiments in a stagnant electrolyte.
The pronounced difference in CO ad saturation coverage obtained upon interaction
with methanol-, formaldehyde-, or formic acid-containing solution under present reac-
tion conditions (see Table 13.1) is attributed to a different influence of adsorbed H upd
species on the probability of dehydrogenation of the three C 1 reactants, which are most
likely related to different spatial requirements for the dehydrogenation step [Desai
et al., 2002; Greeley and Mavrikakis, 2002, 2004; Okamoto et al., 2003; Cao et al.,
2005; Taylor and Neurock, 2005; Hartnig and Spohr, 2005]. Potential-dependent
measurements for C 1 adsorption on Pt/C [Jusys and Behm, 2001] indeed showed a
pronounced increase in CO ad coverage with decreasing H upd coverage for methanol
adsorption, while, for formaldehyde adsorption, the resulting CO ad coverage is prac-
tically independent of the adsorption potential and thus of the H upd coverage
[Chen et al., unpublished results]. Hence, formaldehyde adsorption/decomposition
is not inhibited by a H upd adlayer, and can lead to a displacement of the H upd
adlayer by the resulting CO ad . Formic acid dehydrogenation to CO ad is between the
two cases of methanol dehydrogenation and formaldehyde dehydrogenation (for a
quantitative evaluation and discussion, see [Chen et al., 2006a, b, c]).