Metalloporphyrin Catalysts of Oxygen Reduction - Fuel Cell Catalysis: A Surface Science Approach - page 668

yet to result in an understanding of the structure/activity relationships for metallo-
porphyrin-catalyzed ORR that would allow a rational design of metalloporphyrin-
based catalysts for the operating conditions of practically useful fuel cells, rather
than physiologically relevant ones. Finally, since even some simple Fe porphyrins
catalyze four-electron reduction of O 2 under specific conditions ( physical contact
with an electrode, pH, potentials), achieving goal (i) means demonstrating highly
selective ORR catalysis under conditions where simple Fe porphyrins are clearly
inadequate catalysts.
So far, certain biomimetic catalysts (1 and 2b in Fig. 18.17) have been shown to
reduce O 2 to H 2 O under a slow electron flux at physiologically relevant conditions
(pH 7, 0.2 - 0.05 V potential vs. NHE) and retain their catalytic activity for .10 4 turn-
overs. Probably, only the increased stability of the turning-over catalyst is of relevance
to the development of practical ORR catalysts for fuel cells. In addition, biomimetic
catalysts of series 1, 2, 3, and 5, and catalyst 4b are the only metalloporphyrins studied
in ORR catalysis with well-defined proximal and distal environments. For series 2,
which is by far the most thoroughly studied series of biomimetic ORR catalysts,
these well-defined environments result in an effective catalysis that seems to be the
least sensitive among all metalloporphyrins to the electrode material (whether the
catalyst is adsorbed or in the film) and to chemicals present in the electrolyte or in
the O 2 stream, including typical catalyst poisons (CO and CN 2 ).
18.6.1 Chemistry of O 2 Adducts of Biomimetic Catalysts
Stoichiometric reactions of O 2 with porphyrins 1, 2a - c, 5c, and 8 in organic solvents
were studied, in addition to an analog of 5c with Co replacing Fe, 5cCo (Fig. 18.18).
Dioxygen adducts with 5c, 5cCo, and 8 were reported to be stable at room temperature
in air and to contain a bridging peroxo moiety, based on infrared spectroscopy, in
addition to mass spectroscopy (5cCo) or 1 H-NMR (5c, 8) [Collman et al., 2003a].
Note that 1 H-NMR spectroscopy revealed that the adducts were diamagnetic, but
did not provide information about the binding mode of the O 2 ligand. These adducts
could not be deoxygenated under vacuum, but could be reduced by cobaltocene,
CoCp 2 . For example, titration of a solution of 5cCoO 2 with CoCp 2 resulted in spectral
changes in the 400 - 600 nm region that were interpreted as indicative of the presence
of only two metalloporphyrin species, the oxygenated porphyrin 5cCoO 2 and free,
fully reduced porphyrin 5cCo (Co II /Cu I ), until 4 equivalents of CoCp 2 had been
added. From these results, it was concluded that reduction of 5cCoO 2 to [5cCoO 2 ] 2
was rate-limiting. Similarly, 4 equivalents were required to reduce 8O 2 , and again,
only two species (8O 2 and 8) could be detected by UV - vis when less than 4 equiva-
lents of CoCp 2 were used. In contrast, 5c was regenerated from 5cO 2 upon addition of
only 2 equivalents of CoCp 2 . Unlike experiments in which O 2 reduction by ferrocenes
is catalyzed by cofacial porphyrins in an organic solvent containing an acid (Section
18.5), titrations of 5cO 2 , 5cCoO 2 , and 8O 2 with CoCp 2 were carried out under anhy-
drous conditions excluding protic sources, which poses two questions: (i) What chemi-
cal form are the O 22 atoms derived from reduction of 5cCoO 2 or 8O 2 with 4
equivalents of CoCp 2 (and what is the form of the O 22 species upon reduction of