Time-resolved in situ Raman spectroscopy was used to follow the
decomposition of supported crystalline
BaO2 into defect-rich BaO. This decomposition occurred
under O2 partial pressures which were small
compared
to those expected from a calculated phase diagram. However, the
O2 partial pressures at which this
BaO2
decomposition takes place are much greater than those which are reached
during the catalytic decomposition
of NO over this material. Additionally, supported crystalline
BaO2 was never observed during the actual
catalytic reaction. Therefore, it must be concluded that
crystalline BaO2 does not play a role in the
catalytic
reaction. In addition to crystalline BaO2, with a
major Raman band at 848 cm-1 and components
at 826 and
814 cm-1, another peroxide species which had
a band at 947 cm-1 was identified by
18O isotopic labeling.
Other bands located in this region of the spectrum may be due to
peroxide ions in defect-rich BaO/BaO2;
however, 18O isotope labeling did not provide definitive
evidence for their assignment. The peroxide species
with a Raman band at 947 cm-1 was stable only
in the presence of gas-phase O2 and responded very
rapidly
to changes in the O2 partial pressure. Therefore, it
is likely that these peroxide ions are located at or close
to the surface of defect-rich BaO crystals. It was previously
shown that the species responsible for the Raman
band at 947 cm-1 immediately reacted with NO
to form an active Ba−nitro intermediate. This
peroxide
species on defect-rich BaO may therefore play an important role in the
catalytic decomposition of NO.