90
S. Hikazudani et al. / Journal of Molecular Catalysis A: Chemical 358 (2012) 89–98
content, the authors concluded that Ni atoms controlled the Ag
particle size by suppressing the sintering of Ag particles. Miyaji
et al. reported that propylene oxide was obtained selectively
over Ti-MCM-41 impregnated with metal nitrates. The PO yield
increased with increasing Ti content of Ti-MCM-41 and reached
a maximum at an optimum Si/Ti ratio of 100. It is of interest
that calcium nitrate was the most suitable additive and the acid
treatment of the catalyst improved the PO formation. The presence
of both nitrate and molecular oxygen was essential for the PO
formation [6].
In the epoxidation of propylene by N2O, Zhang et al. [7] studied
the use of alkali metal salt (KCl)-modified Fe-MFI and Fe-MCM-41
and reported that extra-framework Fe species (e.g., FeOx clusters)
PO selectivity of 80% was achieved at a propylene conversion of
3.3% over the KCl-modified Fe-MCM-41. The modification with KCl
increased the dispersion of extra-framework Fe species in tetra-
hedral coordination. Moens et al. [8] investigated the catalytic
activity of alkaline- and alkaline-earth-modified silica-supported
metal oxides for epoxidizing propylene with nitrous oxide, and
Fe and Cr oxides showed the highest activities among the oxides
tested. After modification by Rb2SO4, the oxidation rate increased
significantly, and PO was the principal product, with selectivities as
high as 85–90%. Rb2SO4 severely reduced the acidity of the catalyst,
and thus PO isomerization was drastically reduced. The authors
proposed that well-dispersed, distorted tetrahedral Fe3+ sites were
the active sites for epoxidation and that Fe dispersion was con-
trolled by the promoter salts, with both anions and cations being
essential for catalysis. Epoxidation of propylene in liquid-phase
by H2O2 and organic hydroperoxide was also achieved with these
catalysts.
and O2 has been reported. Haruta and co-workers reported that
nanoscale Au particles on titania supports in the presence of H2
and O2 provide a means for highly selective, vapor-phase propylene
oxide production [17–19]. According to their reports, Au nanopar-
reported that among the noble metals, only Au and Ag were active
for PO formation. Haruta and co-workers also found that nanopar-
ticles of Au-Ti-MCM-41 were effective for propylene epoxidation
in the presence of H2 and O2 [20,21]. Ti was hydrothermally incor-
porated into the mesoporous MCM-41 framework, and the authors
found that the catalyst prepared by hydrothermal incorporation
followed by post-synthesis was higher activity than the catalyst
by hydrothermally during synthesis or by post-synthesis grafting.
The presence of more Ti sites isolated from each other is thought
to be responsible for the enhanced activity of the catalysts com-
prising of titanium incorporated by two-step method [20,21]. In
on nonporous TiO2–SiO2 and concluded that isolated TiO4 sites
incorporated in silica surface layers are an important structural fac-
also investigated, and the authors concluded that quick deactiva-
tion of Au catalysts within hours is mainly affected by the surface
properties rather than the pore structure and diffusion limitation
of the supports [23,24]. Delgass and co-workers [25] studied the
location of Au in TS-1 catalysts and concluded that although a Au
particle diameter of 2–5 nm is essential for PO formation, PO for-
mation rates are not strongly influenced by the TS-1 particle size
and are thus not proportional to the specific external surface area
of the support. These investigators also concluded that PO forma-
tion might have occurred on Au particles residing in the channels
of the TS-1 because the observable gold particles decorating the
TS-1 surface account for only ∼30% of the total Au content of the
catalyst. Extremely small Au clusters of 2–5 nm located near Ti
sites inside the TS-1 pores or on the TS-1 surface were active for
propylene epoxidation. Au/TS-1 catalysts with mesoporous-scale
defects that were induced through the addition of carbon pearls
during synthesis had the highest PO formation rate, despite rela-
tively high Au loading, considerable contamination with octahedral
Ti species, and the presence of a significant portion of Au deposited
on the support surface that might have been inactive for epox-
idation [26]. Explaining the resultant fractional reactant orders
(O2 = 0.31 0.04, H2 = 0.60 0.03, and C3H6 = 0.18 0.04) on Au/TS-
1 using the statistical software package JMP requires a sequence
participating in the rate-determining step. The proposed reac-
tion sequence requires that Ti and Au sites must generate and
use the epoxidation oxidant simultaneously rather than sequen-
tially [27]. After investigating the catalytic activity of a series
of Au/TS-1 catalysts with varying Au and Ti contents prepared
by deposition–precipitation and observing transmission electron
microscopy (TEM) images, Delgass and co-workers reported that Au
loading is closely related to Ti loading and that significant activity
is attributable to Au particles much smaller than 2 nm [28]. Oyama
et al. studied propylene epoxidation in H2 and O2 concentrations as
high as 40% over a highly dispersed Au/TS-1 catalyst in a packed-bed
catalytic membrane reactor and found that the rate of PO produc-
tion was improved by 100–200% compared with that observed in a
conventional packed-bed reactor [29]. The above-mentioned stud-
ies indicate that Au/TiO2 catalysts containing Au particles with
diameters of 2–5 nm supported on TiO2 effectively catalyze PO
formation.
We have investigated noble metals for their ability to pro-
mote the oxidation of isobutane and have found that extremely
small concentrations of noble metals supported on SiO2 and
Bi2Mo3O12 affected the catalytic activity and product distribu-
tion in the partial oxidation of iso-butane [30]. Pt, Pd, and
Rh supported on SiO2 at a concentration of 1 × 10−6 atom/nm2
enhanced the catalytic activity, increased the selectivity of the
partial oxidation products isobutene, methacrolein, acetaldehyde,
and acetone, and decreased COx selectivity. Electron paramag-
netic resonance (EPR) measurements indicated that two valent Rh
ions were monoatomically dispersed on the surface and played
a role in the oxidation of iso-butane. Pd, Ir, Rh, and Au sup-
ported on Bi2Mo3O12 at a concentration of 1.1 × 10−5 atom/nm2
increased the conversion of iso-butane, but Pt and Ru decreased
it. The catalytic activity was increased by decreasing the con-
centration of noble metals to 2 × 10−6 atom/nm2; in particular,
Rh supported on Bi2Mo3O12 at this concentration exhibited
increased catalytic activity without affecting the methacrolein
selectivity. We concluded that supported noble metals interact
with active oxygen ions bridging Mo and Bi on the Bi2Mo3O12
surface.
More recently, we studied the reaction of C3H6 over supported
noble metal catalysts in the presence of O2 and H2 and found that
Pd/TiO2 with extremely low Pd concentration is effective for epoxi-
dation of olefins at ambient temperature [31]. In that study, PO was
obtained from propylene (1–2% yield; 30–60% selectivity) together
with propane. The difference between the Pd/TiO2 catalyst used
in that study and a Au catalyst is that the active site for Pd/TiO2
is so small that is monoatomic. Unfortunately, the amount of Pd
used in the catalyst was too small to detect by any spectroscopic
means.
In this paper, we provide a detailed report of the catalytic activ-
ity of Pd/TiO2 toward oxidation of lower-molecular-weight alkanes
in the presence of H2 and O2, and the structure of the catalyst’s
active site. The effects of catalyst preparation conditions on the
reaction rates and the reaction mechanism of PO formation are
discussed.