Angewandte
Chemie
Abstract: Direct propylene epoxidation by O2 is a challenging
reaction because of the strong tendency for complete combus-
tion. Results from the current study demonstrate that by
generating highly dispersed and stabilized Cu+ active sites in
a TiCuOx mixed oxide the epoxidation selectivity can be tuned.
The TiCuOx surface anchors the key surface intermediate, an
oxametallacycle, leading to higher selectivity for epoxidation of
propylene.
Identifying surface intermediates is a critical step to
understand the reaction mechanism as the surface chemistry
of intermediates governs the epoxidation reaction pathway.
For instance, one of the most important intermediates for
ethylene epoxidation has been identified as an ethylene
oxametallacycle, which has an asymmetric ring structure with
one carbon atom binding directly to the surface and the
second carbon atom indirectly binding to the surface through
an oxygen bridge.[6,12,13] Similarly, propylene oxametallacycle
has been proposed as the reaction intermediate for propylene
epoxidation.[14] Even though these two oxametallacycle
intermediates have a nearly identical structure, the surface
chemistry of them is significantly different, resulting in
different activity and selectivity. Specifically, the g-hydrogen
in the propylene oxametallacycle is very active and readily
reacts with surface oxygen, leading to dehydrogenation and
finally complete oxidation of the allyl species. The formation
of the propylene oxametallacycle intermediate has been
attributed to the higher selectivity for epoxidation on the
copper surface.[14] However, the activity of Cu catalysts is low
with a conversion of less than 4% at below 3008C[15] and
deactivation occurs because of the complete oxidation of
copper to Cu2+.[11] Thus, it is important to improve the activity
by stabilizing Cu in low oxidation states under the reaction
conditions.
P
ropylene oxide (PO) is an important precursor largely
produced for manufacturing numerous commodity chemicals
(e.g. polyurethane and polyols). Current industrial production
of propylene oxide is mainly through the chlorohydrin process
and hydroperoxide mediated process, which are neither cost
effective nor environmentally friendly because of generating
chlorinated or peroxycarboxylic waste.[1] Thus, identifying
appropriate catalysts for direct and selective epoxidation of
propylene with molecular oxygen (C3H6 + 1/2O2!C3H6O)
has received considerable attention.[2–9] For example, Au
nanoparticles supported on TiO2 or titanium silicalite zeolites
exhibit high selectivity (> 90%) for PO.[4] However, this
approach requires co-feeding of large amount of H2 and is
limited by low conversion. In contrast to ethylene epoxidation
utilizing Ag-based catalysts, at present industrial-scale heter-
ogeneous catalytic propylene epoxidation process with O2 has
not been successfully developed.
Mixed oxides, such as spinels or perovskites, are stable,
highly ordered, and are characterized by their unique
electronic structure and physical/chemical properties.[16]
Thus, they have become the focus of numerous research
efforts related to catalytic oxidation reactions.[17] For instance,
a mixed oxide made of TiCuOx demonstrates high catalytic
activity for CO oxidation with copper being stabilized as
Cu+.[18] The stabilization of Cu+ by TiOx strongly suggests that
TiCuOx may show promising selectivity for propylene epox-
idation. By stabilizing and controlling the density of the
surface Cu+ sites, TiCuOx may act as a potential candidate for
catalyzing direct propylene epoxidation. To test this hypoth-
esis, in this study a model Ti–Cu oxide catalyst grown on
Cu(111) has been prepared by physical vapor deposition and
its catalytic performance for direct propylene epoxidation is
measured by temperature-programmed desorption (TPD).
The adsorption of propylene oxide and the formation of
oxametallacycle have been also studied by soft X-ray photo-
emission spectroscopy (XPS) and high-resolution electron
energy loss spectroscopy (HREELS). This study shows that
the model catalysts are active and selective for converting
propylene into propylene oxide by direct partial oxidation,
illustrating the importance of forming stable Cu+ sites in
controlling the epoxidation activity and selectivity.
The main challenge for direct propylene epoxidation with
O2 is that instead of partial oxidation, propylene is readily
oxidized to generate CO2 and H2O. Some attempts have been
made to improve the selectivity for PO, that is, reducing the
Ag catalysts to an extremely small size (three atoms of
silver),[10] or illuminating copper oxide catalysts with visible
light to reduce Cu oxide during reaction.[11] Even though
those studies have provided interesting approaches to alter
the selectivity for epoxidation, they may not yet be practical
or effective enough to be applied in large scale production.
More effort is still needed to search for novel oxidation
catalysts that are stable, economic, and active with signifi-
cantly enhanced selectivity to the desirable epoxidation
product.
[*] Dr. X. Yang, Dr. S. Kattel, Dr. K. Mudiyanselage, Dr. S. D. Senanayake,
Dr. J. A. Rodriguez, Dr. P. Liu, Dr. D. J. Stacchiola, Prof. Dr. J. G. Chen
Chemistry Department
Brookhaven National Laboratory
2 Center St., Upton, NY 11973 (USA)
K. Xiong
Department of Chemical and Biomolecular Engineering, University
of Delaware, 150 Academy St., Newark, DE 19716 (USA)
Dr. S. Rykov
The TiCuOx thin film was synthesized by depositing Ti
oxide onto a Cu2O surface and annealing under an O2
environment. The Cu2O surface was prepared by annealing
a Cu(111) in O2 at 650 K.[19] Deposition of TiOx causes
formation of both TiO2 hexagonal islands and Ti modified row
structures. Figure 1A shows the high-resolution STM image
of the surface structure with the Ti-rich regions showing
depressions. The detailed structure of this TiCuOx thin film
has been described in a previous paper[18] and the STM images
with large area are shown in the Supporting Information
Sergei Rykov, Department of Semiconductors Physics and Nano-
electronics, Peter the Great St. Petersburg Polytechnic University
195251 St. Petersburg (Russia)
Prof. Dr. J. G. Chen
Department of Chemical Engineering
Columbia University
500 W. 120th St., New York, NY 10027 (USA)
E-mail: jgchen@columbia.edu
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2015, 54, 11946 –11951
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim