Isotope labelling study of CO oxidation-assisted epoxidation of propene. Implications for oxygen activation on Au catalysts

Article abstract of DOI:10.1039/c000374c

¹⁸O isotope labelling studies of the CO oxidation-assisted epoxidation of propene, catalyzed by a mixture of Au/TiO₂ and TS-1, using a methanol-H₂O solvent showed the O in the epoxide was exclusively from O₂ and not H₂O or methanol.

Full text of DOI:10.1039/c000374c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Isotope labelling study of CO oxidation-assisted epoxidation of propene.
Implications for oxygen activation on Au catalystsw
Jian Jiang,a Sean M. Oxford,b Baosong Fu,c Mayfair C. Kung,*b Harold H. Kung*b and
Jiantai Mad
Received 7th January 2010, Accepted 18th March 2010
First published as an Advance Article on the web 15th April 2010
DOI: 10.1039/c000374c
18O isotope labelling studies of the CO oxidation-assisted
epoxidation of propene, catalyzed by a mixture of Au/TiO2
and TS-1, using a methanol–H2O solvent showed the O in the
epoxide was exclusively from O2 and not H2O or methanol.
is thermodynamically highly unfavourable (DG1
=
116.8 kJ molÀ1). Thus, for this reaction to occur during
CO oxidation, it must be mechanistically coupled to CO
oxidation so as to achieve a thermodynamically favourable
(DG1 = À140.4 kJ molÀ1) overall reaction of:
Although the high activity of supported Au catalysts for low
temperature CO oxidation has been known for some time,1,2
the detailed mechanism of this reaction has not been resolved.
Adsorbed CO has been detected by IR under conditions
relevant to catalysis,3,4 but information about the adsorbed
oxygen is noticeably lacking. In particular, the manner in
which molecular oxygen is activated remains unknown. Unlike
other noble metals, such as Pt and Pd, that are less active but
nonetheless good oxidation catalysts, there is no evidence of
dissociative adsorption of oxygen on Au under catalytically
relevant conditions. Conventional temperature programmed
desorption experiments could not detect any oxygen desorption
from a sample that had been treated in a flow of O2, and there
is no useful spectroscopic evidence of adsorbed oxygen.
Computational studies offered some insight into the oxygen
activation/adsorption process. Molecular (associative) adsorption
appears feasible,5–7 and it is enhanced by coadsorption of
water8 or the presence of negative charges on the Au
cluster.8–10 However, the role of this form of adsorbed O2 in
catalysis has yet to be established.11,12 Associative adsorption
of O2 was also proposed in the reaction mechanism without
any supporting evidence.13,14
CO + H2O(l) + O2 - H2O2(l) + CO2
(2)
The fact that eqn (1) is thermodynamically highly unfavourable
argues that the mechanism to form H2O2 cannot involve
dissociative adsorption of O2 to form two identical adsorbed
O atoms, which react independently with H2O in two identical,
parallel pathways. This is because each of these two pathways
would be thermodynamically unfavourable. It follows that the
two O atoms of the activated O2 responsible for H2O2 formation
during CO oxidation must react differently.
In our effort to gain insight into the CO oxidation-assisted
peroxide formation process, we coupled it with epoxidation of
alkene and probed the origin of the O in the epoxide with 18
O
isotope. This Au-catalyzed, liquid phase CO oxidation was
conducted by bubbling a gas mixture of CO, O2, propene, and
He through a suspension of Au/TiO2 and TS-1 catalysts. TS-1
is a titanium substituted silicalite that is active for epoxidation
of alkenes using H2O2.19 The rates of CO, O2, and propene
consumption and formation of volatile products, including
propene oxide (PO), acetone, propanediol, and 2-propanol,
were monitored by analyzing the gas phase products at the
reactor exit periodically with an on-line GC-MS. CO2 was also
analyzed, but its rate of formation could not be calculated
accurately from the exit gas due to significant dissolution into
the liquid. At the conclusion of the experiment, the liquid
phase was analyzed for organic products by GC-MS, and
for peroxides by titration. More experimental details are in
Supplementary Information.w
Recent experimental evidence offered some insight into the
form of adsorbed oxygen important in CO oxidation. When
CO oxidation was conducted in an aqueous medium, significant
amounts of H2O2 were formed, even in the absence of H2.15
Whereas the formation of H2O2 from H2 and O2 is thermo-
dynamically feasible and the reaction can be catalyzed by
Au,16–18 the oxidation of H2O to form H2O2:
Table 1 shows some representative results. (See Table S1w
for a more complete set of results and additional experiments.)
When H2O was the solvent, propanediol was the majority
organic product, most probably formed by hydrolysis of
propene oxide (Exp. 1). However, when the liquid contained
>80% methanol, propene oxide was formed selectively, without
any detectable propanediol (Exp. 2 and 5). This could be a
consequence of the hydrophobic nature of the TS-1 pores that
partitions CH3OH preferentially to H2O. Within the uncertainties
(estimated 10%) in both O2 and propene balances over the
course of the experiment, little combustion or hydration of
propene had occurred under these conditions. Titration identified
the presence of peroxide in the liquid (last column, Table 1),
H2O(l) + 1/2O2 - H2O2(l)
(1)
a On leave from the College of Chemistry and Chemical Engineering,
Lanzhou University, Lanzhou 730000, PRC
b Department of Chemical and Biological Engineering,
Northwestern University, Evanston, IL 60208-3120, USA.
E-mail: hkung@northwestern.edu, m-kung@northwestern.edu
c On leave from the School of Chemistry and Chemical Engineering,
Southeast University, Nanjing 211189, PRC
d College of Chemistry and Chemical Engineering,
Lanzhou University, Lanzhou 730000, PRC
w Electronic supplementary information (ESI) available: Catalyst
preparation and characterization, experimental details, reaction
system, product analyses, and more illustrative reaction results. See
DOI: 10.1039/c000374c
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 3791–3793 | 3791

View Article Online
Table 1 Results for liquid phase CO oxidation-assisted epoxidation of propene
Reaction rate,b mmol minÀ1
Quantity,c 10À5 moles
Liquid composition
Exp.a MeOH–H2O, mL : mL Rxn time, min
CO
O2
Alkene
Total PO, Æ1 Other productsd C3H6 cons. Æ2 –OOHe Æ0.2
1
2
5
7
8
0 : 50
40 : 10
292
232
251
>100
240
2.5
11
4.0
0
1.6
5.8
2.5
0
0.83
0.88
0.81
0
2.7
21
23
0
14.6
Trace
Trace
0
24
22
20
0
0.34
0.72
0.37
Nd
47.5 : 2.5
50 : 0
47.5(EtOH) : 2.5f
1.8
1.1
0.08
1.5
Trace
2 Æ 1
Nd
a
Reaction conditions: 100 mg Au/TiO2 + 150 mg TS-1, 40 1C, 480 kPa total pressure; gas feed: 2.5% CO, 1.25% O2, 1.65% propene in He, total
flow rate 40 mL minÀ1; 50 mL liquid (CH3OH + H2O). Steady state rates, determined by analysis of composition of gas exiting the
b
c
reactor. Sum of products formed or propene consumed over the reaction time indicated, determined from both the gas and liquid phase
d
compositions. Only byproduct detected in any significant amount was propanediol in Exp. 1. Total peroxide (H2O2 + ROOH) in the liquid at
e
f
conclusion of experiment. Ethanol was used instead of methanol.
but the technique could not distinguish H2O2 from other
peroxides, such as CH3OOH. When methanol was replaced
by ethanol (Exp. 8), the CO oxidation rate decreased by a
factor of about 2.5, and propene oxide formation by over a
factor of 10. The slower epoxidation rate in ethanol than in
methanol has been reported, although the decrease was less
than that observed here.20 Control experiments showed little
propene oxide formation when either Au/TiO2 (not shown) or
TS-1 (Exp. 8, Table S1w) was missing from the catalyst
mixture, supporting the hypothesis that a stable intermediate
oxidant was formed on Au/TiO2 that diffuses to TS-1 to effect
epoxidation. CO was necessary (Exp. 9, Table S1w), confirming
that this intermediate oxidant cannot be formed by oxidation
of water or methanol directly, but by a mechanism that is
coupled to CO oxidation. Finally, when methanol was replaced
by acetonitrile in the solvent (47.5/2.5 acetonitrile–H2O, Exp. 9),
no propene oxide was formed, no peroxide was detected in the
liquid by titration, and the CO oxidation activity was low.
H218O was used to determine whether the formation of
propene oxide in experiments using a CH3OH–H2O mixture
was from H2O2. We reasoned that oxidation of H218O would
form H18O16OH, and consequently, the epoxide formed would
be 50 : 50 16O : 18O-labeled. Table 2 shows the results of these
experiments (Exps. 12–14). Only 16O-labeled PO was detected
both in the exit gas and the liquid. In contrast, a distribution
of C16O2, C16O18O, and C18O2, enriched in 18O content,
was detected in the gas phase product. The formation of a
distribution of isotopically labelled CO2 indicates oxygen
scrambling between CO2 and H2O, most likely via a bicarbonate
intermediate.
Complementary experiments were conducted using 18O2 in
the feed. The results (Exp. 15 and 16) show that only
18O-labeled PO was formed until at least 1 h into the experiment,
when small amounts of 16O-labeled PO began to appear.
Interestingly, the CO2 was mostly labelled with 16O, consistent
with rapid isotope scrambling with water. A similar observation
was obtained using the ethanol–H2O solvent (Exp. 17). Only
18O-labelled PO was formed when using 18O2. In addition to
the fact that the methanol results have been repeated (Table 2),
other evidence also suggested that these results were not
experimental artifacts due to scrambling in the detection
system. Injection of a solution of unlabelled propene oxide
in H218O into the GC-MS did not change the cracking pattern
of the propene oxide (Fig. S3w), although a huge m/z = 20
peak due to the labelled water was observed. In the experiments
using H218O, trace amounts of 18O-labelled (>90% labelled)
acetone could be detected, which was likely formed by
hydration of propene followed by oxidative dehydrogenation
of the propanol product. Thus, labelled reactive organics
could be detected accurately.
These results show that H2O, although necessary for CO
oxidation, is not involved directly in the epoxidation reaction.
We believe that the role of moisture is to maintain the activity
of the active site on Au for CO oxidation, as suggested in the
literature.21 Alcohol is needed for epoxidation, which does not
occur in the acetonitrile–water solvent. Whereas the data
unequivocally show the formation of a stable intermediate
oxygen carrier that is formed from O2 on Au/TiO2 and diffuses
to TS-1 to effect epoxidation, unfortunately, there is no direct
observation on the nature of this intermediate. Two logical
possibilities are: H2O2 and CH3OOH (or C2H5OOH if ethanol
is used).
H2O2 could be formed according to Scheme 1a. O2 is
adsorbed on the active site on Au/TiO2 as peroxy or
superoxide, which reacts with CO to form CO2 and an
adsorbed O. The adsorbed O may react with another CO in
a nonproductive pathway, or react with methanol to form a
surface OH and a CH2OH radical. Two surface OH combine
to form H2O2 which carries the same isotope label as O2. This
mechanism would postulate HOCH2CH2OH as a byproduct.
Unfortunately, we searched for but could not detect its
presence, although its concentration might be too low for
detection.
Alternatively, the Au-superoxide picks up a proton from
water or methanol to form Au-peroxide (Scheme 1b), which
reacts with methanol or methoxide to form CH3OOH, in
which the O of the OH would have the same isotope label as
O2. The methylhydroperoxide is responsible for propene
epoxidation. In the literature, it has been reported that stable
alkylperoxide can be formed by catalytic oxidation of alkane
with H2O2 under reaction conditions similar to those used
here.22,23 Reaction of Au-superoxide with CO would lead to
nonproductive consumption of CO.
We have provided isotope labelling evidence to show that
the source of oxygen in propene epoxide in the liquid phase,
Au-catalyzed CO oxidation-assisted epoxidation of propene is
from O2 and not H2O. A stable intermediate, formed on Au
with a mechanism that is coupled to CO oxidation, serves as
ꢀc
This journal is The Royal Society of Chemistry 2010
3792 | Chem. Commun., 2010, 46, 3791–3793

View Article Online
Table 2 Results of 18O-labeled experiments
Isotope distribution in productsb
C18O2 : C18O16O : C16O2
P18O : P16
O
Exp.a
H2O isotope ratio, 18O/16
O
O2 isotope ratio, 18O/16
O
Reaction time, min
12
13
14
1/0
1/0
1/0
0/1
0/1
0/1
310
110
120
240
60
180
120
120
180
4.8 : 6.8 : 1
5.2 : 6.4 : 1
4.8 : 6.3 : 1
4.8 : 6.3 : 1
0 : 0.1 : 1
0 : 1
0 : 1
0 : 1
0 : 1
1 : 0
1 : 0
1 : 0
1 : 0
1 : 0
15
0/1
1/0
0 : 0.1 : 1
16
17
0/1
0/1
1/0
1/0
0 : 0.02 : 1
0 : 0.1 : 1
0 : 0.1 : 1
a
Reaction conditions same as Table 1; CH3OH–H2O = 47.5 : 2.5 except Exp. 16, which was 40 : 10, and Exp. 17 which used ethanol instead of
b
methanol. CO2 in the gas product stream, and propene oxide in both gas and liquid products.
Notes and references
1 M. Haruta, N. Yamada, T. Kobayashi and S. Iijima, J. Catal.,
1989, 115, 301–309.
2 M. C. Kung, R. J. Davis and H. H. Kung, J. Phys. Chem. C, 2007,
111, 11767–11775.
3 F. Menegazzo, M. Manzoli, A. Chiorino, F. Boccuzzi,
T. Tabakova, M. Signoretto, F. Pinna and N. Pernicone,
J. Catal., 2006, 237, 431–434.
4 J. D. Henao, T. Caputo, J. H. Yang, M. C. Kung and H. H. Kung,
J. Phys. Chem. B, 2006, 110, 8689–8700.
5 T. Visart de Bocarme, T. D. Chau, F. Tielens, J. Andres,
P. Gaspard, R. L. C. Wang, H. J. Kreuzer and N. Kruse,
J. Chem. Phys., 2006, 125, 054703.
6 F. Tielens, J. Andres, M. Van Brussel, C. Buess-Hermann and
P. Geerlings, J. Phys. Chem. B, 2005, 109, 7624–7630.
7 X. Ding, Z. Li, J. Yang, J. G. Hou and Q. Zhu, J. Chem. Phys.,
2004, 120, 9594–9600.
8 W. T. Wallace, R. B. Wyrwas, R. L. Whetten, R. Mitric and
V. Bonacic-Koutecky, J. Am. Chem. Soc., 2003, 125, 8408–8414.
9 G. Mills, M. S. Gordon and H. Metiu, J. Chem. Phys., 2003, 118,
4198–4205.
10 L. M. Molina and B. Hammer, J. Chem. Phys., 2005, 123, 161104.
11 A. Roldan, S. Gonzalez, J. M. Ricart and F. Illas, ChemPhysChem,
2009, 10, 348–351.
12 S. Yotsuhashi, Y. Yamada, T. Kishi, W. A. Dino, H. Nakanishi
and H. Kasai, Phys. Rev. B: Condens. Matter Mater. Phys., 2008,
77, 115413.
13 J. J. Bravo-Suarez, K. K. Bando, J. Lu, M. Haruta, T. Fujitani and
S. T. Oyama, J. Phys. Chem., 2008, 112, 1115–1123.
14 J. J. Bravo-Suarez, K. K. Bando, T. Fujitani and S. T. Oyama,
J. Catal., 2008, 257, 32–42.
Scheme 1 Proposed mechanisms for formation of oxygen-carrying
intermediate.
15 W. C. Ketchie, M. Murayama and R. J. Davis, Top. Catal., 2007,
44, 307–317.
16 E. E. Stangland, K. B. Stavens, R. P. Andres and W. N. Delgass,
the oxygen carrier to TS-1 where epxoidation occurs. However,
the nature of this oxygen carrier is not established, and
additional experiments are needed to identify this intermediate
and to establish the reaction mechanism. These results offer a
novel method to perform epoxidation using green reagents,
since the reactants (CO, O2, and alcohol) and the byproduct
(CO2) are environmentally much more friendly than the
typical reagents used, such as nitrous oxide, oxychlorides,
and oxometal complexes.
J. Catal., 2000, 191, 332–347.
17 B. S. Uphade, M. Okumura, N. Yamada, S. Tsubota and
M. Haruta, Stud. Surf. Sci. Catal., 2000, 130, 833–838.
18 E. Sacaliuc, A. M. Beale, B. M. Weckhuysen and T. A. Nijhuis,
J. Catal., 2007, 248, 235–248.
19 M. G. Clerici and P. Ingallina, J. Catal., 1993, 140, 71–83.
20 S. M. Danov, A. V. Sulimov and A. V. Sulimova, Russ. J. Appl.
Chem., 2008, 81, 1963–1966.
21 C. K. Costello, J. H. Yang, H. Y. Law, Y. Wang, J. N. Lin,
L. D. Marks, M. C. Kung and H. H. Kung, Appl. Catal., A, 2003,
243, 15–24.
22 G. B. Shul’pin, M. V. Kirillova, T. Sooknoi and A. J. L. Pombeiro,
Catal. Lett., 2008, 123, 135–141.
Work was supported by the U.S. Department of Energy,
Basic Energy Sciences grant DE-FG02-01ER15184. Jian Jiang
was supported in part by the China Scholarship Council. The
mechanism shown in Scheme 1 was suggested by a reviewer.
23 L. S. Shul’pina, M. V. Kirillova, A. J. L. Pombeiro and
G. B. Shul’pin, Tetrahedron, 2009, 65, 2424–2429.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 3791–3793 | 3793