Visible light-induced photoepoxidation of propene by molecular oxygen over chromia-silica catalysts

Article abstract of DOI:10.1039/b108063f

Highly dispersed chromate species on silica catalyse the selective epoxidation of propene to propene oxide (PO) by molecular oxygen under visible light irradiation with the same quantum yield as that under UV light irradiation.

Full text of DOI:10.1039/b108063f

View Article Online / Journal Homepage / Table of Contents for this issue
Visible light-induced photoepoxidation of propene by molecular
oxygen over chromia–silica catalysts
Chizu Murata, Hisao Yoshida* and Tadashi Hattori
Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Nagoya
464-8603, Japan. E-mail: yoshidah@apchem.nagoya-u.ac.jp
Received (in Cambridge, UK) 5th September 2001, Accepted 9th October 2001
First published as an Advance Article on the web 8th November 2001
Highly dispersed chromate species on silica catalyse the
selective epoxidation of propene to propene oxide (PO) by
molecular oxygen under visible light irradiation with the
same quantum yield as that under UV light irradiation.
the intensity of light from the Xe lamp was measured using a
Hamamatsu Photonic Multi-Channel Analyzer C7473 with a
CCD sensor. Diffuse reflectance UV-vis spectra of the
pretreated samples were measured in vacuo on a JASCO V-570
spectrophotometer at room temperature.
Currently propene epoxidation is performed by two liquid phase
processes: the chlorohydrin process and the hydroperoxide
process.1 These processes have some problems of by-products,
wastes, etc. The direct gas phase epoxidation of propene by
molecular oxygen, which is the most simple process without
these problems, is desirable. Recently some workers reported
new approaches for propene epoxidation using O2 and H2.2
However, it is evident that using only O2 is more advantageous,
and many researchers have attempted epoxidation of propene
using only O2.3
On the other hand, ‘photoepoxidation’ of propene using only
O2 over TiO2,4 Ba-Y type zeolite,5 Nb2O5/SiO2,6 MgO/SiO2
and SiO27 has been reported. However, these activities were low
and it was not clear whether the reaction proceeded catalyt-
ically. We have found that isolated tetrahedral Ti species on
titania–silica catalytically promoted the selective photoepoxida-
tion of propene by molecular oxygen.8 However, the isolated
tetrahedral Ti species absorb only UV light (l < 250 nm), and
the reaction proceeded only under UV irradiation. Artificial UV
light sources tend to be expensive, and the UV light component
in sunlight reaching the surface of the earth is relatively small.
Thus, it is significant to develop a visible light-driven
photocatalyst for epoxidation of propene. It is known that Cr
species have LMCT absorption in the visible light region, and
recently NO decomposition and propane oxidation over Cr-
containing mesoporous silica molecular sieves (Cr-HMS) under
visible light irradiation have been reported.9,10 In our previous
screening study of silica-supported metal oxide catalysts for
propene photoepoxidation, CrOx/SiO2 (1.5 mol% as Cr) showed
high propene conversion, but low selectivity for PO.11
Fig. 1 shows diffuse reflectance UV-vis spectra of Cr/Si(0.1),
Cr–Si(0.1) and Cr/Si(1.5). All samples showed three absorption
bands centred around 245, 330 and 460 nm, which were
assigned to the LMCT (from O22 to Cr6+ charge transfer)
transitions of chromate species.9 Cr/Si(1.5) and Cr–Si(0.1)
showed an additional absorption band in the 580–800 nm region
assigned to the d–d transition of octahedral Cr3+ in Cr2O3
clusters.9 Weckhuysen et al. reported that monochromate and
dichromate species existed over Cr/SiO2 0.2 wt% (0.23 mol%),
and, with increasing Cr loading, additional polychromate
(trichromate, tetrachromate, etc.) species and Cr2O3 clusters
were formed.13 Since Cr/Si(0.1) contained 0.09 wt% of Cr and
exhibited a spectrum without an absorption band at 580–800
nm, Cr/Si(0.1) would have dispersed chromate species (mono-
chromate and/or dichromate) most abundantly. The additional
very weak absorption band at 580–800 nm in the spectrum of
Cr–Si(0.1) indicates that a very small amount of polychromate
species and Cr2O3 clusters would also be formed on Cr–Si(0.1).
Over Cr/Si(1.5), larger amounts of polychromate species and
Cr2O3 clusters would exist, illustrated by the stronger band at
580–800 nm in the spectrum.
Table 1 shows the results of photooxidation of propene over
chromia–silica catalysts. The major products were propene
oxide (PO), ethanal, CO and CO2. Small amounts of propanal,
acetone, acrolein, 2-propanol, ethene and butene were also
observed. Cr/Si(1.5) showed very high selectivity for COx and
low selectivity for PO (run 1) as previously reported.11 Cr/
Si(0.1) and Cr–Si(0.1) showed much higher selectivity for PO
than Cr/Si(1.5) when they were compared at similar conversion
(runs 2 and 3). This indicates that chromia–silica catalysts
containing a very small amount (0.1 mol%) of Cr were efficient
for the photoepoxidation of propene. Cr–Si(0.1) showed
slightly lower selectivity for PO and higher selectivity for
ethanal than Cr/Si(0.1). These results on the photoreaction and
the UV-vis spectra indicate that dispersed chromate species on
SiO2 should be effective in the photoepoxidation of propene,
while polychromate species and Cr2O3 clusters should promote
In the present study, we prepared chromia–silica catalysts
containing a very small amount (0.1 mol%) of Cr by the sol–gel
method or impregnation method, and examined the propene
photoepoxidation activity and the dependence of the activity on
the wavelength of photoirradiation.
CrOx–SiO2 binary oxide was prepared by hydrolysis of a
mixed solution of Si(OC2H5)4 and Cr(NO3)3·9H2O dissolved in
ethylene glycol followed by calcination at 773 K in flowing air
for 5 h.12 CrOx/SiO2 supported oxides were prepared by
impregnation method with amorphous silica and
Cr(NO3)3·9H2O aqueous solution, followed by calcination in
the same way.11 CrOx–SiO2 and CrOx/SiO2 are denoted as Cr–
Si(x) and Cr/Si(x) respectively, where x is mol% of Cr; nCr/(nCr
+ nSi) 3 100. Prior to each reaction test and spectroscopic
measurement, the sample was treated with 100 Torr oxygen at
773 K for 1 h, followed by evacuation at 673 K for 1 h. The
photoepoxidation of propene was performed with a conven-
tional closed system (123 cm3) and 200 W Xe lamp in the same
manner as previously reported.8The wavelength of photo-
irradiation light was limited by using TOSHIBA UV-cut glass
filters: UV-31 and Y-43, which allow the transmission of light
with l > 310 nm and 430 nm, respectively. The distribution of
Fig. 1 Diffuse reflectance UV-vis spectra of chromia–silica catalysts. (a) Cr/
Si(0.1), (b) Cr–Si(0.1) (broken line) and (c) Cr/Si(1.5). The samples were
evacuated at 673 K.
2412
Chem. Commun., 2001, 2412–2413
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b108063f

View Article Online
Table 1 Results of the photooxidation of propene over the chromia–silica catalysta
Selectivityf (C%)e
PO
SAb/
m2 g21
Conv.d
Timec/h (C%)
PO yield
(C%)
Run
Catalyst
Filter
Propanal Acetone Acrolein Ethanal HC
COx
1
2
3
4
5
Cr/Si(1.5)g
Cr/Si(0.1)
Cr–Si(0.1)
Cr–Si(0.1)
Cr–Si(0.1)
573
537
382
382
382
no
no
no
UV-31
Y-43
1
2
2
2
2
16.9
16.7
17.8
12.5
7.7
0.6
7.3
5.7
4.2
2.5
3.7
44.0
32.0
33.9
31.8
1.7
2.7
4.6
9.2
9.2
1.9
5.8
5.3
3.5
3.0
4.8
4.3
4.8
5.0
5.8
15.2
17.9
22.2
30.7
30.8
8.9
3.7
5.2
2.4
3.3
61.5
19.6
19.5
14.8
15.7
a Catalyst 0.2 g, propene 100 mmol, O2 200 mmol, reaction time 2 h. b BET surface area. c Reaction time. d Conversion. e Based on introduced propene. f PO,
propene oxide; HC, ethene + butenes; COx, CO + CO2. A very small amount of 2-propanol was also observed, but is not shown here. g From ref. 11.
Cr–Si(0.1) and light intensity irradiated from the Xe lamp
through UV cut-off filters at each wavelength. With increasing
the relative amount of photons absorbed by Cr–Si(0.1), the
conversion and PO yield increased linearly (Fig. 3). This result
means that chromate species excited by UV or visible light
would be equally efficient in the photoepoxidation of propene;
that is, the quantum yield of photoepoxidation under visible
light would be equal to that under UV light. Thus, exclusion of
UV light changed neither the selectivity nor the quantum yield
of propene photoepoxidation over chromate species. Therefore,
it is suggested that chromate species excited by UV or visible
light identically catalyse the photoepoxidation of propene,
Fig. 2 Time course of photooxidation of propene over Cr–Si(0.1).
although the energy of photons was different at each wave-
Conversion (Ω), PO yield (2), and selectivity to propene oxide (5), ethanal
length. This should mean that the energy of visible light is
(!) and CO + CO2 (-).
sufficient to promote photoepoxidation of propene, that is, to
activate oxygen and/or propene on the catalyst.
the oxidation of propene to by-products such as ethanal and
In conclusion, highly dispersed chromate species on SiO2
COx. In addition, the turnover number, TON = (the amount of
were found to catalyse propene epoxidation by molecular
produced PO) / (the amount of Cr on sample), exceeded 2 after
oxygen under photoirradiation, and even under visible light
12 h irradiation over Cr–Si(0.1) (Fig. 2), which indicated that
irradiation. Chromate species excited by visible light would
this reaction proceeded catalytically.
promote propene epoxidation identically to those excited by UV
The effective wavelength for photoepoxidation of propene
light.
over the Cr–Si(0.1) catalyst was examined using UV cut-off
This work was partly supported by a Grant-in-Aid for
filters (Table 1, runs 4 and 5). Even under visible light (l > 430
Encouragement of Young Scientists from the Ministry of
Education, Culture, Sports, Science and Technology (MEXT),
nm) irradiation (run 5), the conversion and PO yield almost
reached half of the run without the filter (run 3). This means that
Japan.
visible light is sufficient to promote photoepoxidation of
propene over chromia–silica catalysts. By using the UV cut-off
filters, selectivity for PO was not significantly changed, but
Notes and references
1 G. Centi, F. Cavani and F. Trifirò, in Selective Oxidation by
Heterogeneous Catalysis, Kluwer Academic/Plenum Publishers, New
York, 2001, p. 101.
2 Y. Wang and K. Otuka, J. Catal., 1995, 157, 450; T. Hayashi, K. Tanaka
and M. Haruta, J. Catal., 1998, 178, 566.
3 G. Lu and X. Zuo, Catal. Lett., 1999, 58, 67; T. A. Nijhuis, S. Musch,
M. Makkee and J. A. Moulijn, Appl. Catal. A, 2000, 196, 217; K. Murata
and Y. Kiyozumi, Chem. Commun., 2001, 1356.
4 P. Pichat, J. Herrmann, J. Disdier and M. Mozzanega, J. Phys. Chem.,
1979, 83, 3122.
5 F. Blatter, H. Sun and H. Frei, Catal. Lett., 1995, 35, 1; F. Blatter, H.
Sun, S. Vasenkov and H. Frei, Catal. Today, 1998, 41, 297; Y. Xiang,
S. C. Larsen and V. H. Grassian, J. Am. Chem. Soc., 1999, 121, 5063.
6 T. Tanaka, H. Nojima, H. Yoshida, H. Nakagawa, T. Funabiki and S.
Yoshida, Catal Today, 1993, 16, 297.
ethanal selectivity increased and COx selectivity decreased. As
shown in Fig. 2, in the range of conversion 7–18% (1–4 h)
ethanal selectivity decreased and COx selectivity increased with
increasing conversion, but selectivity to PO was not so affected
by the conversion. The differences in ethanal and COx
selectivity among runs 3–5 can be attributed to the difference in
conversion, not in the wavelength of light. In other words, the
product distribution would not be affected by the wavelength of
light, which suggests that the chromate species excited by UV or
visible light catalyse the photoepoxidation of propene via the
same mechanism, regardless of the absorbed wavelength. The
relative amount of photons absorbed by Cr–Si(0.1) was
estimated from summing the products of absorption intensity of
7 H. Yoshida, T. Tanaka, M. Yamamoto, T. Funabiki and S. Yoshida,
Chem. Commun., 1996, 2125; H. Yoshida, T. Tanaka, M. Yamamoto, T.
Yoshida, T. Funabiki and S. Yoshida, J. Catal., 1997, 171, 351.
8 H. Yoshida, C. Murata and T. Hattori, Chem. Commun., 1999, 1551.
9 B. M. Weckhuysen, A. A. Verberckmoes, A. L. Buttiens and R. A.
Schoonheydt, J. Phys. Chem., 1994, 98, 579; B. M. Weckhuysen, I. E.
Wachs and R. A. Schoonheydt, Chem. Rev., 1996, 96, 3327.
10 H. Yamashita, K. Yoshizawa, M. Ariyuki, S. Higashimoto, M. Che and
M. Anpo, Chem. Commun., 2001, 435.
11 H. Yoshida, C. Murata and T. Hattori, J. Catal., 2000, 194, 364.
12 A. Ueno, H. Suzuki and Y. Kotera, J. Chem. Soc., Faraday Trans. 1,
1983, 79, 127.
13 B. M. Weckhuysen, R. A. Schoonheydt, J-M Jehng, I. E. Wachs, S. J.
Cho, R. Ryoo, S. Kijlstra and E. Poels, J. Chem. Soc., Faraday Trans.,
1995, 91, 3245.
Fig. 3 The plot of conversion (5) and PO yield (2) in photooxidation of
propene over Cr–Si(0.1) against the relative amount of photons absorbed by
Cr–Si(0.1).
Chem. Commun., 2001, 2412–2413
2413