One-step gas-phase catalytic oxidation of benzene to phenol with molecular oxygen over Cu-supported ZSM-5 zeolites

Article abstract of DOI:10.1039/b209706k

The gas-phase catalytic oxidation of benzene was carried out using molecular oxygen as an oxidant over Cu-supported catalysts. Phenol was formed only when using Cu-supported high-silica-type zeolites as catalysts. The Cu-supported H-ZSM-5 zeolite (CuH-ZSM-5) was the most active among those tested. Deep oxidation products (CO and CO₂) were also produced together with phenol and they were suggested to form either via phenol or directly from benzene, based on the activity test experiments by varying the contact time. The yield and selectivity of phenol had a maximum with respect to the Cu contents in the CuH-ZSM-5, and were optimized with ca. a 0.77 wt.% Cu loading (phenol yield ～4.9%). The calcination of the CuH-ZSM-5 at the higher temperature of around 1123 K was found to be effective for the phenol formation. The H-ZSM-5 with lower Si/A1 atomic ratios was an effective support for the phenol formation. The EPR study revealed that the increase in the square-pyramidal Cu²⁺ species was brought about when the CuH-ZSM-5 was calcined at higher temperatures or when HZSM-5 with lower Si/A1 atomic ratios was used as a support. The isolated square-pyramidal Cu²⁺ species was suggested to be the active species for the phenol formation.

Full text of DOI:10.1039/b209706k

View Article Online / Journal Homepage / Table of Contents for this issue
One-step gas-phase catalytic oxidation of benzene to phenol with
molecular oxygen over Cu-supported ZSM-5 zeolites
Rei Hamada,a Yusuke Shibata,b Satoru Nishiyamab and Shigeru Tsuruya*b
a
Department of Molecular Science, Graduate School of Science and Technology,
Kobe University, Rokkodai, Nada, Kobe 657-8501, Japan
Department of Chemical Science and Engineering, Faculty of Engineering, Kobe University,
Rokkodai, Nada, Kobe 657-8501, Japan. E-mail: tsuruya@cx.kobe-u.ac.jp
b
Received 7th October 2002, Accepted 13th January 2003
First published as an Advance Article on the web 23rd January 2003
The gas-phase catalytic oxidation of benzene was carried out using molecular oxygen as an oxidant over
Cu-supported catalysts. Phenol was formed only when using Cu-supported high-silica-type zeolites as catalysts.
The Cu-supported H-ZSM-5 zeolite (CuH-ZSM-5) was the most active among those tested. Deep oxidation
products (CO and CO2) were also produced together with phenol and they were suggested to form either via
phenol or directly from benzene, based on the activity test experiments by varying the contact time. The yield
and selectivity of phenol had a maximum with respect to the Cu contents in the CuH-ZSM-5, and were
optimized with ca. a 0.77 wt.% Cu loading (phenol yield ꢀ4.9%). The calcination of the CuH-ZSM-5 at the
higher temperature of around 1123 K was found to be effective for the phenol formation. The H-ZSM-5 with
lower Si/Al atomic ratios was an effective support for the phenol formation. The EPR study revealed that the
increase in the square-pyramidal Cu2+ species was brought about when the CuH-ZSM-5 was calcined at higher
temperatures or when HZSM-5 with lower Si/Al atomic ratios was used as a support. The isolated square-
pyramidal Cu2+ species was suggested to be the active species for the phenol formation.
catalytic system using a copper–palladium composite catalyst
supported on silica and thereby obtained phenol by feeding
hydrogen and oxygen under ambient conditions.
1
Introduction
Phenol has been known to be a versatile chemical as an inter-
mediate for the manufacture of various petrochemicals such
as phenolic resins, adipic acid, caprolactum, bisphenol, nitro-
and chlorophenols, phenol sulfonic acid, and so on. At the
present time, phenol is commercially produced via the so-
called ‘‘Cumene Process’’ which accounts for more than
90% of the world output of phenol. This process includes
three steps: (i) alkylation of benzene with propylene to give
cumene (isopropylbenzene), (ii) its oxidation to cumene
hydroperoxide, and (iii) its hydrolysis to phenol and acetone.
The advantage of the ‘‘Cumene Process’’ is that it takes two
inexpensive starting materials, benzene and propylene, and
converts them into two expensive useful products, phenol
and acetone, just using air. However, the economics of the
process significantly depends on the marketability of the acet-
one by-product. The need of a high capital investment due to
its characteristic multi-step process is also a disadvantage.
Furthermore, hazardous compounds that must be appropri-
ately handled are exhausted. For these reasons, a process is
desired whereby phenol can be formed in one step. Many
approaches for this objective have been reported in the litera-
There is currently a considerable interest in the gas-phase
catalytic oxidation process for phenol manufacture because
the gas-phase reaction process has a potential advantage
over the corresponding liquid-phase process from an eco-
nomic point of view. Since the first finding by Iwamoto
and co-workers,14 the gas-phase oxidation of benzene to
phenol with nitrogen monoxide has been intensively studied
by many authors.15–21 However, N2O is expensive for use
as an oxidant on a large scale. Sasaki and co-workers have
demonstrated the benzene oxidation to phenol with high
selectivity in the presence of both O2 and H2 over a sup-
ported Cu–Pd binary catalyst.22,23 However, the use of a
mixture of H2 and O2 have the risk of explosion. Molecular
oxygen is economically favorable as an oxidant because it is
easy to handle and readily available. Nevertheless, a limited
number of reports that describe the gas-phase oxidation of
benzene to phenol with molecular oxygen have already been
published.24–26 For example, Yamanaka and co-workers
obtained phenol through benzene oxidation with molecular
oxygen in the presence of steam over a V–Mo catalyst sup-
ported on SiO2 .27 Up till now, however, the yield and/or
the selectivity are much lower to be competitive with the
‘‘Cumene Process’’.
ture. Classical systems such as Fenton’s reagent (Fe2+
–
H2O2),1 Udenfriend’s reagent (Fe2+–EDTA–O2–L–ascorbic
acid),2 and Hamilton’s reagent (Fe3+–catechol–H2O2)3 have
been known to yield phenol from benzene. Although the yield
based on the added metal ion is rather low, these reaction
systems have received considerable attention up to now, prin-
cipally from the viewpoint of the reaction mechanism.4–7
More recently, Sasaki and co-workers8–11 reported the
liquid-phase oxidation of benzene to phenol using a homoge-
neous CuCl–O2 system in sulfuric acid solution. Sasaki and
co-workers also12,13 reported a liquid-phase heterogeneous
Previously, we reported the liquid-phase oxidation of ben-
zene to phenol with molecular oxygen under atmospheric pres-
sure at room temperature using Cu2+ ion-exchanged zeolites28
and MCM-4129 as well as the co-precipitated CuO–Al2O3
mixed oxide30 in the presence of ascorbic acid as a reducing
reagent for Cu species. Recently we reported the gas-phase cat-
alytic oxidation of benzene to phenol with molecular oxygen
over Cu ion-exchanged ZSM-5 zeolites (Cu-ZSM-5).31
956
Phys. Chem. Chem. Phys., 2003, 5, 956–965
This journal is # The Owner Societies 2003
DOI: 10.1039/b209706k

View Article Online
In the present paper, we demonstrate the gas-phase catalytic
oxidation of benzene with molecular oxygen over a Cu-
supported H-ZSM-5 zeolite (CuH-ZSM-5) prepared by the
solid-state reaction method (SR) and/or conventional impreg-
nation method (IMP). The influence of various parameters
were investigated, such as the reaction conditions, the Cu con-
tents, the calcination temperature of CuH-ZSM-5, and the
Si/Al atomic ratio of the parent ZSM-5. Both the calcination
temperature of CuH-ZSM-5 and the Si/Al atomic ratio sig-
nificanly influenced the yield and the selectivity of phenol.
The CuH-ZSM-5 catalysts were characterized by various tech-
niques including XRD, XAFS, and EPR.
the catalyst was pre-treated in flowing air (N2 ¼ 33 ml minꢁ1
;
O2 ¼ 8.3 ml minꢁ1) at 773 K for 2 h. The reactor was then
purged with flowing N2 flow (80 ml minꢁ1) for 30 min, fol-
lowed by cooling the catalyst bed to the desired reaction tem-
perature (typically to 673 K). The reaction was begun by
feeding benzene at a specific rate using a microfeeder (typically
7.0 ꢂ 10ꢁ5 mol minꢁ1) simultaneously with N2 and O2 (typi-
cally 32 ml minꢁ1, 8 ml minꢁ1, respectively) into the reactor.
The contact time, the ratio of catalyst weight to the whole feed
rate (W/F), is typically 2.36 ꢂ 102 g-cat min molꢁ1. The reac-
tion products were collected every 1 h using a gastight syringe
and a glass trap cooled with a refrigerant (liquid N2 + diethyl
malonate) to 223 K for the gaseous products and other organic
products, respectively. The organic products were analyzed
using a GC equipped with an FID (Shimadzu GC-14B) using
a 4 m stainless steel column packed with silicon OV-17 at
453–493 K under a N2 carrier. A 1.5 m stainless steel column
packed with Thermon-3000 5% on SHINCARBON A at
473 K was used under a N2 carrier for hydroquinone. The gas-
eous products were analyzed by the intermediate cell method
with a GC equipped with a TCD (Shimadzu GC-8A) using
stainless steel columns containing activated charcoal (1 m)
and 5A molecular sieves (1 m) at 393 K and 413 K, respec-
tively, under a H2 carrier. The yield of the oxidation products
and the selectivity of phenol were obtained as the average
values of the time-on stream of 2, 3, and 4 h. The yield and
selectivity of each of the products were defined as follows:
2
Experimental
2.1 Catalysts
The ZSM-5 zeolites (Si/Al ¼ 32, 75, 100, 123, and 160) were
prepared by a conventional hydrothermal synthesis method
according to a patent.32 The as-synthesized ZSM-5 zeolite
was ion-exchanged in a 1 M NaNO3 aqueous solution at 353
K for 2 h, followed by washing with deionized water, dried
at 393 K overnight in an oven, and finally calcined at 773 K
for 5 h in flowing air to obtain the Na-ZSM-5 zeolite. To
obtain the H-ZSM-5 zeolite, the calcined NaZSM-5 zeolite
was ion-exchanged three times in a 1 M NH4NO3 aqueous
solution at 353 K for 6 h, washed, dried at 393 K overnight,
and calcined at 773 K for 5 h in flowing air. Na-X (Tosoh),
Na-Y (Tosoh), H-mordenite (Tosoh, JRC-Z-M2O) and
H-beta (Tosoh, JRC-Z-HB25) were ion-exchanged in an
NH4NO3 aqueous solution to prepare the H-type counterpart
of each zeolite. Al2O3 (Nikki Chemical, JRC-ALO-6), SiO2
(Nikki Chemical, JRC-SIO-4) and SiO2–Al2O3 (Nikki Chemi-
cal, N631HN) were used after being calcined at 773 K for 5 h
in flowing air. H-Al-MCM-41 was prepared by a hydrothermal
method as previously described in detail.8 The compositions of
the zeolites (Si/Al atomic ratio) were determined by atomic
absorption spectroscopy (AAS) after homogeneously being
dissolved using a few drops of hydrofluoric acid. The Cu-con-
taining catalysts were prepared by the so-called solid-state
reaction (SR) method.33 In the preparation, a mixture of Cu
halide (CuCl was used unless otherwise noted, guaranteed
reagent, Nacalai tesque) and a support powder such as H-
ZSM-5 was carefully ground in an agate mortar. The ground
mixture was heated in flowing N2 (100 ml minꢁ1) for 17 h at
a given temperature (773–1273 K). The Cu-supported catalysts
prepared by the solid-state reaction method will be designated
in the present paper as, for instance, SR (1123)-Cu-H-ZSM-5,
in which the figure in parenthesis indicates the calcination tem-
perature in K unit. The Cu-H-ZSM-5 catalysts prepared by a
conventional impregnation method using copper acetate were
calcined at 1123 K in flowing N2 for 5 h. They are designated
as the IMP-Cu-H-ZSM-5 catalysts.
½phenolꢃ
Yield of phenol ð%Þ ¼
ꢂ 100
ð1Þ
ð2Þ
ð3Þ
½benzeneꢃ
ð1=6Þ ꢂ ½CO2ꢃ
Yield of CO2 ð%Þ ¼
ꢂ 100
½benzeneꢃ
ð1=6Þ ꢂ ½COꢃ
Yield of CO ð%Þ ¼
ꢂ 100
½benzeneꢃ
Selectivity of phenol ð%Þ
½phenolꢃ
¼
ꢂ 100 ð4Þ
½phenolꢃ þ ð1=6Þ ½CO2ꢃ þ ð1=6Þ ½COꢃ
where [ ] is the number of moles produced or fed. The obtained
carbon balances were usually more than 90%.
2.3 Measurement of X-ray diffraction profiles for
ZSM-5 zeolites
The X-ray powder diffraction (XRD) patterns were recorded
on a Rigaku RINT 2001 XRD diffractometer using the Cu-
Ka line in order to confirm the preservation of the crystallinity
of the zeolite after the preparations.
2.4 Measurement of BET surface area of catalysts
The surface areas of the parent H-ZSM-5 zeolite and CuH-
ZSM-5 catalysts were measured by the BET method based
on the nitrogen adsorption at 77 K in a micro-adsorption
vacuum apparatus (residual pressure below 10ꢁ3 Pa).
2.2 Gas-phase catalytic oxidation of benzene with
molecular oxygen
The gas-phase catalytic oxidation of benzene was carried out
at atmospheric pressure in a continuous flow fixed-bed reactor
made from Pyrex glass (internal diameter ¼ 15 mm). Benzene
(guaranteed reagent, Nacalai Tesuque) was checked by gas
chromatography and used without further purification. The
oxygen and nitrogen gases (Kobe Oxygen Co.) were used as
received. Prior to the experiments, all catalysts were pelletized
without any binder and applied, after crushing and fractionat-
ing, in the form of grains (80–160 mesh). The typical procedure
for the benzene oxidation was as follows. The reactor contain-
ing 0.50 g of catalyst was electrically heated. The temperature
of the catalyst bed was measured with a thermocouple located
in the middle of the catalyst bed. Prior to starting the reaction,
2.5 Measurement of electron paramagnetic resonance
spectra (EPR) for CuH-ZSM-5
The X-band EPR spectra (n ¼ 9.43 GHz, microwave power:
1.00 mW) of the catalyst samples were recorded at room tem-
perature at a modulation frequency of 100 kHz (modulation
width: 0.32 mT) on a JES-TE300 EPR spectrometer (JEOL).
The EPR signals of Cu2+ were obtained at room temperature
and registered in the field region from 2350 to 3850 G with a
sweep time of 8 min. The g values were evaluated in relation
to a Mn2+ external standard, which was also used to normalize
the EPR signal intensities. Before the EPR measurement, a
0.20 g sample placed in a Pyrex glass vessel connected to a
Phys. Chem. Chem. Phys., 2003, 5, 956–965
957

View Article Online
had no activity for phenol formation, although they all exhib-
quartz tubular ampoule (inner diameter ¼ 3 mm) was heated
under vacuum (10ꢁ3 Pa) to 773 K and exposed to O2 (20
kPa) at the same temperature for 1 h. The sample was then
cooled to room temperature, and evacuated to 10ꢁ3 Pa for
15 min. The treated zeolites were transferred into the quartz
ampoule without exposure to air and sealed off.
ited a slight activity for the deep oxidation to CO2 and CO.
The reason for this apparent lack of activity is not totally clear.
It is suggested that the Cu-supported high-silica zeolites may
be active for phenol formation because ZSM-5, ZSM-11 and
zeolite beta contain a low concentration of Al in their frame-
work. The relatively lower activity of CuH-ZSM-11 and
CuH-beta compared to Cu-HZSM-5 may be caused either by
deviation from an optimum amount for phenol formation of
both the Cu-loading and the Si/Al ratio (namely, Cu/Al ratio)
of these zeolites or by the differences in their framework struc-
tures (e.g., pore size and pore arrangement). The differences in
the nature of the copper species on these zeolites may also be a
likely cause. The Cu-supported H-mordenite (SR(923)-CuH-
Mordenite) was inactive despite its high-silica feature (Si/
Al ¼ 13). This is presumably due to the difference in the chan-
nel structure of each zeolite.34 ZSM-5, ZSM-11, and the zeolite
beta possess 3-dimensional channel systems through which
benzene could easily diffuse. Although mordenite also has a
3-dimensional channel system, one of its channels consists of
2.6 Cu K-edge X-ray absorption fine structure spectra (XAFS)
The Cu K-edge XAFS spectra of the SR-Cu-H-ZSM-5 sam-
ples were obtained at the beamline BL01B1 of the SPring-8 syn-
chrotron radiation facility at the Japan Synchrotron Radiation
Research Institute (JASRI). All samples were pressed into self-
supported disks without any binder. No chemical pre-treat-
ment was carried out for the samples and all measurements
were conducted with exposure of the samples to air. The Cu
K-edge absorption spectra were recorded in the transmission
mode at room temperature under atmospheric pressure. The
storage ring energy was 8.0 GeV and the ring current was
60–100 mA. A Si (311) double crystal monochromator was
used. The photon energy was calibrated by the characteristic
pre-edge peaks due to the 3d ! 4p* transition in the absorp-
tion spectrum of a Cu foil (8984.0 eV).
˚
smaller 8-member oxygen windows (2.6 ꢂ 5.7 A) through
which the benzene molecule could have difficulty in diffusing
through.
3.1.2 Dependence of copper precursor. A series of SR(923)-
CuH-ZSM-5 zeolites were prepared by the solid-state reaction
method using various copper halides as copper precursors. The
oxidation of benzene was performed over these catalysts
(Table 2). The activity for phenol formation was higher in
the sequence of CuCl > CuCl2 > CuI > CuBr > CuBr2 ꢄ
CuF2 . SR(923)-CuH-ZSM-5 prepared using CuF2 showed a
considerably low phenol yield and selectivity (0.18% and
2.3%, respectively), presumably due to rather higher melting
point of CuF2 (1223 K) than the calcination temperature
(923 K) for the preparation. The higher melting point of
CuF2 probably prevents copper species from migrating into
the channel system of H-ZSM-5. Accordingly, the dispersion
of Cu species in ZSM-5 was suggested to be important for
the effective formation of phenol as well as for the suppression
of the formation of CO2 and CO. On the other hand, the oxi-
dation states of the copper precursor seem to play only a minor
role in the catalytic activity for phenol formation, particularly
with respect to phenol selectivity, which was higher in the
sequence of CuCl ꢄ CuCl2 ꢀ CuBr2 > CuI ꢀ CuBr ꢄ CuF2 .
3
Results
3.1 Gas-phase catalytic oxidation of benzene over SR- and
IMP-CuH-ZSM-5
3.1.1 Dependence of supports. The oxidation of benzene
with O2 was performed at 673 K over a variety of supported
Cu catalysts prepared at 923 K by the solid-state reaction
method (Table 1). Phenol was produced when using the Cu-
supported Na-ZSM-5, H-ZSM-5, H-ZSM-11 and H-Beta as
catalysts. Among the catalysts tested here, the Cu-supported
H-ZSM-5 (SR(923)-CuH-ZSM-5) showed the highest phenol
yield and selectivity. In contrast, SR(923)-CuNa-ZSM-5, the
Na-type counterpart of the CuH-ZSM-5, produced a rather
small amount of phenol and simultaneously produced a large
amount of CO2 , suggesting that the Na species in the Na-type
zeolites will suppress the phenol formation and/or accelerate
the deep oxidation of benzene. When the SR(923)-CuH-
ZSM-5 was employed as a catalyst, a trace amount of biphenyl
was also formed together with phenol. The Cu-supported H-
ZSM-11 and H-Beta showed rather low activities for phenol
formation. The Cu-supported H-X zeolite (SR(923)-CuH-X),
Cu-supported H-Y zeolite (SR(923)-CuH-Y), Cu-supported
H-mordenite (SR(923)-CuH-mordenite), and Cu-supported
mesoporous H-[Al]MCM-41 (SR(923)-CuH-Al-MCM-41)
3.1.3 Effect of reaction temperature. The gas-phase cataly-
tic oxidation of benzene with molecular oxygen was performed
at various temperatures over SR(923)-CuH-ZSM-5 (Cu ¼
0.70 wt.%, Si/Al ¼ 75). Fig. 1 shows the dependence of the
yield of products and the selectivity of phenol on the reaction
temperature. Phenol, CO2 , and CO were formed at around
600 K and higher temperatures. The yield of phenol increased
with increasing temperature up to around 673 K, but a further
Table 1 Oxidation of benzene to phenol over Cu-supported on
various zeolites and oxidesa
Yield (%)
Catalystb
Cu (wt.%) Phenol CO2 CO
Table 2 Gas-phase catalytic oxidation of benzene over SR(923)-Cu-
H-ZSM-5 catalysts prepared by the solid-state reaction method using
various copper halides as precursorsa
SR(923)-CuH-Mordenite (13)
SR(923)-CuH-X (2)
0.57
0.61
0
0.4 0.13
0
1.9
4.3
0
1
SR(923)-CuH-Al-MCM-41 (98) 0.65
0
Yield (%)
Selectivity
SR(923)-CuH-ZSM-11 (210)
SR(923)-CuH-beta (12)
SR(923)-CuNa-ZSM-5 (100)
SR(923)-Cu-H-ZSM-5 (100)
SR(923)-Cu/SiO2
0.58
0.61
0.72
0.65
0.27
0.29
0.31
0.2
0.1
0.02
1.3
0
5.2 2.6
1.7 0.8
Precursor
Cu (wt.%)
Phenol
CO2
CO
of phenol (%)
21.1
1
2.1 2.1
4.2 0.06
1.1 0.22
CuF2
CuCl
CuCl2
CuBr
CuBr2
CuI
0.58
0.61
0.61
0.63
0.56
0.61
0.18
1.05
0.84
0.67
0.62
0.82
6.6
1.7
3.8
3.6
2.5
4.4
1.2
1.6
3.3
3.5
2.9
3.8
2.3
24.0
10.6
8.6
SR(923)-Cu/SiO2-Al2O3
SR(923)-Cu/Al2O3
0
0
11.8
2
10.1
9.1
a
Catalyst, 0.50 g; benzene/O2/N2 ¼ 0.14/1/4; W/F ¼ 236 g-cat min
molꢁ1; reaction temperature, 673 K; all catalysts were prepared at
923 K by the solid-state reaction (SR) method. b Figure in parenthesis
shows a Si/Al atomic ratio of support material.
a
Catalyst, 0.50 g; benzene/O2/N2 ¼ 0.18/1/4; W/F ¼ 236 g-cat min
molꢁ1; reaction temperature, 673 K.
958
Phys. Chem. Chem. Phys., 2003, 5, 956–965

View Article Online
Fig. 1 Dependence of reaction temperature on the oxidation of ben-
zene over SR(923)-Cu-H-ZSM-5. Yields of phenol (S), CO2 (/), CO
(L) and selectivity of phenol (X); catalyst, SR(923)-Cu-H-ZSM-5,
Fig. 3 Influence of the Cu loading on the yields of phenol (S), CO2
(/), CO (L) and the selectivity of phenol (X). Catalysts, SR(1123)-
Cu-H-ZSM-5 (0.50 g); benzene/O2/N2 ¼ 0.18/1/4; W/F ¼ 236
g-cat min molꢁ1; reaction temperature, 673 K.
0.50 g; benzene/O2/N2 ¼ 0.18/1/4; W/F ¼ 236 g-cat min molꢁ1
.
increase in the reaction temperature caused a decrease in the
phenol yield. The yields of CO2 and CO also increased with
the increasing reaction temperature and they steeply increased
at temperatures higher than 673 K. Consequently, the selectiv-
ity of phenol decreased almost linearly with increasing reaction
temperature. It is considered that the deep oxidation of the
once formed phenol is accelerated at higher temperatures. As
will be mentioned below, CO2 and CO could be formed via
the deep oxidation of phenol, leading to the reduced yield of
phenol. Based on these results, in this paper, the reaction
was usually conducted at 673 K.
direct deep oxidation of benzene, namely, independent of the
formation of phenol.
3.1.5 Influence of the amount of copper loading. The gas-
phase catalytic oxidation of benzene with molecular oxygen
was carried out at 673 K over SR(1123)-CuH-ZSM-5 (Si/Al ¼
75) having different amounts of copper (Cu ¼ 0 ꢀ 1.14
wt.%). As shown in Fig. 3, the yields of products and the selec-
tivity of phenol were highly influenced by the amount of Cu in
CuH-ZSM-5. The phenol yield sharply increased with the
increasing amount of Cu loading in the range of the lower
amounts of Cu loading. A further increase in copper loading
resulted in a decreased yield of phenol. The phenol yield was
optimized at the Cu loading of around 0.77 wt.% (phenol yield:
4.9%). The yield of CO2 and CO gradually increased with the
rise in the copper loading in the lower Cu loading region
(Cu < 0.7 wt.%), while they steeply increased in the higher
Cu loading region (Cu > 0.7 wt.%). The maximum selectivity
of phenol (40.2%) was obtained with a 0.63 wt.% Cu loading.
The parent H-ZSM-5 (Cu ¼ 0 wt.%) had no catalytic activity
for phenol formation but produced only a small amount of
CO2 and CO (0.62% and 0.39%, respectively). This indicates
that the active species for phenol formation should be the
Cu-containing species. The formation of small amounts of
CO2 and CO on the parent H-ZSM-5 suggests the participa-
tion of the Brønsted acid site of the H-ZSM-5 during the for-
mations of both CO2 and CO. In general, the Cu species in the
Cu-zeolites are considered to be present in various forms such
as Cu2+, [Cu–O–Cu]2+, Cu+, [Cu(OH)]+ and CuO depending
on the Cu/Al value of the zeolites and the treatments condi-
tions,35 although their exact local coordination are not yet
clear, especially for the Cu-ZSM-5 zeolites. An increase in
the amount of CuO species are observed when the Cu/Al2
value of Cu-ZSM-5 exceeded 1.0 (over ion-exchange), whereas
the isolated Cu2+ or Cu+ ions located near the Al3+ site of
ZSM-5 are predominant when the Cu/Al2 values are smaller
than 1.0.36 The results obtained in Fig. 3 indicate that the
isolated Cu ions species should be active for the formation
of phenol, and the aggregated Cu species such as CuO for
CO2 and CO.
3.1.4 Influence of the contact time (W/F). The gas-phase
catalytic oxidation of benzene with molecular oxygen was car-
ried out over SR(923)-CuH-ZSM-5 (Cu ¼ 0.70 wt.%, Si/
Al ¼ 75) under different contact times (W/F) using different
amounts of catalysts (0.1–1.0 g). Fig. 2 illustrates the yield of
products and the selectivity of phenol with respect to the con-
tact time. The yield of phenol sharply increased in the lower
contact time region (W/F < ca. 150), while it kept almost con-
stant value at the higher contact times. The yields of CO2 and
CO increased almost linearly over the entire W/F region inves-
tigated. The selectivity of phenol linearly decreased with the
increasing contact time. The extrapolation of the linear plot
of the phenol selectivity to the intercept (W/F ¼ 0) was not
100% (ca. 37%). It is concluded from these observations that
CO2 and CO should be formed through two distinct pathways:
(a) consecutively through the deep oxidation of phenol once
formed on the catalyst surface, and (b) directly through the
3.1.6 Effect of the partial pressure of both benzene and oxy-
gen. The effect of the partial pressures of both benzene and
oxygen on the benzene oxidation to phenol was investigated.
The oxidation of benzene was performed under a variety of
O2 or benzene partial pressures at 673 K over IMP(1123)-
CuH-ZSM-5 (Cu ¼ 0.70 wt.%, Si/Al ¼ 75, calcined at
1123 K). The effect of the partial pressure of benzene was
Fig. 2 Relationship between contact time (W/F) vs. yields of phenol
(S), CO2 (/), CO (L) and selectivity of phenol (X). Catalysts, SR(923)-
Cu-H-ZSM-5, 0.1–1.0 g; benzene/O2/N2 ¼ 0.18/1/4, reaction tem-
perature, 673 K.
Phys. Chem. Chem. Phys., 2003, 5, 956–965
959

View Article Online
Fig. 6 Dependence of the partial pressure of O2 on the yields of phe-
nol (S), CO2 (/), CO (L) and the selectivity of phenol (X). Catalyst,
IMP-Cu-H-ZSM-5 (Cu ¼ 0.70 wt.%, Si/Al ¼ 75) (0.50 g); W/F ¼
236 g-cat min molꢁ1; Pbenzene , 4.5 kPa; reaction temperature, 673 K.
Fig. 4 Dependence of the partial pressure of benzene on the yields of
phenol (S), CO2 (/), CO (L) and the selectivity of phenol (X). Cata-
lyst, IMP-Cu-H-ZSM-5 (Cu ¼ 0.70 wt.%, Si/Al ¼ 75) (0.50 g);
W/F ¼ 236 g-cat min molꢁ1; reaction temperature, 673 K.
O2 partial pressure of ca. 4 kPa and a further increase in
the partial pressure of O2 caused a considerable decrease
in the selectivity of phenol. The space-time yield (STY) of phe-
nol was logarithmically plotted as a function of the partial
pressure of O2 as illustrated in Fig. 7. A linear correlation
between the space-time yield of phenol and the partial pressure
of O2 was observed. The kinetic order for the partial pressure
of O2 was 0.18.
examined by varying the benzene partial pressure while keep-
ing the O2 partial pressure at 19.5 kPa. At the same time,
the partial pressure of N2 was also changed to keep W/F con-
stant (236 g-cat min molꢁ1). Fig. 4 illustrates the dependence
of the product yield and the selectivity of phenol with respect
to the partial pressure of benzene. The yield of phenol mono-
tonously decreased with an increase in the partial pressure of
benzene, while the selectivity of phenol steadily increased up
to the benzene pressure of ca. 20 kPa (benzene/O2 ꢀ 1) and
then it remained almost constant (phenol selectivity: ca.
55%). Fig. 5 represents the logarithmic plots of the space-time
yield (STY) of phenol as a function of the partial pressure of
benzene, exhibiting two different linear lines. Thus, the kinetic
order in benzene can be estimated from these slopes to be 0.70
and ꢁ0.27 for the lower-partial pressure (< ca. 15 kPa) and
higher partial pressure regions ( > ca. 15 kPa), respectively.
Assuming that the oxidation of benzene proceeds via the redox
cycle of Cu2+/Cu+,31 the observed negative effect of the ben-
zene partial pressure at higher-pressure region is likely due to
the increasing difficulty in the re-oxidation of Cu+ to Cu2+
with O2 .
3.1.7 Dependence of the calcination temperature of CuH-
ZSM-5. The gas-phase catalytic oxidation of benzene with
molecular oxygen was carried out over SR-CuH-ZSM-5
(Cu ¼ 0.63 wt.%, Si/Al ¼ 75) calcined at various tempera-
tures (773–1273 K). As illustrated in Fig. 8, the yield of phenol
as well as those of CO2 and CO increased with the increasing
calcination temperature up to 1123 K, while a further increase
in the calcination temperature caused a significant decrease in
the yield of each product. The yield of phenol was optimized
for the SR-CuH-ZSM-5 calcined at 1123 K. When the calcina-
tion temperature was elevated from 773 K to 1123 K, the yield
of phenol rose by ca. 5 times from 0.7% to 3.7%. The selectivity
of phenol also exhibited a behavior similar to the product
yield, except that it remained approximately constant between
1123 K and 1223 K. Accordingly, the calcination of the
SR-CuH-ZSM-5 at higher temperatures (ꢀ1123 K) was found
to be favorable for the oxidation of benzene to phenol with
respect to both yield and selectivity. A similar depend-
ence of the calcination temperature was observed for the
The dependence of the partial pressure of O2 was also exam-
ined in a similar manner. Fig. 6 depicts the yield of each pro-
duct and the selectivity of phenol for various partial pressures
of O2 and a constant benzene partial pressure (4.5 kPa). The
yield of phenol increased with the increasing partial pressure
of O2 . The yields of both CO2 and CO also increased. The
selectivity of phenol had a maximum value (ca. 75%) at the
Fig. 5 Logarithmic plots of the space-time yield (STY) of phenol as a
function of the partial pressure of benzene (corresponds to Fig. 4).
Fig. 7 Logarithmic plots of the space-time yield (STY) of phenol as a
function of the partial pressure of O2 (corresponds to Fig. 6).
960
Phys. Chem. Chem. Phys., 2003, 5, 956–965

View Article Online
Fig. 8 Variation in the yields of phenol (S), CO2 (/), CO (L) and
selectivity of phenol (X) with respect to the calcination temperature
of the Cu-H-ZSM-5. Catalysts, SR-Cu-H-ZSM-5 (0.50 g); benzene/
O2/N2 ¼ 0.18/1/4; W/F ¼ 236 g-cat min molꢁ1; reaction tempera-
ture, 673 K.
Fig. 10 Dependence of BET surface area of SR-Cu-H-ZSM-5
(Cu ¼ 0.63 wt.%, Si/Al ¼ 75) on the calcination temperature.
that the original surface area of the H-ZSM-5 (375 m2 gꢁ1
)
support was approximately maintained even after being cal-
cined at 1223 K. However, when the SR-Cu-H-ZSM-5 was cal-
cined at 1273 K, its surface area was reduced almost by half to
180 m2 gꢁ1. Fig. 11 shows the BET surface areas of IMP-CuH-
ZSM-5 with various Si/Al atomic ratios and a 0.70 wt.% Cu
loading calcined at 1123 K (S) and the corresponding parent
HZSM-5 supports calcined at 773 K (X). The differences in
the BET surface areas were almost negligible among the sam-
ples with the different Si/Al atomic ratios. Also, no significant
change in the BET surface area was observed among the
samples before and after Cu incorporation.
Cu-impregnated H-ZSM-5 zeolites (IMP-CuH-ZSM-5) (figure
not shown).
3.1.8 Dependence of the Si/Al atomic ratio of IMP-CuH-
ZSM-5 during the oxidation of benzene. A series of IMP-
CuH-ZSM-5s with different Si/Al atomic ratios (32, 75, 100,
123, 160) and the same copper loading (0.70 wt.%) were pre-
pared by the impregnation method and calcined at 1123 K
for 5 h. The gas-phase catalytic oxidation of benzene was per-
formed over these IMP-CuH-ZSM-5 catalysts. As shown in
Fig. 9, the yield of phenol increased as the Si/Al atomic ratio
of the IMP-CuH-ZSM-5 became lower. On the contrary, the
yield of CO and CO2 increased with an increase in the Si/Al
atomic ratio of the IMP-CuH-ZSM-5 (Si/Al, yield of CO2 ,
yield of CO ¼ 32, 3.4%, 2.6%; 160, 12.3%, 5.9%, respectively).
Thus, the selectivity of phenol increased with a decrease in the
Si/Al atomic ratio of the IMP-CuH-ZSM-5.
3.3 XRD profiles of the SR- and IMP-CuH-ZSM-5 catalysts
Fig. 12 indicates the XRD profiles of the SR-Cu-H-ZSM-5 cal-
cined at various temperatures and those of the parent H-ZSM-
5 calcined at 773 K. Although a slight decrease in the intensity
was noted after the introduction of Cu, no new peaks
appeared, suggesting that Cu was highly dispersed in the H-
ZSM-5 channel and the crystal structure of the ZSM-5 zeolites
remained practically unperturbed by the calcinations. For the
SR-CuH-ZSM-5 zeolite calcined at 1273 K (Fig. 12g), how-
ever, the background of its diffraction profile was slightly
raised in the 2y range between 15 and 30ꢅ, indicating the for-
mation of an amorphous phase, or the partial destruction of
the zeolite structure. In addition, the relatively higher intensity
of the (100) peak (2y ¼ ca. 8ꢅ) than that of the (200) peak
(2y ¼ ca. 23ꢅ) for the SR-CuH-ZSM-5 zeolite calcined at
3.2 BET surface area measurement of the SR- and
IMP-CuH-ZSM-5 catalysts
The BET surface areas of the SR-Cu-H-ZSM-5 zeolites cal-
cined at different temperature are shown in Fig. 10, indicating
Fig. 9 Dependence of the Si/Al atomic ratio of the HZSM-5 support
on the yield of phenol (S) and the selectivity of phenol (X). Catalysts,
IMP-CuH-ZSM-5 (Cu ¼ 0.70 wt.%, Si/Al ¼ 32, 75, 100, 120, 160)
Fig. 11 Dependence of BET surface area of both the IMP-CuH-
ZSM-5 (S) (Cu ¼ 0.70 wt.%) calcined at 1123 K and the parent
HZSM-5 (X) on the Si/Al atomic ratio.
(0.50 g); benzene/O2/N2 , 0.18/1/4; W/F ¼ 236 g-cat min molꢁ1
;
reaction temperature, 673 K.
Phys. Chem. Chem. Phys., 2003, 5, 956–965
961

View Article Online
of SR(1273)-CuH-ZSM-5 had two maxima at 8992 eV and
8997 eV (Figs. 13d–f), which resemble that of CuO (Fig.
13a), suggesting that the Cu in these samples were in the
Cu2+ state. On the other hand, a new peak appeared in the
spectrum of the SR(1273)-Cu-H-ZSM-5 at 8986 eV (Fig.
13g), which is an intermediate value between the pre-edge peak
of Cu2O (8989 eV) and Cu foil (8984 eV), although the above
mentioned two peaks (8991 and 8996 eV) were virtually invar-
iant. Haller et al.37 have measured the Cu K-edge XANES
spectra of Cu-ZSM-5 catalyst under NO + C3H6 + O2/He at
increasing temperatures of 523–766 K and reported that
the isolated Cu2+ and Cu1+ ions, in addition to CuO (or its
cluster) and Cu2O could be co-present from the analysis of
the XANES spectra using a mathematical procedure based
on evolving factor analysis. A shoulder at ca. 8983–8984 eV
was identified37,38 as 2- and 3-coordinated Cu1+ species char-
acteristic of the 1s ! 4p transition. These results and our pre-
sent observation indicate that a mixture of Cu0 and Cu1+
species could be present in this sample besides the Cu2+
,
CuO and Cu2O. Presumably, isolated exchanged Cu2+ species
and aggregated CuO may coexist at the calcination tempera-
tures of less than 1223 K, and the isolated Cu2+ (and also
the aggregated CuO) may be reduced to Cu1+ and Cu0 due
to auto-reduction during the calcination at 1273 K, and thus
the contents of the Cu2+ species were diminished at the
temperature of 1273 K as revealed by the EPR measurement
mentioned below.
Fig. 12 XRD profiles of (a) parent H-ZSM-5 (Si/Al ¼ 75); SR-Cu-
H-ZSM-5 (Cu ¼ 0.63 wt.%, Si/Al ¼ 75) calcined at (b) 923 K, (c)
1023 K, (d) 1123 K, (e) 1173 K, (f) 1223 K and (g) 1273 K.
1273 K indicates the occurrence of the dealuminization of the
ZSM-5 framework. The IMP-CuH-ZSM-5 with a different
Si/Al atomic ratio and the same copper loading (0.70 wt.%)
exhibited almost the same diffraction profiles as that of the
pure ZSM-5 zeolites and no differences were observed (figures
not shown).
3.5 EPR measurement
Fig. 14 shows the X-band EPR spectra of the SR-Cu-H-ZSM-
5 (Cu ¼ 0.63 wt.%) calcined at different temperatures. Fig. 15
illustrates the EPR spectra of the IMP-CuH-ZSM-5 with dif-
ferent Si/Al atomic ratios and the same Cu loading (0.70
wt.%). They all exhibited the characteristic hyperfine splitting
(HFS) on the parallel components of the spectra which results
from the hyperfine coupling between the 3d unpaired electron
and the Cu nuclear spin (I ¼ 3/2), indicating the presence of
the isolated Cu2+ species in these examined Cu-H-ZSM-5s.37
Apparently, the intensities and profiles of the EPR signals sig-
nificantly varied with the variation in either the calcination
temperatures or the Si/Al atomic ratio. Accordingly, the
3.4 X-ray absorption near edge spectra (XANES) of the
SR-CuH-ZSM-5 catalysts
The differential XANES profiles of the hydrated (recorded
while exposed to air) SR-CuH-ZSM-5 samples calcined at dif-
ferent temperatures (773–1123 K) are shown in Fig. 13,
together with that of the reference samples (CuO, Cu2O, and
Cu foil). All profiles of the catalyst samples other than that
Fig. 14 EPR spectra of the SR-Cu-H-ZSM-5 (Cu ¼ 0.63 wt.%, Si/
Al ¼ 75). Calcined at (a) 923 K, (b) 1023 K, (c) 1123 K, (d) 1173 K,
(e) 1223 K and (f) 1273 K. Prior to measurement, each sample was cal-
cined at 773 K in O2 (20 kPa) for 1 h, followed by evacuation at 300 K
for 15 min.
Fig. 13 Differential curves of XANES spectra of (a) CuO, (b) Cu2O,
(c) Cu foil; SR-Cu-H-ZSM-5 (Cu ¼ 0.63 wt.%, Si/Al ¼ 75) calcined at
(d) 773 K, (e) 1123 K, (f) 1223 K and (g) 1273 K.
962
Phys. Chem. Chem. Phys., 2003, 5, 956–965

View Article Online
Fig. 16 The g part of the EPR spectra of the SR-Cu-H-ZSM-5
k
Fig. 15 EPR spectra of the IMP-CuH-ZSM-5 with different Si/Al
ratio and uniform copper loading (0.70 wt.%) calcined at 1123 K.
Si/Al atomic ratio: (a) 32, (b) 75, (c) 100, (d) 123 and (e) 160. Prior
to measurement, each sample was calcined at 773 K in O2 (20 kPa)
for 1 h, followed by evacuation at 300 K for 15 min.
(Cu ¼ 0.63 wt.%, Si/Al ¼ 75) calcined at (a) 923 K, (b) 1023 K,
(c) 1123 K, (d) 1173 K, (e) 1223 K and (f) 1273 K. (corresponds to
Fig. 14).
depending on the calcination temperature, namely, species I
1
k
1
k
2
k
2
(g ¼ 2.33, A ¼ 158), species II (g ¼ 2.31, A ¼ 159),
k
change in the calcination temperature or the Si/Al atomic
ratio was found to cause a change in the Cu coordination
and/or the concentration of the isolated Cu2+ species.
3
k
3
and species III (g ¼ 2.28, A ¼ 169). According to litera-
k
ture,39–43 species I and II were assigned as Cu2+ ions in the
square-pyramidal coordination and species III to Cu2+ ions
in the square-planar coordination. The integrated intensity of
each signal, which is a measure of the relative concentration
of corresponding copper species, was estimated (Fig. 17).
The possible existence of a third contribution in spectrum f
of Fig. 16 will be due to an effect of the different hyperfine split-
ting of the two EPR-active Cu isotopes. The relative concen-
tration of the square-pyramidal Cu2+ increased with the
increasing calcination temperature, while that of the square-
planar Cu2+ decreased and almost completely disappeared at
the calcination temperatures of 1123 K or higher. A similar
observation already has been reported39,42,43 where the EPR
signal corresponding to the square-planar Cu2+ decreased
4
Discussion
The results of both the BET surface area measurements (Fig.
10 and Fig. 11) and XRD analysis (Fig. 12) indicated that
the partial destruction of the zeolite structure occurred when
the SR-CuH-ZSM-5 was calcined at 1273 K. This partial
destruction of the ZSM-5 structure will cause the lowered
activity of the SR-CuH-ZSM-5 calcined at 1273 K (Fig. 8).
The XANES analysis (Fig. 13) indicates the stabilization of
the Cu species on the SR(1273)-CuH-ZSM-5 in the reduced
state (Cu+ or Cu0). Such an irreversible change in the some
of the copper species may possibly be another cause of the
observed reduced catalytic activity of the SR(1273)-CuH-
ZSM-5. Despite the significant improvement in the catalytic
activity of the SR-CuH-ZSM-5 calcined at a higher tempera-
ture (ꢀ1123 K), the BET surface area, the XRD, and the
XANES techniques showed almost no variation in the zeolite
structure or oxidation state of the copper species in these SR-
CuH-ZSM-5s calcined at 773–1223 K. Similarly, the variation
in the Si/Al atomic ratio of the IMP-CuH-ZSM-5 had little
influence on either their specific surface area (Fig. 11) or the
crystallinity of the zeolites (figures not shown). Therefore,
the observed variation in the catalytic activity of the benzene
oxidation should be explained in terms of the electronic prop-
erties and/or microscopic change in the local structure of the
active Cu sites. The EPR analysis was performed in order to
investigate the local state of copper, which is considered to
be the active species for the phenol formation.
4.1 EPR spectra of SR-Cu-H-ZSM-5 calcined at different
temperatures
Fig. 17 Variation in the relative concentration of Cu2+ ions in the
SR-CuH-ZSM-5 calcined at different temperatures (923–1273 K) with
the same copper loading (0.63 wt.%). Data were calculated from the
integrated intensity of EPR signal corresponding to each species:
(S), square-pyramidal Cu2+ ions (I + II); (X), square-planar Cu2+
ions (III).
The g part of the EPR spectra of the SR-CuH-ZSM-5 shown
k
in Fig. 14 are magnified in order to more precisely analyze
these spectra (Fig. 16). It clearly indicates the presence of at
least three configurations of Cu2+ ions in the Cu-H-ZSM-5
Phys. Chem. Chem. Phys., 2003, 5, 956–965
963

View Article Online
after treatment at the higher temperature of ꢀ1023 K,
accounting for the activity loss for NO decomposition. For
such an intensity change of the EPR signal during heat treat-
ment, Kucherov and co-workers proposed a spin–lattice inter-
action between Cu2+ and the zeolite framework.42 On the
other hand, a paramagnetic but EPR inactive cupric species,
such as the oligomeric CuO species, binuclear [CuOCu]2+
,
have been assumed to be present in Cu-ZSM-5 and trans-
formed into the EPR active Cu2+ species.44,45 Hall et al. have
reported46 that the loss of the detectable Cu2+ from the EPR
signal of the Cu-NaZSM-5 during the variation of the dehy-
dration temperatures from room temperature to 773 K reflects
the decreasing symmetry around these Cu2+ ions, not conver-
sion into other EPR silent species, because the exposure to
H2O vapor causes the return of the initial value of the EPR sig-
nal. In our opinion, the observed increase in the EPR signal
upon increasing calcination temperature (923–1273 K) should
result from the transformation of such an EPR inactive species
into the EPR active isolated Cu2+ ions. Comparing Fig. 8 with
Fig. 17, the variation of the relative concentration of square-
pyramidal Cu2+ with respect to the calcination temperature
are very similar to that of the phenol yield as well as that of
the COx (CO2 + CO) yields, suggesting that the square-pyrami-
dal Cu2+ species are active for the formation of phenol, CO2
and CO. In addition, a similar dependence in the yield of each
product with respect to the calcination temperature suggests
that the same sites may participate in the oxidation of benzene
to phenol and also to the deep oxidation products. It should
be kept in mind, however, that CO2 and CO should also
be formed over the aggregated Cu species such as CuO,
as previously mentioned.
Fig. 19 Variation in the relative concentration of Cu2+ ions in the
SR-CuH-ZSM-5 with different Si/Al atomic ratios and uniform cop-
per loading (0.70 wt.%). Data were calculated from the integrated
intensity of EPR signal corresponding to each species: (S), square-
pyramidal Cu2+ ions (I + II); (X), square-planar Cu2+ ions (III).
the variation in the Si/Al atomic ratio were observed, indicat-
ing that these ‘‘EPR active’’ copper species in the present sam-
ples were the same in quality. Fig. 19 indicates the intensity
change of the signal with the variation in the Si/Al atomic
ratio. The concentration of the square-pyramidal Cu2+ (spe-
cies I + II) increase with the decrease in the Si/Al atomic ratio,
whereas that of the square-planar Cu2+ (species III) was
almost unchanged. Thus, it is confirmed that both the concen-
tration and the coordination of Cu2+ ions significantly varied
with the variation in the Si/Al atomic ratio. Such variation in
the square-pyramidal Cu2+ species accords well with the varia-
tion in the catalytic activity for benzene oxidation, as proved if
Fig. 9 and Fig. 19 are compared. Consequently, it is suggested
again, that the observed increase in phenol yield with the
decrease in the Si/Al atomic ratio is due to the increase in
the concentration of the isolated Cu2+ ions with a square-
pyramidal coordination and, therefore, that the active species
for the phenol formation should be the square-pyramidal
Cu2+ ions.
4.2 EPR spectra of IMP-CuH-ZSM-5 with different
Si/Al atomic ratio
Fig. 18 indicates the magnified g part of the EPR spectra (Fig.
k
15) of the IMP-CuH-ZSM-5 with different Si/Al atomic ratios
and uniform copper loading. In each of the spectra, at least
three series of hyperfine splittings (HFS) were clearly observed,
1
k
1
k
2
namely, species I (g ¼ 2.33, A ¼ 157), species II (g
¼
k
2
k
3
3
2.31, A ¼ 159), and species III (g ¼ 2.28, A ¼ 171), as
k
k
is the case of the EPR spectra (Fig. 16) for the SR-CuH-
ZSM-5. No shifts in these HFS parameters accompanied by
In contrast to the effect of the calcination temperature (Fig.
8), where the variation in the yields of both CO2 and CO were
similar to that of phenol, the variation in the yields of both
phenol and (CO2 + CO) had an opposite trend with respect
to the variation in the Si/Al atomic ratio (see Fig. 9 and Sec-
tion 3.1.8). Assuming the ‘‘EPR active’’ square-pyramidal
Cu2+ as the active species for phenol formation, such a differ-
ence may be due to the participation of the ‘‘EPR inactive’’
copper species in the benzene oxidation. This consideration
agrees with the effects of both the contact time (W/F) and
the amount of copper loading in the benzene oxidation
(Figs. 2 and 3), where two distinct active species, the isolated
and aggregated Cu species, were suggested for phenol forma-
tion and for CO and CO2 formation, respectively. Though
there is no direct evidence for the presence of aggregated Cu
species (such as CuO) in the present study, CuO was con-
firmed46 to present on Cu-NaZSM-5 calcined at 773 K based
on the reflectance spectra, and the CO-TPR method showed46
that about 15% of the total Cu was present as CuO.
5
Conclusions
Phenol was produced by the gas-phase catalytic oxidation of
benzene with molecular oxygen using Cu-supported ZSM-5
type zeolites prepared by both the solid state reaction and
impregnation method. Among these investigated in the present
Fig. 18 The g part of the EPR spectra of IMP-Cu-H-ZSM-5 with
k
different Si/Al atomic ratios of (a) 32, (b) 75, (c) 100, (d) 123, (e)
160. (corresponds to Fig. 15).
964
Phys. Chem. Chem. Phys., 2003, 5, 956–965

View Article Online
14 M. Iwamoto, J. Hirata, K. Matsukami and S. Kagawa, J. Phys.
Chem., 1983, 87, 903.
15 G. I. Panov, V. I. Sobolev and A. S. Kharitonov, J. Mol. Catal.,
1990, 61, 85.
16 R. Burch and C. Howitt, Appl. Catal. A: General, 1993, 103, 135.
17 L. V. Pirutko, K. A. Dubkov, L. P. Solovyeva and G. I. Pavov,
React. Kinet. Catal. Lett., 1996, 58(1), 105.
18 J. L. Motz, H. Heinchen and W. F. Ho¨lderich, J. Mol. Catal. A:
Chemical, 1998, 136, 175.
19 Jin S. Yoo, A. R. Sohail, S. Scott Grimmer and C. Choi-Feng,
Catal. Lett., 1994, 29, 299.
20 R. Leanza, I. Rossetti, I. Mazzola and L. Forni, Appl. Catal. A:
General, 2001, 205, 93.
21 L. M. Kustov, A. L. Tarasov, V. I. Bogdan, A. A. Tyrlov and
J. W. Fulmer, Catal. Today, 2001, 61, 123.
22 T. Kitano, T. Wani, T. Ohnishi, J. Li-Fen, Y. Kuroda, A. Kunai
and K. Sasaki, Catal. Lett., 1991, 11, 11.
study, CuH-ZSM-5 was the most active for phenol formation.
The amount of copper loading in the CuH-ZSM-5 significantly
influenced the yield and the selectivity of phenol, which were
optimized at the Cu/Al2 ratio of approximately 1. The kinetic
order of 0.18 in the partial pressure of O2 was obtained and
those of 0.70 and ꢁ0.27 in the partial pressure of benzene were
obtained for the lower and higher benzene partial pressure
regions, respectively. The calcination of the CuH-ZSM-5 at
higher temperature (ꢀ1123 K) was found to be effective for
higher phenol yield and selectivity, while the calcination at
much higher temperatures ( > 1273 K) resulted in a significant
loss of catalytic activity. Also, the utilization of CuH-ZSM-5
with a lower Si/Al atomic ratio (Si/Al ¼ 32) resulted in the
significant increase in the phenol yield. The EPR study sug-
gested that the observed activity enhancement was due to the
increase in the concentration of the isolated Cu2+ ions with
square-pyramidal coordination and, therefore, that the active
species for phenol formation should be the square-pyramidal
Cu2+ ions.
23 T. Kitano, T. Nakai, M. Nitta, M. Mori, S. Ito and K. Sasaki,
Bull. Chem. Soc. Jpn., 1994, 67, 2850.
24 T. Kiyoura, Jpn. Patent 48-61439, 1973.
25 M. Okamoto and M. Ono, Jpn. Patent 49-19271, 1974.
26 R. Onigawara, Jpn. Patent 56-87527 (1981); Jpn Patent 56-77234,
1981.
27 I. Yamanaka, M. Katagiri, S. Takenaka and K. Otsuka, Shoku-
bai, 1999, 41(6), 438.
28 T. Ohtani, S. Nishiyama, S. Tsuruya and M. Masai, J. Catal.,
1995, 155, 158.
Acknowledgements
29 J. Okamura, S. Nishiyama, S. Tsuruya and M. Masai, J. Mol.
Catal. A: Chemical, 1998, 135, 133.
The authors thank Mr Kenji Nomura of Kobe University for
his technical assistance during this study. The authors also
express their thanks to the Japan Synchrotron Radiation
Research Institute (JASRI) for our synchrotron radiation
experiment at beam line BL01B1 of the Spring-8.
30 T. Miyahara, H. Kanzaki, R. Hamada, S. Kuroiwa, S. Nishiyama
and S. Tsuruya, J. Mol. Catal. A: Chemical, 2001, 176, 141.
31 H. Yamanaka, R. Hamada, H. Nibuta, S. Mototani, S.
Nishiyama and S. Tsuruya, J. Mol. Catal. A: Chemical, 2002,
178, 89.
32 R. J. Argauer and G. R. Landolt, U.S. Patent 3702886, 1988.
33 C. Jia, P. Beaunier and P. Massaiani, Microporous Mesoporous
Mater., 1998, 24, 69.
34 W. M. Meier and D. H. Olson, Atlas of Zeolite Structure Types,
1992.
References
´ ´
35 J. Sarkany, J. L. d’Itri and W. M. H. Sachtler, Catal. Lett., 1992,
1
2
3
4
5
6
7
8
9
C. F. Cross, E. J. Bevan and T. Herberg, Ber. Dtsch. Chem. Ges.,
1900, 33, 2015.
S. Udenfriend, C. T. Clark, J. Axelrod and B. B. Brodie, J. Biol.
Chem., 1954, 208.
G. A. Hamilton, R. J. Workman and L. Woo, J. Am. Chem. Soc.,
1964, 86, 3390.
S. Ito, K. Ueno, A. Mitarai and K. Sasaki, J. Chem. Soc., Perkin
Trans. 2, 1993, 255.
A. Kunai, S. Hata, S. Ito and K. Sasaki, J. Am. Chem. Soc., 1986,
108, 6012.
T. Kurata, Y. Watanabe, M. Katoh and Y. Sawaki, J. Am. Chem.
Soc., 1988, 110, 7472.
S. Tamagaki, K. Suzuki and W. Tagaki, Bull. Chem. Soc. Jpn.,
1989, 62, 148.
K. Sasaki, S. Ito, Y. Saheki, T. Kinoshita, T. Yamasaki and
J. Harada, Chem. Lett., 1983, 37.
´
´
16, 241; T. Beutel, J. Sarkany, G.-D. Lei, J. Y. Yan and W. M. H.
Sachtler, J. Phys. Chem., 1996, 100, 845.
36 C. T. Abreu, M. F. Ribeiro, C. Henriques and G. Delahay, Appl.
Catal., B, 1997, 12, 262.
37 C. Marquez-Alvarez, I. Rodriguez-Ramos, A. Guerrero-Ruiz,
G. L. Haller and M. Fernandez-Garcia, J. Am. Chem. Soc.,
1997, 119, 2905.
38 Lung-Shan Kau, J. Spira-Splomon, J. E. Penner-Hahn, K. O.
Hodgson and E. I. Solomon, J. Am. Chem. Soc., 1987, 109,
6433.
39 A. V. Kucherov, C. P. Hubbard and M. Shelef, J. Catal., 1995,
157, 603.
40 A. V. Kucherov, H. G. Karge and R. Schlo¨gl, Microporous Meso-
porous Mater., 1998, 25, 7.
41 J. Soria, A. M. Arias, A. M. Chaparro, J. C. Conesa and Z. Schay,
J. Catal., 2000, 190, 352.
42 A. V. Kucherov and A. A. Slinkin, Kinet. Catal., 1997, 38(5), 703.
43 A. V. Kucherov, A. N. Shigapov and M. Shelef, J. Catal., 1999,
186, 334.
A. Kunai, S. Hata, S. Ito and K. Sasaki, J. Org. Chem., 1986, 51,
3471.
10 S. Ito, A. Kunai, H. Okada and K. Sasaki, J. Org. Chem., 1988,
53, 296.
11 S. Ito, T. Yamasaki, H. Okada, S. Okino and K. Sasaki, J. Chem.
Soc., Perkin Trans. 2, 1988, 285.
44 S. C. Larsen, A. Aylor, A. T. Bell and J. A. Reimer, J. Phys.
Chem., 1994, 98, 11 533.
12 A. Kunai, T. Wani, Y. Uehara, F. Iwasaki, Y. Kuroda, S. Ito and
K. Sasaki, Bull. Chem. Soc. Jpn., 1989, 62, 2613.
13 T. Kitano, Y. Kuroda, A. Itoh, J. Li-Fen, A. Kunai and
K. Sasaki, J. Chem. Soc., Perkin Trans. 2, 1990, 1991.
45 G. T. Palomino, P. Fisicaro, S. Bordiga and A. Zecchina, J. Phys.
Chem. B, 2000, 104, 4064.
46 M. L. Jacono, G. Fierro, R. Dragone, X. Feng, J. d’Itri and
W. K. Hall, J. Phys. Chem. B, 1997, 101, 1979.
Phys. Chem. Chem. Phys., 2003, 5, 956–965
965