Deactivation of Au/TiO2 catalyst in the hydrogenation of o-chloronitrobenzene in the presence of CO2

Article abstract of DOI:10.1016/j.molcata.2010.11.029

The hydrogenation of o-chloronitrobenzene over Au/TiO₂ was investigated in supercritical carbon dioxide (scCO₂), ethanol, H ₂O, and H₂O/CO₂ at 140 °C. The reaction rate followed the order of H₂O > H₂O/CO₂ > ethanol > scCO₂. Au/TiO₂ was deactivated in the systems containing CO₂. Changes of Au/TiO₂ catalysts were investigated by means of transmission electron microscopy (TEM), in situ infrared (IR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and diffuse reflectance ultraviolet-visible spectra (DR-UV/Vis). Aggregation of Au particles was excluded from origins of deactivation by TEM. Carbonate-like species formed on gold in the presence of CO₂ and H₂ were detected by in situ IR spectroscopy. The results of XPS and DR-UV/Vis reveal that the active species of Au⁰ was oxidized in the presence of CO₂. The possible formation of CO and its influence on the activity of Au/TiO₂ were also discussed. Accordingly, the formation of carbonates, the oxidation of surface Au⁰ by CO₂, and the CO formation are proposed to be the possible factors for the deactivation of Au/TiO₂.

Full text of DOI:10.1016/j.molcata.2010.11.029

Journal of Molecular Catalysis A: Chemical 335 (2011) 183–188
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Deactivation of Au/TiO2 catalyst in the hydrogenation of o-chloronitrobenzene in
the presence of CO2
Yufen Haoa,b, Ruixia Liua,b, Xiangchun Menga, Haiyang Chenga,b, Fengyu Zhaoa,∗
a State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, PR China
b Graduate School of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 April 2010
Received in revised form 23 October 2010
Accepted 26 November 2010
Available online 10 December 2010
The hydrogenation of o-chloronitrobenzene over Au/TiO2 was investigated in supercritical carbon
dioxide (scCO2), ethanol, H2O, and H2O/CO2 at 140 ◦C. The reaction rate followed the order of
H2O > H2O/CO2 > ethanol > scCO2. Au/TiO2 was deactivated in the systems containing CO2. Changes of
Au/TiO2 catalysts were investigated by means of transmission electron microscopy (TEM), in situ infrared
(IR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and diffuse reflectance ultraviolet-visible
spectra (DR-UV/Vis). Aggregation of Au particles was excluded from origins of deactivation by TEM.
Carbonate-like species formed on gold in the presence of CO2 and H2 were detected by in situ IR spec-
troscopy. The results of XPS and DR-UV/Vis reveal that the active species of Au0 was oxidized in the
presence of CO2. The possible formation of CO and its influence on the activity of Au/TiO2 were also
discussed. Accordingly, the formation of carbonates, the oxidation of surface Au0 by CO2, and the CO
formation are proposed to be the possible factors for the deactivation of Au/TiO2.
Keywords:
Au/TiO2
Deactivation
o-Chloronitrobenzene
Hydrogenation
CO2
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
p-chloronitrobenzene (p-CNB) over Au/Al2O3 in ethanol, propanol,
butanol, and pentanol, and obtained the activity in the order of
Catalysis by gold is attracting increasing interest in the pre-
ceding few years because gold, when highly dispersed on various
supports, is catalytically active for a number of reactions, includ-
ing CO oxidation [1]; alcohol oxidation [2]; olefin epoxidation
[3]; selective hydrogenation of unsaturated hydrocarbons, etc.
Recently, Corma et al. reported that the supported gold catalysts
were highly chemoselective for hydrogenation of aromatic nitro-
compounds due to the cooperation between gold nanoparticles and
support which lead to preferentially activation of the nitro group
[4]. To dissolve the substrate, organic solvents such as methanol,
ethanol, cyclohexane, ethyl acetate and toluene are often used in
both the liquid and conventional gas phase hydrogenation [4–7].
But these organic solvent are volatile, toxic and associate with
serious health hazards and environmental problems. To avoid the
drawbacks of organic solvents, supercritical carbon dioxide (scCO2)
was proposed to be a green alternative for its environmental benign
nature such as non-flammability, relative inertness, complete mis-
cibility with gas and easy separation from liquid/solid products
after reaction [8–11].
ethanol > propanol > butanol > pentanol. They attribute the lower
hydrogenation rate in pentanol to a more facile polarization of
pentanol relative to ethanol that results in a stronger interaction
with the catalyst and inhibits p-CNB hydrogenation to a greater
degree [12]. We have investigated the hydrogenation of nitrocom-
pounds in scCO2 over Pd/C, Pt/C, Ni/Al2O3 and Ni/TiO2, and found
the selectivity to anilines in scCO2 is higher than that in ethanol
[13–16]. The interaction of compressed CO2 with reacting species
(i.e., nitrobenzene, nitrosobenzene, and N-phenylhydroxylamine)
is one of the important factors for this improved selectivity. The
reactivity of nitro group is decreased but that of nitroso group is
increased and the transformation of N-phenylhydroxylamine to
aniline is promoted [13,14].
The combination of supported gold catalyst and scCO2 seems
to be worthy of studying, but up to now there are limited reports
about it. For the selective oxidation reactions, Wang et al. [17] and
Baiker et al. [18] found scCO2 was superior to other organic solvents
when supported gold catalysts was used to catalyze the alcohols
oxidation. For the hydrogenation reactions, Chaterjee et al. in situ
inlayed the gold nanoparticles into the pores of MCM-48 in scCO2
and found special selectivity in the crotonaldehyde hydrogenation
[19]; and the physically and chemically mixed TiO2-supported Pd
and Au catalysts presented dramatically improved reaction rate for
the hydrogenation of citral in scCO2 in our previous work [20]. In
addition, a direct interaction between gold particles and CO2 was
suggested by the results that the optical absorption spectrum of
Besides those environmental considerations, the use of
solvent also has large effects on the reaction rate and selec-
tivity. Keane et al. studied the gas-phase hydrogenation of
∗
Corresponding author. Tel.: +86 431 8526 2410; fax: +86 431 8526 2410.
E-mail address: zhaofy@ciac.jl.cn (F. Zhao).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.11.029

184
Y. Hao et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 183–188
small gold particles in scCO2 significantly changed with CO2 pres-
sure in the literature [21]. Therefore, it is significant to discuss the
catalytic performance of Au catalyst in the presence of CO2, it was
expected to obtain some unexpected results by the variation of CO2
pressure.
The present work was undertaken to study the hydrogenation
of o-CNB over Au/TiO2 in scCO2 or H2O/CO2. The Au catalyst deac-
tivated in the presence of CO2. Possible factors which may induce
the deactivation were discussed in detail by using the transmis-
sion electron microscopy (TEM), in situ infrared (IR) spectroscopy,
X-ray photoelectron spectroscopy (XPS), and diffuse reflectance
ultraviolet–visible spectra (DR-UV/Vis).
2.3. Catalyst characterization
The particle sizes of Au on Au/TiO2 catalysts were examined by
transmission electron microscopy (TEM, JEOL JEM-2000 EX). The
TEM samples were prepared by dispersing the catalyst powder in
ethanol under ultrasonic and then allowing a drop of the suspen-
sion to evaporate on copper grids covered with holey carbon film.
An area containing about 50 particles was selected from the TEM
pictures to calculate the Au particle size distribution.
Diffuse reflectance UV/Vis spectra were recorded with
a
Shimadzu UV-3100 PC spectrophotometer. X-ray photoelectron
spectroscopy (XPS, VG Microtech 3000 Multilab) was used to exam-
ine the electronic properties of Au on the surface of the Au/TiO2
catalysts. The C 1s peak at 284.6 eV arising from adventitious car-
bon was used as reference. This reference gives binding energy
values with a precision of 0.2 eV. The surface composition of
the samples was determined from the peak areas of the corre-
sponding lines using a Shirley type background and empirical cross
section factors for XPS. The freshly prepared catalysts were char-
acterized directly. The used catalysts, prior to characterization,
were washed thoroughly with diethyl ether 6 times, separated
by centrifugation at 10,000 rmp, and dried in vacuum at 60 ◦C
overnight.
The metal loadings were determined by inductively coupled
plasma-optical emission spectrometry (ICP-OES; Thermo Scien-
tific ICAP6000, USA) after the extraction of metal species from
the support by aqua regia. To detect the carbonate-like species on
gold, diffuse reflectance infrared Fourier transform spectroscopy
(DRIFTS) was carried out on a JASCO FT/IR-620 spectrometer
equipped with a MCT detector. Prior to the measurement, the sam-
ple was heated to 50 ◦C for 0.5 h under a gas flow of 90% He and
10% H2 to get rid of the oxide on the gold surface. The temperature
was raised to 140 ◦C and the chamber was purged with high purity
helium for 5 min. A background spectrum was collected. Then, a
flow of 90% H2 + 10% CO2 was introduced into the chamber and the
spectra were collected in 5, 15, and 30 min, respectively.
2. Experimental
2.1. Catalyst preparation
The anatase TiO2-supported (Nanjing Haitai Nano material
Co.) gold catalysts with different particle sizes were prepared by
deposition–precipitation as described in the literature [22]. The
required amount of an aqueous solution of HAuCl4 (0.0025 mol l−1
,
Sinopharm Chemical Reagent Co. Ltd.) was added into beaker, and
NaOH solution (5 mol l−1) was added dropwise to HAuCl4 under
vigorous stirring. After the desired pH value was reached, the
required amount of anatase was added to this solution with keep-
ing vigorous stirring and the resulted suspension was aged at room
temperature for 2 h while stirring. A solution of NaBH4, prepared
in ice water, was added to the suspension, the color of the mix-
ture turned to dark purple from yellow, indicating the reduction of
auric ion. After aging for 8 h, the formed solid samples were washed
with deionized water and ethanol, and separated by centrifugation
at 10,000 rpm. The supernatant was detected with AgNO3 solution
to ensure there was no residual Cl− on the catalyst. Then the solid
material was dried in vacuum at 60 ◦C overnight. The gold par-
ticle sizes were adjusted by controlling the pH value of HAuCl4
with NaOH. The higher was the pH value, less fraction of gold
deposited on the support, thus smaller the Au particles would be.
In this process the pH value of HAuCl4 was controlled with NaOH
at 10 and 6.2 and the Au particle size was dispersed at 2–5 nm
(denoted as Au/TiO2 (I)) and 8–12 nm (denoted as Au/TiO2 (II)),
respectively.
3. Results and discussion
3.1. Catalytic performances of Au/TiO2 in different reaction media
The catalytic performances of Au/TiO2 were evaluated for the
hydrogenation of o-CNB in several media such as scCO2, ethanol
and water. As hydrogen is completely miscible with scCO2 and
this is beneficial to the hydrogenation reactions, the hydrogena-
tion of o-CNB over Au/TiO2 was performed in scCO2 and compared
with that in ethanol. The conversion of o-CNB in scCO2 was about
half of that in ethanol (entries 1 and 2 in Table 1), which is dif-
ferent from our previous results obtained with Ni/TiO2 catalyst.
Over Ni/TiO2, the conversion of o-CNB in scCO2 is much higher than
that in ethanol at 35 ◦C [14]. Moreover, the conversion of o-CNB in
H2O was about 3 times larger than that in ethanol. It was reported
that the hydrogenation rate of p-nitrophenol in water–ethanol
and water–dioxane increased with water content and became the
highest in pure water [23]. The promoting effect of H2O on the
hydrogenation of nitroaromatics was also reported over supported
Pt, Ni, Ag, and Cu catalysts [24,25]. o-CNB is soluble in ethanol and
sparingly soluble in water (0.44 g l−1 at 20 ◦C [26]). The promoting
effect of H2O might be partially attributed to the H-bond inter-
action of interfacial H2O molecules with hydrophobic reactant or
transition state intermediates [27]. When CO2 was introduced into
the H2O system, the conversion decreased with the increase of CO2
pressure (entries 3–5 in Table 1). It seems that CO2 molecule plays
a negative effect on the catalytic activity of Au/TiO2 in the present
hydrogenation.
2.2. Hydrogenation of o-chloronitrobenzene
The hydrogenation of o-CNB was carried out in a 50 ml high-
pressure stainless steel batch reactor. A certain amount of Au/TiO2,
deionized water and o-CNB were loaded in the reactor. After that
the reactor was sealed, flushed three times with N2 to remove
the air, and heated to 140 ◦C for 15 min. Hydrogen (4 MPa) and
then liquid CO2 was introduced into the reactor to the desired
pressure. The reaction mixture was stirred continuously with a
magnetic stirrer during reaction. The stirring speed was 900 rpm
to make sure the reaction was free from any significant mass trans-
port limitations. After reaction the reactor was cooled in ice-water
bath, depressurized carefully, and the liquid product was extracted
with 10 ml diethyl ether. The reaction mixture was analyzed with
a gas chromatograph (GC, Shimadzu GC-2010, FID, Rtx-5 capil-
lary column, 30 m × 0.25 mm × 0.25 ␮m), and was identified by gas
chromatography/mass spectrometry (GC/121 MS, Agilent 5890).
Hydrogenation reactions in ethanol, water, and H2O/CO2 were con-
ducted in the same reactor using similar procedures. The extent of
error for the conversion was in the range of 2%.
The phase behavior was examined by the naked eye through the
transparent sapphire windows attached to an 85 cm3 high-pressure
reactor [13].

Y. Hao et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 183–188
185
Table 1
Hydrogenation of o-CNB over Au/TiO2 in different reaction media.
Entry
Medium
Conversion
(%)
Selectivity (%)
No. of
phases
o-CAN
Others
1
2
3
4
5
scCO2 (12 MPa)
Ethanol (5 ml)
H2O (5 ml)
H2O (5 ml) + 6 MPa CO2
H2O (5 ml) + 8.3 MPa CO2
14.3
27.8
83.7
76.9
71.1
98.7
96.6
92.7
94.1
88.7
1.3
3.4
7.4
5.9
11.3
1
1
2
2
2
Reaction conditions: o-CNB 0.2 g, Au/TiO2 (I) 0.02 g, H2 4.0 MPa, 140 ◦C, reaction time 1 h.
The selectivity to o-CAN in scCO2 was better than those in other
systems containing liquid H2O and ethanol. The main byproducts
were o-chloronitrisobenzene, N-chlorophenylhydroxylamine and
aniline. The higher selectivity in scCO2 might be related with the
lower concentration of reacting species in the CO2 gas phase than
that in the liquid phase. A nearly 100% selectivity to CAN has been
achieved over Au/TiO2 in the gas phase hydrogenation of CNB at
180 ◦C [12].
3.2. Influence of CO2 pressure and reaction time
To further study the effect of CO2 on the catalytic feature of
Au/TiO2 in o-CNB hydrogenation, Au/TiO2 catalysts with different
particle sizes were investigated in scCO2 and H2O/CO2. In scCO2
(Fig. 1), the conversion of o-CNB and the selectivity to o-CAN did
not change with CO2 pressure. This trend is totally different to our
previous results that the conversion of o-CNB increases with CO2
pressure and reaches the maximum at 10 MPa when Pt/C was used
as catalyst [16]. In H2O/CO2 (Fig. 2), the o-CNB conversion decreased
with increasing CO2 pressure. In both scCO2 and H2O/CO2 reaction
media, Au/TiO2 catalysts with smaller gold particles were more
active than that with the larger ones. Claus et al. reported that
the TOF value increased with decreasing of particle size when sup-
ported gold catalysts were used for the selective hydrogenation
of acrolein [28]. Keane et al. also reported that the hydrogenation
of m-dinitrobenzene over supported Au/TiO2 exhibited a particle
size sensitivity where higher specific rates were obtained with
smaller Au particles; the catalysts bearing the smaller Au particles
(dp = 3.4 nm) delivered an order of magnitude higher specific rate
when compared with larger ones (dp = 6.1 nm) [29].
Fig. 2. Hydrogenation of o-CNB over Au/TiO2 with different particle sizes in
H2O/CO2. Reaction conditions: o-CNB 0.2 g, Au/TiO2 0.02 g, H2O 5 ml, H2 4 MPa,
140 ◦C, 1 h.
ity to o-CAN remained at nearly 100% through the whole reaction
process, but the o-CNB conversion increased with reaction time in
the first 8 h and did not change anymore. This indicates that the
Au/TiO2 catalysts deactivated during the reaction.
3.3. Possible factors resulting in the deactivation of Au/TiO2
Since deactivation of supported catalysts usually relates with
agglomeration of metal particles, the change of the particle sizes
before and after reaction in scCO2 at 140 ◦C was examined with XRD
and TEM. But no diffraction peaks corresponding to Au were found,
which is mainly due to the highly dispersity of gold particles and the
low loading of gold. TEM results (Fig. 4) demonstrate that there was
no significant agglomeration of gold particles after reaction. It was
reported that the calcination of Au/TiO2 catalysts at 200 ◦C could
The influence of reaction time was also investigated in present
work (Fig. 3). For both Au/TiO2 (I) and Au/TiO2 (II), the selectiv-
Fig. 1. Hydrogenation of o-CNB over Au/TiO2 with different particle sizes in scCO2.
Reaction conditions: o-CNB 1.5 g, Au/TiO2 0.15 g, H2 4 MPa, 140 ◦C, 1 h. Au/TiO2 (I),
the Au particle sizes are in the range of 2–5 nm; Au/TiO2 (II), the Au particle sizes
are in the range of 8–12 nm.
Fig. 3. Effect of reaction time in scCO2. Reaction conditions: o-CNB 1.5 g, Au/TiO2
0.04 g, H2 4 MPa, CO2 12 MPa, 140 ◦C.

186
Y. Hao et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 183–188
Fig. 4. TEM images of the fresh (a) and used (b) Au/TiO2 catalysts in scCO2.
not lead to the agglomeration of the Au particles [22]. Therefore,
the agglomeration of Au particles was not the main factor causing
the deactivation of catalysts.
species on gold. The intensity of the bands did not change with
increasing contact time with CO2 and H2. Hence, carbonate-like
species are formed on gold during the hydrogenation in the pres-
ence of CO2 and this might be one of reasons that causing the
deactivation of Au/TiO2 in scCO2.
Because carbonate formation is one of common factors that
cause the deactivation of supported metal catalysts in the pres-
ence of CO or CO2 [30,31], DRIFTS experiment was made to detect
whether carbonate-like species were formed on Au/TiO2 under
the gas flow of CO2 and H2. Fig. 5 shows the IR difference spec-
tra of Au/TiO2 in the 1000–1800 cm−1 region with increasing time
on stream. Carbonate-like species on gold are characterized by IR
bands at 1277, 1315, 1343, 1534, 1566, and 1606 cm−1 [31]. Thus,
among the bands at 1270–1655 cm−1 in Fig. 5, the peaks at 1270,
1353, 1534 and 1610 cm−1 might be attributed to carbonate-like
To further investigate the factors causing the deactivation of
Au/TiO2 catalyst, XPS was used to investigate the electronic state of
gold on the surface of the fresh and used catalysts. The C 1s peak of
contaminant carbon at 284.6 eV was used as a reference. Analysis
of Cl 2p region shows that the amount of Cl on the activated cata-
lyst is negligible. As shown in Fig. 6, the peaks with binding energy
of Au 4f7/2 at 83.0 eV and Au 4f5/2 at 86.7 eV are characteristic of
Au0 [32], which indicates the complete reduction of gold catalyst
on the fresh Au/TiO2. It is worth to note that there is a shift of 1 eV
to lower binding energy from standard values (84.0 and 87.7 eV)
due to the strong metal–support interaction [32–34]. For the used
Au/TiO2 catalyst in scCO2, new bands at 84.5 (Au 4f7/2) and 89.1 eV
(Au 4f5/2) were found, which indicates that oxidized Au species are
present. The band at 84.5 eV (Au 4f7/2) was possibly assigned to Au
4f7/2 of Auı+, and the one at 89.1 eV assigned to Au 4f5/2 of Au3+ [30].
For the used catalyst in H2O/CO2, Au3+ and Auı+ were also detected
together with metallic gold species (Fig. 6(iii)). For the used cata-
lyst in H2O (Fig. 6(iv)), only two peaks corresponding to Au0 were
found, which is the same as the fresh catalyst. The results of UV/Vis
measurement are shown in Fig. 7. The peak position of nanosized
Au metal particles is between 500 and 600 nm; small clusters such
ı+
as (Au)n exhibit a band around 390 nm [35,36]. Fig. 7 shows that
the absorption spectrum of the used Au/TiO2 in H2O is similar to
that of the fresh catalyst; however, for the used Au/TiO2 catalysts
in scCO2 or in H2O/CO2, new absorption bands appeared around
430 nm, indicating the presence of oxidized Au species. The above
results suggest that the original Au0 species was oxidized by CO2
rather than by H2O and reacting species. Mihaylov et al. reported
the same phenomenon that the Au0 species was oxidized by CO2
during the catalytic oxidation of CO at room temperature [37–39].
Fig. 5. IR difference spectra in the 1000–1800 cm−1 region recorded during 10%
CO2/90% H2 treatment of Au/TiO2 at 140 ◦C with increasing time on stream (min):
(a) 5; (b) 15; (c) 30.

Y. Hao et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 183–188
187
Table 2
Results for the hydrogenation of o-CNB over Au/TiO2 pretreated under different
conditions.
Entry
Conversion (%)
Selectivity (%)
1a
2b
3c
4d
8.2
6.1
2.9
1.5
98.5
99.0
94.8
96.5
Reaction conditions: o-CNB 1.5 g, Au/TiO2 (II) 0.04 g, H2 4 MPa, CO2 12 MPa, 140 ◦C,
reaction time 1 h.
a
Fresh Au/TiO2 (II) catalyst.
b
Au/TiO2 (II) was treated in 12 MPa CO2 at 140 ◦C for 8 h.
c
Au/TiO2 (II) was treated in 12 MPa CO2 and 4 MPa H2 at 140 ◦C for 8 h.
Charged CO (0.1 MPa) into the reaction system.
d
that with scCO2 and H2. This indicates that there might be other
factors affecting the activity of Au/TiO2.
As CO could be generated from CO2 during the hydrogenation
process, and may block the active sites [41], we have made attempts
to detect CO adsorbed on gold by DRIFTS measurement, but CO
was not found under a gas flow of 90% H2 and 10% CO2 at 140 ◦C.
In another DRIFTS study, CO was first introduced into the cham-
ber and the CO adsorbed on Au/TiO2 was detected; however, the
adsorbed CO was removed from the catalyst surface by He purge.
This indicates that the adsorption of CO on Au/TiO2 is weak and
reversible. In addition, CO could not be detected by GC analysis of
the reactor effluent gas mixture using a TCD detector, due to the
relatively high concentration of CO2 and less amount of CO formed.
However, when a trace amount of CO (0.1 MPa) was introduced into
the reaction system, the o-CNB conversion decreased significantly
(entry 4 in Table 2), implying that CO produced from CO2 and H2
might be another possible factor for the deactivation of Au/TiO2. ICP
analysis was also made to investigate the possibility of Au leaching
during the reaction. The ratio of the leached to the original amount
was 0.59%, implying the Au leaching is negligible even though the
H2O/CO2 reaction system presents somewhat acidity (pH = 3) [42].
Thus, besides the molecular interactions of CO2 with react-
ing species as proposed in previous work [13,14], compressed
CO2 plays the following roles in the present work: (1) increas-
ing the solubility of H2 in the liquid phase (positive); (2) forming
carbonate-like species over Au/TiO2 (negative); (3) oxidizing the
active species of Au0 (negative); and (4) acting as the source of
CO (negative). The counterbalance of these positive and negative
effects resulted in the constant conversion of o-CNB with increas-
ing CO2 pressure in scCO2. For the reaction in H2O/CO2, besides
the above factors, the amount of o-CNB dissolved into CO2 may
increase and the concentration of o-CNB in H2O may decrease with
Fig. 6. Au 4f XPS spectra of Au/TiO2 (I). Fresh (i), after reaction in scCO2 (ii), in
H2O/CO2 (iii), and in H2O (iv). Reaction conditions: o-CNB 1.5 g, Au/TiO2 0.15 g, H2
4 MPa, 140 ◦C, 1 h.
Metallic gold Au0 is the main active species for hydrogenation reac-
tions. Therefore, the oxidation of Au0 in the presence of CO2 should
be another reason for the deactivation of catalyst. The oxidized
gold species could not be reduced into metallic state in the present
hydrogenation temperature of 140 ◦C. It is consistent with the fol-
lowing temperature programmed reduction results: oxidized gold
should be reduced at >170 ◦C for the 1 wt% Au/TiO2 [40].
Moreover, to get a further insight of the influence of CO2 on the
gold catalysts, Au/TiO2 was pretreated under different conditions
before reaction (Table 2). The conversion of o-CNB over the cata-
lyst pretreated in scCO2 was slightly lower than that over the fresh
catalyst (entries 1 and 2), but higher than that over the Au/TiO2
treated under CO2 and H2 (entries 2 and 3). This suggests that the
presence of H2 would accelerate the deactivation of the catalyst.
XPS results (Fig. 8) show that both the Au/TiO2 catalysts pre-
treated with scCO2 or with scCO2 and H2 were partially oxidized.
The ratio of Auı+ was larger for the catalyst treated with scCO2 than
(iii)
(ii)
(iv)
(i)
200
400
Wavelength (nm)
600
Fig. 7. Diffuse reflectance UV/Vis spectra of different Au/TiO2 catalysts: fresh (i),
after reaction in scCO2 (ii), in H2O/CO2 (iii), and in H2O (iv). Reaction conditions:
o-CNB 1.5 g, Au/TiO2 (I) 0.15 g, H2 4 MPa, 140 ◦C, 1 h.
Fig. 8. Au 4f XPS spectra of Au/TiO2 (II): (i) fresh, (ii) treated in 12 MPa CO2 at 140 ◦
C
for 8 h, (iii) treated in 12 MPa CO2 and 4 MPa H2 at 140 ◦C for 8 h.

188
Y. Hao et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 183–188
increasing CO2 pressure. Hence, the conversion in H2O/CO2 linearly
decreased.
[13] X.C. Meng, H.Y. Cheng, Y. Akiyama, Y.F. Hao, W.B. Qiao, Y.C. Yu, F.Y. Zhao, S.
Fujita, M. Arai, J. Catal. 264 (2009) 1.
[14] X.C. Meng, H.Y. Cheng, S. Fujita, Y.F. Hao, Y.J. Shang, Y.C. Yu, S.X. Cai, F.Y. Zhao,
M. Arai, J. Catal. 269 (2010) 131.
4. Conclusion
[15] C.Y. Xi, H.Y. Cheng, J.M. Hao, S.X. Cai, F.Y. Zhao, J. Mol. Catal. A 282 (2008) 80.
[16] F. Zhao, Y. Ikushima, M. Arai, J. Catal. 224 (2004) 479.
[17] X.G. Wang, H. Kawanami, S.E. Dapurkar, N.S. Venkataramanan, M. Chatterjee,
T. Yokoyama, Y. Ikushima, Appl. Catal. A 349 (2008) 86.
[18] P. Haider, B. Kimmerle, F. Krumeich, W. Kleist, J.D. Grunwaldt, A. Baiker, Catal.
Lett. 125 (2008) 169.
[19] M. Chatterjee, Y. Ikushima, Y. Hakuta, H. Kawanami, Adv. Synth. Catal. 348
(2006) 1580.
[20] R.X. Liu, Y.C. Yu, K. Yoshida, G.M. Li, H.X. Jiang, M.H. Zhang, F.Y. Zhao, S. Fujita,
M. Arai, J. Catal. 269 (2010) 191.
[21] M. Arai, Y. Nishiyama, Y. Ikushima, J. Supercrit. Fluids 13 (1998) 149.
[22] A. Wolf, F. Schuth, Appl. Catal. A 226 (2002) 1.
[23] S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation for Organic
Synthesis, John Wiley and Sons, 2001, p. 336.
[24] J. Ning, J. Xu, J. Liu, H. Miao, H. Ma, C. Chen, X. Li, L. Zhou, W. Yu, Catal. Commun.
8 (2007) 1763.
The conversion of o-CNB during hydrogenation over Au/TiO2 in
scCO2 changes little with CO2 pressure. The Au/TiO2 catalyst deacti-
vates during the hydrogenation in scCO2. Carbonate-like species are
formed on gold and may cover some of active sites. The original Au0
species was partially oxidized by CO2 even in the presence of H2. It
seems likely that CO might be formed in the hydrogenation process
and poisons some active sites. The above three factors, formation
of carbonate-like species, the oxidation of surface metallic Au, and
the generation of CO might be responsible for the deactivation of
Au/TiO2 during the hydrogenation of o-CNB in scCO2.
[25] P. Maity, S. Basu, S. Bhaduri, G.K. Lahiria, Adv. Synth. Catal. 349 (2007) 1955.
[26] M.H. Abraham, J. Le, J. Pharm. Sci. 88 (1999) 868.
[27] Y.S. Jung, R.A. Marcus, J. Am. Chem. Soc. 129 (2007) 5492.
[28] C. Mohr, H. Hofmeister, P. Claus, J. Catal. 213 (2003) 86.
[29] F. Cardenas-Lizana, S. Gomez-Quero, H. Idriss, M.A. Keane, J. Catal. 268 (2009)
223.
Acknowledgments
The authors gratefully acknowledge the financial support from
the One Hundred Talent Program, NSFC 20873139 and KJCX2, YW.
H16.
[30] H. Wang, H.Q. Zhu, Z.F. Qin, F.X. Liang, G.F. Wang, J.G. Wang, J. Catal. 264 (2009)
154.
[31] Y. Hao, M. Mihaylov, E. Ivanova, K. Hadjiivanovic, H. Knozinger, B.C. Gates, J.
Catal. 261 (2009) 137.
[32] R. Radnik, C. Mohr, P. Claus, Phys. Chem. Chem. Phys. 5 (2003) 172.
[33] S. Arrii, F. Morfin, A.J. Renouprez, J.L. Rousset, J. Am. Chem. Soc. 126 (2004)
1199.
[34] L. Ilieva, J.W. Sobczak, M. Manzoli, B.L. Su, D. Andreeva, Appl. Catal. A 291 (2005)
85.
[35] A.C. Gluhoi, X. Tang, P. Marginean, B.E. Nieuwenhuys, Top. Catal. 39 (2006) 101.
[36] J.L. Margitfalvi, A. Fasi, M. Hegedus, F. Lonyi, S. Gobolos, N. Bogdanchikova, Catal.
Today 72 (2002) 157.
[37] M. Mihaylov, E. Ivanova, Y. Hao, K. Hadjiivanov, B.C. Gates, H. Knozinger, Chem.
Commun. (2008) 175.
[38] M. Mihaylov, H. Knozinger, K. Hadjiivanov, B.C. Gates, Chem. Ing. Tech. 79
(2007) 795.
[39] M. Mihaylov, E. Ivanova, Y. Hao, K. Hadjiivanov, H. Knozinger, B.C. Gates, J. Phys.
Chem. C 112 (2008) 1897.
[40] F. Cardenas-Lizana, S. Gomez-Quero, A. Hugon, L. Delannoy, C. Louis, M.A.
Keane, J. Catal. 262 (2009) 235.
References
[1] M. Haruta, Catal. Today 36 (1997) 153.
[2] C.H. Christensen, B. Jorgensen, J. Rass-Hansen, K. Egeblad, R. Madsen, S.K. Klit-
gaard, S.M. Hansen, M.R. Hansen, H.C. Andersen, A. Riisager, Angew. Chem. Int.
Ed. 45 (2006) 4648.
[3] M.D. Hughes, Y.J. Xu, P. Jenkins, P. McMorn, P. Landon, D.I. Enache, A.F. Carley,
G.A. Attard, G.J. Hutchings, F. King, E.H. Stitt, P. Johnston, K. Griffin, C.J. Kiely,
Nature 437 (2005) 1132.
[4] M. Boronat, P. Concepcion, A. Corma, S. Gonzalez, F. Illas, P. Serna, J. Am. Chem.
Soc. 129 (2007) 16230.
[5] A. Corma, P. Serna, Science 313 (2006) 332.
[6] D.P. He, H. Shi, Y. Wu, B.Q. Xu, Green Chem. 9 (2007) 849.
[7] B.J. Zuo, Y. Wang, Q.L. Wang, J.L. Zhang, N.Z. Wu, L.D. Peng, L.L. Gui, X.D. Wang,
R.M. Wang, D.P. Yu, J. Catal. 222 (2004) 493.
[8] A. Baiker, Chem. Rev. 99 (1999) 453.
[9] C.A. Eckert, B.L. Knutson, P.G. Debenedetti, Nature 383 (1996) 313.
[10] J.R. Hyde, P. Licence, D. Carter, M. Poliakoff, Appl. Catal. A 222 (2001) 119.
[11] W. Leitner, Acc. Chem. Res. 35 (2002) 746.
[41] S. Ichikawa, M. Tada, Y. Iwasawea, T. Ikariya, Chem. Commun. (2005) 924.
[42] C. Roosen, M. Ansorge-Schumacher, T. Mang, W. Leitner, L. Greiner, Green
Chem. 9 (2007) 455.
[12] F. Cárdenas-Lizana, S. Gómez-Quero, M.A. Keane, ChemSusChem 1 (2008) 215.