Hydroxylation of benzene by activated carbon catalyst

Article abstract of DOI:10.1016/S1872-2067(11)60444-0

Phenol is widely used in industry. The hydroxylation of benzene to phenol can meet the increasing requirement for phenol and green chemistry. Commercial wood-based activated carbon was treated by different methods, including oxidation by nitric acid and h

Full text of DOI:10.1016/S1872-2067(11)60444-0

CHINESE JOURNAL OF CATALYSIS
Volume 33, Issue 10, 2012
Online English edition of the Chinese language journal
Cite this article as: Chin. J. Catal., 2012, 33: 1622–1630.
ARTICLE
Hydroxylation of Benzene by Activated Carbon Catalyst
a
b
XU Jiaquan, LIU Huihui, YANG Ruiguang, LI Guiying , HU Changwei
Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064,
Sichuan, China
Abstract: Phenol is widely used in industry. The hydroxylation of benzene to phenol can meet the increasing requirement for phenol and
green chemistry. Commercial wood-based activated carbon was treated by different methods, including oxidation by nitric acid and hy-
drogen peroxide and thermal treatment at different temperatures, and used as the catalyst for the hydroxylation of benzene to phenol in
2 2
acetonitrile using H O as oxidant. The mechanism of the hydroxylation was also investigated. The activated carbon samples were char-
acterized by Boehm titration, nitrogen adsorption, and X-ray photoelectron spectroscopy. The pH had a significant effect on the phenol
yield. The effect of benzene adsorption was also investigated. Several model compounds with oxygen groups were used as catalysts for the
hydroxylation of benzene. The reaction between phenolic hydroxyl and quinone on the surface of activated carbon was responsible for the
2 2
hydroxylation of benzene to phenol when it activated H O to form the hydroxyl radical. Activated carbon was an efficient and stable
catalyst and a maximum phenol yield of 14.4% was obtained under the optimal reaction conditions.
Key words: activated carbon; benzene hydroxylation; oxygen group; model compound; mechanism
Phenol is an intermediate for preparing chemicals such as
phenol resins, fibers, caprolactam, adipic acid, dyestuffs, and
medicine. More than 90% of the phenol produced in industry is
based on the cumene route, which produces acetone as a by-
product [1]. The direct hydroxylation of benzene to phenol
with a clean oxidant is an alternative route to the cumene
process. Oxidants employed in the hydroxylation of benzene
groups [15,16]. With a metal/activated carbon catalyst, the
metal species is usually the active phase for the hydroxylation
of benzene, while the activated carbon is just a support [12,13].
Activated carbon has good catalytic activity in oxidation reac-
tions because of its ability to accept electrons. Car-
bonyl/quinone groups play an important role in the oxidation
reaction [17í19]. Hydroxyl radicals can be produced by the
activation of hydrogen peroxide on activated carbon [20,21],
and the hydroxyl radicals were stabilized by resonance on the
carbon surface [22í24].
are hydrogen peroxide, molecular oxygen, and N O. Using
2
nitrous oxide as oxidant can give a high phenol yield and se-
lectivity [2,3] but the process has an economic advantage only
when nitrous oxide is a byproduct from adipic acid production.
When molecular oxygen is used as an economic and clean
oxidant, there is only low activity, even in complex systems
Separation in a homogeneous catalytic system and leaching
of active components from supported catalysts are still chal-
lenges for the hydroxylation of benzene using H O as the
2
2
[
1,4í6]. For the hydroxylation of benzene with H O as oxi-
oxidant [25]. In the present work, a cheap and easily prepared
catalyst was obtained by the modification of wood-based ac-
tivated carbon. The role of surface oxygen groups on the acti-
vated carbon and effect of pH on the yield of phenol were
investigated.
2
2
TM
dant, many catalysts have been used, such as TS-PQ [7],
NaY zeolite [8], heteropoly acid [9], Fe/MWCNTS [10],
VO -defects/MWCNT [11], Fe/activated carbon [12,13], and
2
activated carbon [14], where the oxygen groups on the surface
was reported to have an effect on the activity.
Activated carbon is an important catalyst and catalyst sup-
port because of its physical properties and surface functional
1 Experimental
Received 10 May 2012. Accepted 13 June 2012.
a
Corresponding author. Tel/Fax: +86-28-85415745; E-mail: gchem@scu.edu.cn
b
Corresponding author. Tel/Fax: +86-28-85411105; E-mail: chwehu@mail.sc.cninfo.net
This work was supported by the National Natural Science Foundation of China (20502017, 20872102).
Copyright © 2012, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier BV. All rights reserved.
DOI: 10.1016/S1872-2067(11)60444-0

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
1
.1 Catalyst preparation
(1486.6 eV). The source was operated at 12 kV and 12 mA.
The binding energy was calibrated with C 1s at 284.8 eV. A
linear background was subtracted from all spectra and then
peak fitting was performed with a 80/20 Lorentz-Gauss func-
tion.
Commercially available wood-based activated carbon from
Jiangsu Nantong Activated Carbon Cooperation (China) with a
particle size of 0.42í0.85 mm was washed by distilled water
and then dried at 110 ºC to obtain the sample named WAC.
WAC (1.0 g) was immersed in a series of nitric acid solutions
1.3 Activity test
(
10.0 ml) of different concentrations at room temperature for
4 h, and then washed by boiled distilled water until it reached
2
To estimate the adsorption capacity of benzene, 0.5 g sample
a constant pH and finally dried at 110 ºC for 12 h. The samples
were denoted as WAC-xN (x was the concentration of nitric
acid solution, mol/L).
was added to a solution of 15.0 ml acetonitrile, 6.0 ml H O, and
2
0.5 ml benzene. Adsorption was performed at 30 ºC for 48 h.
The amount of benzene adsorbed on the activated carbon
sample was calculated as q = [21.5 × (C íC )]/m, where q is the
The sample WAC-0.5N was placed in a quartz tube in N
2
0
1
3
flow (50.0 cm /min) and heated at 10.0 ºC/min from room
amount of benzene adsorbed, C is the initial concentration of
0
temperature to the desired temperature which was held for 4 h.
After that, the system was cooled down to room temperature in
benzene in the solution, C is the benzene concentration after
1
48 h, and m is the amount of sample used.
3
N2 flow (50 cm /min). The samples were labeled as
The hydroxylation of benzene was carried out in a
two-necked 50 ml round flask placed in a water bath. Typical
reaction conditions were 0.60 g catalyst, 2.0 ml (22.5 mmol)
benzene, 15.0 ml acetonitrile, and 6.0 ml (58.8 mmol) hydro-
gen peroxide (30 wt%) kept at 70 ºC for 6 h. The reaction
mixture was filtered and analyzed by high performance liquid
chromatography (Waters 1525P) using a reverse phase C18
column and an ultraviolet detector set at 225 and 254 nm. The
main products and byproducts were quantified using o-cresol
as internal standard. 50 vol% water and 50 vol% acetonitrile
were used as the mobile phase. All products were first identi-
fied by coupled gas chromatography and mass spectroscopy
(Agilent 5973 Network 6890N). The amount of dihy-
droxy-benzene formed was small and no tar was detected in the
crude reaction mixture. The yield of phenol was calculated as
(mmol phenol)/(mmol initial benzene). The selectivity for
phenol was calculated as (mmol phenol)/(mmol phenol + mmol
benzoquinone).
WAC-0.5N-y (y was the temperature of thermal treatment, ºC).
Part of WAC-0.5N was treated further with 30% H O (1.0 g
2
2
WAC-0.5N with 4.0, 8.0, or 12.0 ml H O ) at 70 ºC for 6 h. The
2
2
samples were named WAC-0.5N-zH (z was the ratio between
the volume of 30% H O solution and the amount of
2
2
WAC-0.5N sample, ml/g).
1
.2 Catalyst characterization
Titration of the samples was carried out using the Boehm
3
method [26í29]. Aqueous solutions of NaHCO , NaOH, and
Na CO with a concentration of 0.05 mol/L were used in the
2
3
titration. 0.05 mol/L of CH CH ONa was freshly prepared
3
2
using chromatography pure ethanol and sodium. Each sample
0.50 g) was placed in 50 ml of the above alkaline solutions for
4 h at room temperature. Excess hydrochloric acid solution
0.10 mol/L) was added to the filtrate, and this was then back
(
2
(
titrated with NaOH using methyl red as indicator.
In situ infrared (IR) spectroscopy was used to monitor the
oxidation of hydroquinone by H O in acetonitrile at different
2 Results and discussion
2
2
pHs at 70 ºC. In situ IR was carried out on a ReactIR ic10
Mettler Toledo). The absorption intensity of hydroquinone
2.1 Textural characteristics of the samples
(
and quinone in the reaction was detected using their charac-
The nitrogen adsorption isotherms and pore size distribution
of different activated carbon samples are shown in Fig. 1. All
the isotherms were type I [30], which meant that all the samples
were microporous materials. The pore size distribution showed
that the activated carbon samples were microporous materials.
The appearance of hysteresis loops in the nitrogen isotherms
indicated the existence of some mesoporosity. The hysteresis
loops were type H4 with parallel adsorption and desorption
branches, which suggested slit-shaped pores.
í1
teristic absorption at 1517 and 1661 cm , respectively.
The BET surface area (ABET) of the samples was measured
by N adsorption at í196 ºC on a Micromeritics Instrument
2
(
Tri-star II). Prior to measurement, the sample was evacuated at
í2
1
00 ºC to a pressure of less than 10 Pa. The BET equation was
used to get the surface area from the N adsorption isotherm.
2
The micropore volume (Vmic), micropores surface area (Amic),
and external surface area (A ) were obtained using the t-plot
ext
method.
The details of the texture of the samples are shown in Table
1. The results showed that the oxidation of the activated carbon
sample with dilute nitric acid (0.50 mol/L) increased ABET and
Amic, which was probably because this treatment made more
pore openings [12], that is, ashes and other impurities in the
To determine the surface oxygen groups and corresponding
carbon species on the activated carbon, X-ray photoelectron
spectroscopy (XPS) characterization was carried out on an
XSAM800 spectrometer (KRATOS) using Al K radiation
Į

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
Table 1 Textural characteristic of the activated carbon samples
(a)
50
3
2
2
2
Sample
V
mic/(m /g)
A
BET/(m /g)
A
mic/(m /g)
A
Ext/(m /g)
WAC
0.196
0.212
0.199
0.179
0.183
0.158
0.235
0.237
0.261
0.238
570
603
582
627
601
619
682
688
762
699
423
460
429
389
397
342
508
511
563
513
147
143
152
238
204
276
174
176
199
186
WAC-0.5N
WAC-10N
WAC-0.5N-4H
WAC-0.5N-8H
WAC-0.5N-12H
WAC-0.5N-250
WAC-0.5N-450
WAC-0.5N-650
WAC-0.5N-900
WAC-0.5N-900
WAC-5N-8H
WAC-0.5N
WAC-10N
WAC
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure (p/p
0
)
the oxidized sample was treated at different temperatures, the
surface area was increased and it reached a maximum with 650
ºC. A slight decomposition of the sample during the treatment
(b)
0.05
WAC
resulting in the removal of oxygen groups as CO can explain
2
this [33].
WAC-0.5N
2
.2 Boehm titration of the samples
WAC-0.5N-900
Table 2 summarizes the results of the Boehm titration of the
WAC-0.5N-8H
samples. For the fresh samples, with the increase of the nitric
acid concentration in the treatment, the amount of oxygen
groups on the sample increased, which was in agreement with
previous reports [15,34,35]. Plentiful oxygen groups were
created on the surface of WAC-0.5N after it was treated with
WAC-10N
0
2
4
6
8
10
Pore width (nm)
Fig. 1.
N adsorption-desorption isotherms and pore size distribution of
2
H O .
For
WAC-0.5N-4H,
WAC-0.5N-8H,
and
2
2
different activated carbon samples.
WAC-0.5N-12H, with increasing amounts of H O used, the
2
2
amount of carboxyl, lactone, and carbonyl groups increased.
When the oxidized sample was treated at 250 ºC, the amount of
carboxyl groups decreased while the amounts of other oxygen
groups did not change, which was in agreement with previous
reports [34,36]. When the oxidized sample was heated at 450
ºC, the amount of carboxyl and phenolic hydroxyl groups
decreased while the amount of lactone increased, which indi-
cated that dehydration to form lactone occurred during the
thermal treatment [37]. Carboxyl groups on the surface of
WAC-0.5N were removed when the sample was treated at 650
ºC. After being treated at 900 ºC for 4 h, carboxyl groups and
pore channels were removed by the nitric acid. The anchoring
of oxygen groups on the inner surface of the pores would also
contribute to the increase of the micropores. However, as the
concentration of nitric acid was increased, ABET decreased
because some macropores collapsed or because there were
more oxygen groups extending from the wall of the slit-shaped
mesopores [14,31].
ABET and AExt increased while Amic decreased when
WAC-0.5N was further treated with 30% H O , which was
2
2
consistent with the results reported in the literature [32]. When
Table 2 Results of Boehm titration of the fresh and used activated carbon samples
Carboxyl (mmol/g) Lactone (mmol/g) Phenolic hydroxyl (mmol/g)
Fresh Fresh Used
0.116 0.151
Carbonyl (mmol/g)
Sample
Fresh
0
Used
0.230
0.335
0.340
0.395
0.426
0.651
0.356
0.172
0.131
0.235
Used
0.032
0.204
0.315
0.176
0.325
0.334
0.268
0.294
0.119
0.039
Fresh
0.213
0.266
0.286
0.334
0.453
0.670
0.263
0.191
0.106
0.081
Used
0.266
0.358
0.226
0.417
0.372
0.597
0.235
0.268
0.160
0.084
WAC
0.040
0.097
0.197
0.103
0.205
0.288
0.098
0.153
0.019
0
WAC-0.5N
0.179
0.257
0.232
0.321
0.482
0.085
0.027
0
0.183
0.262
0.276
0.277
0.227
0.187
0.109
0.049
0.005
0.132
0.111
0.107
0.260
0.152
0.193
0.146
0.196
0.161
WAC-10N
WAC-0.5N-4H
WAC-0.5N-8H
WAC-0.5N-12H
WAC-0.5N-250
WAC-0.5N-450
WAC-0.5N-650
WAC-0.5N-900
0

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
1
6
lactone on the surface of WAC-0.5N disappeared, while phe-
nolic hydroxyl also nearly disappeared. The results were con-
sistent with those in the literature [36].
14
(2)
(
1)
1
2
0
8
6
4
2
0
The surface chemistry of the activated carbon after being
used in the hydroxylation reaction was also investigated (Table
1
2). The amounts of carboxyl groups and lactone had increased
distinctly during the reaction, except with WAC on which the
amount of lactone had decreased slightly. For the samples
treated above 450 ºC, the amount of both phenolic hydroxyl
and carbonyl groups were increased, while for the other sam-
ples, the phenolic and carbonyl groups were changed. It was
reported that the surface chemistry of multi-walled carbon
nanotubes can be changed by an oxidation reaction [10]. The
surface chemistry of the samples in this work was easily
changed by the oxidation reaction. The changes were larger for
samples with a lower amount of oxygen groups before it was
used in the reaction.
-
1
0
1
2
3
4
5
6
7
+
ꢀ
lg[H ]
+
Fig. 2. Relationship between [H ] in the reaction system and the phenol
yield over WAC-0.5N-650 (1) and other samples used in our work (2).
position into hydroxyl radicals was dependent on the aqueous
phase pH [38] when the reaction was catalyzed by activated
carbon, while the oxidation reactions were also affected by pH
[39]. In order to investigate the effect of the aqueous phase pH
on the activities of the samples, the hydroxylation of benzene
over WAC-0.5N-650 was performed with various aqueous
phase pH values, which were adjusted by using solutions of
perchloric acid and sodium hydroxide. Figure 2 shows that the
2
.3 Activity of the catalysts
The catalytic activity in terms of phenol yield (average value
from at least three measurements) is shown in Table 3. The
differences in activity of the activated carbon catalysts were
small for the majority of the samples. The unmodified WAC
sample showed low activity for the hydroxylation of benzene.
The activity of sample WAC-0.5N which was treated with
nitric acid was increased to a high level. However, the activity
of the activated carbon samples did not increase with the in-
crease of nitric acid concentration. The yield of phenol from
WAC-10N (11.5%) was slightly lower than that from
+
+
concentration of H ([H ]) had a significant effect on the yield
+
of phenol. The phenol yield increased with [H
]
from 2.63
+
mol/L to 0.22 mol/L (pH = 0.66), and then decreased when [H ]
ꢂ7
was decreased from 0.22 mol/L (pH = 0.66) to 3.31×10 mol/L
+
(
pH = 6.48). A blank experiment was carried out by using [H ]
=
0.22 mol/L (pH = 0.66) in the absence of a catalyst and only a
trace amount of phenol (0.3%) was detected. Another control
experiment was also carried out employing WAC-0.5N-NaOH,
which was WAC-0.5N treated with NaOH, as catalyst and only
1.1% phenol yield was obtained. The results indicated that both
WAC-0.5N (12.5%). After being treated further with H O , the
2
2
samples became more active and the highest phenol yield of
3.5% was obtained with WAC-0.5N-8H and WAC-0.5N-12H.
1
+
The heat treatment of WAC-0.5N at 900 ºC for 4 h resulted in a
drastic decrease in the yield of phenol from 12.5% to 7.6%.
This showed that the hydroxylation activity was related to the
surface oxygen groups of the sample.
H and the catalyst were necessary to the reaction.
XPS was employed to characterize the surface chemistry of
+
sample WAC-0.5N-650 after reaction with different [H ]. The
results are shown in Fig. 3 and Table 4. By reference to the
literature [40], for surface oxygen, the deconvoluted peaks at
the binding energies of 530.3í531.8, 531.9í533.4,
+
2
.4 Effect of H on the reaction
5
33.4í533.6, and 534.8í535.6 eV corresponded to carbonyl
It was reported that the rate of hydrogen peroxide decom-
groups, alcohol and/or ether groups, carboxyl and/or ester
groups, and chemisorbed oxygen and/or water, respectively. As
shown in Fig. 3 and Table 4, the relative amounts of the four
kinds of oxygen groups on WAC-0.5N-650 after reaction at
Table 3 Yield of phenol over the catalysts
Sample
Yield of phenol (%)
Selectivity for phenol (%)
WAC
1.3
12.5
11.5
12.7
13.5
13.5
12.0
12.1
11.1
7.6
99.7
86.3
89.2
86.9
86.3
84.8
86.6
86.6
86.9
89.0
+
different [H ] were almost the same, indicating that the samples
WAC-0.5N
+
kept under different [H ] had almost the same surface chemis-
WAC-10N
try. Therefore the difference in phenol yield shown in Fig. 2(1)
WAC-0.5N-4H
WAC-0.5N-8H
WAC-0.5N-12H
WAC-0.5N-250
WAC-0.5N-450
WAC-0.5N-650
WAC-0.5N-900
+
was mainly caused by the variation of [H ]. The surface
+
chemistry during the reaction under different [H ] was similar.
The results shown in the Fig. 4 indicated that the pH of the
reaction solution decreased with increasing amount of oxygen
groups on the activated carbon samples. The relationship be-
+
tween phenol yield over different catalysts and [H ] of the

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
3400
3200
3000
2800
2600
2400
2200
2000
1800
(
a)
(b)
(c)
5
42 540 538 536 534 532 530 528 526 542 540 538 536 534 532 530 528 526 542 540 538 536 534 532 530 528 526
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
+
+
+
Fig. 3. O 1s XPS spectra of sample WAC-0.5N-650 after reaction with different [H ]. (a) [H ] = 0.22 mol/L (pH = 0.66); (b) [H ] = 0.01 mol/L (pH =
+
-4
1
.95); (c) [H ] = 3.2×10 mol/L (pH = 3.49).
reaction system is shown in Fig. 2(2), where the variation of
1.6
+
[
H ] of the system was caused by the structure of the catalyst.
Total oxygen
1.4
1.2
1.0
0.8
0.6
As shown in Fig. 2, Fig. 2(2) was not the same as Fig. 2(1),
which confirmed that the structure of the activated carbon also
affected the activity of the catalyst besides the pH. It can be
seen that the oxygen groups on the activated carbon contrib-
uted to the activity by both affecting the pH of the system and
by structural effects. Chen et al. [14] had reported that the
carboxyl had a large effect on the phenol yield, which may be
Phenolic hydroxyl
and carbonyl
+
0.4
.2
caused by the change of [H ] in the system.
Carboxyl
0
2
.5 Catalytic performance with model compounds
1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3
pH
In order to identify the roles of the oxygen-containing spe-
Fig. 4. Relationship between pH and amount of oxygen groups on the
cies, homogeneous reactions were performed under the same
reaction conditions using model compounds containing car-
boxyl, quinone, or hydroquinone groups. The structural envi-
ronment of these groups in the model compounds was similar
to that of these groups on activated carbon [17]. As shown in
Table 5, 0.3% phenol yield was obtained when the hydroxyla-
tion of benzene was carried out using H O as oxidant in the
activated carbon after reaction.
hand, a higher phenol yield was obtained using hydroquinone
(
0.6%) and catechol (0.7%) as homogeneous catalyst, implying
that the carbonyl and hydroxyl groups on the surface of the
activated carbon samples were the active species. The phenol
yield increased remarkably when hydrochloric acid was used to
adjust the pH of the reaction system to 0.86 with hydroquinone
2
2
absence of catalyst, and a similar phenol yield (0.2%) was
+
obtained when [H ] was adjusted to 0.14 mol/L (pH = 0.86). As
compared with the reaction without a catalyst (0.3% phenol
yield), the reaction catalyzed by benzoic acid had a similar
phenol yield (0.4%). This indicated that the carboxylic acid
groups on the activated carbon samples were not essential for
Table 5 Activity of model compounds as the catalyst in the hydroxyla-
tion of benzene
+
Sample
[H ]/(mol/L)
NA
Yield of phenol (%)
Quinone
0.3
1.3
0.7
2.8
0.6
2.7
0.4
0.4
0.3
0.2
+
the catalysis, and its function was to provide H . On the other
0
0
0
0
.14 (pH = 0.86)
NA
Catechol
Table 4 Deconvolution of O 1s XPS spectra for WAC-0.5N-650 after
.14 (pH = 0.86)
NA
+
reaction with different [H ]
Hydroquinone
Benzoic acid
Surface oxygen (%)
.14 (pH = 0.86)
NA
+
[
H ]/(mol/L)
Alcohol/ Carboxyl/
Carbonyl
Chemisorbed
ether
31.58
32.64
32.73
ester
15.59
16.44
15.74
.14 (pH = 0.86)
NA
0
0
3
.22 (pH 0.66)
.01 (pH 1.95)
31.18
29.92
28.97
21.65
21.00
22.55
Blank
0.14 (pH = 0.86)
ꢂ4
.2×10 (pH 3.49)
+
NA: [H ] of the system was not adjusted by acid or alkali.

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
3.0
2.5
2.0
1.5
1.0
0.5
0.0
70
60
50
40
30
20
-0.5
0.0
0.5
1.0
lg[H ]
1.5
2.0
0
.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
+
ꢀ
Amount of total oxygen (mmol/g)
+
Fig. 5. Effect of H on phenol yield using hydroquinone as catalyst.
Fig. 7. Dependence of amount of benzene adsorbed on the amount of
oxygen groups.
(
(
2.7% phenol yield), catechol (2.8% phenol yield), and quinone
+
1.3% phenol yield) as catalysts, which confirmed that H
2+
lic hydroxyl and quinone is similar to the cycle between Fe
facilitated the reaction. Possibly the carbonyl pairs at the edge
of the carbon layers behaved similarly to quinones, and the
hydroxyl pairs acted similarly to hydroquinone [41], and this
redox cycle was involved in the hydroxylation of benzene over
the activated carbon catalyst.
3+
and Fe .
2.6 Benzene adsorption
The amount of benzene adsorbed on different catalysts was
+
To investigate the effect of [H ] on phenol yield, hydro-
also investigated. A close relationship between the amount of
benzene adsorbed and the amount of total oxygen groups on the
activated carbon is shown in Fig. 7. The amount of benzene
adsorbed decreased with increased total oxygen groups. The
results showed that benzene adsorption was influenced by the
surface chemistry of the catalyst, which was consistent with a
previous report [42], where the amount of benzene adsorbed
decreased with decreasing pH at the point of zero charge of
activated carbon samples. It is known that the adsorption of the
substrate has a significant effect on the heterogeneously cata-
lyzed reaction. The activity of the catalyst usually decreases
with decreased substrate adsorption. Thus, increased total
oxygen would decrease the activity of catalyst by decreasing
the amount of benzene adsorbed, which was in accord with the
results obtained by Chen et al. [14].
+
quinone was used as the catalyst with various [H ]. The results
are shown in Fig. 5. The highest phenol yield was obtained at a
+
[
H ] of 0.14 mol/L (pH = 0.86).
+
To study how [H ] affected the reaction, in situ IR was used
to monitor the oxidization of hydroquinone by H O . The result
2
2
was shown in Fig. 6. When the reaction was carried out at a
+
[
H ] of 0.01 mol/L (pH = 2.0), it was difficult for hydroquinone
to be oxidized to quinone. When the reaction was carried at a
+
[
H ] of 2.75 mol/L, hydroquinone was oxidized to quinone
+
quickly. When the reaction was carried at a [H ] of 0.14 mol/L
pH = 0.86), hydroquinone was oxidized to quinone, and an
(
equilibrium was established between the hydroquinone and
quinone. Thus the equilibrium between hydroquinone and
quinone plays an important role in the hydroxylation of ben-
+
zene. The function of H in this system may be similar to its
role in the Fenton system. The transformation between pheno-
2
.7 Active site on the activated carbon catalyst
0.04
Inductive coupled plasma analysis indicated that WAC
+
+
possessed some Na , K , and other impurities. To probe if these
impurities also affected the catalytic performance, a control
experiment was carried out employing WAC-0.5N-900 as
catalyst. When 2.0 mmol NaCl was added into the reaction
system only 2.3% phenol yield was obtained. As compared to
the result in Table 3, the phenol yield in the presence of NaCl
was drastically decreased (7.6% vs 2.3%). The results sug-
gested that the presence of the salt in the reaction system af-
fected the catalytic performance of that sample. Therefore,
impurities from WAC dissolved in the system may also reduce
the hydroxylation activity. Thus, in the following discussion
about the effect of oxygen groups on the catalytic performance,
WAC is not included because of its high impurity content.
Hydroquinone
Quinone
+
[
H ] = 0.01 mol/L (pH = 2.0)
Quinone
+
[
H ] = 0.14 mol/L (pH = 0.86)
Hydroquinone
Quinone
+
[
H ] = 2.75 mol/L
Hydroquinone
0
30
60
90
120
150
Time (min)
+
Fig. 6. IR spectra of hydroquinone for various [H ].

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
Table 6 Reusability of WAC-0.5N and WAC-0.5N-8H for the hy-
droxylation of benzene
14
12
10
8
6
Yield of phenol (%)
WAC-0.5N-8H WAC-0.5N
Selectivity for phenol (%)
Run
WAC-0.5N-8H WAC-0.5N
1
2
3
4
5
13.5
12.9
12.2
11.9
11.7
12.5
12.7
10.5
10.9
10.8
86.3
83.6
86.6
86.8
87.9
86.3
85.8
89.2
88.1
88.5
0.2
0.3
0.4
0.5
0.6
0.7
0.8
and affecting the pH of the system and amount of benzene
adsorption. It was reported that hydroxyl radicals were pro-
duced during the activation of hydrogen peroxide by activated
carbon [20,21], and quinone/phenolic groups act as electron
transmitter [22,43]. Salicylic acid (SA, hydroxybenzoic acid)
was used as a molecular probe to detect the hydroxyl radical
[44]. When 22.5 mmol SA was added into the system with
WAC-0.5N, 6.2% phenol yield was obtained, which was only
half of the result shown in Table 3 (12.5%). Meanwhile,
2,3-dihydroxybenzoic acid and 2,5-dihydroxybenzoic were
produced. The result indicated that the hydroxyl radical was
involved in the hydroxylation of benzene. It can be proposed
that the equilibrium between phenolic groups and quinone on
Phenolic hydroxyl and carbonyl (mmol/g)
Fig. 8. Relationship between the yield of phenol and the sum of pheno-
lic hydroxyl and carbonyl on the activated carbon after reaction.
To probe the relationship between the chemistry of the
samples and catalytic performance, the amount of oxygen
groups on activated carbon before and after reaction were
measured. The results indicated that the oxygen groups were
related to the catalytic activity. Figure 8 showed relationship
between the sum of phenolic hydroxyl and carbonyl groups of
the activated carbon samples after reaction and the phenol
yield. The yield of phenol increased as the sum of phenolic
hydroxyl and carbonyl increased, and then became constant.
The dependence of phenol yield on the amount of other surface
oxygen groups of the activated carbon before and after reaction
was also studied, but the correlation was not as good. It was
deduced that the phenolic hydroxyl and carbonyl groups on the
activated carbon samples were the active sites for the hy-
droxylation of benzene. In the literature [19], these phenolic
hydroxyl and carbonyl groups were shown to be active for
related oxidation reactions.
the activated carbon activated H
acted with benzene. This is illustrated in Fig. 9. The
o-hydroquinone species was initially oxidized by H to
o-semiquinone and subsequently quinone (o-quinone) while
got protons and electrons and formed ·OH. Then, ·OH
2 2
O to form ·OH, which re-
2 2
O
H O
2 2
reacted with benzene producing the hydroxybenzene radical.
Next, quinone was reduced by the hydroxybenzene radical to
o-hydroquinone to produce phenol.
On the basis of the above results, we can propose how the
oxygen groups affected the catalytic performance. As dis-
cussed above, the oxygen groups of the activated carbon af-
fected the hydroxylation of benzene by providing active sites,
2.8 Stability of the activated carbon catalyst
To examine the recyclability of the activated carbon cata-
lysts in the hydroxylation of benzene, WAC-0.5N and
WAC-0.5N-8H were used for five cycles. As shown in Table 6,
the activity was kept almost constant after the first two runs.
The results indicated that the catalyst was stable for the reac-
tion.
OH
H
OH
3
Conclusions
O
O
O
O
HO
OH
ꢀ
ꢀ
ꢀ
ꢀ
Modified activated carbon showed good catalytic activity
+
e
+ e
ꢀ
ꢀ
for the hydroxylation of benzene. Phenolic hydroxyl groups
and quinone on the activated carbon were the active sites. The
reaction between phenolic hydroxyl groups and quinone acti-
vated H O to form ·OH, which reacted with benzene to form
e
e
o-Quinone
o-Semiquinone
o-Hydroquinone
2
2
phenol. The pH of the system is an important factor. Oxygen
groups on the activated carbon have the roles in the hydroxy-
lation of benzene that they provide active sites for the reaction,
and affect the pH of the system and the amount of benzene
adsorption.
OH
H O
2
2
Fig. 9. Reaction pathway for the hydroxylation of benzene over acti-
vated carbon.

XU Jiaquan et al. / Chinese Journal of Catalysis, 2012, 33: 1622–1630
23 Boehm H P. Carbon, 1994, 32: 759
References
2
2
2
2
2
2
3
4
5
6
7
8
9
0
Takaoka M, Yokokawa H, Takeda N. Appl Catal B, 2007, 74:
79
1
1
2
3
4
Niwa S I, Eswaramoorthy M, Nair J, Raj A, Itoh N, Shoji H,
Guo B, Zhu L F, Hu X K, Zhang Q, Tong D M, Li G Y, Hu C
W. Catal Sci Technol, 2011, 1: 1060
Namba T, Mizukami F. Science, 2002, 295: 105
Zhang F M, Chen X, Zhuang J, Xiao Q, Zhong Y J, Zhu W D.
Catal Sci Technol, 2011, 1: 1250
Boehm H P. In: Eley D D, Herman P, Weisz P B eds. Advances
in Catalysis. New York: Academic Press, 1966. 179
Goertzen S L, Thériault K D, Oickle A M, Tarasuk A C, Andreas
H A. Carbon, 2010, 48: 1252
Yuranov I, Bulushev D A, Renken A, Kiwi-Minsker L. Appl
Catal A, 2007, 319: 128
Tani M, Sakamoto T, Mita S, Sakaguchi S, Ishii Y. Angew
Chem, Int Ed, 2005, 44: 2586
Oickle A M, Goertzen S L, Hopper K R, Abdalla Y O, Andreas
H A. Carbon, 2010, 48: 3313
5
6
7
Dittmeyer R, Bortolotto L. Appl Catal A, 2011, 391: 311
Chen J Q, Gao S, Li J, Lv Y. Chin J Catal, 2011, 32: 1446
Chammingkwan P, Hoelderich W F, Mongkhonsi T, Kanchana-
wanichakul P. Appl Catal A, 2009, 352: 1
Boehm H P, Diehl E, Heck W. Proceedings of the 2nd London
Carbon and Graphite Conference. London, UK, 1965. 369
Brunauer S, Deming L S, Deming W E, Teller E. J Am Chem
Soc, 1940, 62: 1723
8
9
Abbo H S, Titinchi S J J. Appl Catal A, 2009, 356: 167
Zhang J, Tang Y, Li G Y, Hu C W. Appl Catal A, 2005, 278:
3
1
2
Zhu S, He X Y, Zhao L, Li H P. Chemistry, 2010, 3: 284
3
Moreno-Castilla C, López-Ramón M V, Carrasco-MarÕ
Carbon, 2000, 38: 1995
Zerda T W, Yuan X, Moore S M, Leon y Leon C A. Carbon,
999, 37: 1999
ғ
n F.
2
51
1
1
1
0
1
2
Song S Q, Yang H X, Rao R C, Liu H D, Zhang A M. Appl
Catal A, 2010, 375: 265
3
3
3
3
4
5
1
Song S Q, Jiang S J, Rao R C, Yang H X, Zhang A M. Appl
Catal A, 2011, 401: 215
Figueiredo J L, Pereira M F R, Freitas M M A, Órfão J J M.
Carbon, 1999, 37: 1379
Choi J S, Kim T H, Choo K Y, Sung J S, Saidutta M B, Ryu S O,
Song S D, Ramachandra B, Rhee Y W. Appl Catal A, 2005, 290:
1
Jia J G, Li Ch, Zhu Ch L, Xiao Ch Y, Xu D, Yan X, Liang Ch H.
Carbon Techniques, 2009, 28(6): 11
3
6
7
Zielke U, Hüttinger K J, Hoffman W P. Carbon, 1996, 34: 983
Moreno-Castilla C, Carrasco-Marín F, Mueden A. Carbon, 1997,
1
3
4
Zhong Y K, Li G Y, Zhu L F, Yan Y, Wu G, Hu C W. J Mol
Catal A, 2007, 272: 169
3
3
5: 1619
1
Chen C H, Xu J Q, Jin M M, Li G Y, Hu C W. Chin J Chem
Phys, 2011, 24: 358
3
3
8
9
Larsen E C, Walton J H. J Phys Chem, 1940, 44: 70
León y León C A, Radovic L R. In: Thrower P A ed. Chemistry
and Physics of Carbon. New York: Marcel Dekker, 1994. 213
Biniak S, SzymaĔski G, Siedlewski J, Swiatkowski A. Carbon,
1
5
6
Rodríguez-Reinoso F. Carbon, 1998, 36: 159
Chen Ch, Lin X Y, Yu Zh X, Chen X H, Zheng Q, Wei K M.
Petrochem Technol, 2004, 33: 1024
1
40
41
42
43
44
1
997, 35: 1799
1
7
Song W, Li Y, Guo X H, Li J, Huang X M, Shen W J. J Mol
Catal A, 2010, 328: 53
Serp P, Figueiredo J L. Carbon Materials for Catalysis. New
Jersey: Wiley, 2009. 177
1
8
9
Yu G X, Lu S X, Chen H, Zhu Z N. Carbon, 2005, 43: 2285
Pereira M F R, Órfão J J M, Figueiredo J L. Appl Catal A, 1999,
Wibowo N, Setyadhi L, Wibowo D, Setiawan J, Ismadji S. J
Hazard Mater, 2007, 146: 237
1
1
84: 153
Van der Zee F P, Bisschops I A E, Lettinga G, Field J A.
Environ Sci Technol, 2003, 37: 402
2
0
1
Georgi A, Kopinke F D. Appl Catal B, 2005, 58: 9
2
Ono Y, Matsumura T, Kitajima N, Fukuzumi S I. J Phys Chem,
Diez L, Livertoux M H, Stark A A, Wellman-Rousseau M,
Leroy P. J Chromatogr B, 2001, 763: 185
1977, 81: 1307
22
Donnet J B. Carbon, 1968, 6: 161