Carbon Catalyzed Hydroxylation of Benzene with Dioxygen to Phenol over Surface Carbonyl Groups

Article abstract of DOI:10.1002/cctc.201801668

Carbon materials are promising environmentally-benign heterogeneous catalysts but design of effective carbon catalysts with comparable or even superior performance to metal-based ones is highly challengeable. Herein, N-doped mesoporous carbons with the su

Full text of DOI:10.1002/cctc.201801668

Accepted Article
Title: Carbon catalyzed hydroxylation of benzene with dioxygen to
phenol over surface carbonyl groups
Authors: Wanjian Shan, Shuai Li, Xiaochun Cai, Jie Zhu, Yu Zhou, and
Jun Wang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201801668
Link to VoR: http://dx.doi.org/10.1002/cctc.201801668
A Journal of
www.chemcatchem.org

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
Carbon catalyzed hydroxylation of benzene with dioxygen to
phenol over surface carbonyl groups
Wanjian Shan, Shuai Li, Xiaochun Cai, Jie Zhu, Yu Zhou*, Jun Wang*[a]
Abstract: Carbon materials are promising environmentally-benign
heterogeneous catalysts but design of effective carbon catalysts with
comparable or even superior performance to metal-based ones is
highly challengeable. Herein, N-doped mesoporous carbons with the
surface enriched with abundant oxygen species were synthesized by
a facile hydrothermal doping strategy in the self-assembly of phenolic
resin, followed with a carbonization process. Direct hydroxylation of
benzene to phenol with O2 as the sole oxidant was effectively
catalyzed by these carbons, affording a yield superior or comparable
to previous efficient transition metal and even noble metal-based
catalysts. The catalyst is facilely recovered and stably reused. The
surface carbonyl groups are the active sites for O2 activation (rate
determining step) to generate hydroxyl radicals, which are the reactive
oxygen species to oxidize the benzene. This methodology is readily
extendable to the oxidation of various other benzene derivatives.
still remain the subject of intense debate due to the related to
polymers, pharmaceuticals and agrochemicals. In addition, this
process has the drawbacks of low phenol yield (~5%) and
coproduction of acetone.[3, 6] In a bid to surmount these problems,
direct hydroxylation of benzene with an economically and
environmentally benign oxidant such as hydrogen peroxide
(H2O2) and molecular oxygen (O2) is among the most important
alternatives and has been extensively studied.[6a-c] Both metal-
bearing and metal-free catalysts were developed for the benzene
hydroxylation with H2O2, and the former normally exhibited better
6d-f]
performance.[3,
For example, commercial wood-based
activated carbon oxidised by nitric acid was made as an efficient
and stable carbocatalyst, giving a phenol yield of 14.4%.[7] The
yield was 18% using chemically converted graphene from
graphite in H2O2 based benzene hydroxylation.[8] By contrast,
many metal-based catalysts usually exhibited higher phenol yield
(>30%) in the benzene hydroxylation with H2O2. Though O2 as
oxidant has high atomic utilization and economic superiority,
oxidation of benzene to phenol is highly challengeable because
of the inertness of benzene and O2.[6g, 9] This oxidation usually
relied on the employment of either a sacrificial reducing agent
coupled with a transition metal-based catalyst or noble metals to
generate reactive oxygen species. Significantly, rapid decline of
the activity was observable in the reusability tests of these metal
catalysts due to the metal leaching.[9a, 10] Only several V or Fe
based catalysts were prepared for the oxidation of benzene with
O2 to phenol in the absence of wasteful reductant, but still suffered
from apparent deactivation during the recycling.[9c, 10] To the best
of our knowledge, none of the efficient metal-free catalysts has
been developed for the O2-mediated benzene hydroxylation so far.
Great efforts have been made to unravel the active sites of
carbon catalyzed oxidations with O2.[11] It is reported that
boundary defects (edges and vacancies), heteroatom doping and
oxygen-containing functional groups of carbon materials are the
possible active sites for O2 involved oxidation reactions such as
alcohol oxidation,[11b] ethylbenzene oxidation,[5a] cyclohexane
oxidation,[12] epoxidation of styrene,[2g] and alkane oxidative
dehydrogenation (ODH).[2c, 11a] For example, the active sites for
the oxidative coupling of amines to imines are the boundary
defects with unpaired electrons and carboxylic acid groups. The
former activates O2 to generate superoxide radicals, and the latter
traps the reactants aided by the localized unpaired electrons.[13]
The ketonic groups of carbon nanotubes are regarded as critical
ingredient of the active sites for ODH reactions.[1c] Various
heteroatoms such as the graphitic sp2 N species with an unpaired
electron that delocalizes over the basal plane of graphene act as
the catalytically active centers in the oxidation of alcohols via
reducing O2 to surface oxygen species.[14] The surface oxygen
groups like P=O and P−OH functionalities of P-doped carbon
Introduction
Carbon catalysts have attracted growing attentions for the
environmentally friendly and sustainable production of chemicals,
due to their low cost, non-toxicity, natural abundance, easy
functionalization, and tunable porosity and chemical structure.[1]
Many approaches have been proposed, affording numerous
carbon catalysts (e.g. fullerene, carbon nanotubes, graphene,
nanodiamond and porous carbons) with potential applications in
acid/base, oxidation, dehydrogenation, hydrogenation, and
photo/electro-catalysis.[2] In general, most of carbon catalysts
exhibited inferior performance to homogeneous and metal-
complex surface structures of carbons.[1b, 3] Hence, carbocatalysis
with high and stable performance is full of challenges.
Catalytic oxidation reactions have been widely applied in the
synthesis of petrochemicals and fine chemicals, but many
industrial processes are energy consuming and environmental
unfavourable.[4] This is exemplified by the three-step cumene
route to produce phenol, an important organic intermediate based
heterogeneous ones, particularly in many oxidations.[4-5] To
develop active carbon catalysts, it is imperative to identify the
active sites and gain insight into the mechanism, but many details
[a]
W. Shan, S. Li, X. Cai, J. Zhu, Dr.Y. Zhou, Prof. Dr. J. Wang
State Key Laboratory of Materials-Oriented Chemical Engineering
College of Chemical Engineering
Nanjing Tech University
Nanjing, Jiangsu 210009 (P.R. China)
E-mail: junwang@njtech.edu.cn (J. Wang),
njutzhouyu@njtech.edu.cn (Y. Zhou)
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
materials are involved in the oxidation of benzyl alcohol, while O2
regenerated the active sites.[15] Despite of these early efforts, rare
attention is paid on the direct O2 activation over the surface
oxygen groups of carbons to generate reactive oxygen species
such as hydroxyl radicals, which are the widely observable
oxygen transfer agent in the benzene hydroxylation.
a
c
b
In this work, we report the synthesis of N-doped and surface O-
enriched mesoporous carbons (NOCs) that behave as efficient
heterogeneous metal-free catalysts for the hydroxylation of
benzene with O2 to phenol. Carbon precursors are hydrothermally
synthesized via an acid-assisted organic-organic self-assembly
approach by using triblock copolymer Pluronic F127 as the soft-
template and melamine as nitrogen resource. Varying the N
content in the precursor and carbonization temperature affords
carbons with modulated surface oxygen species. These carbon
catalysts are highly effective in the oxidation of benzene to phenol
with O2. The catalyst is easily recovered and stably reused, and
such pleasurable reusability has not been observed before.
Systematically studies are carried out to investigate the catalytic
process by small molecular model catalysts, radical quenching
and trapping experiments, 18O-isotope tracer experiment,
measurement of kinetic isotope effect, and theoretical
calculations. On the basis of these, we propose the most probable
active sites and the reaction route. The scope is explored and
various benzene derivatives are oxidized by O2 to the
corresponding O-containing chemicals with good yields using this
carbon catalyst.
5 μm
5 μm
e
d
f
200 nm
C
O
g
200 nm
N
Figure 1.SEM images of (a) NOC-0 and (b) NOC-0.15. (c) TEM image and (d)
HAADF-STEM images of NOC-0.15 and the corresponding EDX mapping
images for (e) C, (f) N and (g) O elements.
the resultant carbons are composed of spheres with the size of
several micrometers. The spherical particles of NOC-0 had
smooth surface, whereas distorted spheres were observed over
the samples with elevated N content (Figures 1 a, b and Figures
S1a-d). The spherical morphology of typical sample NOC-0.15
was also reflected by TEM images (Figure 1 c). HAADF-STEM
image and EDS elemental imaging of NOC-0.15 (Figure 1 d)
showed the homogeneous distribution of C, O and N.
Results and Discussion
Quantitative analyses of the porosities were obtained by
nitrogen sorption at 77 K. The isotherms of NOC-m (Figure S2)
are type I+IV, indicating the existence of micro and
mesopores.[10a] The relatively sharp and high uptakes at low
pressure range is corresponding to the abundant micropores. The
textural properties are listed in Table 1. All these carbons present
high surface areas (591~708 m2 g-1) and pore volumes (0.28~0.38
cm3 g-1). In the wide angle XRD patterns (Figure S3), each NOC-
m exhibited a broad reflection peak at 2θ≈23° corresponding to
the interlayer distance of 0.384 nm, index of (002) reflection for
pure graphitic lattice. These samples also showed the broad
Characterization of carbons
Scheme
1 displays the hydrothermal synthesis of carbon
precursors, the melamine modified phenolic resins. Carbonization
of these precursors at 650 °C provided a series of NOCs. They
are termed NOC-m (m=0, 0.05, 0.10, 0.15 and 0.20), in which m
denotes the melamine mass in the initial synthetic gel of resin
precursor (weight ratio: F127/H2O/ethanol/resorcinol/melamine/
HCl/formaldehyde=4/20/20/4.4/m/1/4). The sample NOC-0 is the
N-free counterpart obtained through carbonizing the melamine-
free resin. The chemical compositions of NOC-m were measured
by CHN elemental analysis (Table S1). N content is 1.31 wt% for
m=0.05, and continuously increases along with the melamine
mass (m), reaching a maximum value of 10.4 wt% at m=0.20. This
result indicates that melamine acts as an efficient nitrogen source
to modulate the N content in carbons. SEM images revealed that
reflection at 43.1° for the (100) crystal face of graphitic carbon[16]
,
suggesting the graphitization. 13C NMR spectrum of NOC-0.15 is
close to that of NOC-0 (Figure S4), implying that they have almost
same carbon skeleton. Sharp and intensive derivative peak at
100~160 ppm for sp2 carbon atoms in C=C double bonds
indicates the formation of mainly aromatic carbons.[17] TG curves
(Figure S5) showed a slight weight loss below 100 °C attributable
to the water desorption. Apparent weight loss happened above
500 °C, indicating high thermal stability.[18] Two signals at 1352
cm−1 (D-band) and 1546 cm−1 (G-band) were observed in the
Raman spectra of NOC-m (Figure S6). The presence of D-band
corresponds to the disordered graphite structure. The intensity
ratio (ID/IG) of D and G band indicates the density of defects. The
ID/IG values of NOC-m are in the range of 0.83~0.88, suggesting
that their defects density is close to each other.[19] ESR spectra
(Figure S7) showed a broad peak attributable to the unpaired
spins at the edges of graphitic carbon network.[2b, 13]
Scheme 1. Schematic illustration for the preparation of N-doped and oxygen-
enriched carbons. MRF resin: melamine-resorcinol-formaldehyde resin.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
Table 1. Textural properties
Table 2. Content of oxygen groups in typical NOC-m samples
Content (mmol g-1)
[a]
(m2g-1)
708
[b]
[c]
SBET
Vp
Dav
N content
(wt%)[e]
[d]
Entry
Sample
ID/IG
(cm3g-1)
0.48
0.30
0.31
0.29
0.28
0.32
0.35
0.36
0.43
0.57
0.67
0.60
0.30
0.32
(nm)
2.71
1.85
1.98
1.90
1.86
2.12
1.95
1.78
1.63
1.68
2.40
2.46
2.21
1.86
Carboxyl
Lactone
Phenol
Carbonyl
1
NOC-0
0.83
0.84
0.88
0.86
0.85
0.82
0.84
0.87
0.98
1.12
----
----
NOC-0
NOC-0.05
NOC-0.15
0.150
0.150
0.200
0.125
0.125
0.175
0.125
0.200
0.075
0.975
2.175
6.500
2
NOC-0.05
NOC-0.1
NOC-0.15
NOC-0.2
NOC-500
NOC-600
NOC-700
NOC-800
NOC-900
AC
653
1.31
3.55
4.82
10.35
5.89
4.71
3.32
2.97
2.25
----
3
626
4
610
5
603
6
599
b
a
Total oxygen content
NOC-0
C-O-C
ketonic C=O
7
665
30
Surface oxygen content
25.1
8
799
25
20
15
10
5
C-OH
9
1195
1354
1110
974
16.5
10
11
12
13[f]
14[g]
NOC-0.05
15.1
ketonic C=O
13.5
C-O-C
11.4
10.2
C-OH
9
8.6
7.3
6.9
NAC-900
NOC-0.15
NOC-0.15
----
3.82
4.26
4.08
580
0.85
----
NOC-0.1
C-O-C
0
ketonic C=O
625
0
-
1
.
5
5
1
2
.
0
-
0
0
-
.
.
0
quinonic C=O
0
OC
-
-
N
OC
OC
OC
N
OC
N
C-OH
N
N
c
[a] BET surface area. [b] Total pore volume. [c] Average pore size. [d] Detected
by Raman spectroscopy. [e] Total N content determined by CHN elemental
analysis. [f] Recovered NOC-0.15 after the 6th run. [g] Recovered NOC-0.15
under anaerobic conditions.
Total nitrogen content
Surface nitrogen content
ketonic C=O
12
NOC-0.15
quinonic C=O
C-O-C
9.5
9
6
3
0
ketonic C=O
NOC-0.2
C-O-C
C-OH
The surface chemical composition was measured by XPS
analyses. In the full scan spectra, NOC-0.15 demonstrated the
largest peak intensity of O 1s signal (Figure S8), implying that it
possesses the highest surface oxygen content of 25.1 at%. The
O 1s spectra of NOC-0.1, NOC-0.15 and NOC-0.2 are fitted with
four peaks at 531.1, 532.3, 533.3 and 534.2 eV for quinone C=O
(O1), ketonic C=O (O2), ether-type (C–O–C) (O3), and hydroxyl
(C–O–H) (O4), respectively (Figure 2a).[20] The O 1s spectra of
NOC-0 and NOC-0.05 are divided into three peaks at 532.3, 533.3
and 534.2 eV for ketonic C=O (O2), C–O–C (O3) and C–O–H
(O4), respectively. The dispersions of these O species are
summarized in Table S2, showing that the C=O groups are the
major O species for these carbons. N-doping causes the
formation of quinone C=O and increasing the N content promotes
the formation of more surface O species. The highest content of
C=O groups (14.6 at%) is found for NOC-0.15 with moderate N
content. The C 1s spectra (Figure S9) are resolved into four
individual component peaks at 284.7, 285.5, 286.5 and 289.3 eV
corresponding to the C–C, C–OH, C=O and O=C-O bonds,
respectively. These signals additionally reveal the existence of
surface C=O groups. The N 1s XPS signal is weak for these
carbons (Figure S8), implying that their N species exist rarely on
the surface. Additionally, Boem titration of typical samples
including NOC-0, NOC-0.05 and NOC-0.15 was conducted to
measure the content of O-containing groups (Table 2). The
content of carboxyl, lactone and phenol groups for these samples
were low and close to each other. The content of carbonyl groups
increased rapidly with increasing N content, suggesting that these
carbons have abundant carbonyl groups. This phenomenon
observed in Boem titration is in line with the XPS analysis.[7]
Figure 2b correlates the elemental composition from CHN
elemental analyses and XPS spectra. The total oxygen content
was tentatively evaluated by CHN elemental analysis, according
to the following formula: O%=1-C%-H%-N% by assuming that
there were only C, H, O and N element in the NOCs. From this
preliminarily measurement, it was found that at the low and
moderate N contents (m=0.05, 0.1 and 0.15), and the total O
contents of N-doped carbons are close to that of N-free NOC-0.
4.4
quinonic C=O
3.2
1.5
1.2
0.8
0.7
0.4
538
536
534
532
530
528
1
5
1
.
0
-
2
5
.
.
0
0
-
.
0
-
0
-
OC
N
OC
OC
OC
N
Binding Energy (eV)
N
N
Figure 2. (a) O 1s XPS spectra of NOC-m series. (b) Total and surface oxygen
content of NOC-m series. (c) Total and surface nitrogen content of NOC-m
series. The total nitrogen and oxygen content were detected by CHN elemental
analysis; the surface nitrogen and oxygen content were measured by XPS
spectra.
NOC-0.2 with the high N content affords slightly higher total O
content of 13.5 at%. Surface O content of each NOC-m is higher
than the total O content, visualizing their surface is enriched with
O species. Notably, NOC-0.15 possesses the highest ratio of
surface to total O species. Figure 2c compares the total and
a
b
12.7
12.5
12.1
11.9
15
11.6
11.5
12
12.5
10.8
9
6
3
0
10
5
9.1
6.8
5.1
1.2
0.8
0
0
1
2
15
0.
05
1
2
3
4
5
6
0.
00
0.
AC
0.
NOC-
C-9
NA
Recycling run
NOC-
NOC-
NOC-
NOC-
d
c
NOC-0.15
15
NOC-0.1
NOC-500
NOC-700
NOC-0.2
10
NOC-500
NOC-700
NOC-0.1
NOC-0.15
10
5
NOC-0.2
NOC-0.05
NOC-0.05
NOC-0
NOC-900
NOC-0
5
NOC-900
AC
AC
NAC-900
NAC-900
0
0
5
10
15
5
10
15
20
25
Surface carbonyl and hydroxyl content (at%)
Total surface oxygen content (at%)
Figure 3. (a) Phenol yield of different carbons in the oxidation of benzene to
phenol with O2. Reaction conditions: benzene (45 mmol, 4 mL), catalyst (0.2 g),
LiOAc (0.6 g), aqueous solution of acetic acid (60 vol. %, 25 mL), 2.2 MPa O2,
150 °C. (b) Reusability of NOC-0.15. Reaction conditions: benzene (45 mmol,
4 mL), catalyst (0.2 g), LiOAc (0.6 g), aqueous solution of acetic acid (60 vol. %,
25 mL), 2.2 MPa O2, 150 °C, 48 h. Plot of phenol yield vs. (c) the amount of the
total surface oxygen content (O1+O2+O3+O4) and the amount of the surface
carbonyl and hydroxyl content (O1+O2+O4) of carbons.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
surface N content. Their total N content gradually increases along
with melamine amount in the synthetic gel of the resin precursor.
By contrast, their surface N content remains in a much low level
even in the case of high total N content. These phenomena
indicate that the present synthetic strategy provides a series of
surface O-enriched carbons and N-doping by using melamine is
an effective method to manage the surface O content.
Carbonization temperature is an important parameter for the
formation of carbons (Figures S10–S11). By fixing m=0.15 (same
to that of NOC-0.15 obtained at 650 °C), varying the temperature
offered carbons NOC-T (T=500, 600, 700, 800 and 900), in which
T denotes the carbonization temperature. Along with the increase
of the temperature, obtained carbons exhibited higher surface
area, larger pore volume, enhanced ID/IG (characteristic of more
defects), and decreased total O content. XPS spectra of selected
samples NOC-500, 700 and 900 were collected and compared
with that of NOC-0.15 (Figures S12–S13). All these three samples
have rarely N species on the surface, same as NOC-m series. For
NOC-500, the total O content of (26.4 at%) is higher than the
surface O content (18.6 at%) (Table S1 and Table S3), indicating
that low temperature is not suitable for the formation of surface O-
enriched carbons. Nonetheless, high carbonization temperatures
(e.g. 700 and 900 °C) resulted in low total O content and are also
unfavorable towards carbons with enriched O species on the
surface. Briefly, the carbonization temperature and melamine
content in the initial gel dominate the surface O content in the
present route.
Figure 4. Hydroxylation of benzene to phenol catalyzed by small molecular
model catalysts. Reaction conditions: benzene (45 mmol, 4 mL), catalyst (5.7
mmol), LiOAc (0.6 g), aqueous solution of acetic acid (60 vol. %, 25 mL), 2.2
MPa O2, 150 °C. The top ones of the figure mimic the carbonyl group; the left
ones of the figure mimic lactone and carboxyl groups; the right ones of the figure
mimic the ether group; the bottom ones of the figure mimic the hydroxyl group.
(Figure 3b). Such exceeding reusability is attributable to the well-
preserved structure during the catalysis process (Figures S15-
S20; Table S1 and S3). By contrast, significant decline of the
activity was observed in previous metal-catalyzed benzene
hydroxylation with O2. To the best of our knowledge, it is for the
first time to reach an efficient carbon catalyzed hydroxylation of
benzene to phenol with O2. The phenol yield of the present system
is superior to majority of the transition metal or even noble metal
catalysts and comparable to the most efficient ones (Table S5).
Previously, many carbon-catalysed hydroxylation of benzene with
H2O2 afforded the yields above 10%.[7-8, 22] Compared with these
systems, NOC-0.15 exhibited comparable phenol yield using O2
as a facile available and atom-efficient oxidant. Notably, this
carbon catalyst avoids the metal leaching derived deactivation
during the recycling process and affords stable reusability, which
has not been achieved before. In short, the present
heterogeneous catalyst shows remarkable efficiency and stability.
Hydroxylation of benzene to phenol with oxygen
The aforementioned NOCs are employed as the heterogeneous
catalysts in the hydroxylation of benzene to phenol by using O2
as the sole oxygen source. Figure 3a displays the phenol yield
over different catalysts. The phenol yield was 5.1% over NOC-0,
whereas NOC-m (m=0.05-0.20) gave improved activity. The
highest yield of 12.5% was found over NOC-0.15. This activity
greatly outperforms those of activated carbon (AC, yield: 1.2%)
and N-doped activated carbon (NAC-900, yield: 0.8%), and is also
superior to those of carbons obtained under other carbonization
temperatures (NOC-T, T=500, 600, 700, 800 and 900; yields:
4.8~11.9%, Figure S14). Various influencing factors including the
solvent, amount of additive LiOAc, benzene concentration,
reaction time and O2 pressure were studied. The activities under
different reaction conditions are summarized in Table S4. The
yield of phenol was 1.5% by using water as the solvent and
apparently improved by using aqueous acetic acid solution. This
phenomenon demonstrates that the solvent plays a vital role in
the reaction and acidic environment promotes the phenol
formation. The investigation of the solvent composition indicates
that a suitable aqueous acetic acid solution benefits high activity.
The yield of phenol was 2.5% in the absence of LiOAc, suggesting
that LiOAc is not indispensable to trigger this reaction and acts as
an efficient additive. The concentration of LiOAc significantly
affects the activity by acting as the buffer agent to provide and
maintaining a suitable acidic environment.[21] A rough survey of
other reaction conditions suggests that the benzene
concentration, reaction time and O2 pressure are also important
for the formation of phenol.
Active sites and reactive oxygen species
Various factors affect the catalytic performance of carbons, such
as the surface area, chemical composition (e.g. N, P or B
substitution), defects (e.g. zig-zag and armchair edges), and
surface groups (e.g. oxygen groups).[2b, 3, 12, 15] Correlation of the
structural information and the activities of the carbons above
suggest that the surface area, N content and defects are not the
determining parameters for the activity (Table 1 and Figure S21).
For example, NOC-900 with the largest surface area of 1354 m2
g-1 and highest defects density (ID/IG=1.12) results in a low yield
of 4.8%; NOC-0.2 with the most abundant N content (10.35 wt. %)
gave moderate yield of 9.1%. Generally, the reaction happens on
the surface of a solid catalyst in a heterogeneous catalysis. XPS
analyses indicate that the surface of these carbons is enriched
with oxygen groups but lack of N species. The plot of the phenol
yield vs. the total surface oxygen content (Figure 3c) gives a
positive relationship between these two parameters, implying that
the surface O species are responsive to the activity.
After reaction, the catalyst was facilely recovered and reused.
Stable activity of NOC-0.15 was found in a six-run recycling test
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
Aerobic oxidations usually involve the formation of reactive
oxygen species as the oxygen transfer agent to give the oxidative
product. For instance, species such as superoxide (·O2-),
peroxide (·OOH) and hydroxyl radicals (·OH) have been found in
many previous metal-catalyzed benzene hydroxylation
systems.[24] Radical quenching and trapping experiments were
conducted to measure the exact active species in our
hydroxylation system.[25] The typical carbon catalyst NOC-0.15
and a small molecular model catalyst cyclohexanone were
employed in these tests. NOC-0.15 catalyzed benzene
hydroxylation gave the yields close to each other in the absence
and presence of butylated hydroxytoluene (BHT, scavenger for
·O 22-, BHT/benzene molar ratio=1:1), ruling out the possibility of
peroxide species as the reactive oxygen species (Table S4).
Significant deactivation was observed by adding additional tert-
butyl alcohol (TBA, scavenger for ·OH, TBA/benzene molar
ratio=1:1) or DMSO (DMSO, scavenger for ·OH, DMSO/benzene
molar ratio=1:1) into NOC-0.15 catalyzed hydroxylation system
(Figure 5a and Table S4). Similar decline of the phenol yield
happened by using the model catalyst cyclohexanone in the
presence of TBA (Figure 5b). These phenomena suggest that
hydroxyl radicals are involved in the oxygen transformation, which
is additionally confirmed by the result that higher concentration of
TBA resulted in more significant deactivation (Figure 5c). The
kinetic curve of cyclohexanone in the presence of relatively low
TBA dosage (TBA/benzene molar ratio=1:2) was compared with
that of a scavenger-free system (Figure 5d). Similar yield was
achieved in the TBA-involved system by elongating the reaction
time. The reason can be assigned to that small amount of
scavenger was consumed along with increasing reaction time and
the additional formed hydroxyl radicals trigger the oxygen
transformation. Excessive scavenger (TBA/benzene molar
ratio=1:1 or higher) dramatically inhibited the generation of
hydroxyl radicals and caused low activity (Figures5a-c).
Additional evidence for the formation of hydroxyl radicals was
provided by phenyl-α-tert-butyl nitrone (PBN) trapping experiment
(Figure 5e). The signals derived from the hyperfine structure of
corresponding PBN–(·OH) adduct radical were observed in the
EPR spectra of NOC-0.15+PBN and cyclohexanone+PBN. By
contrast, inactive acetone gave a silent signal in this test. The
combination of the results above shows that the hydroxyl radicals
are the reactive oxygen agents.
Though many researches indicated that hydroxyl radicals can
directly oxidize the benzene to phenol,[26] there are still various
other possible reaction routes for the carbon catalyzed
oxidations.[1b, 11b] For example, some reactions happened via the
interaction of the surface oxygen groups with the substrate to give
the product, and then these oxygen groups are regenerated by
O2.[2c, 11a] In many cases, water was involved in the reaction and
the O in water provided (or partly) the terminal O resources.[27] To
exclude these possibilities, we performed the anaerobic oxidation
and 18O-isotope tracer experiment, respectively. NOC-0.15
catalyzed reaction under N2 atmosphere led to only trace amount
of phenol (yield<0.2%) even with amplified catalyst dosage
(Figure S23). The content of surface O species, especially the
active sites O1+O2+O4, of the spent catalyst in this N2-mediated
reaction was close to that of the fresh one (Table S3). These
exclude the probable formation of hydroxyl radicals from the
surface oxygen species, in line with the preservation of the O
a
b
none
none
12
16
tert-butyl alcohol (45 mmol)
tert-butyl alcohol (45 mmol)
12
8
8
4
0
4
0
0
10
20
30
40
50
0
5
10
15
20
Time (h)
Time (h)
c
12
d
none
tert-butyl alcohol (22.5 mmol)
12
8
4
0
8
4
Molar ratio of tert-butyl alcoh1o.l5to benzene
0.0
0.5
1.0
2.0
4
8
12
16
20
Time (h)
e
f
16
cyclohexanone
12
8
NOC-0.15
acetone
4
0
3450
3460
3470
3480
3490
3500
3510
0
5
10
15
20
Magnetic field (G)
Reaction time (h)
Figure 5.(a-d) Influence of radical quencher tert-butyl alcohol on the production
of phenol. Reaction conditions: benzene (45 mmol, 4 mL), catalyst (0.2 g of
NOC-0.15 for Figure 5a; 0.56 g of cyclohexanone for Figure 5b-d), LiOAc (0.6
g), aqueous solution of acetic acid (60 vol. %, 25 mL), 2.2 MPa O2, 150 °C; the
reaction time was 20 h for Figure 5c. (e) EPR spectra of PBN+cyclohexanone,
PBN+NOC-0.15 and PBN+acetone. (f) Primary isotope effect in the
cyclohexanone catalyzed benzene hydroxylation under the reaction conditions
similar to that of Figure 5d.
To identify the role of different O species, we performed the
benzene hydroxylation catalyzed by homogenous organic
molecules with hydroxyl, carbonyl, lactone, ether, or carboxyl
groups, which mimic different surface oxygen species of carbons
(Figure 4). No product was detected by using benzyl ether,
phthalide and terephthalic acid, excluding the possible activity
from
ether,
lactone
and
carboxylic
acid
anthrone,
groups.
and
Phenanthraquinone,
1,4-benzoquinone,
cyclohexanone offered yields of 13.6~16.4%, whereas acetone
was inactive. This indicates that possible active sites are quinone
and cyclic ketonic C=O groups rather than linear chain aliphatic
ketone (Csp3=O). Yields were 12.5~14.4% when the hydroxylation
was catalyzed by hydroquinone, resorcinol and phloroglucinol
phenol. The phenolic hydroxyl groups can be easily oxidized by
O2 to form quinone C=O and there exists quinone/hydroquinone
tautomerism in carbons, enabling these hydroxyl groups are also
the active sites.[23] The evaluation of these small molecular model
catalysts indicates that the activity of carbons comes from O1, O2
and O4 species corresponding to quinone C=O, cyclic ketonic
C=O and Csp2–OH, respectively. This is visualized by the plots of
the phenol yield over all the carbons presented vs. the content of
O1+O2+O4 species (Figure 3d). By contrast, no such relationship
is observed in the plots of phenol yield vs. the content of each O
species (Figure S22). The percentages of (O1+O2)/(O1+O2+O4)
are 0.56~1.0 for these carbons, suggesting that the carbonyl
groups are the major active sites.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
Scheme 2. (a-c) DFT models of possible adsorption structures of O2 on the carbonyl groups of small molecular model catalysts. (d) Proposed reaction
pathway for carbon-catalyzed oxidation of benzene to phenol.
content of the spent catalyst during the recycling test. Also, phenol
production via the reaction of the surface O species with benzene
did not happen in this system. No Ph18OH was detected by using
H218O (Figure S24). Thus the O atom in the product phenol
directly comes from O2 rather than H2O and surface O species. In
other words, O atom transfers into the hydroxyl radicals through
the O2 activation by the surface active O species, and then into
phenol through benzene oxidation by hydroxyl radicals.
adsorption barrier of O2 on acetone (94.2 kcal mol-1) is extremely
high. These indicate that O2 activation by acetone is unfavorable,
in line with its inert activity. The interaction of O2 with the C=O
groups of cyclohexanone and 1,4-benzoquinone weakens the O-
O bond with significant elongated IO-O of 1.557 Å and 1.561 Å,
suggesting the tendency to cleave O-O bond. The adsorption
barriers of O2 on cyclohexanone and 1,4-benzoquinone are 49.4
and 56.9 kcal mol-1, respectively. Both of them are greatly lower
than that on acetone. Lower Eads of cyclohexanone relative to
that of 1,4-benzoquinone responds to the higher activity of the
former than that of latter (yield: 16.4% vs. 13.6%). The
calculations above highlight the vital role of cyclic ketonic and
quinone C=O in O2 activation by prolonging the O-O bond and
providing a relatively low adsorption barrier.
The findings above demonstrate that 1) the active sites are the
surface O species; 2) the hydroxyl radicals are the reactive O
species; 3) O2 is the terminal oxidant; 4) O2 activation is the rate-
determining step. By virtue of these, we propose a possible
reaction route for the present hydroxylation reaction (Scheme 2d):
O2 is chemosorbed on the surface of carbon and then activated
by the surface O species (mainly carbonyl groups) to produce the
reactive hydroxyl radicals, which oxidize the benzene to furnish
the product phenol. Previously, several active sites such as
defects, edge sites (zig-zag and armchair edges), heteroatoms
and surface functional groups have been proposed for O2
activation via the following mechanisms: 1) O2 was reduced to
peroxo and superoxide species by carbon π electrons and this
Reaction pathway
Kinetic isotope effect (KIE) was determined by competitive
hydroxylation of C6H6 and deuterated benzene (C6D6) with O2
(Figure 5f). The low intramolecular KIE value of 1.2 excludes the
involvement of the C-H bond cleavage in the kinetic relevant step.
This phenomenon also suggests that H-abstraction mechanism
with high KIE value of 4.9 is not included in the oxidation.[6e, 26a]
Alternatively, the formation of a complex through the interaction
of hydroxyl radical and benzene is the most probable
hydroxylation way. Therefore, the key step in the hydroxylation is
the O2 activation by surface O species to produce the reactivate
O species (·OH). This is in consistent with the kinetic curve, in
which the initial reaction rate is low and explosive hydroxylation
happened after a relatively long induction period (Figures 5a-b).
Density functional theory (DFT) calculations were conducted to
probe the insights of O2 activation on the surface O species of
carbons (Scheme 2a-c). Small molecular model catalysts were
subjected to the theoretical simulations for the sake of simplicity.
Cyclohexanone and 1,4-benzoquinone were used to resemble
the major surface active sites of carbonyl groups, the cyclic
ketonic and quinone C=O. Acetone was used to mimic the
inactive sites, the linear chain aliphatic ketone (Csp3=O). Such
modeling has been widely employed to afford a favorable
compromise between precision and computational effort.[3, 25a, 28]
Scheme 2a-c illustrates the variation of O2 interacting with these
activation
mode
was
enhanced
after
incorporating
heteroatoms;[29] 2) superoxide species was formed through the O2
reduction by the edge sites with unpaired electrons and acted as
the oxidant;[2b] 3) the surface oxygen groups such as quinone
were the active sites to react with the substrate and then
regenerated by O2.[11a, 30] In our hydroxylation reaction, O2 was
activated by the surface O species rather than carbon π electrons
to generate the hydroxyl radicals as the reactive oxygen species;
models and the corresponding adsorption barrier energies (Eads). superoxide species was not involved in the reaction, as illustrated
The O-O bond length (IO-O) is 1.376 Å for the free O2 and keeps
unchanged when O2 interacts with the C=O group of acetone. The
by the radical quenching experiment; the inert activity in the
anaerobic reaction excludes the inclusion of the surface O
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
species in the first step of the reaction. It has been reported that
activation of O2 on metal sites produced superoxide that
subsequently oxidized acetic acid to generate peracetic acid
(CH3COOOH).[31] Because of the exclusion of superoxide radical
in this work, the possible formation of CH3COOOH in the reaction
system can be eliminated. Taken these together, we propose the
most probable O2 activation in our hydroxylation system that is
different from those previous ones.[2b, 12, 15, 29] O2 is chemosorbed
on the surface of carbon and then activated through interaction
with cyclic ketonic or quinone C=O. The prolonged O-O bond of
activated O2 promotes their disassociation. Because the
superoxide species is not involved, we tentatively propose that
the activated O2 is disassociated into hydroxyl radicals in the
presence of proton H+ and electron donor. The acid reaction
environment provides sufficient H+. There are various electro
donors in carbons, such as the edges with unpaired electron,
nitrogen p-orbital with unpaired electron delocalized in the
preferentially oxidized by hydroxyl radicals. As expected,
oxidation of toluene gave the benzaldehyde yield of 8.4% and
cresols yield of 3.1% (1.9% and 1.2% for o- and p- product,
respectively), higher than that over mesoporous graphitic carbon
nitride with the benzaldehyde yield of 3%.[21b] A higher activity was
observed in the oxidation of ethylbenzene with the yield of 12.8%
for acetyl benzene and 4.0% for benzaldehyde. The above
catalysis evaluation, though rough, allows suggesting that NOC-
0.15 is active in the aerobic oxidations through direct activation of
C-H bonds. Compared to previous carbon nanotubes catalyzed
oxidation of ethylbenzene, NOC-0.15 exhibited higher selectivity
and comparable yield.[5a] Thus, this kind of surface O-enriched
carbon material is efficient heterogeneous carbon catalyst for
aerobic oxidations through the generation of robust hydroxyl
radicals via straightforward O2 activation.
conjugated sp2-hybridized structure, carbon
quinone/hydroquinone tautomerism and carbene-type free site
generated from C-H disproportionation.[1c,
π
orbital,
Conclusions
23,
30,
32]
Surface O-enriched mesoporous carbons were prepared through
the hydrothermal synthesis of melamine modified phenolic resin
together with a carbonization step. They have large surface area
and their surface is enriched with abundant oxygen groups. The
content and dispersion of these surface oxygen species are
facilely managed through the present synthetic strategy. These
carbons exhibit high yield and stable reusability in the benzene
hydroxylation to phenol with O2. The surface oxygen groups,
mainly the cyclic ketonic and quinone C=O, are the active sites,
activating O2 to hydroxyl radicals that oxidize benzene to phenol.
Aerobic oxidation of multi benzene derivatives proceeded
Disproportionation and tautomerism of the small molecular model
catalysts provide similar electron donors. All of these promote the
conversion of activated O2 into hydroxyl radicals. The interaction
of hydroxyl radicals with benzene ring produces an adduct
hydroxybenzene radical, the cleavage of which affords the phenol
and releases the H+ to the solution and the electron to the
corresponding electron receptors. The electron donors and
receptors of carbons resemble the low and high valent state of the
metals, and thus carbons exhibit similar behavior to the metal in
the oxidations. The stronger ESR signal of NOC-0.15 than that of
NAC-900 reflects that the former catalyst has more unpaired
electron. This promotes the electron transfer and thus responds
to the higher activity of NOC-0.15 relative to NAC-900 (Figure
S25). The carbonyl groups in the form of cyclic ketonic and
quinone C=O play a vital role in the present hydroxylation. On the
one hand, they activate O2 through efficient chemo-sorption to
elongate the O-O bond. On the other hand, their
disproportionation and tautomerism afford efficient electron
transmitter such as quinone/phenolic groups.
effectively with the same strategy. This work provides
a
fascinating opportunity to rationally design carbon catalyst for
aerobic oxidations.
Experimental Section
Materials and method
All reagents were commercially available and used without additional
purification. Pluronic F127 (EO106PO70EO106, average molecular weight:
12600) and activated carbon (AC) were purchased from Sigma Aldrich.
Melamine and resorcinol were purchased from Sinopharm Chemical
Reagent CO., Ltd. Formaldehyde solution was purchased from West Long
Chemical CO., Ltd.
Functionality of hydroxyl radicals in oxidation of C-H bond
Hydroxyl radicals are effective reactive oxygen species to oxidize
various organic compounds. The catalytic potential of our
established carbons was investigated by engaging NOC-0.15 in
the oxidation of benzene derivatives with O2 (Table S6).
Hydroxylation of naphthalene produced naphthol with the yield of
3.1% (0.8% and 2.3% for 1- and 2-product, respectively), like that
of hydroxylation of benzene. Previously, hydroxylation of
naphthalene with O2 has been carried out by using a metal
catalyst prepared by immobilizing H5PMo10V2O40 on protonated g-
C3N4, giving the yield of 3.72 %.[33] A yield of 1.5% was found in
the carbon catalysed hydroxylation of naphthalene with H2O2
using chemically converted graphene from graphite.[6f] However,
there was no report related to the carbon catalysed hydroxylation
of naphthalene with O2 as oxidant. Herein, NOC-0.15 served as a
carbon catalyst in the hydroxylation of naphthalene with O2, and
the resulting yield of 3.1% is close to that of the metal-catalysed
system. Side-chain Csp3-H bonds in alkyl benzene with lower bond
dissociation energies than that of Csp2-H bonds are normally
Elemental analyses were carried on a CHN elemental analyser Vario EL
cube. Brunauer−Emmett−Teller (BET) surface areas and pore volumes
were calculated from the nitrogen sorption isotherms determined on a
BELSORP-MINI analyzer at 77 K. The pore size distribution curves were
calculated from the nitrogen sorption isotherms using Barret-Joyner-
Halenda (BJH) method. Thermogravimetric (TG) analyses were performed
on an STA409 instrument under nitrogen atmosphere at a heating rate of
10 °C min−1. Field emission scanning electron microscope (FESEM;
Hitachi S-4800, accelerated voltage, 5 kV) accompanied by Energy
Dispersive X-ray spectrometry (EDX; accelerated voltage, 20 kV) was
used to study the morphology and the element distribution. Transmission
electron microscopy (TEM) images, high-resolution high-angle annular
dark-field scanning transmission electron microscopy (HAADF-STEM)
images and the corresponding energy-dispersive X-ray (EDX) spectral
imaging were recorded on a JEM-2100 F field-emission transmission
electron microscope (200 KV). X-ray photoelectron spectra (XPS) was
conducted on a PHI 5000 Versa Probe X-ray photoelectron spectrometer
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
equipped with Al Kα radiation (1486.6 eV). Raman spectra were collected
for carboxylic groups and lactones, sodium hydroxide (NaOH) for lactones
and phenols, and sodium ethoxide (CH3CH2ONa) for phenols and carbonyl
groups. The concentration of these aqueous solution was 0.05 mol/L. The
carbon sample (1 g) was placed in 50 mL solution of NaHCO3, Na2CO3,
NaOH, and CH3CH2ONa, respectively. They were shaken for 1 h and then
kept at room temperature for 24 h. After that, back titration by HCl was
performed to calibrate the content of different surface groups.[7]
on
Argon/Krypton Ion Laser system (excitation line: 514 nm). Electron spin-
resonance (ESR) spectra were recorded on Bruker EMX-10/12
a Horiba HR 800 spectrometer with a Spectra Physics 2018
a
spectrometer at the X-band. Electron paramagnetic resonance (EPR)
spectra were recorded on a Bruker E580. X-ray diffraction (XRD) patterns
were measured with a Smart Lab diffractometer (Rigaku Corporation)
equipped with a 9 kW rotating anode Cu source at 40 kV and 200 mA, from
5 to 50° (0.2° s−1).
Kinetic Isotope Effect
Synthesis of carbon catalysts
Kinetic isotope effect (KIE) was studied by using benzene and benzene-
d6 as the substrate, respectively. KIE value was calculated by (moles of
phenol)/(moles of phenol-d5).
Melamine modified phenolic resin precursors were hydrothermally
prepared under an acidic condition using melamine as nitrogen source[20a]
.
18O-isotope tracer experiment
In a typical synthesis, F127 (4 g) was dissolved in the mixture solvents of
water (20 g) and ethanol (20 g) to form a clear solution. Subsequently,
resorcinol (4.4 g) and melamine (0.15 g) were added into the above
solution and stirred at room temperature for 30 min. After that, HCl (1.0 g,
37 wt.%) and aqueous formaldehyde solution (4.0 g, 35 wt.%) was added
and the resulting mixture was stirred for 60 min. The transparent
homogeneous mixture was transferred into a Teflon-lined autoclave and
heated at 100 °C for 10 h. The orange-yellow product was separated by
filtration, washed with ethanol and dried at 80 °C for 8 h. Modulation of the
melamine amount in the synthetic gel provided carbon precursors with
adjustable N content. N-doped and surface O-enriched mesoporous
carbons (NOCs) were prepared via the carbonization of the above
precursors. Typically, the calcination was carried out in a tubular furnace
under N2 atmosphere at 400 °C for 2 h and then 650 °C for 2 h with a ramp
rate of 1 °C min-1. The resulting carbons were denoted as either NOC-m
Hydroxylation of benzene was carried out by using mixture of acetic acid
(60 vol. %) and H218O as the solvent. The distribution of products (phenol
and phenol-18O) was determined by isotope pattern on the gas
chromatography-mass spectrometry (GC-MS, Agilent 7890A/5975C).
Radical quenching and trapping experiments.
Selective radical quenching experiments were carried out by using the
scavengers of tert-butyl alcohol (TA), dimethylsulfoxide (DMSO) and
butylated hydroxytoluene (BHT). The reaction proceeded in the presence
of scavenger under the conditions of benzene (45 mmol, 4 mL), scavenger
(22.5~45 mmol), catalyst (0.2 g), LiOAc (0.6 g), aqueous solution of acetic
acid (60 vol. %, 25 mL), 2.2 MPa O2, 150 °C, 12~48 h. Radical trapping
experiment was performed by measuring the electron paramagnetic
resonance (EPR) spectra of the reaction solution in the presence of radical
trapper phenyl-α-tert-butyl nitrone (an efficient trapper for hydroxyl
radicals). EPR reaction conditions: NOC-0.15 (7 mg L-1), PBN (25 mg L-1),
pH 2.0. A Bruker E580 instrument was employed to collect the EPR
spectra.
or NOC-T, in which
T and m denoted the melamine mass and
carbonization temperature, respectively. For comparison, N-doped
activated carbon (NAC-900) was prepared. AC (2 g), melamine (0.5 g) and
ethanol (20 g) were mixed in a beaker and stirred at room temperature for
30 min. Subsequently, the mixture was transferred into an oven and then
heated at 100 °C for 6 h. The solvent was removed by evaporation and the
solid was calcined in a tubular furnace under N2 atmosphere (900 °C, 2 h,
3 °C min-1).
Theoretical Calculations
Catalysis Test
Structure optimization was carried out at B3LYP/6-31G level by using
density functional theory (DFT) implemented in Gaussian09 program. The
adsorption energy (Eads) was calculated by the following equation: Eads=E1-
(E2+E3), where E1, E2 and E3 are the total energy of the O2-catalyst adduct,
O2 and catalyst, respectively.
Hydroxylation of benzene with dioxygen (O2) was carried out in 100 mL
stainless steel autoclave equipped with a mechanical stirrer and an
automatic temperature controller. Typically, benzene (45 mmol, 4 mL),
catalyst (0.2 g), lithium acetate (LiOAc, 0.6 g), aqueous solution of acetic
acid (60 vol.%, 25 mL) were added into the reactor, which was flushed with
O2 and then pressurized to 2.2 MPa O2 at room temperature. The reaction
was conducted at 150 °C for 48 h with vigorous stirring. After reaction,
apparent pressure drop was observable. Product was analyzed by a gas
chromatograph (GC Agilent 7890B) equipped with a flame ionization
Acknowledgements
The authors thank greatly the National Natural Science
Foundation of China (Nos. 21476109, U1662107, 21136005 and
21303084) and PAPD.
detector
(FID)
detector
and
a
capillary
column
(HP-5,
30m×0.25mm×0.25μm). 1,4-Dioxane was used as the internal standard
for the GC analysis to calculate the yield of phenol according to the
formula: mmol phenol / mmol initial benzene (seen in the supplementary
experimental for details). Under the employed conditions, phenol was the
only product detected by GC. By products such as catechol, hydroquinone,
and benzoquinone were undetectable. A six-run recycling test was
performed to investigate the reusability. After each run, the solid catalyst
was separated by centrifugation, washed with aqueous acetic acid, dried
in the vacuum, and then charged in the next run. Oxidation of other arenes
including toluene, ethylbenzene and naphthalene were carried out with the
same procedure to that of benzene hydroxylation by using target substrate.
Keywords: carbon materials• metal-free catalyst• surface
oxygen groups• hydroxylation of benzene to phenol•
heterogeneous catalysis
[1]
a) X. Liu, L. Dai, Nat. Rev. Mater. 2016, 1, 16064-16075; b) P. Tang,
G. Hu, M. Li, D. Ma, ACS Catal. 2016, 6, 6948-6958; c) J. Zhang, X.
Liu, R. Blume, A. Zhang, R. Schlögl, D. S. Su, Science 2008, 322,
73-77.
[2]
a) A. Villa, J. P. Tessonnier, O. Majoulet, D. S. Su, R. Schlögl, Chem.
Commun. 2009, 29, 4405-4407; b) G. Lv, H. Wang, Y. Yang, T.
Deng, C. Chen, Y. Zhu, X. Hou, ACS Catal. 2015, 5, 5636; c) C.
Liang, H. Xie, V. Schwartz, J. Howe, S. Dai, S. H. Overbury, J. Am.
Chem. Soc. 2009, 131, 7735-7741; d) A. Primo, F. Neatu, M. Florea,
V. Parvulescu, H. Garcia, Nat. Commun. 2014, 5, 5291-5291; e) F.
Boem titration
Boem titration was carried out using aqueous solution of sodium
bicarbonate (NaHCO3) for carboxylic groups, sodium carbonate (Na2CO3)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
Su, S. C. Mathew, G. Lipner, X. Fu, M. Antonietti, S. Blechert, X.
[15]
M. A. Patel, F. Luo, M. R. Khoshi, E. Rabie, Q. Zhang, C. R. Flach,
R. Mendelsohn, E. Garfunkel, M. Szostak, H. He, ACS Nano 2016,
10, 2305-2315.
A. Vinu, S. Anandan, C. Anand, P. Srinivasu, K. Ariga, T. Mori,
Microporous Mesoporous Mater. 2008, 109, 398-404.
H. N. Pham, A. E. Anderson, R. L. Johnson, T. J. Schwartz, B. J.
O’Neill, P. Duan, K. Schmidtrohr, J. A. Dumesic, A. K. Datye, ACS
Catal. 2015, 5, 4546-4555.
V. M. Gun’Ko, O. P. Kozynchenko, S. R. Tennison, R. Leboda, J.
Skubiszewska-Zięba, S. V. Mikhalovsky, Carbon 2012, 50, 3146-
3153.
Z. Zafar, Z. H. Ni, X. Wu, Z. X. Shi, H. Y. Nan, J. Bai, L. T. Sun,
Carbon 2013, 61, 57-62.
a) X. Zhao, Q. Zhang, B. Zhang, C. M. Chen, A. Wang, T. Zhang, D.
S. Su, J. Mater. Chem. 2012, 22, 4963-4969; b) R. Arrigo, M.
Hävecker, S. Wrabetz, R. Blume, M. Lerch, J. Mcgregor, E. P. J.
Parrott, J. A. Zeitler, L. F. Gladden, A. Knopgericke, J. Am. Chem.
Soc. 2010, 132, 9616-9630.
a) X. Cai, Q. Wang, Y. Liu, J. Xie, Z. Long, Y. Zhou, J. Wang, ACS
Sustainable Chem. Eng. 2016, 4, 4984-4996; b) X. H. Li, X. Wang,
M. Antonietti, ACS Catal. 2012, 2, 2082–2086.
a) L. Zhang, H.-h. Liu, G.-y. Li, C.-w. Hu, Chin. J. Chem. Phys. 2012,
25, 585-591; b) Y.-J. Lyu, T. Qi, H.-Q. Yang, C.-W. Hu, Catal. Sci.
Technol. 2018, 8, 176-186.
L. R. Radovic, A. J. A. Salgado-Casanova, Carbon 2018, 126, 443-
451.
M. M. Montemore, M. A. van Spronsen, R. J. Madix, C. M. Friend,
Chem. Rev. 2017, 118, 2816-2862.
a) J. C. Espinosa, S. Navalón, A. Primo, M. Moral, J. F. Sanz, Á. M,
H. García, Chem. - Eur. J. 2015, 21, 11966-11971; b) W. Schilling,
D. Riemer, Y. Zhang, N. Hatami, S. Das, ACS Catal. 2018, 8, 5425-
5430.
a) A. Raba, M. Cokoja, W. A. Herrmann, F. E. Kühn, Chem.
Commun. 2014, 50, 11454-11457; b) D. Wang, M. Wang, Z. Li, ACS
Catal. 2015, 6852-6857; c) P. Chen, L. Chen, Y. Zeng, F. Ding, X.
Jiang, N. Liu, C.-T. Au, S.-F. Yin, Appl. Catal., B 2018, 234, 311-
317; d) A. M. Alsabagh, F. Z. Yehia, G. Eshaq, A. E. Elmetwally,
ACS Sustainable Chem. Eng. 2017, 5, 4811-4819.
X. Duan, H. Sun, J. Kang, Y. Wang, S. Indrawirawan, S. Wang, ACS
Catal. 2015, 5, 4629-4636.
X. Duan, H. Sun, Z. Ao, L. Zhou, G. Wang, S. Wang, Carbon 2016,
107, 371-378.
a) X. Hu, Z. Zhou, Q. Lin, Y. Wu, Z. Zhang, Chem. Phys. Lett. 2011,
503, 287-291; b) X. Guo, W. Qi, W. Liu, P. Yan, F. Li, C. Liang, D.
Su, ACS Catal 2017, 7, 1424-1427.
N. Walpen, G. J. Getzinger, M. H. Schroth, M. Sander, Environ. Sci.
Technol. 2018, 52,5236-5245.
G. Luo, X. Lv, X. Wang, S. Yan, X. Gao, J. Xu, H. Ma, Y. Jiao, F. Li,
J. Chen, RSC Adv. 2015, 5, 94164-94170.
Wang, J. Am. Chem. Soc. 2010, 132, 16299-16301; f) X. Zou, L.
Wang, B. I. Yakobson, Nanoscale 2018, 10, 1129-1134; g) Y. Gao,
X. Chen, J. Zhang, N. Yan, ChemPlusChem 2015, 80, 1556-1564;
h) S. Wu, G. Wen, X. Liu, B. Zhong, D. S. Su, ChemCatChem 2014,
6, 1558-1561; i) B. Yuan, P. Li, M. Tian, Q. Qin, X. Liu,
ChemCatChem 2018, https://doi.org/10.1002/cctc.201801406.; j) S.
Li, G. Li, G. Li, G. Wu, C. Hu, Microporous Mesoporous Mater. 2011,
143, 22-29.
[16]
[17]
[18]
[3]
[4]
[5]
G. Wen, S. Wu, B. Li, C. Dai, D. S. Su, Angew. Chem., Int. Ed. 2015,
54, 4105-4109.
M. Hong, X. Jie, C. Chen, Q. Zhang, J. Ning, M. Hong, L. Zhou, X.
Li, Catal. Lett. 2007, 113, 104-108.
[19]
[20]
a) J. Luo, F. Peng, H. Yu, H. Wang, W. Zheng, ChemCatChem 2013,
5, 1578-1586; b) A. Dhakshinamoorthy, A. Primo, P. Concepcion, M.
Alvaro, H. Garcia, Chem.Eur.J. 2013, 19, 7547-7554.
a) O. Shoji, T. Kunimatsu, N. Kawakami, Y. Watanabe, Angew.
Chem. Int. Ed. 2013, 52, 6606-6610; b) M. Yamada, K. D. Karlin, S.
Fukuzumi, Chem. Sci. 2016, 7, 2856-2863; c) X. Ye, Y. Cui, X. Qiu,
X. Wang, Appl. Catal. B: Environ. 2014, 152-153, 383-389; d) Y.
Morimoto, S. Bunno, N. Fujieda, H. Sugimoto, S. Itoh, J. Am. Chem.
Soc. 2015, 46, 5867-5870; e) T. Tsuji, A. A. Zaoputra, Y. Hitomi, K.
Mieda, T. Ogura, Y. Shiota, K. Yoshizawa, H. Sato, M. Kodera,
Angew. Chem. Int. Ed. 2017, 129, 7779-7782; f) J. H. Yang, G. Sun,
Y. Gao, H. Zhao, P. Tang, J. Tan, A. H. Lu, D. Ma, Energy Environ.
Sci. 2013, 6, 793-798; g) S. I. Niwa, M. Eswaramoorthy, J. Nair, A.
Raj, N. Itoh, H. Shoji, T. Namba, F. Mizukami, Science 2002, 295,
105-107.
[6]
[21]
[22]
[23]
[24]
[25]
[7]
[8]
[9]
J. Xu, H. Liu, R. Yang, G. Li, C. Hu, Chin. J. Catal. 2012, 33, 1622-
1630.
J.-H. Yang, G. Sun, Y. Gao, H. Zhao, P. Tang, J. Tan, A.-H. Lu, D.
Ma, Energy Environ. Sci. 2013, 6, 793-798.
a) W. Wang, G. Ding, T. Jiang, P. Zhang, T. Wu, B. Han, Green
Chem. 2013, 15, 1150-1154; b) B. B. Sarma, R. Carmieli, A.
Collauto, I. Efremenko, J. M. L. Martin, R. Neumann, ACS Catal.
2016, 6, 6403-6407; c) Z. Long, Y. Zhou, G. Chen, W. Ge, J. Wang,
Sci. Rep. 2014, 4, 3651-3656.
[26]
[10]
a) Q. Qin, Y. Liu, W. Shan, W. Hou, K. Wang, X. Ling, Y. Zhou, J.
Wang, Ind. Eng. Chem. Res. 2017, 56, 12289-12296; b) Z. Long, G.
Chen, S. Liu, F. Huang, L. Sun, Z. Qin, Q. Wang, Y. Zhou, J. Wang,
Chem. Eng. J. 2018, 334, 873-881.
[27]
[28]
[29]
[11]
[12]
a) W. Qi, D. Su, ACS Catal. 2014, 4, 415-429; b) D. S. Su, G. Wen,
S. Wu, F. Peng, R. Schlogl, Angew. Chem. Int. Ed. 2017, 56, 936-
964.
V. N. Chi, Y. T. Liao, T. C. Kang, J. E. Chen, T. Yoshikawa, Y.
Nakasaka, T. Masuda, C. W. Wu, Green Chem. 2016, 18. 5959-
5961.
[30]
[31]
[32]
[33]
[13]
[14]
C. Su, M. Acik, K. Takai, J. Lu, S. Hao, Y. Zheng, P. Wu, Q. Bao, T.
Enoki, Y. J. Chabal, Nat. Commun. 2012, 3, 1298-1307.
V. S. Jeyaraj, M. Kamaraj, V. Subramanian, J. Phys. Chem. C 2015,
119, 619-636.
a) L. R. Radovic, B. Bockrath, J. Am. Chem. Soc. 2005, 127, 5917-
5927; b) L. R. Radovic, J. Am. Chem. Soc. 2009, 131, 17166-17175.
H. Wang, M. Zhao, Q. Zhao, Y. Yang, C. Wang, Y. Wang, Ind. Eng.
Chem. Res. 2017, 56, 2711-2721.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801668
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Mesoporous carbons with the surface
enriched with abundant oxygen
species afforded high yield and stable
reusability in the hydroxylation of
benzene to phenol with O2. The
surface carbonyl groups are the active
sites for O2 activation to generate
hydroxyl radicals to oxidize benzene.
Wanjian Shan, Shuai Li, Xiaochun Cai,
Jie Zhu, Yu Zhou*, Jun Wang*
Page No. – Page No.
Carbon catalyzed hydroxylation of
benzene with dioxygen to phenol
over surface carbonyl groups
This article is protected by copyright. All rights reserved.