Direct selective hydrogenation of phenol and derivatives over polyaniline-functionalized carbon-nanotube-supported palladium

Article abstract of DOI:10.1002/cplu.201200276

Cyclohexanone is an industrially important intermediate in the synthesis of materials such as nylon and polyamides, but direct selective hydrogenation of phenol to cyclohexanone under green conditions is a challenge owing to the over-reduction of cyclohexanone to cyclohexanol. A catalyst made of palladium nanoparticles supported on polyaniline-functionalized carbon nanotubes, Pd-PANI/CNT, which was shown to be highly active towards the direct hydrogenation of phenol to cyclohexanone, is reported. Phenol conversion exceeding 99 % was achieved with a cyclohexanone selectivity of >99 % under atmospheric pressure of hydrogen in aqueous media. The generality of the catalyst for this reaction was demonstrated by selective hydrogenation of other hydroxylated aromatic compounds with similar performance, again under green and mild conditions. It is suggested the Pd-N interactions and polymeric stabilization play a key role in the formation of stable and highly dispersed palladium nanoparticles on the conducting composite material PANI/CNT. The results also indicate that the phenol conversion is related presumably to the conductive property of PANI/CNT, whereas the cyclohexanone selectivity is attributed to the nitrogen-containing nature of PANI/CNT. Support group: Direct selective hydrogenation of phenol to cyclohexanone was achieved over a polyaniline (PANI)-functionalized carbon-nanotube (CNT)-supported palladium catalyst under atmospheric hydrogen pressure in aqueous media (see scheme). The results indicate that the phenol conversion is related to the conductive property of composite material PANI/CNT, whereas the cyclohexanone selectivity is attributed to the nitrogen-containing nature of PANI/CNT.

Full text of DOI:10.1002/cplu.201200276

CHEMPLUSCHEM
FULL PAPERS
DOI: 10.1002/cplu.201200276
Direct Selective Hydrogenation of Phenol and Derivatives
over Polyaniline-Functionalized Carbon-Nanotube-
Supported Palladium
Jinzhu Chen,*[a] Wei Zhang,[a, c] Limin Chen,[b] Longlong Ma,*[a] Hui Gao,[a] and Tiejun Wang[a]
Cyclohexanone is an industrially important intermediate in the
synthesis of materials such as nylon and polyamides, but direct
selective hydrogenation of phenol to cyclohexanone under
green conditions is a challenge owing to the over-reduction of
cyclohexanone to cyclohexanol. A catalyst made of palladium
nanoparticles supported on polyaniline-functionalized carbon
nanotubes, Pd–PANI/CNT, which was shown to be highly active
towards the direct hydrogenation of phenol to cyclohexanone,
is reported. Phenol conversion exceeding 99% was achieved
with a cyclohexanone selectivity of >99% under atmospheric
pressure of hydrogen in aqueous media. The generality of the
catalyst for this reaction was demonstrated by selective hydro-
genation of other hydroxylated aromatic compounds with sim-
ilar performance, again under green and mild conditions. It is
suggested the Pd–N interactions and polymeric stabilization
play a key role in the formation of stable and highly dispersed
palladium nanoparticles on the conducting composite material
PANI/CNT. The results also indicate that the phenol conversion
is related presumably to the conductive property of PANI/CNT,
whereas the cyclohexanone selectivity is attributed to the ni-
trogen-containing nature of PANI/CNT.
Introduction
Cyclohexanone is well known as
one of the key intermediates in
the synthesis of caprolactam and
adipic acid, which are the main
ingredients for the preparation of
Nylon 6, Nylon 66, and polyamide
resin.[1] The industrial production
of cyclohexanone involves typical-
ly either the oxidation of cyclo-
hexane[2] or the hydrogenation of
phenol.[3] The former route re-
quires high temperature and high
generally obtained through one- or two-step processes.[5] The
two-step process involves hydrogenation of phenol to cyclo-
hexanol followed by dehydrogenation to cyclohexanone.[6] The
one-step, selective direct hydrogenation of phenol to cyclohex-
anone is certainly preferable, as it avoids the endothermic de-
hydrogenation step, and the reaction can be conducted in
either the gas phase or liquid phase.
Traditionally, phenol hydrogenation has been carried out in
the gas phase, with supported Pd catalysts as the most effi-
cient systems for this reaction.[7] However, high temperatures
are required normally for this process (150–3008C), and the
generation of carbonaceous deposits (coking) in the course of
the reaction leads to deactivation of the catalyst. Liquid-phase
phenol hydrogenation offers cost and energy savings, because
the reaction can be performed at relatively low tempera-
tures.[5,8,9,11,12] Many researchers have contributed to this area,
and multiple catalysts have been screened, however, another
important and inherent drawback of current liquid-phase hy-
drogenation protocols relates to the use of environmentally
unfriendly solvents (e.g., dichloromethane)[8] and/or expensive
and more sophisticated conditions (supercritical carbon diox-
ide as solvent, which requires high H2 and CO2 pressures
>7.0 MPa).[5,9] In this regard, a selective phenol hydrogenation
in aqueous media is highly desirable, because water is environ-
mentally benign and easily available.[10] However, to date, only
a few studies have reported aqueous phenol hydrogenation
protocols, which provided cyclohexanol/cyclohexanone mix-
tures instead of being completely selective to cyclohexa-
none,[11] because the cyclohexanone product is readily hydro-
genated further to cyclohexanol under the reaction conditions
(Scheme 1a).[11a]
pressure and generates a large
COVER
number of byproducts.[4] In the
latter route, cyclohexanone is
[a] Prof. Dr. J. Chen, W. Zhang, Prof. Dr. L. Ma, Dr. H. Gao, Prof. Dr. T. Wang
Key Laboratory of Renewable Energy and Natural Gas Hydrate
Guangzhou Institute of Energy Conversion
Chinese Academy of Sciences
Guangzhou 510640 (P. R. China)
Fax: (+86)20-3722-3380
E-mail: chenjz@ms.giec.ac.cn
mall@ms.giec.ac.cn
[b] Dr. L. Chen
College of Environmental Science and Engineering
South China University of Technology
Guangzhou 510006 (P. R. China)
[c] W. Zhang
Graduate University of Chinese Academy of Sciences
Beijing 100049 (P. R. China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201200276.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 142

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
By having in mind the potential applications of PANI/CNT as
a special support, we focused on supported Pd as the catalytic
agent and nitrogen-rich PANI/CNT as a basic host to construct
a binary Pd–PANI/CNT catalytic metal-conducting support to fa-
cilitate the electron transfer during the redox-reaction process
for phenol hydrogenation (Figure 1a). Moreover, highly dis-
persed active Pd nanoparticles on a CNT with enhanced nucle-
ation and stability are expected to be developed in this study
through the introduction of PANI to bridge the Pd nanoparti-
cles and CNT walls with the presence of palladium–nitride (Pd–
Scheme 1. a) General reaction pathway for the hydrogenation of phenol.
Structures of b) PANI and c) PVI/CNT.
Most recently, Wang and co-workers achieved both excellent
conversion and selectivity in the hydrogenation of phenol in
aqueous media by using a catalyst made of Pd supported on
mesoporous graphitic carbon nitride (Pd/mpg-C3N4).[12] Never-
theless, the preparation of a mpg-C3N4 support involved the
use of aqueous ammonium bifluoride (NH4HF2) or hydrogen
fluoride (HF), which are hazardous and not environmentally
friendly. Therefore, the development of greener processes with
the attainment of high cyclohexanone selectivity with elevated
phenol conversion is still a great challenge.
In contrast to the fact that mpg-C3N4 is currently limited
only to laboratory-scale preparation,[13] commercially available
carbon materials such as carbon nanotubes (CNTs), carbon
nanofibers (CNFs), and acetylene carbon black (CB) have re-
ceived growing attention as supports as well as for heteroge-
neous catalysts because of their high mechanical strength, tun-
able surface properties, high surface area, conductivity, and so
forth. However, the graphenelike structure of such carbon ma-
terial has hindered the high dispersion of precious metals with
uniform size distributions onto such surfaces.[14] In addition,
the graphitized surfaces of CNTs are destroyed easily during
the introduction of active functional groups by oxidative acid
treatment. As a result, the above process also reduces the sta-
bility and durability of CNTs. The functionalization of CNT surfa-
ces, such as nitrogen-doped carbon (CNx) nanotubes, is
a common practice in circumventing such a deficiency.[15]
Moreover, without destroying the graphitized surfaces, wrap-
ping CNTs with conducting polymer polyaniline (PANI,
Scheme 1b) was recently explored to disperse metallic parti-
cles.[16] These one-dimensional nanofibrillar PANI/CNT compo-
site materials, consisting generally of a core–shell structure of
PANI coated on CNTs through the strong p–p interactions be-
tween them, are promising for use as sensors and electro-
des.[17]
Figure 1. a) Molecular interactions in the hydrogenation of phenol over Pd–
PANI/CNT catalyst; b) FTIR spectra of CNT, PANI, and PANI/CNT; c) N2 adsorp-
tion/desorption isotherms and d) XRD patterns of PANI/CNT and Pd–PANI/
CNT.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 143

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
N) interactions and p–p bonding. In addition, the nitrogen
functionalities on the surface of PANI/CNT, which provide a hy-
drophilic microenvironment and act as Lewis base sites, are ex-
pected to anchor phenol through OÀH···N hydrogen bonds
(Figure 1a). To the best of our knowledge, this is the first study
demonstrating the delicately designed binary catalyst of PANI-
functionalized CNT-supported Pd for the purpose of selective
hydrogenation in aqueous media.
Results and Discussion
Synthesis and characterization of binary catalyst Pd–PANI/
CNT
A PANI-functionalized CNT support was prepared by the poly-
merization of aniline in the presence of a multiwalled carbon
nanotube suspension.[16b] Catalyst Pd–PANI/CNT was prepared
by reducing K2PdCl4 with NaBH4 on a PANI/CNT support.
Elemental analysis confirmed the presence of PANI in the
PANI/CNT composite material. The PANI content in the compo-
site material was calculated to be 33.3% based on the nitrogen
component. FTIR spectra of the CNTs, PANI nanofibers, and
PANI/CNT support are presented in Figure 1b. Like graphite,
the FTIR spectrum of pure CNTs is practically featureless with
extremely low infrared absorption intensities (Figure 1b).[17e]
For PANI/CNT, two characteristic bands at n˜ =1226 and
1289 cmÀ1 can be assigned to C=N and CÀN stretching vibra-
tions, respectively, corresponding to the bands at n˜ =1220 and
1298 cmÀ1 in PANI.[18] The peak at n˜ =1576 cmÀ1 is associated
with a quinoid-ring vibration, whereas the band at n˜ =
1493 cmÀ1 is from a benzene-ring vibration in PANI/CNT, which
corresponds to the bands at n˜ =1583 and 1496 cmÀ1 in PANI,
respectively (Figure 1b). Furthermore, compared with PANI,
these vibrations of PANI/CNT are downshifted 3–
9 cmÀ1 owing to the strong p–p interactions be-
tween CNTs and PANI in PANI/CNT in which the CNTs
reduce the PANI through charge transfer. These re-
sults also verify the presence of PANI in the as-pre-
pared PANI/CNT composite materials.
Figure 2. a,b) TEM images and c) particle size distribution of Pd–PANI/CNT.
(fcc) crystalline Pd, respectively.[19c] The average size of the crys-
tallites was calculated to be approximately 6 nm from the
major diffraction peak (111) by using Scherrer’s formula. The
HRTEM images revealed highly dispersed Pd particles with
a mean size of 4.1 nm, which is a little bit smaller than the
value predicted by XRD, for which the narrowly distributed
particle sizes range from 3.0 to 5.0 nm owing to the PANI/CNT
stabilization.
Pd–PANI/CNT was evaluated further by X-ray photoelectron
spectroscopy (XPS). The existence of a N 1s signal originating
from PANI further verifies the presence of PANI in the Pd–PANI/
CNT catalyst (Figure 3a). The quantification analysis revealed
the N content is about 4.77% in Pd–PANI/CNT, which is in ac-
cordance with the elemental analysis results (4.66%). The XPS
spectrum of Pd 3d can be deconvolved into two main doublet
As determined from the nitrogen adsorption/de-
sorption isotherm, Brunauer–Emmett–Teller (BET) sur-
face areas of PANI/CNT and Pd–PANI/CNT are 95 and
121 m2 gÀ1, respectively (Figure 1c). The increased
surface area of Pd–PANI/CNT can be attributed to the
formation of new nanopores by means of the partial
deposition of Pd nanoparticles in the matrix of poly-
mer PANI.
The formation of Pd nanoparticles was proven by
powder XRD (Figure 1d) and high-resolution trans-
mission electron microscopy (HRTEM; Figure 2). The
diffraction peaks of PANI/CNT and Pd–PANI/CNT lo-
cated at a 2q value of about 26 and 438 were as-
signed to the characteristic peaks of the (002) and
(100) packing of graphitic CNTs, respectively.[19a–b] The
diffraction peaks of Pd–PANI/CNT at 39.9, 46.7, and
68.28 in the XRD patterns can be assigned to the
Figure 3. a) XPS scan survey for PANI/CNT and Pd–PANI/CNT; b) Pd 3d XPS spectra of Pd–
(111), (200), and (220) planes of face-centered cubic
PANI/CNT; c) N 1s XPS spectra of PANI/CNT and Pd–PANI/CNT.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 144

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
peaks, as illustrated in Figure 3b. The doublet with a binding
energy at 335.6 (assigned to Pd 3d5/2) and 340.8 eV (assigned
to Pd 3d3/2) can be indexed to the Pd0 state. The other set of
peaks at 337.2 (assigned to Pd 3d5/2) and 342.6 eV (assigned to
Pd 3d3/2) are ascribed to the PdII state.[20a–b] By calculating the
integration areas of these two doublets, it can be estimated
that Pd0 was formed as the major phase on the surface of the
catalyst (75–78%).
Table 2. Hydrogenation of phenol with various catalysts in water under
various conditions.[a]
Run
Catalyst[b]
T [8C]
t [h]
Conv. [%]
Selectivity[c] [%]
C=O
CÀOH
As shown in Figure 3c, the N 1s line of Pd–PANI/CNT was de-
convoluted into three superimposed peaks at 398.1, 399.1, and
402.3 eV. The peaks at 398.1 and 399.1 eV are characteristic
peaks for the imine (ÀN=) and amine (ÀNHÀ), respectively;
whereas the peaks at 402.3 eV are related to the positively
charged nitrogen atoms (ÀNH+À).[20c–d,16a] Furthermore, the N
1s peaks of PANI/CNT are shifted to lower binding energies rel-
1
2
3
4
5
6
7
8
Pd/AC
Pd/CNT
Pd/PANI
80
80
80
80
80
80
80
80
60
45
80
80
80
80
9
9
9
9
9
9
9
9
11
19
8
8
8
55
98
51
>99
>99
>99
60
73
98
98
86
89
87
11
13
>99
>99
>99
>99
>99
>99
>99
98
>99
>99
>99
>99
<1
<1
<1
<1
<1
<1
<1
2
<1
<1
<1
<1
Pd–PANI/CNT
Pd–PANI/CNF
Pd–PANI/CB
Pd–PANI/AC
Pd–PVI/CNT
Pd–PANI/CNT
Pd–PANI/CNT
Pd–PANI/CNT
Pd–PANI/CNT
Pd–PANI/CNT
Pd–PANI/CNT
9
10
11
12
13
14
Table 1. XPS analysis results of Pd–PANI/CNT and related materials.
85
84
82
Run
Material
Pd 3d/N 1s, Binding energy [eV] (Share [%])
8
1
2
3
4
5
Pd/CNT
Pd0: 3d5/2, 335.3; 3d3/2, 340.6[20b]
[a] Conditions: phenol (40 mg), catalyst (10 wt% Pd, 5 mol% Pd relative
to phenol), water (3 mL), H2 (0.1 MPa), unless otherwise noted. [b] Activat-
ed carbon (AC), carbon nanotubes (CNT), polyaniline (PANI), carbon nano-
fibers (CNF), acetylene carbon black (CB), poly(1-vinylimidazole) (PVI).
[c] C=O indicates cyclohexanone, and CÀOH denotes cyclohexanol.
Pd–PANI/CNT
Pd–PANI/CNT
PANI/CNT
PANI
Pd0: 3d5/2, 335.6; 3d3/2, 340.8
ÀN=, 398.1 (39.3); ÀNHÀ, 399.1 (39.3)
ÀN=, 398.3 (58.1); ÀNHÀ, 399.4 (26.6)
ÀN=, 398.6 (45.0); ÀNHÀ, 399.6 (45.3)[20e]
ative to PANI (Table 1, runs 4 and 5), which is in accordance
with the IR results; this suggests the existence of strong p–p
interactions between the CNTs and PANI in PANI/CNT. It is
noteworthy that XPS data show a shift to lower binding ener-
gies in the N 1s region for Pd–PANI/CNT when compared with
PANI/CNT (Figure 3c; Table 1, runs 3 and 4), whereas a shift to
higher binding energies of Pd 3d3/2 peaks in Pd–PANI/CNT is
observed simultaneously, contrary to Pd/CNT (Table 1, runs 1
and 2). The above XPS data indicate that nitrogen atoms of
PANI/CNT, amine (ÀNHÀ) or imine (=NÀ), are capable of partici-
pating in the interactions with Pd nanoparticles. Therefore, the
bifunctional stabilizing mechanism of PANI/CNT on Pd nano-
particles, that is, Pd–N interactions and polymeric immobiliza-
tion, is suggested for stabilizing highly dispersed Pd0 particles
and preventing their further oxidation upon exposure to the
air.
a moderate phenol conversion of 51% (Table 2, run 3). Surpris-
ingly, both excellent phenol conversion (higher than 99%) and
cyclohexanone selectivity (exceeding 99%) were achieved,
when PANI-functionalized CNTs, CNFs, and CB were introduced
to this catalyst system as a support (Table 2, runs 4–6). Never-
theless, PANI-functionalized AC-supported palladium catalysts
afforded cyclohexanone selectivity >99% with a phenol con-
version of only 60% (Table 2, run 7). The above results indicate
clearly that the observed selectivity towards the formation of
cyclohexanone relies on the microenvironment of PANI (Fig-
ure 1a; Table 2, runs 3–7). Whereas, the conductive property of
CNTs, CNFs, and CB with an intrinsic graphite structure might
be responsible for the excellent conversion towards electro-in-
active AC (Table 2, runs 1–2, 4–7).
To further investigate the roles of the p-conjugated conduct-
ing polymer PANI on the catalytic performance of Pd–PANI/
CNT, the electro-inactive polymer poly(1-vinylimidazole) (PVI)
was grafted covalently to CNTs (PVI/CNT, Scheme 1c) by in situ
radical polymerization of 1-vinylimidazole with CNTs.[21] It was
found that PVI/CNT-supported palladium catalysts afforded cy-
clohexanone selectivity exceeding 99% with a phenol conver-
sion of 73% (Table 2, run 8). The above results thus reveal that
the catalyst, with PANI-functionalized conducting materials
(CNT, CNF, CB) as supports, generally achieved excellent phenol
conversions (Table 2, runs 4–6). Whereas both PANI- and PVI-
containing catalysts generally afforded excellent cyclohexa-
none selectivities (Table 2, runs 3–8). This finding suggests the
phenol conversion is most likely related to the conductive
property of PANI which can transfer electrons[17a–b] between
the conductive support (CNT, CNF, and CB) and Pd nanoparti-
Direct hydrogenation of phenol and derivatives over Pd–
PANI/CNT in aqueous media
The catalytic performance of Pd–PANI/CNT for the phenol hy-
drogenation was then investigated. For comparison purposes,
Table 2 displays the catalytic properties of differently support-
ed Pd for the hydrogenation of phenol at an atmospheric hy-
drogen pressure in aqueous media. Pd/AC (activated carbon)
and Pd/CNT catalysts showed a similar catalytic selectivity, gen-
erally producing a mixture of cyclohexanone and cyclohexanol
with phenol conversions of 55 (Table 2, run 1) and 98%
(Table 2, run 2), respectively. In the case of palladium support-
ed on PANI, the hydrogenation is highly selective. Cyclohexa-
none selectivity was obtained higher than 99%, however, with
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 145

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
cles, and therefore accelerate the redox-reaction process.
Whereas, the cyclohexanone selectivity is presumably attribut-
ed to the nitrogen-containing nature of PANI and PVI.
cyclohexanone and PANI. Thus, being replaced by a more
strongly binding new phenol molecule, the cyclohexanone
leaves the surface of the PANI quickly (Scheme 2c), and avoids
further hydrogenation to cyclohexanol.
The influence of the reaction temperature on the phenol hy-
drogenation revealed that the reaction was accelerated at
higher temperature; for example, full conversion in 9 h was ob-
tained at 808C with selectivity exceeding 99% (Table 2, run 4).
As the temperature decreased to 458C, both excellent conver-
sion and selectivity could still be reached; however, only after
a prolonged reaction time to 19 h (Table 2, runs 9 and 10). The
slight decrease of the conversion and selectivity may be
caused by hydrogen leaking from a hydrogen-filled balloon
over the reaction vessel and further hydrogenation of cyclo-
hexanone to cyclohexanol, respectively, after a long reaction
time.
Thus, PANI brings at least four advantages to this study.
First, it is electron conductive, so it can transfer electrons be-
tween the CNT support and Pd nanoparticles, which acceler-
ates the redox-reaction process.[16,17a–b] Second, it prevents Pd
nanoparticles from aggregating with each other through poly-
meric stabilization and therefore leads to a uniform dispersion
of Pd nanoparticles on the CNT surfaces (Figure 2). Third, Pd
nanoparticles can be tightly anchored onto the surfaces of the
CNTs through Pd–N interactions with the N atoms of PANI as
shown in Figures 1a and 3c. Finally, owing to the nitrogen-con-
taining nature and strong hydrogen bonding with phenol,
PANI possesses a molecular-recognition function towards reac-
tant phenol and product cyclohexanone molecules, which pre-
vents further hydrogenation of cyclohexanone to cyclohexanol
under the reaction conditions, and thus improves the selectivi-
ty of the desired cyclohexanone production.
The reusability of the Pd–PANI/CNT catalyst was also tested
for phenol hydrogenation in water. The reaction was stopped
at each time point of 8 h as described in Table 2 (runs 11–14);
the catalyst was recovered and reused directly for the next run
under the same conditions. It is important to highlight that no
changes regarding the selectivity to cyclohexanone were ob-
served with respect to the fresh catalyst after three recycles.
However, a slightly reduced phenol conversion from 86 to
82% was observed, which can reasonably be related to the
catalyst mass loss during its recovery.
To explore the scope of the application of the Pd–PANI/CNT
catalyst, we investigated the hydrogenation of various hydroxy
aromatic compounds under similar conditions. Table 3 shows
the results of these hydrogenations. As can be seen, >99%
conversions were achieved in most cases, and the overall se-
lectivities to cycloketones exceeded 96% in all cases. As ex-
pected, methylphenol derivatives gave methylcyclohexanones
with both excellent conversions and selectivities (Table 3,
runs 1–3). Owing to the fact that the steric-hindrance effect of
the methyl group on the phenol ring interferes with crucial hy-
drogen-bond formation between the hydroxyl group of
phenol and the nitrogen atom of PANI, it takes ortho-cresol
a prolonged time to achieve complete conversion. Moreover,
4-isopropylphenol, with a bulky isopropyl group inhibiting
attack of the palladium hydride species towards para-C of the
phenol ring, was selectively hydrogenated to the ketone prod-
uct with excellent selectivity as well (Table 3, run 4). Interest-
ingly, hydrogenation of pyrocatechol produced both the re-
ductive product hydroxylcyclohexanone and the reductive de-
hydration product cyclohexanone in 83 and 13% yields, re-
spectively (Table 3, run 5), and the overall cycloketone selectivi-
ties exceeded 96%. In the case of resorcinol, cyclohexanone
was the only product detected, again with high conversion
and selectivity (Table 3, run 6). Similarly, hydrogenation of 4-
methoxyphenol led to a mixture of methoxycyclohexanone
and cyclohexanone with overall cycloketone selectivities ex-
ceeding 99% (Table 3, run 7). The formation of cyclohexanone
is presumably derived from hy-
Cyclohexanone and cyclohexanol were the only reaction
products observed over the entire range of conditions studied.
Furthermore, cyclohexanol was detected only beyond a distinct
reaction time, which suggests that the hydrogenation of
phenol is stepwise with cyclohexanone as the “intermediate”
(Scheme 1a). Previous investigations of the mechanism have
shown that the support plays a key role in the hydrogenation
of phenol. Generally, phenol adsorbed in a nonplanar fashion
over basic sites of support gives rise to cyclohexanone where-
as that adsorbed on acidic sites in a coplanar fashion leads to
the formation of cyclohexanol and cyclohexane.[9b,12] In our
case, Scheme 2 demonstrates a special function of the PANI/
CNT support and illustrates the step sequence of the selective
hydrogenation of phenol. Phenol, a strongly hydrogen-bridg-
ing interacting system,[11b] can interact strongly with the PANI
surface through hydrogen bonds to form strong OÀH···N inter-
actions. Therefore, in the initial stage, phenol can be easily ab-
sorbed in a nonplanar fashion over basic nitrogen sites of the
surface, and H2 is activated by the supported Pd (Scheme 2a).
The benzene ring of phenol is then hydrogenated partially to
the enol (Scheme 2b), which can isomerize rapidly to give cy-
clohexanone. There is, however, a weaker interaction between
drogenolytic cleavage of the CÀ
OÀC bond followed by phenol
hydrogenation. In the case of 4-
chlorophenol, the cyclohexa-
none product was obtained in
high yield as a result of reduc-
tive dechlorination (Table 3,
run 8). The above results there-
Scheme 2. Proposed reaction mechanism of the selective hydrogenation of phenol over Pd–PANI/CNT.
fore indicate that catalyst Pd–
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 146

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
nanoparticles but also to additional functions of the “nonpla-
nar” adsorption of phenol, which finally gives rise to the fast
and selective hydrogenation to cyclohexanone.
Table 3. Hydrogenation of hydroxyl aromatic derivatives with Pd–PANI/
CNT catalyst.[a]
Aqueous-phase hydrogenation under atmospheric pressure
avoids the hazards and toxicity associated with the use of or-
ganic solvents, reduces energy and waste costs, and simplifies
separation procedures at the end of the reaction. For these
reasons, this novel process should give an economically viable
and environmentally attractive route to several cycloketones.
The research thus highlights an efficient and novel catalytic
material for transforming phenols into cycloketones under
green and sustainable conditions.
Run
1
Substrate
t [h]
Conv. [%]
Product selectivity [%]
15
>99
>99
2
3
17
24
>99
>99
97
98
Experimental Section
Preparation of PANI
Polyaniline was synthesized at 0–58C in a solution of HCl using am-
monium persulfate as the oxidizing agent according to a literature
method.[17f] Typically, an aqueous solution of aniline (3 mL,
32 mmol) in 1m hydrochloric acid (100 mL) and another solution
of ammonium peroxydisulfate (1.83 g, 8 mmol) in hydrochloric acid
(1m, 100 mL) was prepared and rapidly mixed by pouring the two
solutions together and immediately stirring to ensure sufficient
mixing before polymerization began. The reaction mixture was
constantly stirred in an ice bath with a temperature range of 0–
58C for 8 h, then the crude fibers were filtered and purified by
washing with deionized water, NaOH solution, and three more
times with water, and then drying the powder at 808C overnight
under vacuum.
4
12
81
98
5
6
26
11
>99
>99
83
13
>99
7
8
25
18
>99
>99
93
98
7
Preparation of composite material PANI/CNT
The PANI/CNT composite was synthesized by the polymerization of
aniline in the presence of
a multiwalled carbon nanotube
(MWCNT) suspension.[16b] In a typical synthesis, a HCl solution (2m,
75 mL) containing a MWCNT suspension (0.75 g) was sonicated at
room temperature for about half an hour then mixed with a two-
fold amount of aniline monomer (1.5 mL). FeSO4 (2–3 drops,
5 wt%) in aqueous solution was later added, followed by addition
of (NH4)2S2O8 (0.8 g, dissolved in 10 mL deionized water), which
served as the oxidant. The reaction mixture was constantly stirred
in an ice bath (temperature range of 0–58C) for 1 h and was then
reacted at room temperature for 3 h. The black powder was fil-
tered and washed with distilled water/ethanol until the filtrate
became colorless, and then the powder was dried at 808C over-
night under vacuum. This composite was denoted as PANI/CNT.
Other carbon materials such as activated carbon, acetylene carbon
black, and carbon nanofibers functionized with polyaniline were
prepared by means of the same procedure as for PANI/CNT, and
were denoted as PANI/AC, PANI/CB, and PANI/CNF, respectively.
[a] Conditions: substrate (0.4 mmol), 10 wt% Pd–PANI/CNT (10 mol% Pd
relative to substrate), water (3 mL), H2 (0.1 MPa), 1008C, unless otherwise
noted.
PANI/CNT can be applied in the selective hydrogenation of var-
ious phenol derivatives into cycloketones.
Conclusion
Highly dispersed palladium nanoparticles with narrow size dis-
tribution have been supported on PANI-functionalized CNTs
owing to the Pd–N interactions and polymeric stabilization.
The novel Pd–PANI/CNT catalyst exhibits a high activity for the
selective hydrogenation of phenol and derivatives to the corre-
sponding cycloketones under atmospheric pressure of hydro-
gen using water as a clean solvent. The high catalytic perfor-
mance has been attributed to the special structure of the
PANI/CNT composite material, that is, the phenol conversion is
related to the conductive property of PANI/CNT, whereas the
cyclohexanone selectivity is attributed to the nitrogen-contain-
ing nature of PANI/CNT. The PANI/CNT composite material
leads not only to a very stable and uniform dispersion of Pd
Preparation of composite material PVI/CNT
The poly(1-vinylimidazole)-functionalized MWCNT was prepared ac-
cording to a literature method.[21] The detailed process was as fol-
lows: A 200 mL dried Schlenk flask was charged with a magnetic
stirrer, MWCNT (200 mg), and DMF (100 mL). After the solution had
stirred for 6 h at room temperature, AIBN (60 mg) and 1-vinylimida-
zole (4.1 mL) were added, and the mixture was degassed by four
freeze–pump–thaw cycles. The flask was placed in a thermostated
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 147

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
[5] M. Chatterjee, H. Kawanami, M. Sato, A. Chatterjee, T. Yokoyama, T.
Suzuki, Adv. Synth. Catal. 2009, 351, 1912–1924.
oil bath at 658C under a nitrogen atmosphere and was stirred for
48 h. After polymerization, the mixture was cooled to room tem-
perature, diluted to 200 mL with fresh DMF, bath sonicated for
0.5 h, and centrifuged for 2 h. After removal of the homogeneous
black supernatant, the black sediment was washed thoroughly
with ethanol (50 mL) twice, and vacuum-filtered through a 0.2 mm
PTFE membrane. The solid was finally dried overnight at 808C
under vacuum. The obtained composites were denoted as PVI/
CNT.
[6] V. Z. Fridman, A. A. Davydov, J. Catal. 2000, 195, 20–30.
[7] a) L. M. Sikhwivhilu, N. J. Coville, D. Naresh, K. V. R. Chary, V. Vishwana-
than, Appl. Catal. A 2007, 324, 52–61; b) P. Claus, H. Berndt, C. Mohr, J.
Radnik, E. J. Shin, M. A. Keane, J. Catal. 2000, 192, 88–97.
[8] a) H. Z. Liu, T. Jiang, B. X. Han, S. G. Liang, Y. X. Zhou, Science 2009, 326,
1250–2152; b) H. Li, J. Liu, S. H. Xie, M. H. Qiao, W. L. Dai, Y. F. Lu, H. X.
Li, Adv. Funct. Mater. 2008, 18, 3235–3241.
[9] a) H. J. Wang, F. Y. Zhao, S.-I. Fujita, M. Arai, Catal. Commun. 2008, 9,
362–368; b) C. V. Rode, U. D. Joshi, O. Sato, M. Shirai, Chem. Commun.
2003, 1960–1961.
[10] L. ꢁ. de Cienfuegos, R. Robles, D. Miguel, J. Justicia, J. M. Cuerva, Chem-
SusChem 2011, 4, 1035–1048.
Preparation of catalyst Pd–PANI/CNT
PANI/CNT (500 mg) was suspended in THF/H2O (1:1 v/v, 100 mL)
with ultrasonic treatment for 30 min and was stirred mechanically
for another 30 min. K2PdCl4 (154 mg) was then added to the solu-
tion. Pd nanoparticles formed and were anchored on the support
after adding a mixture of NaBH4 (85 mg, 2.25 mmol) and Na2CO3
(85 mg, 0.8 mmol) in water (20 mL) at room temperature and was
stirred for another 1 h. The resulting samples were filtered, washed
with distilled water, and dried at 808C overnight under vacuum.
Other catalysts Pd–PANI, Pd–PANI/AC, Pd–PANI/CB, Pd–PANI/CNF,
and Pd–PVI/CNT were prepared following the same procedures as
for Pd–PANI/CNT.
[11] a) Y. Z. Xiang, L. Ma, C. S. Lu, Q. F. Zhang, X. N. Li, Green Chem. 2008, 10,
939–943; b) P. Makowski, R. D. Cakan, M. Antonietti, F. Goettmann,
M. M. Titirici, Chem. Commun. 2008, 999–1001; c) H. L. Liu, Y. W. Li, R.
Luque, H. F. Jiang, Adv. Synth. Catal. 2011, 353, 3107–3113; d) F. Lu, J.
Liu, J. Xu, Mater. Chem. Phys. 2008, 108, 369–374; e) H. Y. Cheng, R. X.
Liu, Q. Wang, C. Y. Wu, Y. C. Yu, F. Y. Zhao, New J. Chem. 2012, 36, 1085–
1090.
[12] Y. Wang, J. Yao, H. R. Li, D. S. Su, M. Antonietti, J. Am. Chem. Soc. 2011,
133, 2362–2365.
[13] a) Y. Wang, X. C. Wang, M. Antonietti, Angew. Chem. 2012, 124, 70–92;
Angew. Chem. Int. Ed. 2012, 51, 68–89; b) N. Stock, S. Biswas, Chem. Rev.
2012, 112, 933–969.
[14] E. Dujardin, T. W. Ebbesen, H. Hiura, K. Tanigaki, Science 1994, 265,
1850–1852.
Typical procedure for hydrogenation of phenol
[15] a) P. Ayala, R. Arenal, M. Rꢂmmeli, A. Rubio, T. Pichler, Carbon 2010, 48,
575–586; b) S. R. Stoyanov, A. V. Titov, P. Krꢃla, Coord. Chem. Rev. 2009,
253, 2852–2871.
[16] a) D. He, C. Zeng, C. Xu, N. Cheng, H. Li, S. Mu, M. Pan, Langmuir 2011,
27, 5582–5588; b) H.-Y. Lee, W. Vogel, P. P-J. Chu, Langmuir 2011, 27,
14654–14661.
[17] a) F. Huang, F. L. Lou, D. Chen, ChemSusChem 2012, 5, 888–895; b) H. X.
Zhong, H. M. Zhang, Z. Xu, Y. F. Tang, J. X. Mao, ChemSusChem 2012, 5,
1698–1702; c) M. Ding, Y. Tang, P. Gou, M. J. Reber, A. Star, Adv. Mater.
2011, 23, 536–540; d) M. Cochet, W. K. Maser, A. M. Benito, M. A. Calle-
jas, M. T. Martꢄnez, J-M. Benoit, J. Schreiber, O. Chauvet, Chem. Commun.
2001, 1450–1451; e) Y. Liao, C. Zhang, Y. Zhang, V. Strong, J. Tang, X.-G.
Li, K. Kalantar-zadeh, E. M. V. Hoek, K. L. Wang, R. B. Kaner, Nano Lett.
2011, 11, 954–959; f) J. Huang, R. B. Kaner, Angew. Chem. 2004, 116,
5941–5945; Angew. Chem. Int. Ed. 2004, 43, 5817–5821.
[18] a) H. Zengin, W. Zhou, J. Jin, R. Czerw, D. W. Smith, L. Echegoyen, D. L.
Carroll, S. H. Foulger, J. Ballato, Adv. Mater. 2002, 14, 1480–1483; b) S.
Quillard, G. Louarn, S. Lefrant, A. G. Macdiarmid, Phys. Rev. B 1994, 50,
12496–12508.
The reaction was conducted in a Schlenk flask (20 mL) with a mag-
netic stirrer under a hydrogen balloon at atmospheric pressure. A
typical procedure was as follows: phenol (40 mg, 0.426 mmol), cat-
alyst Pd–PANI/CNT (Pd 10 wt%, 23 mg, 5 mol% Pd relative to
phenol), and water (3 mL) were placed in a flask. The flask was
purged with H2 to remove the air 3 times. The reaction was then
stirred at 600 rpm under a hydrogen balloon under atmospheric
pressure at 808C. After the reaction, the reactor was placed in ice
water to quench the reaction and the products were subsequently
extracted with ethyl acetate and analyzed by GC and GC-MS. Cy-
clohexanone and cyclohexanol were the only reaction products ob-
served. Selectivities and conversions were determined according to
the literature methods (see the Supporting Information for
details).[8a]
[19] a) M. Terrones, W. K. Hsu, A. Schilder, H. Terrones, N. Grobert, J. P. Hare,
Y. Q. Zhu, M. Schwoerer, K. Prassides, H. W. Kroto, D. R. M. Walton, Appl.
Phys. A 1998, 66, 307–317; b) W. Feng, X. D. Bai, Y. Q. Lian, J. Liang,
X. G. Wang, K. Yoshino, Carbon 2003, 41, 1551–1557; c) F. B. Zhang,
H. L. Li, Carbon 2006, 44, 3195–3198.
[20] a) B. M. Choudary, S. Madhi, N. S. Chowdari, M. L. Kantam, B. Sreedhar, J.
Am. Chem. Soc. 2002, 124, 14127–14136; b) X.-R. Ye, Y. H. Lin, C. M.
Wang, M. H. Engelhard, Y. Wang, C. M. Wai, J. Mater. Chem. 2004, 14,
908–913; c) J.-E. Park, S.-G. Park, A. Koukitu, O. Hatozaki, N. Oyama,
Synth. Met. 2004, 141, 265–269; d) M. Hasik, E. Wenda, A. Bernasik, K.
Acknowledgements
This study was supported by National Basic Research Program of
China (973 Program, 2012CB215304), National Natural Science
Foundation of China (21172219, 21250110061, 21207039, and
21101053),
Guangdong
Natural
Science
Foundation
(S2011010002274), and Guangzhou Science and Technology Proj-
ect (12C64061560).
´
Kowalski, J. W. Sobczak, E. Sobczak, E. Bielanska, Polymer 2003, 44,
7809–7809; e) A. Drelinkiewicz, A. Zie˛ba, J. W. Sobczak, M. Bonarowska,
Keywords: hydrogenation · nanotubes · palladium · phenol ·
´
Z. Karpinski, A. Waksmundzka-Gꢅra, J. Stejskal, React. Funct. Polym.
2009, 69, 630–642.
polyaniline
[21] a) S. H. Qin, D. Q. Qin, W. T. Ford, J. E. Herrera, D. E. Resasco, S. M. Bachi-
lo, R. B. Weisman, Macromolecules 2004, 37, 3965–3967; b) S. H. Qin,
D. Q. Qin, W. T. Ford, J. E. Herrera, D. E. Resasco, Macromolecules 2004,
37, 9963–9967.
[1] World Nylon 6 and 66 Supply/Demand Report, PCI Fibers and Raw Mate-
rials, Seaford, 1998.
[2] Y. Wang, J. S. Zhang, X. C. Wang, M. Antonietti, H. R. Li, Angew. Chem.
2010, 122, 3428–3431; Angew. Chem. Int. Ed. 2010, 49, 3356–3359.
[3] S. G. Shore, E. Ding, C. Park, M. A. Keane, J. Mol. Catal. A 2004, 212,
291–300.
Received: October 19, 2012
Revised: November 16, 2012
[4] T. F. S. Silva, G. S. Mishra, M. F. G. da Silva, R. Wanke, L. Martins, A. J. L.
Pombeiro, Dalton Trans. 2009, 9207–9215.
Published online on December 12, 2012
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 142 – 148 148