Highly selective hydrogenation of phenol and derivatives over a pd@carbon nitride catalyst in aqueous media

Article abstract of DOI:10.1021/ja109856y

Cyclohexanone is an important intermediate in the manufacture of polyamides in chemical industry, but direct selective hydrogenation of phenol to cyclohexanone under mild conditions is a challenge. We report here a catalyst made of Pd nanoparticles supported on a mesoporous graphitic carbon nitride, Pd@mpg-C₃N₄, which was shown to be highly active and promoted the selective formation of cyclohexanone under atmospheric pressure of hydrogen in aqueous media without additives. Conversion of 99% and a selectivity higher than 99% were achieved within 2 h at 65 °C. The reaction can be accelerated at higher temperature, but even at room temperature, 99% conversion and 96% selectivity could still be obtained. The generality of the Pd@mpg-C ₃N₄ catalyst for this reaction was demonstrated by selective hydrogenation of other hydroxylated aromatic compounds with similar performance.

Full text of DOI:10.1021/ja109856y

COMMUNICATION
pubs.acs.org/JACS
Highly Selective Hydrogenation of Phenol and Derivatives over a
Pd@Carbon Nitride Catalyst in Aqueous Media
Yong Wang,*,† Jia Yao,‡ Haoran Li,*,‡ Dangsheng Su,§ and Markus Antonietti†
†Department of Colloid Chemistry, Max-Planck Institute of Colloids and Interfaces, Research Campus Golm, D-14424 Potsdam,
Germany
‡Department of Chemistry, Zhejiang University, Hangzhou 310027, China
§Department of Inorganic Chemistry, Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, 14195 Berlin, Germany
S Supporting Information
b
selectivity, while changes in the reaction temperature have had
a considerable impact on the ultimate product composition.13
ABSTRACT: Cyclohexanone is an important intermediate
in the manufacture of polyamides in chemical industry, but
direct selective hydrogenation of phenol to cyclohexanone
under mild conditions is a challenge. We report here a
catalyst made of Pd nanoparticles supported on a mesopor-
ous graphitic carbon nitride, Pd@mpg-C3N4, which was
shown to be highly active and promoted the selective
formation of cyclohexanone under atmospheric pressure
of hydrogen in aqueous media without additives. Conver-
sion of 99% and a selectivity higher than 99% were achieved
within 2 h at 65 °C. The reaction can be accelerated at higher
temperature, but even at room temperature, 99% conver-
sion and 96% selectivity could still be obtained. The general-
ity of the Pd@mpg-C3N4 catalyst for this reaction was
demonstrated by selective hydrogenation of other hydro-
xylated aromatic compounds with similar performance.
Al2O3-supported Pd catalysts, however, have a low resistance to
deactivation by coke deposition, and the use of MgO as a catalyst
support is not suitable for industrial applications because of its
poor mechanical performance. Carbons, including activated
carbon, carbon nanofibers, and hydrophilic carbon, etc., are also
good supports because of their high surface area.16-18 Pd@C
catalysts can catalyze the hydrogenation reaction in the liquid
phase, but even if the required reaction temperature in a batch
process is lower than for gas-phase reactions, the selectivity of
phenol hydrogenation toward cyclohexanone remains poor,
which imposes further separation steps and strongly limits the
usefulness of these catalysts. Other catalysts systems, such as
Pt@C, Ru@C, and Rh@C, are not for general use and entail
important practical drawbacks regarding low conversion, low
selectivity, and harsh reaction conditions. For instance, when
Ru@C was used as the catalyst in the hydrogenation of phenol,
only 30% conversion and 5% selectivity toward cyclohexanone
were achieved, evne under a hydrogen pressure of 10 MPa.18
Because of that, attaining high selectivity (>95%) at elevated
conversions (>80%) remains a challenging catalytic problem.
Most recently, Han and co-workers achieved both excellent
conversion and selectivity in the hydrogenation of phenol by
using a dual-supported Pd Lewis acid catalyst.19 However, the
fact that these hosts contain Lewis acids such as AlCl3 imposes
severe limitations on their use in hydrogenation applications in
general and adds another chemical sensitivity restricting sub-
strates, purities, and reaction conditions. Therefore, alternative
heterogeneous catalysts for selective hydrogenation of phenol
under ambient conditions are desirable.
yclohexanone is of high industrial importance because of its
C
use as an intermediate in the production of caprolactame
and adipic acid, which are used to manufacture nylon 6 and nylon
66, respectively.1 Cyclohexanone production on a commercial
scale typically involves either the oxidation of cyclohexane2-4 or
the hydrogenation of phenol.5 The former route requires high
temperatures and high pressures and generates undesirable
byproducts that lower the product yield and complicate the
recovery/separation steps.3,4 In the latter route, phenol can be
hydrogenated to cyclohexanone in a “one-step” or a “two-step”
process.6 The two-step process involves hydrogenation of phenol
to cyclohexanol followed by dehydrogenation to cyclohexa-
none.7 The one-step, selective direct hydrogenation of phenol
to cyclohexanone is certainly preferable, as it avoids the endo-
thermic dehydrogenation step. However, cyclohexanone is an
active intermediate and is easily hydrogenated to cyclohexanol
and other byproducts. The distribution of products from phenol
hydrogenation is highly dependent on the type of catalyst and the
properties of the support,8 and the development of an effective
catalyst for this process has attracted broad research activity.9-11
The hydrogenation of phenol is generally carried out in the vapor
phase with supported palladium catalysts such as Pd@Al2O3 and
Pd@MgO.12-15 The incorporation of promoters such as alkali
and alkaline-earth metals has been shown to enhance the
Previous investigations have in fact shown that the support
plays a key role in the hydrogenation of phenol. Modern views
suggest that the reaction occurs mainly between phenol chemi-
sorbed on the support and hydrogen activated on the metal
(normally Pd) sites. Generally, phenol adsorbed in nonplanar
fashion over basic sites gives rise to cyclohexanone while that
adsorbed on acidic sites in coplanar fashion leads to the forma-
tion of cyclohexanol and cyclohexane.13,20 We have recently
shown that an abundant and easily accessible polymeric semi-
conductor, polymeric mesoporous graphitic carbon nitride
(mpg-C3N4), exhibits extreme chemical and thermal stability,
Received: November 10, 2010
Published: February 4, 2011
r
2011 American Chemical Society
2362
dx.doi.org/10.1021/ja109856y J. Am. Chem. Soc. 2011, 133, 2362–2365
|

Journal of the American Chemical Society
COMMUNICATION
Figure 1. (a) XRD patterns of the parent mpg-C3N4 and Pd@mpg-
C3N4. (b) N2 adsorption/desorption isotherms. (c) XPS spectra of
Pd@mpg-C3N4.
Scheme 1. Phenol Hydrogenation over Pd@mpg-C3N4
Figure 2. (a) STEM, (b) HRTEM, and (c) TEM images and (d)
particle size distribution of Pd@mpg-C3N4. The inset in (b) is the local
fast Fourier transform.
respectively.25 The average size of the crystallites was calculated
to be ∼7 nm from the major diffraction peak (111) using
Scherrer’s formula. The morphologies of Pd@mpg-C3N4 com-
posite materials were also investigated by HRTEM (Figure 2b).
The HRTEM images revealed well-dispersed Pd particles with a
mean size of 3.3 nm (Figure 2d), which is a little bit smaller than
that the value predicted by XRD. The Pd X-ray photoelectron
spectroscopy (XPS) of the catalyst presents a doublet corre-
sponding to Pd 3d5/2 and Pd 3d3/2. The Pd 3d5/2 peak at 335.9
eV is attributed to Pd0 (metallic palladium), while the Pd 3d5/2
peak at 337.3 eV is related to Pd2þ (palladium oxide).26
According to XPS data, Pd0 was formed as the major phase on
the surface of the catalyst (∼70%). A previous report showed
that the introduction of amine groups in the carbon carrier
resulted in increased proportions of Pd0 that were resistant to
reoxidation during air contact.26 In our case, mpg-C3N4 has a
high nitrogen content of 60 wt %. These π-bonded planar
C-N-C-layers along with their incompletely condensed amino
groups in the carrier are suitable for stabilizing highly dispersed
Pd0 particles and prevent their oxidation.
can be chemically shaped to a variety of nanostructures, and can
be directly used in heterogeneous catalysis.2,21-24 The most
active system in mpg-C3N4 is in fact presumably an incompletely
condensed, N-bridged “poly(tri-s-triazine)” polymer. The nitro-
gen functionalities on the surface might act as strong Lewis base
sites, while the π-bonded planar layered configurations are
expected to anchor phenol via special O-H N or O-H
π
3 3 3
3 3 3
interactions. In this work, we focused on supported Pd as the
catalytic agent and considered the use of this nitrogen-rich meso-
porous material as a basic host to construct binary Pd@mpg-
C3N4 catalytic metal-semiconductor heterojunctions for the
hydrogenation of phenol (Scheme 1).
Mesoporous carbon nitride was prepared by thermally in-
duced self-condensation of cyanamide using Ludox HS40 as a
hard template, which resulted in a yellow powder with graphitic
packing structure [as revealed by X-ray diffraction (XRD);
Figure 1a] and a Brunauer-Emmett-Teller (BET) surface area
of 179 m2 g-1 (as determined from the desorption branch of the
nitrogen adsorption/desorption isotherm; Table S1 in the Sup-
porting Information). A deposition-precipitation (DP) route
was then employed to deposit Pd on mpg-C3N4. In Figure 1b,
nitrogen adsorption isotherms of the parent and modified carbon
nitride materials are presented. The isotherm of the parent mpg-
C3N4 shows a small hysteresis (typical of the existence of
randomly connected spherical pores) but only weak micropor-
osity. After the modification, the isotherm shape was preserved,
indicating that the Pd nanoparticles do not block or alter the pore
system.
The activity of Pd@mpg-C3N4 for the hydrogenation of
phenol was then investigated. For our purpose, we first investi-
gated the effect of various solvents for the reaction. The
corresponding conversions and selectivities are listed in Table
S2. With Pd@mpg-C3N4 under the primary conditions, we could
readily achieve selectivities toward cyclohexanone in the range
57-92%, with a conversion of phenol of 30-100% in 6 h at
100 °C and 3 MPa hydrogen pressure (Table S2). Among the
solvents tested, water and CH2Cl2 exhibited the highest activity.
Ethanol also gave high activity but with a low selectivity toward
cyclohexanone. It seems that the catalyst is very active even in
water under atmospheric pressure of hydrogen (Table S2);
accordingly, we afterward limited the subsequent catalytic study
to an aqueous phenol feed at atmospheric hydrogen pressure and
temperatures less than 100 °C.
The formation of Pd nanoparticles was proven by powder
XRD (Figure 1 a) and high-resolution transmission electron
microscopy (HRTEM) (Figure 2). The XRD patterns show the
characteristic peak of (002) packing of graphitic carbon nitride.
The diffraction peaks at 40.0 and 46.6° in the XRD patterns can
be assigned to the (111) and (200) planes of Pd metal particles,
Under similar operating conditions, the catalytic performance
of Pd@mpg-C3N4 was found to be far better than that of any
2363
dx.doi.org/10.1021/ja109856y |J. Am. Chem. Soc. 2011, 133, 2362–2365

Journal of the American Chemical Society
COMMUNICATION
Table 1. Hydrogenation of Phenol with Pd@mpg-C3N4
Catalyst in Water under Various Conditionsa
Scheme 2. Possible Reaction Mechanism of Phenol over
Pd@mpg-C3N4
selectivity (%)
entry PH (MPa) T (°C) time (h) conv. (%) CdO
C-OH
2
1b,c
2c
3
1
30
30
30
100
65
65
45
45
45
45
45
25
25
25
25
25
25
12
12
12
1
13
45
94
>99
>99
>99
>99
>99
>99
>99
99
6
<1
<1
<1
<1
<1
<1
<1
1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
1
46
4
>99
91
Furthermore, cyclohexanol was detected at all only beyond a
distinct reaction time, which suggests that the hydrogenation of
phenol is going stepwise with cyclohexanone as the “intermedi-
ate” (Scheme 1). Depending on the nature of the catalyst’s active
site, phenol can interact with the surface through the hydroxy
5
1.2
6
2
3
>99
57
7
8
6
74
9
9
86
group to form strong O-H N or O-H π interactions.
3 3 3
3 3 3
10
11d
12
12
24
24
24
24
48
72
99
98.5
1.5
<1
<1
2
Recently, Jhung and co-workers reported the efficient use of
the NH and NH2 groups on the surface of carbon nitride for the
effective adsorption of phenol.27 Although an interaction of the
aromatic ring of phenol with the π electron of the carbon nitride
81
>99
>99
98
12
62
13
66
14c
15e
16
0.1
0.1
0.1
0.1
60
>99
>99
97
<1
<1
3
could not be excluded, we expected that such a π π interac-
3 3 3
tion is weaker and that the phenol therefore adsorbs in nonplanar
fashion over basic sites. The transfer of electron density in the
heterojunction from the semiconductor g-C3N4 enriches
the electron density of the metallic Pd and accelerates the
hydrogenation reaction, as compared to Pd@C. Scheme 2
illustrates the step sequence of the hydrogenation of phenol. In
the initial stage, phenol can be easily absorbed on the surface of
the catalyst, and H2 is activated by the electronically supported
Pd (Scheme 2 a). The benzene ring of phenol is then partially
hydrogenated to the enol (Scheme 2 b), which can isomerize
rapidly to give cyclohexanone. There is only a weaker H-bridge
donor, and the cyclohexanone leaves the surface of the catalyst
quickly, being replaced by a more strongly binding new phenol
molecule (Scheme 2 c) and avoiding further hydrogenation to
cyclohexanol.
To illustrate the general applicability of Pd@mpg-C3N4, the
method was extended to ring hydrogenation of other hydro-
xylated aromatic compounds. Table 2 shows the results of these
hydrogenations. As can be seen, 99% conversion was achieved in
all cases, and the selectivitvies toward substituted cyclohexa-
nones were excellent (>96%). As expected, methylphenol deri-
vatives gave methylcyclohexanones with both excellent conver-
sions and selectivities (Table 2, entries 1-3). Hydrogenation of
hydroquinone (Table 2, entry 4) and pyrocatechol (Table 2,
entry 6) yielded the synthetically very valuable hydroxylcyclo-
hexanones, again with high conversions and selectivities. Inter-
estingly, in the case of resorcinol, the obtained product was not
hydroxylcyclohexanone but the reductive dehydration product
cyclohexanone (Table 2, entry 5). Furthermore, this Pd@mpg-
C3N4 catalyst is predicted to be active in the hydrogenation of
other phenol derivatives, such as naphthols, methoxyphenols,
aminophenols, etc. Work focused on more detailed hydrogena-
tion studies of Pd@mpg-C3N4 material is ongoing.
56
81
17
>99
96
4
a Reaction conditions: phenol (0.5 mmol), Pd (5 mol % relative to
phenol), water solvent (2 mL), unless otherwise noted. b Data taken
from ref 19; Pd@C was used as the catalyst. c Using 2 mL of CH2Cl2 as
the solvent. d VH /VN = 7:3. e Using 6 mL of water as the solvent.
2
2
commercial Pd@C catalyst. It is evident that the Pd@C catalyst
(without a Lewis acid as cocatalyst) generally produces a mixture
of cyclohexanone and cyclohexanol with low phenol conversion
(13%; Table 1, entry 1), even using CH2Cl2 as the solvent at a
hydrogen pressure of 1 MPa.19 In contrast to that, Pd@mpg-
C3N4 produced cyclohexanone as the major product with very
high selectivity (>99%) and good conversion (45%) under 0.1
MPa hydrogen pressure (Table 1, entry 2). Even in the case of
water as the solvent, selective hydrogenation could be achieved
(Table 1, entry 3). The use of water as a solvent complies with the
rules of green chemistry.
Variation of the reaction temperature had a considerable effect
on the reaction conversion, as illustrated in Table 1. Conversion
of 99% was achieved with a selectivity exceeding 99% within 2 h
at an ambient hydrogen pressure and 65 °C. The reaction was
accelerated at higher temperature; for example, full conversion in
1 h was obtained at 100 °C with a high selectivity of 99%. Even at
room temperature, 99% conversion and 96% selectivity could be
reached after a longer reaction time of 72 h. The partial loss of the
reaction selectivity may be caused by the hydrogenation of cyclo-
hexanone to cyclohexanol after a long reaction time, which could
also be observed when the hydrogen pressure was increased
during the reaction process (Table 1, entries 13 and 14). The
Pd@mpg-C3N4 could be easily recovered several times by simple
washing using acetonitrile. However, a slightly decrease in the
conversion was observed, which may be caused by leaching of
the Pd nanoparticles (Figure S1). Moreover, control experiments
indicated that a high purity of hydrogen is not necessary. The
reaction could be conducted even under a gas mixture of hydro-
In conclusion, palladium nanoparticles have been heteroge-
neously supported on mesoporous carbon nitrides with their
special base properties and electronic behavior. The novel Pd
catalyst exhibits a high activity for the direct hydrogenation of
phenol to cyclohexanone. The hydrogenation proceeds efficiently,
with highest selectivity found under atmospheric pressure of
hydrogen at room temperature using water as a clean solvent.
The high catalytic performance has been attributed to the special
gen and nitrogen (VH /VN = 7:3), which is a prerequisite for
2
low-price practical app2lications (Table 1, entry 11).
Cyclohexanone and cyclohexanol were the only reaction
products observed over the entire range of conditions studied.
2364
dx.doi.org/10.1021/ja109856y |J. Am. Chem. Soc. 2011, 133, 2362–2365

Journal of the American Chemical Society
COMMUNICATION
Table 2. Hydrogenation of Hydroxyl Aromatic Derivatives
with Pd@mpg-C3N4 Catalyst in Watera
(5) Shore, S. G.; Ding, E.; Park, C.; Keane, M. A. J. Mol. Catal. A:
Chem. 2004, 212, 291–300.
(6) Chatterjee, M.; Kawanami, H.; Sato, M.; Chatterjee, A.; Yokoyama,
T.; Suzuki, T. Adv. Synth. Catal. 2009, 351, 1912–1924.
(7) Fridman, V. Z.; Davydov, A. A. J. Catal. 2000, 195, 20–30.
(8) Watanabe, S.; Arunajatesan, V. Top. Catal. 2010, 53, 1150–1152.
(9) Wang, H. J.; Zhao, F. Y.; Fujita, S. I.; Arai, M. Catal. Commun.
2008, 9, 362–368.
(10) Mahata, N.; Vishwanathan, V. J. Catal. 2000, 196, 262–270.
(11) Morales, J.; Hutcheson, R.; Noradoun, C.; Cheng, I. F. Ind. Eng.
Chem. Res. 2002, 41, 3071–3074.
(12) Park, C.; Keane, M. A. J. Colloid Interface Sci. 2003, 266, 183–
194.
(13) Neri, G.; Visco, A. M.; Donato, A.; Milone, C.; Malentacchi, M.;
Gubitosa, G. Appl. Catal., A 1994, 110, 49–59.
(14) Shore, S. G.; Ding, E. R.; Park, C.; Keane, M. A. Catal. Commun.
2002, 3, 77–84.
(15) Claus, P.; Berndt, H.; Mohr, C.; Radnik, J.; Shin, E. J.; Keane,
M. A. J. Catal. 2000, 192, 88–97.
(16) Makowski, P.; Cakan, R. D.; Antonietti, M.; Goettmann, F.;
Titirici, M. M. Chem. Commun. 2008, 999–1001.
(17) Diaz, E.; Mohedano, A. F.; Calvo, L.; Gilarranz, M. A.; Casas,
J. A.; Rodriguez, J. J. Chem. Eng. J. 2007, 131, 65–71.
(18) Rode, C. V.; Joshi, U. D.; Sato, O.; Shirai, M. Chem. Commun.
2003, 1960–1961.
(19) Liu, H. Z.; Jiang, T.; Han, B. X.; Liang, S. G.; Zhou, Y. X. Science
2009, 326, 1250–1252.
(20) Li, H.; Liu, J.; Xie, S. H.; Qiao, M. H.; Dai, W. L.; Lu, Y. F.; Li,
H. X. Adv. Funct. Mater. 2008, 18, 3235–3241.
a Reaction conditions: Pd@mpg-C3N4 (25 mg), substrate (0.5 mmol),
H2 (1 bar pressure).
(21) Wang, Y.; Wang, X. C.; Antonietti, M.; Zhang, Y. J. ChemSusChem
2010, 3, 435–439.
structure of the catalytic semiconductor-metal heterojunc-
tion, which leads not only to a very stable and uniform dispersion
of Pd but also to additional electronic activation of the metal
nanoparticles and a “nonplanar” adsorption of phenol, which
finally gives rise to the fast and selective hydrogenation to
cyclohexanone.
(22) Thomas, A.; Fischer, A.; Goettmann, F.; Antonietti, M.; Muller,
J. O.; Schlogl, R.; Carlsson, J. M. J. Mater. Chem. 2008, 18, 4893–4908.
(23) Wang, Y.; Li, H. R.; Yao, J.; Wang, H. J.; Antonietti, M. Chem.
Sci. [Online early access]. DOI: 10.1039/c0sc00475h. Published Online:
Dec 3, 2010.
(24) Wang, Y.; Di, Y.; Antonietti, M.; Li, H. R.; Chen, X. F.; Wang,
X. C. Chem. Mater. 2010, 22, 5119–5121.
(25) Zhang, F. B.; Li, H. L. Carbon 2006, 44, 3195–3198.
(26) Radkevich, V. Z.; Senko, T. L.; Wilson, K.; Grishenko, L. M.;
Zaderko, A. N.; Diyuk, V. Y. Appl. Catal., A 2008, 335, 241–251.
(27) Haque, E.; Jun, J. W.; Talapaneni, S. N.; Vinu, A.; Jhung, S. H.
J. Mater. Chem. 2010, 20, 10801–10803.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental details and more
b
characterization results. This material is available free of charge
via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
chemwy@zju.edu.cn; lihr@zju.edu.cn
’ ACKNOWLEDGMENT
We thank Dr. Bingsen Zhang for the measurement of
HRTEM images. This work was supported by the ENERCHEM
Project of MPI and the National Science Foundation of China
(20806065, 20873121, and 20990221).
’ REFERENCES
(1) Dodgson, I.; Griffin, K.; Barberis, G.; Pignataro, F.; Tauszik, G.
Chem. Ind. 1989, 830–833.
(2) Wang, Y.; Zhang, J. S.; Wang, X. C.; Antonietti, M.; Li, H. R.
Angew. Chem., Int. Ed. 2010, 49, 3356–3359.
(3) Silva, T. F. S.; Mishra, G. S.; da Silva, M. F. G.; Wanke, R.;
Martins, L.; Pombeiro, A. J. L. Dalton Trans. 2009, 9207–9215.
(4) Schuchardt, U.; Cardoso, D.; Sercheli, R.; Pereira, R.; de Cruz,
R. S.; Guerreiro, M. C.; Mandelli, D.; Spinace, E. V.; Fires, E. L. Appl.
Catal., A 2001, 211, 1–17.
2365
dx.doi.org/10.1021/ja109856y |J. Am. Chem. Soc. 2011, 133, 2362–2365