Catalytic ring hydrogenation of phenol under supercritical carbon dioxide

Article abstract of DOI:10.1039/b304344d

A charcoal-supported rhodium catalyst was highly active for the ring hydrogenation of phenol and cresols under supercritical carbon dioxide.

Full text of DOI:10.1039/b304344d

Catalytic ring hydrogenation of phenol under supercritical carbon
dioxide
Chandrashekhar V. Rode, Uday D. Joshi, Osamu Sato and Masayuki Shirai*
Supercritical Fluid Research Center, National Institute of Advanced Industrial Science and Technology,
4-2-1, Nigatake, Miyagino, Sendai, 983-8551, Japan. E-mail: m.shirai@aist.go.jp
Received (in Cambridge, UK) 22nd April 2003, Accepted 20th June 2003
First published as an Advance Article on the web 30th June 2003
A charcoal-supported rhodium catalyst was highly active for
the ring hydrogenation of phenol and cresols under super-
critical carbon dioxide.
Pd catalyst under supercritical carbon dioxide at very high
temperature (523 K). However, maximum selectivity to the
corresponding cyclohexanol derivative was only 70%, with
formation of several other side products.14 In our work, at very
low temperature (328 K) rhodium was found to be the most
active for the ring hydrogenation of phenol, while for gas and
liquid phase reactions at > 453 K palladium is reported to be a
suitable catalyst for this hydrogenation.2–9 We used commercial
supported metal catalysts and did not add any promoter to, or
perform any special treatment on, the catalysts. The catalyst
performances observed for two systems (i.e. under gas/liquid
phase and supercritical carbon dioxide conditions) can not be
compared directly because the reaction conditions such as
temperature range and solvent system are totally different.
However, it is expected that the reaction mechanism under
supercritical carbon dioxide is totally different from that under
gas and liquid phase conditions. Not only does the enhancement
of hydrogen solubility in supercritical carbon dioxide contribute
to the higher catalyst activity but also there is an additional role
played by the supercritical carbon dioxide suggesting i) the
adsorption characteristics of phenol on active sites on metal
particles and/or ii) the electronic structures of the metal particles
are different under supercritical carbon dioxide from non-
supercritical conditions. Bhanage et al. reported that the
activities and selectivities for hydrogenation of a,b-unsaturated
aldehydes over a supported platinum catalyst under super-
critical carbon dioxide were higher than those in ethanol
solvent.15 They explained the enhancement not only by the
activation of reactants and but also by a change in electronic
state of the supported metal particles in supercritical carbon
dioxide which has a large dielectric constant. Arai et al. also
reported that the electron density of gold particles decreased in
supercritical carbon dioxide.16 The catalytic activities of metals
under supercritical carbon dioxide conditions could be different
due to modification by electron transfer to carbon dioxide which
would affect the adsorption characteristics of phenol as
compared with gas and liquid phase conditions. The lower
electron density of rhodium under supercritical carbon dioxide
would cause strong adsorption of phenol leading to the highest
catalyst activity. The apparent activation energy of ring
Adipic acid is an important chemical intermediate, for the
manufacture of nylon-6,6. The worldwide industrial processes
of adipic acid production are up to 2.2 million tons per year.1,2
Most industrial adipic acid production processes use nitric acid
oxidation of cyclohexanol, cyclohexanone, or both. Also,
cyclohexanol and cyclohexanone are produced by the ring
hydrogenation of phenol in industry. Generally, the hydro-
genation of phenol is carried out in the vapour phase and
supported palladium catalysts have been reported as the best
catalysts for this reaction.2–9 There has been hardly any work
reported for liquid phase hydrogenation of phenol except that by
Velasco et al.10 Supported metal catalysts are useful for
hydrogenation processes in industry. However, in the case of
vapor phase hydrogenation, high temperatures are needed and
coking during reaction causes catalyst deactivation. Hydro-
genation reactions in the liquid phase using supported metal
catalysts have wide ranging applications in pharmaceutical and
fine chemical processes, however, reaction velocities are not so
high and the separation of pure products from solvents is
critical. Moreover, eliminating the use of organic solvents is
highly desirable for environmentally benign processing.11
Organic gas–liquid reactions with solid catalysts under super-
critical carbon dioxide have several other advantages: i) high
reaction rates due to increased solubility of reactant gases in
supercritical fluids, thereby eliminating mass transfer re-
sistance; ii) easy separation of catalysts and products.12,13 Here
we report the catalytic ring hydrogenation of phenol and its
derivatives with supported metal catalysts under supercritical
carbon dioxide.
Commercially available catalysts were used in this work, viz.
5 wt% carbon supported palladium (5% Pd/C), rhodium (5%
Rh/C), platinum (5% Pt/C), and ruthenium (5% Ru/C), from
Wako Chemicals, Japan. The dispersion values of metal
particles on charcoal were determined by a hydrogen adsorption
method. Before hydrogen adsorption all catalysts were treated
with flowing hydrogen at 573 K for 10 min. The adsorption
temperature was 323 K. Hydrogenation reactions were carried
out in stainless-steel reactors (50 ml capacity). The weighed
amount of catalyst and reactants were placed in a reactor and
flushed twice with carbon dioxide. After the required tem-
perature was attained with a hot air circulating oven, hydrogen
and carbon dioxide were introduced into the reactor. After
reactions, the pressure was released slowly and the contents
were discharged to separate the catalyst by simple filtration. The
reaction mixture was diluted with methanol and a sample was
analyzed with GC-FID and GC-MS.
Table 1 Catalyst screening for the hydrogenation of phenola
Selectivity (%)
Disper-
Catalyst Conver- Cyclo-
Cyclo-
Catalyst
sion (%)b amount/g sion (%) hexanone hexanol TONc
5% Rh/C
5% Ru/C
5% Pd/C
5% Pt/C
5% Pt/Cd
12
25
3
8
8
0.0228
0.0639
0.1400
0.0780
0.0780
53
30
3
0
1
17
5
54
—
97
83
95
46
—
3
8244
759
304
—
The catalyst screening results for phenol hydrogenation are
shown in Table 1. Charcoal-supported rhodium, ruthenium and
palladium catalysts were active for the ring hydrogenation of
phenol to cyclohexanone and cyclohexanol at 328 K under
supercritical carbon dioxide conditions and the turnover
numbers for different catalysts were in the following order:
rhodium > ruthenium > palladium > platinum. Hitzler et al.
have reported ring hydrogenation of cresols over a commercial
125
a Reaction temperature 328 K; reaction time 2 h; hydrogen pressure 10
MPa; carbon dioxide pressure 10 MPa; initial phenol 0.02 mol. b The
dispersion was determined by a hydrogen adsorption method. c TON =
(moles of phenol reacted)/(moles of surface metal atoms). d Reaction
temperature 353 K.
1960
CHEM. COMMUN., 2003, 1960–1961
This journal is © The Royal Society of Chemistry 2003

View Online
Table 3 Hydrogenation of cresol over 5% Rh/Ca
Selectivity (%)
hydrogenation over the 5% Rh/C catalyst under supercritical
carbon dioxide conditions evaluated in our work was 32 kJ
mol21. In order to compare the performance under the same
conditions but without supercritical carbon dioxide (i.e. 328 K)
a separate experiment was conducted which showed no
hydrogenation. This indicates the reaction mechanism is
different in the two cases. The apparent activation energies
previously reported for liquid and gas phase phenol hydro-
genation (Pd/Al2O3 and Pd/MgO catalysts) were 57 and 65 kJ
mol21, respectively.2,17 The low apparent activation energy
under mild temperature in supercritical carbon dioxide suggests
the different adsorption characteristic of phenol and/or the
electronic interactions of metal particles with carbon dioxide
under supercritical conditions. We do not have any direct
evidence about the electronic state of the metal particles,
however, there is a strong possibility that carbon dioxide
molecules are more strongly adsorbed on the surface of
palladium metal than on other metal particles, making the
palladium surface inaccessible so that it exhibits lower activities
than rhodium and ruthenium in supercritical carbon dioxide.
Further study such as in-situ investigations of the structure and
electronic state of metal particles under supercritical carbon
dioxide and determination of the surface concentrations of
hydrogen and phenol adsorbed are needed.
The conversion over the Rh/C catalyst increased with
increasing reaction temperature and time. For example, com-
plete conversion of phenol to cyclohexanol was achieved at 353
K in 2 h over Rh/C. The reuse of the Rh/C catalyst was also
studied. No deactivation was observed after use 3 times. 100%
conversion (10% selectivity for cyclohexanone and 90% for
cyclohexanol) was obtained over the Rh/C catalyst at 328 K
after 4 h for each run.
In addition, we also examined the influence of the pressures
of both hydrogen and supercritical carbon dioxide on the
catalyst activity and selectivity and these results are presented in
Table 2. It was found that both hydrogenation activity and
selectivity to cyclohexanol increased with increasing hydrogen
pressure at 10 MPa carbon dioxide, caused by an increase in the
concentration of surface hydrogen with increasing hydrogen
pressure. Phenol hydrogenation is a successive reaction in
which first phenol is hydrogenated to cyclohexanone followed
by its hydrogenation to cyclohexanol. However, cyclohexanol
was observed at low phenol conversion under high hydrogen
pressure, indicating that cyclohexanol would be formed not
only via the cyclohexanone intermediate but also directly from
phenol. At high concentration of surface hydrogen, the
hydrogenation of phenol to cyclohexanol would easily occur,
resulting in higher selectivity to cyclohexanol (Table 2).
Methycyclo-
hexanone
Methycyclo-
hexanol
Reactant
Conversion (%)
m-Cresol
o-Cresol
p-Cresol
99
88
47
83
31
55
17
69
45
a Reaction temp. 328 K; hydrogen pressure 9 MPa; carbon dioxide pressure
11 MPa; reaction time 2 h; initial cresol 0.0185 mol; weight of catalyst
0.0455 g.
The hydrogenation activity also increased with increasing
carbon dioxide pressure. Bhanage et al. also observed higher
conversions with increasing carbon dioxide pressure in the case
of cinnamaldehyde hydrogenation under supercritical carbon
dioxide.15 Enhanced mass transfer may contribute to the
increased activity at higher carbon dioxide concentration. On
the other hand, the selectivity to cyclohexanol did not increase
with increasing carbon dioxide pressure. The surface hydrogen
concentration would decrease at higher carbon dioxide pressure
because of the dilution of carbon dioxide, reducing the
possibility of direct transformation from phenol to cyclohex-
anol, thus the selectivity to cyclohexanol did not increase.
The hydrogenation activities for cresols, derivatives of
phenol, were also investigated over 5% Rh/C (Table 3). The 5%
Rh/C catalyst also showed high activities for ring hydrogenation
of cresols and corresponding methylcyclohexanone and me-
thylcyclohexanol were formed. Among cresols, the catalysts
showed the lowest activity for p-cresol hydrogenation, which
may be due to the position of the methyl group with respect to
the hydroxy group. The position of the substituent has also
affected the selectivity pattern substantially.
Notes and references
1 World Nylon 6 & 66 Supply/Demand Report 1998, PCI-Fibers & Raw
Materials.
2 S. Narayanan and K. Krishna, Appl. Catal. A, 1998, 174, 221–229.
3 Y. Z. Chen, C. W. Liaw and I. Lee, Appl. Catal. A, 1999, 177, 1–8.
4 N. Mahata and V. Vishwanathan, Catal. Today, 1999, 49, 65–69.
5 N. Mahata and V. Vishwanathan, J. Catal., 2000, 196, 262–270.
6 S. Narayanan and K. Krishna, Appl. Catal. A, 2000, 198, 13–21.
7 P. Claus, H. Berndt, C. Mohr, J. Radnik, E. Shin and M. A. Keane, J.
Catal., 2000, 192, 88–97.
8 S. Scire, S. Minico and C. Crisafulli, Appl. Catal. A, 2002, 235,
21–31.
9 S. G. Shore, E. Ding, C. Park and M. A. Keane, Catal. Commun., 2002,
3, 77–84.
Table 2 Effect of hydrogen and carbon dioxide pressures on the
hydrogenation of phenol over 5% Rh/Ca
10 J. R. Gonzales-Velasco, M. P. Gonzalez-Marcos, S. Arnaiz, J. I.
Gutierrez-Ortiz and M. A. Gutierrez-Ortiz, Ind. Eng. Chem. Res., 1995,
34, 1031–1036.
Selectivity (%)
11 B. Subramanian, C. J. Lyon and V. Arunajatesan, Appl. Catal. B, 2002,
37, 272–292.
H2 pressure/ CO2 pressure/ Conversion
Cyclo-
hexanone
Cyclo-
hexanol
12 Chemical Synthesis Using Supercritical Fluids, ed. P. G. Jessop and W.
Leitner, Wiley, New York, 1999.
MPa
MPa
(%)
13 M. G. Hitzler and M. Poliakoff, Chem. Commun., 1997, 1667–1668.
14 M. G. Hitzler, F. R. Smail, S. K. Ross and M. Poliakoff, Org. Proc. Res.
Dev., 1998, 2, 137–146.
15 B. M. Bhanage, Y. Ikushima, M. Shirai and M. Arai, Catal. Lett., 1999,
62, 175–177.
16 M. Arai, Y. Nishiyama and Y. Ikushima, J. Supercrit. Fluids, 1998, 13,
149–153.
17 N. Mahata and V. Vishwanathan, J. Mol. Catal., 1997, 120, 267–270.
6
7.5
10
10
10
10
10
10
14
20
39
42
53
72
87
40
31
17
22
34
60
69
83
77
65
a Reaction temperature 328 K; reaction time 2 h; initial phenol 0.02 mol.
CHEM. COMMUN., 2003, 1960–1961
1961