New direct hydroxylation of benzene with oxygen in the presence of hydrogen over bifunctional ion-exchange resins

Article abstract of DOI:10.1039/b204033f

Benzene was hydroxylated directly to phenol with oxygen in the presence of hydrogen over Pt/Pd containing acid resins such as Amberlyst and Nafion/silica composites; over the latter materials the best results were obtained.

Full text of DOI:10.1039/b204033f

New direct hydroxylation of benzene with oxygen in the presence of
hydrogen over bifunctional ion-exchange resins
Wilhelm Laufer and Wolfgang F. Hoelderich*
Department of Chemical Technology and Heterogeneous Catalysis, University of Technology RWTH
Aachen, Worringerweg 1, 52074 Aachen, Germany. E-mail: Hoelderich@rwth-aachen.de;
Fax: +49-(0)241-8022291; Tel: +49-(0)241-8026560
Received (in Cambridge, UK) 26th April 2002, Accepted 26th June 2002
First published as an Advance Article on the web 8th July 2002
Benzene was hydroxylated directly to phenol with oxygen in
the presence of hydrogen over Pt/Pd containing acid resins
such as Amberlyst and Nafion/silica composites; over the
latter materials the best results were obtained.
the reactor was cooled down to 15 °C. Samples were analyzed
with HPLC (Merck/Hitachi L-6200A equipped with a L-4500
2
Diode Array Detector) using MeOH–H O mixtures as the
eluent.
The best results were obtained over the Nafion based material
SAC 13, with a maximum benzene conversion of 4.25% (see
Table 1 and Fig. 1; over metal free SAC 13 no conversion could
be found). As side products, resulting from the oxidation of
phenol, hydroquinone, quinone and catechol were found. Over
the strong acid resin Amberlyst there was some phenol
formation. The conversion to phenol was also found to be
dependent on the amounts of palladium and platinum.
The conventional industrial process for the production of the
important intermediate phenol, the 3 step Hock process, suffers
from many disadvantages as the formation of acetone as an
inevitable couple product, the production of waste-water as well
as a low overall yield. The direct hydroxylation of benzene has
therefore been attracting great interest and is one of the more
challenging subjects in catalysis research at the present time.
One route is the direct gas phase oxidation with O
2
or air;
The performance of the Pd/Pt/Nafion-silica catalyst was
probably a result from the strong acidic properties, combined
with the hydrophobic/hydrophilic properties. The SAC 13
based catalyst was superior compared to the other composite
materials (see Fig. 1). Its higher activity could not be explained
with a higher surface area or a larger cumulative pore volume,
unfortunately this process leads to low selectivities because of
the formation of maleic acid and the oxidation to CO
2
. The
oxidation over Cu or Cu/Pd catalysts in an acidic medium
resulted in low conversions, and the catalysts had to be
1
regenerated under H
2
. Better results were obtained using O
2
2
2
21
over palladium and heteropolyacids. An alternative is the use
of N O over H-[Fe]-MFI or steamed H-[Al]-MFI zeolites. The
O, reacts
as the BET surface area ranged from 15 m g for SAC 80 to
2
21
3
21
2
21
2
158 m g for SAC 15 (pore volume 36 cm g ), 92 m g
3
21
2
21
surface a-oxygen, formed upon decomposition of N
2
for SAC 13 (pore volume 21 cm g ), up to 270 m g for
SAC 5. The better performance of the SAC 13 based catalyst
was possibly caused by a higher dispersion of the Nafion resin
over the silica matrix, resulting in an optimal distribution of the
sulfonic acid groups over the catalyst.
with benzene to phenol, obtaining conversions of ca. 30% and
selectivities of over 95%.3 The activity seems to be related to
the Lewis acidity. These catalysts deactivate rapidly due to
,4
4
5
6
strong coking. Sasaki et al. and Miyake et al. used Pd/SiO
Pd or Pt/V /SiO based systems in acetic acid, and Tatsumi et
al. investigated the use of Pd containing TS-1 in strong acids
such as HCl in the direct oxidation with O and H , forming
in situ. With these systems only low phenol yields were
obtained.
2
,
2
O
5
2
The addition of small amounts of NaBr resulted in an increase
in the phenol yield. Without NaBr yields of only 0.5% were
7
2
1
2
2
obtained, whereas the addition of 0.5 mmol l NaBr resulted in
yields of up to 4.2%. This was probably a result of a promoting
2 2
H O
1
2
2 2
or stabilizing effect of the bromide on the H O formation.
Here the use of Pd and Pt containing acid resins such as
Amberlyst 15 and Nafion-silica composites for the direct
Table 1 Catalyst compositions and results of the hydroxylation of benzene
with O and H . Conditions: solvent water, batch wise, 1000 kPa O , 1000
, 1000 kPa N
hydroxylation of benzene with an O
2
–H
2
gas mixture was
2
2
2
kPa H
2
2
. Catalyst reduction: 5 vol% H
2
at 150 °C
investigated. Nafion is an ionomer with an inert polymeric
backbone and highly acidic sulfonic groups, and has hydro-
phobic (–CF CF –) and hydrophilic (–SO H) regions. Nafion
2 2 3
Carrier
Metal content [wt%]
Phenol [mmol]
silica nanocomposites, produced by the entrapment of Nafion
particles in a porous silica network, seem promising replace-
ments for hazardous conventional strong homogeneous acids.8
The choice of the type and the amounts of metal was based on
previous experiments on the direct epoxidation of propylene
Nafion/SiO
Amberlyst 15
Amberlyst 15
2
SAC 13
0.6% Pd + 0.6% Pt
1.0% Pd + 0.01% Pt
2.5% Pd + 1.2% Pt
1.0881
0.0619
0.2274
,9
2 2
–O
over Pd/Pt/TS-1 type materials.10,11
with H
g of Nafion-silica composites with various Nafion contents
e.g. SAC 13 contained 13 wt% Nafion), kindly provided by Du
Pont de Nemours & Company, Wilmington and Amberlyst 15
2
(
(
Aldrich) were suspended for 24 h at 80 °C in 20 g of an aqueous
Pd(NH ](NO and [Pt(NH ]Cl solution. The materials
[
3
)
4
3
)
2
3
)
4
2
2
1
were reduced at 150 °C (heating rate 1 °C min ) under H
vol% H in N , or N and characterized by ICP AES
Spectroflame D) and nitrogen sorption (Micromeretics ASAP
010).
.2 g of catalyst, 20 g solvent containing 0.5 mmol l21 NaBr
2
, 5
2
2
2
(
2
0
and 2 g benzene were charged into a 200 ml autoclave. For batch
experiments the autoclave was subsequently pressurized
(
700–3000 kPa) with O
2
, H
2
2
and N
2
2
with MFC’s. For semi-
were added con-
Fig. 1 Influence of the Nafion content in the composite on phenol formation.
Conditions: solvent 15 g methanol + 5 g H O, batchwise, 1000 kPa O , 1000
kPa H , 1000 kPa N . Catalyst: Pd/Pt/Nafion/SiO , 0.5 wt% Pd + 0.5 wt%
continuous experiments O
, H and N
2
2
2
tinuously, maintaining a constant pressure with an overflow
valve. The slurry was heated to 43 °C while stirring. After 3 h
2
2
2
2
Pt, reduced under 5 vol% H .
1
684
CHEM. COMMUN., 2002, 1684–1685
This journal is © The Royal Society of Chemistry 2002

The addition of higher amounts of NaBr resulted in a strong
Table 4 Influence of the reduction atmosphere on the catalytic performance.
Conditions: semi-continuous, 1.3 l h H , 1.3 l h O , 1.4 l h N , 1400
2 2 2
kPa
21
21
21
2
1
decrease of the yield: upon addition of 60 mmol l NaBr the
yield dropped to only 0.2%, which was probably due to a
decrease of the number of acid sites by the ion-exchange with
sodium.
Q
Phenol Phenol HQ sel. CAT sel.
yield [%] sel. [%] [%] sel. [%] [%]
The reaction parameters were optimized for an SAC 13 based
catalyst containing 0.5 wt% Pd and 0.5 wt% Pt, reduced under
Reduction method
Unreduced
5
vol% H
2
. The highest amounts of phenol were formed
0.2
6.9
7.6
7.5
100
59
56
—
31
36
31
—
9
8
—
—
—
0.1
between 30–40 °C, with increasing amounts of higher oxidized
compounds, especially hydroquinone, at increasing tempera-
tures (see Fig. 2). At temperatures below 20 °C only very low
phenol yields were found.
N
2
5 vol% H
2
in N
2
H
2
57
11
O
2
2
and H starting concentrations. When keeping the gas
concentrations low there was less side product formation.
Prior to the reaction the noble metal catalysts were reduced.
In earlier investigations on the direct epoxidation of propene an
autoreduction at 150 °C under nitrogen resulted in the most
active catalysts.1
0,11,13
As Nafion is destroyed at ca. 250 °C a
similar reduction temperature was used, as at 150 °C the Pd- and
Pt-tetramine complexes are already partially decomposed. The
influence of the reduction atmosphere was investigated; the
catalysts were calcined under H
2
2 2 2
, a 5% H in N mixture, and
N
. In the latter method the metals were autoreduced by the
ammonia resulting from the decomposition of the tetramine
complexes. Although the catalysts have to be reduced in order
to obtain an active material (over the unreduced material only
traces of phenol were found), there was little influence of the
different reduction methods on the catalytic performance (see
Fig. 2 Influence of the reaction temperature on the catalytic performance;
HQ: hydroquinone, Q: quinone, CAT: catechol. Conditions: semi-con-
2
Table 4). The yields over the under H reduced materials were
tinuous, 1.3 l h2
1
21
21
2
H , 1.3 l h
2
O , 1.4 l h
2
N , 700 kPa.
somewhat higher than that over the autoreduced material, which
might be due to an incomplete decomposition of the amines.
The selectivities to phenol and the side products hydroquinone,
catechol and quinone were comparable. After regenerating the
spent catalyst with a methanol wash the materials regained their
activity for at least two recycles, although the yield dropped to
ca. 3% with comparable selectivities.
Benzene could be hydroxylated directly to phenol over Pt/Pd
impregnated acid resins like Amberlyst or Nafion/silica com-
posites with oxygen in the presence of hydrogen. The best
results were obtained over a Nafion/silica composite containing
13% Nafion between 30–40 °C in a 3+1 water/methanol
mixture. The catalyst had to be reduced at 150 °C; however, the
reduction atmosphere had little influence on the performance. A
The phenol yield increased with the polarity of the solvents in
the order acetone < 2-propanol < ethanol < methanol <
water, which was somewhat surprising as the solubility of
benzene decreases in these solvents (see Table 2). With these
increasing phenol yields the solvents were more protic. This
again is an indication that strong acid sites are needed for this
hydroxylation. While using water, hydroquinone was formed in
high quantities as well. Without a solvent only very low yields
could be obtained. A 3+1 water/methanol mixture was found to
result in the highest phenol selectivities.
Table 2 Influence of the solvent on the phenol formation. Conditions: batch
2 2 2
wise, 1000 kPa O , 1000 kPa H , 1000 kPa N
semi-continuous reactor system keeping the O
2 2
and H
concentration constant at 700–1400 kPa resulted in higher
Solvent
Yield [%]
Selectivity [%]
phenol yields than when using a batch reactor.
No solvent
Water
Water/methanol (3+1)
Water/methanol (1+3)
Methanol
Ethanol
Isopropanol
Acetone
< 0.1
4.2
4.2
2.5
0.5
0.2
< 0.1
—
100
41
56
54
70
100
100
—
Notes and references
1
2
3
4
5
6
A. Kunai, T. Kitano, Y. Kuroda, J. LiFen and K. Sasaki, Catal. Lett.,
990, 4, 139.
L. C. Passoni, F. J. Luna, M. Wallau, R. Buffon and U. Schuchardt, J.
Mol. Catal., 1998, 134, 129.
G. I. Panov, A. K. Uriarte, M. A. Rodkin and V. I. Sobolev, Catal.
Today, 1998, 41, 365.
J. L. Motz, H. Heinichen and W. F. Hölderich, J. Mol. Catal., 1998, 136,
1
1
75.
Using a low constant pressure of 700 or 1400 kPa with O
2
+H
2
K. Sasaki, A. Kunai, S. Ito, F. Iwasaki and M. Hamada (Tosoh Corp.),
JP 02138233, 1990.
T. Miyake, M. Hamada, Y. Sasaki and K. Sekizawa (Tosoh Corp.), JP
ratios of 1+1 up to 2+1 in a semi continuous mode less higher
oxidized side products were formed. Starting with a pressure of
3
000 kPa in a batch mode resulted in an increased side product
0
3178946, 1991.
formation (see Table 3), which was probably due to the higher
7 T. Tatsumi, K. Yuasa and H. Tominaga, J. Chem. Soc., Chem.
Commun., 1992, 1446.
8
9
F. J. Waller and R. W. van Scoyoc, Chemtech., 1987, 17, 438.
M. A. Harmer, Q. Sun, A. J. Vega, W. E. Farneth, A. Heidekum and W.
F. Hölderich, Green Chem., 2000, 2, 7.
Table 3 Influence of the oxidation procedure on the catalytic performance.
2 2 2
Conditions: batch wise, 1000 kPa O , 1000 kPa H , 1000 kPa N , semi-
continuous: 1.3 l h2
1
21
21
2
H , 1.3 l h
2
O , 1.4 l h
N
2
at 700 or 1400 kPa
1
1
0 W. Laufer, R. Meiers and W. F. Hölderich, in Proc. 12th Int. Zeolite
Conf., July 1998, eds. M. M. J. Treacy, B. K. Marcus, M. E. Bisher and
J. B. Higgins, Material Research Society, Warrendale, Pennsylvania,
999, 1351.
1 W. Laufer and W. F. Hölderich, Appl. Catal. A: General, 2001, 213,
63.
Q
Phenol Phenol HQ sel. CAT sel.
yield [%] sel. [%] [%] sel. [%] [%]
1
Procedure
1
Batch, 3000 kPa
Semi-continuous, 700 kPa
Semi-continuous, 1400 kPa
4.2
3.2
5.6
40
63
45
52
26
49
5
10
5
2.5
0.5
0.1
1
1
2 L. W. Gosser (E.I. Du Pont de Nemours Co.), EP 0342047, 1989.
3 R. Meiers, U. Dingerdissen and W. F. Hölderich, J. Catal., 1998, 176,
3
76.
CHEM. COMMUN., 2002, 1684–1685
1685