Highly active and selective Pd-Cu-TiO2 catalyst for the direct synthesis of benzyl acetate by gas phase acetoxylation of toluene

Article abstract of DOI:10.1246/cl.2004.1238

Extremely high benzyl acetate selectivity (ca. 94%) at a toluene conversion of ca. 60% was achieved over Pd-Cu-TiO₂ catalyst for the first time by gas phase acetoxylation of toluene with good long-term stability.

Full text of DOI:10.1246/cl.2004.1238

1238
Chemistry Letters Vol.33, No.10 (2004)
Highly Active and Selective Pd–Cu–TiO2 Catalyst for the Direct Synthesis
of Benzyl Acetate by Gas Phase Acetoxylation of Toluene
A. Benhmid, K. V. Narayana, A. Martin,ꢀ B. Lucke, and M.-M. Pohl
¨
Institut fur Angewandte Chemie Berlin-Adlershof e.V.#, Richard-Willstatter-Str. 12, D-12489 Berlin, Germany
¨
¨
(Received June 30, 2004; CL-040765)
Extremely high benzyl acetate selectivity (ca. 94%) at a
toluene conversion of ca. 60% was achieved over Pd–Cu–TiO2
catalyst for the first time by gas phase acetoxylation of toluene
with good long-term stability.
The product stream was analyzed on line by GC.
The BET surface areas and pore volumes of the fresh cata-
lysts are varied in the range from 44 to 78 m2/g and 0.109 to
0.127 cm3/g depending upon the nature of promoter used.
TEM analysis of fresh catalysts with different promoters showed
that the size of Pd particles is found to be dependent on the type
of promoter used. Bigger Pd particles exhibited by Sb (ca.
10 nm) and smaller by Bi (ꢁ1 nm). The size of Pd particles dis-
played by different promoters are in the following order:
Sb > Sn > Cu > Bi. The morphology of Pd particles is mostly
spherical. The fresh Pd–Cu–TiO2 catalyst showed Pd particles
in the range of 1 to 3 nm, with narrow size distribution.
The direct oxidation of methyl aromatics to their corre-
sponding aldehydes/alcohols is often unselective as these prod-
ucts undergo further oxidation. This problem can be overcome
using suitable catalysts and appropriate auxiliary reagents to pro-
duce stable products. Vapor phase acetoxylation is one such ex-
ample for producing stable end products, particularly to produce
various industrially important esters. No suitable method is
available till date for the direct conversion of alkyl benzenes
in general and toluene to benzyl acetate (BA) in particular, in
a single step by vapor phase acetoxylation. BA is used mainly
in the perfumery, food, and chemical industry. The most of the
work reported so far on the acetoxylation of methyl aromatics
(e.g., toluene to BA) was carried out under liquid phase condi-
tions and in batch reactors.1–6 Literature survey reveals that Pd
based catalysts are being widely used for toluene acetoxylation
both in liquid and vapor phase. However, no apparent success
is achieved in terms of obtaining higher yields of BA. Further-
more, all these processes/catalysts reported so far suffer from
various drawbacks like easy deactivation, leaching problems
100
80
60
40
20
0
X-Tol
Y-BA
Y-BAL
S-BA
etc. An attempt by Eberson and Jonsson7 achieved only very
¨
Sb
Cu
Sn
Bi
Pd-TiO2
low yields of acetoxylated products (ꢁ1% per pass). Very recent
report by Komatsu et al.8 on the gas phase acetoxylation of tol-
uene over SiO2 supported different intermetallic Pd compounds
like Pd2Ge, Pd5Ga2, etc., a maximum yield of BA of around 7%
could only be achieved.
Figure 1. Influence of promoter on catalytic performance of
Pd-M-TiO2 catalysts (M = Sb, Bi, Sn, Cu); X-Tol: toluene con-
version, Y-BA: benzyl acetate yield, Y-BAL: benzaldehyde
yield, S-BA: benzyl acetate selectivity.
In this communication, we explore the influence of various
promoters on the catalytic performance of Pd acetoxylation cat-
alysts and report for the first time Pd–Cu–TiO2 as highly effi-
cient catalyst system.
The preparation of catalyst involves mainly two steps. The
first step deals with the impregnation of the promoter (e.g., Sb,
Sn, Bi, Cu) on to the TiO2 (anatase), followed by oven drying
and calcination at 400 ꢂC for 3 h. The second step involves the
impregnation of Pd source (PdCl2) to the above solid in a desired
amount followed by drying of the resulting solid mass in an oven
at 120 ꢂC for 16 h as described elsewhere.9 The sources of pro-
Our preliminary catalytic tests showed poor performance of
monometallics (i.e., Pd alone supported on TiO2)10 while the
combination of Pd with any of the promoter used in the present
study proved an amazing effect on their catalytic performance
due to synergetic effects between Pd and promoter. The influ-
ence of promoter on the catalytic performance is shown in
Figure 1, which clearly indicates that the nature of promoter
has a strong influence on the activity, selectivity, and the life
of the catalyst. Among all the promoters studied, Sb is found
to display best performance giving 68% conversion of toluene
and 85% selectivity of BA. The order of the reactivity of these
promoters is Sb > Cu > Sn > Bi. It is noteworthy that both
Sb and Sn promoted catalysts though they exhibit considerably
good performance, they were found to deactivate after some
hours of catalytic tests owing to coke deposition. However, no
such deactivation problem was observed in case of Cu and Bi
promoted catalysts, which additionally exhibit extremely high
selectivity of BA (ca. 95%) compared to the other two catalysts.
Between these two advantageous systems, Cu promoted catalyst
.
.
moters are SbCl3, SnCl2 2H2O, BiCl3, and CuCl2 2H2O. The
content of Pd is kept constant at 10 wt % in every case, while
the contents of all these promoters are selected to be 8 wt % ex-
cept Bi (7 wt %). The catalytic tests were carried out in a fixed
bed stainless steel reactor. 1 mL of catalyst is loaded in the reac-
tor and the reaction is performed at 2 bars. The catalyst was ac-
tivated in situ under airflow at 300 ꢂC for 2 h prior to the activity
tests. The organic feed mixture of toluene and acetic acid in the
mole ratio of 1:4 was pumped to the reactor using a HPLC pump.
Copyright Ó 2004 The Chemical Society of Japan

Chemistry Letters Vol.33, No.10 (2004)
1239
displayed better performance giving higher activity and hence
this Pd–Cu–TiO2 catalyst has become the main focus of the pres-
ent study. This catalyst gave 60% conversion of toluene and 94%
selectivity of BA without any deactivation. The major by-prod-
uct is benzaldehyde (BAL) and the balance is total oxidation
products (COx). In addition, the catalytic performance of pure
fresh
Pd–TiO2 catalyst without promoter (XTol ¼ 2:4% & YBA
¼
2:2%) is also given for better comparison on the influence of
promoter.
used
50 nm
20 nm
Figure 2 demonstrates the time-on-stream behavior of
Pd–Cu–TiO2 catalyst over a period of 160 h. It is evident from
the figure that the catalyst is found to exhibit very low initial
activity (<5%), which is increased significantly with time-on-
stream reaching maximum at 60% after about 60 h and then
displayed consistent performance throughout the catalytic runs.
It is seen from TEM analysis of used catalysts that the size of
Pd particles is significantly growing during the course of
reaction and it is also observed in earlier experiments that
bigger Pd particles are much favorable for better performance.
This might be the reason for attaining steady-state conditions
only after a certain period of time, which is the required time
for Pd particle growth to an appropriate size.
Figure 3. TEM images of the fresh and used Pd–Cu–TiO2
catalysts.
associated with bigger Pd particles, which is in good agreement
with earlier investigations.10
In summary, this is a first report on the direct synthesis of
BA with higher yield (ca. 55%) and extremely high selectivity
of BA (94%) by gas phase acetoxylation. Furthermore, incorpo-
ration of Cu and Bi has successfully solved the problem of the
catalyst deactivation, while Cu is found to be more effective than
Bi. However, additional experiments are necessary for better un-
derstanding of the nature of the active sites and the precise role
of the Cu that prevents catalyst deactivation.
100
80
References and Notes
#
A member of the EU-funded Coordination Action of
Nanostructured Catalytic Oxide Research and Development
in Europe (CONCORDE).
60
1
D. R. Bryant, J. E. McKeon, and B. C. Ream, J. Org. Chem.,
33, 4123 (1968).
2
3
J. M. Davidson and C. Triggs, J. Chem. Soc. A, 1968, 1331.
S. K. Ivanov and S. K. Tanielyan, Oxid. Commun., 7, 69
(1984).
E. Benazzi, C. J. Cameron, and H. Mimoun, J. Mol. Catal.,
69, 299 (1991).
E. Benazzi, C. J. Cameron, and H. Mimoun, J. Catal., 140,
311 (1993).
S. Tanielyan and R. Augustine, J. Mol. Catal., 87, 311
(1994).
X-Tol
Y-BA
S-BA
40
20
0
4
5
6
7
8
9
0
30
60
90
120
150
180
Time / h
Figure 2. Time-on-stream analysis of 10% Pd–Cu–TiO2 cata-
lyst (denotation as for Figure 1).
L. Eberson and L. Jonsson, Acta Chem. Scand., Ser. B, 28,
597 (1974).
¨
Used Pd–Cu–TiO2 catalyst exhibited significantly bigger Pd
particles even up to 150 nm indicating particle growth during the
course of reaction due to agglomeration (Figure 3). Similar such
growth of Pd particles is also observed in case of the catalysts
with other promoters such as Sb, Sn, Bi. Based on all these ob-
servations it seems more probable that higher catalyst activity is
T. Komatsu, K. Inaba, T. Uezono, A. Onda, and T. Yashima,
Appl. Catal., A, 251, 315 (2003).
A. Benhmid, K. V. Narayana, A. Martin, and B. Lucke,
¨
Patent submitted to German patent office., 2004.
10 A. Benhmid, K. V. Narayana, A. Martin, and B. Lucke,
¨
Chem. Commun., 2004, in press.
Published on the web (Advance View) August 28, 2004; DOI 10.1246/cl.2004.1238