Direct formation of hydrogen peroxide from H2/O2 using a gold catalyst

Article abstract of DOI:10.1039/b205248m

Supported Au catalysts are very selective for the direct formation of hydrogen peroxide from H₂/O₂ mixtures at 2°C; the rate of H₂O₂ synthesis is markedly increased if Au-Pd alloy nanoparticles are generated by

Full text of DOI:10.1039/b205248m

Direct formation of hydrogen peroxide from H /O using a gold catalyst
2
2
a
a
c
c
Philip Landon, Paul J. Collier, Adam J. Papworth, Christopher J. Kiely and Graham J.
Hutchings*
a
a
Department of Chemistry, Cardiff University, P.O. Box 912, Cardiff, UK CF10 3TB.
E-mail: hutch@cardiff.ac.uk
Johnson Matthey Technology Centre, Blounts Court, Sonning Common, Reading, UK RG4 9NH
Departments of Engineering and Chemistry, University of Liverpool, Liverpool, UK L69 3BX
b
c
Received (in Cambridge, UK) 29th May 2002, Accepted 1st August 2002
First published as an Advance Article on the web 14th August 2002
Supported Au catalysts are very selective for the direct
formation of hydrogen peroxide from H /O mixtures at 2
overcome by using supercritical media due to the enhanced
14
2
2
solubility of H
containing 5 wt% metal were prepared by coprecipitation.
These catalysts were evaluated for the synthesis of hydrogen
peroxide using supercritical CO (35 °C, 9.7 MPa) using a
2
.
Au/ZnO, Pd/ZnO and Au:Pd/ZnO catalysts
1
5
°
2 2
C; the rate of H O synthesis is markedly increased if Au–
Pd alloy nanoparticles are generated by the addition of Pd.
2
Selective oxidation is important in the synthesis of fine
chemicals and intermediates. Recently, there has been much
standard procedure.16 The results are shown in Table 1 and it is
apparent that the Au/ZnO and Au:Pd/ZnO catalysts exhibit
some hydrogen peroxide synthesis, albeit at a low rate.
However, the Pd catalyst only generated water as a product. At
this temperature it was considered that the hydrogen peroxide
formed was relatively unstable with respect to decomposition or
1
interest in the design of new heterogeneous catalysts for
selective oxidation under ambient conditions, and these typi-
cally use hydrogen peroxide as the oxidant.2 At present,
hydrogen peroxide is produced by the sequential hydrogenation
,3
4
and oxidation of alkyl anthraquinone and global production is
hydrogenation (see Scheme 1). This was confirmed in a separate
6
25
ca. 1.9 3 10 tonnes per annum. This process is currently only
2 2
experiment in which H O (16.8 3 10 mol, 0.5 wt%) was
4
economic on a large scale (4–6 3 10 tpa), whereas it is often
stirred with and without the Pd catalyst under the same
conditions in supercritical CO . When the Pd catalyst was
required practically on a much smaller scale. In view of this,
there is considerable interest in the direct manufacture of
hydrogen peroxide from the catalysed reaction of hydrogen and
oxygen.5 At present, some success has been achieved using Pd
as a catalyst, especially when halides are used as promoters.7
Typically, dilute solutions of hydrogen peroxide are produced,
and earlier studies indicate that the Pd catalyst can be combined
with an oxidation catalyst, TS-1, so that the hydrogen peroxide
2
present ca. 30% of the hydrogen peroxide decomposed in 1 h
and, even in the absence of a catalyst, ca 10% was decomposed.
Hence, although using supercritical CO as the reaction medium
2
may have overcome the diffusion limitation, the inherent
instability of hydrogen peroxide at the elevated temperature
required to achieve supercritical conditions mitigates against
the use of this medium.
Subsequently, experiments were conducted at significantly
lower temperatures (2 °C) and the results, also shown in Table
1, indicate that hydrogen peroxide can be formed at a high rate
,6
,8
9
produced is used in situ. A further recent development in
catalysis has been the marked interest in using supported Au as
an active oxidation10 and hydrogenation catalyst, and very
recently we have been shown that Au supported on carbon can
selectively oxidise glycerol to glyceric acid.12 Supported Au
catalysts have been investigated for the oxidation of propene to
11
for the supported Au catalyst. The selectivity for H
2 2
O for the
Au/Al catalyst was determined to be 53%. In this case, the
2 3
O
17
catalysts were prepared by impregnation and they were tested
2 2
propene oxide using O /H
mixtures,13 and it is considered that
a surface hydroperoxy species may be formed as the oxidant. To
date, there are no reported studies concerning the use of
supported Au catalysts for the direct synthesis of hydrogen
2 2
peroxide from O /H . We have now addressed this point and, in
this communication, we show that supported Au catalysts can
be very effective and, furthermore, the rate of hydrogen
peroxide formation can be significantly enhanced by the use of
a supported Au/Pd alloy.
Our initial approach to the design of a direct hydrogen
oxidation process for the synthesis of hydrogen peroxide, was to
use supercritical CO
earlier studies with Pd catalysts have indicated that H
is a significant problem and this can be expected to be largely
2
as a reaction medium. This is because
2
diffusion
Scheme 1
2 2 2 2
Table 1 Formation of H O from the reaction of H /O over Au and Pd catalysts
b
Temperature/
°C
Pressure/
MPa
O
2
/H
2
2 2
H O
Catalyst
Solventa
mol ratio
mmol g(catalyst) h21
21
Au/Al
Au+Pd (1+1)/Al
Pd/Al
Au/ZnO
Au+Pd (1+3)/ZnO
Au+Pd (1+1)/ZnO
Au+Pd (3+1)/ZnO
Pd/ZnO
2
O
3
CH
CH
CH
SCCO
SCCO
SCCO
SCCO
SCCO
3
3
3
OH
OH
OH
2
2
2
2
2
2
2
2
35
35
35
35
35
3.7
3.7
3.7
9.2
9.2
9.2
9.2
9.2
1.2
1.2
1.2
1.0
1.1
0.8
0.9
1.3
1530
4460
370
9
7
12
8
2 3
O
2
O
3
0
a
2 2 2 2
SCCO = supercritical CO O formation averaged over 30 min experiment.
, details of experimental methods given in refs. 16 and 18 b Rate of H
2
058
CHEM. COMMUN., 2002, 2058–2059
This journal is © The Royal Society of Chemistry 2002

View Online
for hydrogen peroxide formation at 2 °C using methanol as
solvent.18 The supported Au catalyst produces more hydrogen
peroxide than the supported Pd catalyst. However, more
interestingly, the supported Au+Pd (1+1 by wt) catalyst
produces significantly more than the pure Au catalyst. This
indicates a synergistic effect of Pd acting as a promoter for the
Au catalyst. To demonstrate that the supported gold catalysts
functioned as wholly heterogeneous catalysts, an experiment
was carried out using a supported gold catalyst at 2 °C and the
observed with the Au catalysts results from either a decrease in
the rate of H decomposition or an enhancement in the rate of
formation and this will be the subject of further study.
This work formed part of a research project set up and
executed under the Institute of Applied Catalysis Foresight
Challenge Research Programme and we thank the Institute of
Applied Catalysis and the EPSRC for funding.
2 2
O
2 2
H O
yield of H
catalyst was removed by careful filtration and the solution was
used for a second experiment using O /H . No further H was
formed and this confirms that the formation of hydrogen
peroxide involves gold acting as a wholly heterogeneous
catalyst. To determine if the supported Au+Pd catalyst comprise
AuPd alloy particles or if the two components exist separately,
the material was examined by scanning transmission electron
microscopy (STEM) as shown in Fig. 1.19 In the bright field
image (top left) it is difficult to distinguish the metal particles
2 2
O was determined. Following this reaction, the gold
Notes and references
1
2
3
M. Benson and P. Gallezot, Catal. Today, 2000, 57, 127.
R. A. Sheldon, Stud. Surf. Sci. Catal., 1997, 110, 151.
G. Jenzer, T. Mallet, M. Maciejewski, F. Eigenmann and A. Baiker,
Appl. Catal. A, 2001, 208, 125.
2
2
2 2
O
4
H. T. Hess, in Kirk-Othmer Encyclopedia of Chemical Engineering, ed.
I. Kroschwitz and M. Howe-Grant, Wiley, New York, 1995, ch. 13, p.
961.
5 J. Van Weynbergh, J.-P. Schoebrichts and J.-C. Colery, US Pat.,
5447706, 1995.
6
7
8
9
B. Zhou and L.-K. Lee, US Pat., 6168775, 2001.
J. Wanngard, Eur. Pat., 0816286 Al, 1998.
K. T. Chuang and B. Zhou, US Pat., 5338531, 1984.
T. Tatsumi, K. Yuasa and H. Tominaga, J. Chem. Soc., Chem.
Commun., 1992, 1446.
from the Al
2 3
O support. However, the annular dark field image
(
top right) gives strong atomic number (Z) contrast and shows
the metal particles as bright spots which range from 2 to 9 nm
in diameter. Energy dispersive X-ray (EDS) maps of the same
a a
area using the Al K (1.486 eV) O K (0.525 eV), Au L (9.712
1
1
0 L. Prati and M. Rossi, J. Catal., 1998, 176, 552.
1 J. E. Bailie and G. J. Hutchings, Chem. Commun., 1999, 2151.
eV) and Pd L (2.838 eV) signals are also shown as a montage in
Fig. 1. It is clear that the Au and Pd signals are spatially
superimposed indicating that the metal nanoparticles are in fact
AuPd alloys. EDS point analysis from more than 200 particles
have been carried out which show that all the particles are alloys
and although there is some compositional variation from
particle-to-particle, the average composition is 50 wt% Au+50
wt% Pd as expected.
12 S. Carrettin, P. McMorn, P. Johnston, K. Griffin and G. J. Hutchings,
Chem. Commun., 2002, 696.
13 T. Hayashi, K. Tanaka and M. Haruta, J. Catal., 1998, 178, 566.
1
4 G. J. Hutchings, I. H. Stewart and E. G. Derouane, ‘Catalytic reactions
at supercritical conditions’, Curr. Top. Catal., 1999, 2, 17.
1
5 Au/ZnO, Pd/ZnO and Au-Pd/ZnO catalysts containing 5 wt% metal
were prepared by co-precipitation. An aqueous solution of
Pd(NO
3 2 2
) ·6H O
(Strem, 99.999%) and/or HAuCl
4 2
·3H O (Strem,
This study demonstrates that supported Au catalysts are
effective for the direct synthesis of hydrogen peroxide from
hydrogen and oxygen. In particular, Au catalysts may provide a
significant improvement over Pd catalysts that have been
9
9.9%) and Zn(NO
3
)
2
·H O (Aldrich, 99.999%) which contained the
2
calculated amounts of Pd and/or Au to give the desired loading on ZnO
was heated to 80 °C. Na CO3 (Aldrich, 1 mol 121) was added with
2
stirring, until pH 9 was reached and the mixture was stirred for a further
45 min. The catalyst was washed with hot deionised water and
recovered by filtration, dried at 120 °C and calcined (400 °C; 3 h) and
investigated previously.4–9 At the present time, it is not possible
to comment on whether the improved hydrogen peroxide yield
2
then reduced in a flow of 5% H in Au at 400 °C.
1
6 Catalyst testing was carried out in a Parr stainless steel autoclave with
a nominal volume of 50 ml. Typically the catalyst was charged with
catalyst (0.05 g), solvent, purged three times with CO
filled with 5% H /CO and 25% O /CO to give the required O
ratio. For the experiments in supercritical CO , additional CO was then
added to a total pressure of 9.2 MPa using a Milton Roy dosing pump.
The reaction mixture was stirred (1200 rpm) at 35 °C for 1 h. H and O
were analysed by on-line gas chromatography and H was deter-
mined at the end of the experiment using a titration method with
Ce(SO
7 Au/Al
metal were prepared using incipient wetness impregnation of g-Al
Condea SCF-140) with Pd(NO ·6H O (Strem, 99.999%) and/or
HAuCl ·3H O (Strem 99.9%). The catalysts were dried, calcined and
2
(3 MPa) and then
2
2
2
2
/H
2 2
2
2
2
2
2 2
O
4 2
) .
1
2 3 2 3 2 3
O , Pd/Al O and Au:Pd/Al O , catalysts containing 5 wt%
2 3
O
(
3
)
2
2
4
2
reduced as specified in ref. 15.
1
1
3
8 Experiments were carried out at 2 °C using aqueous methanol (CH OH,
5
2
.6 g; H O, 2.9 g) as solvent as otherwise described in ref. 16.
9 Samples were analysed in a VG systems HB601 UX scanning
transmission electron microscope operating at 100 kV. The microscope
was fitted with an Oxford Instruments INCA TEM 300 system for
energy dispersive X-ray (EDS) analysis. Samples for STEM examina-
tion were prepared by dispersing the catalyst powder in high purity
ethanol, then allowing a drop of the suspension to evaporate on a holey
carbon microscope grid.
Fig. 1 Bright field (top left) and annular dark field (top right) STEM images
of the Au Pd/Al
maps of the same area taken with Al K
2
O
3
catalyst. The lower montage shows a series of X-ray
, O K , Au L and Pd L signals.
a
a
CHEM. COMMUN., 2002, 2058–2059
2059