Catalysis Communications 8 (2007) 247250

www.elsevier.com/locate/catcom

Direct synthesis of hydrogen peroxide from H2 and O2 and

in situ oxidation using zeolite-supported catalysts
a,b,*
Gang Li , Jennifer Edwards b, Albert F. Carley b, Graham J. Hutchings b

a
Department of Catalysis Chemistry and Engineering, State Key Laboratory of Fine Chemicals, Dalian University of Technology,
158-39, Zhongshan Road, Dalian 116012, China
b
Department of Chemistry, Cardi University, Cardi CF10 3AT, United Kingdom

Received 21 January 2006; received in revised form 13 June 2006; accepted 14 June 2006
Available online 27 June 2006

Abstract

Four microporous materials, zeolites HZSM-5, Y, Beta and TS-1, were used as the supports to prepare supported gold catalysts using
impregnation or deposition precipitation. The gold catalysts were tested in the direct synthesis of hydrogen peroxide from H2 and O2 and
for CO oxidation. The eect on the catalytic activity of dierent metal (e.g., Pd, Pt, Cu, Ag, Rh or Ru) on the synthesis of hydrogen
peroxide was also tested. Organic substrates, such as cyclohexane or cyclooctene, were introduced to investigate the possibility of
in situ H2O2 oxidation with these catalysts.
 2006 Elsevier B.V. All rights reserved.

Keywords: Gold catalysis; Hydrogen peroxide; In situ oxidation

1. Introduction solutions of hydrogen peroxide when required for use in

Hydrogen peroxide is an environmentally friendly oxi-
dant that plays an important role in green chemistry.
The commercial production of H2O2 is a circuitous process
that involves the catalytic hydrogenation of an alkylan-
throquinone to the corresponding hydroquinone, followed
by its treatment with O2 to produce H2O2 and the original
anthroquinone. However, there are problems associated
with the anthraquinone route, including the cost of the qui-
none solvent system and the requirement for periodic
replacement of anthraquinone due to hydrogenation. In
addition, the process is economically viable only on a rela-
tively large scale (46 104 tonne/annum). This necessi-
tates the transportation and storage of concentrated
*
Corresponding author. Address: Department of Catalysis Chemistry
and Engineering, State Key Laboratory of Fine Chemicals, Dalian
University of Technology, 158-39, Zhongshan Road, Dalian 116012,
China. Tel./fax: +86 411 8368 9065.
E-mail address: liganghg@dlut.edu.cn (G. Li).
the ne chemicals industry, because only relatively small
amounts are required at any one time. Hence, the develop-
ment of a new, highly ecient, and smaller-scale manufac-
turing process for H2O2 is of signicant commercial
interest. The direct synthesis of hydrogen peroxide from
the oxidation of H2 with O2 has been studied for many
years and until recently the catalysts used in these investi-
gations have been based on Pd [16]. However, it has been
shown recently that AuPd alloys supported on alumina
can give signicant improvements in the rate of hydrogen
peroxide formation when compared with the Pd only cata-
lyst [7,8]. Subsequently, Okumura et al. [9] reported that
the Au/AC, Au/MCM-41, and Au/SiO2 catalysts prepared
by gas-phase grafting (GG) methods can be very active for
the direct production of H2O2 from H2 and O2. The rate of
H2O2 formation over Au/SiO2 is 1.601 mmol/h/g-cat when
the reaction is carried out at 288 K and 1.0 MPa. Ishihara
et al. [10] studied the eects of additives on Au/SiO2 cata-
lyst and found that the addition of a small amount of Pd is
eective for increasing the H2O2 formation rate. Edwards

1566-7367/$ - see front matter  2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2006.06.021
248 G. Li et al. / Catalysis Communications 8 (2007) 247250

et al. [11] reported that the H2O2 productivity of 5% Au/ stirrer (02000 rpm) and provision for measurement of
a-Fe2O3 and 2.5% Au2.5% Pd/a-Fe2O3 prepared by temperature and pressure. Typically, the autoclave was
impregnation is 0.54 and 16 mol/h/kg-cat, respectively. charged with catalyst (0.01 g unless otherwise stated), sol-
Subsequently, TiO2 supported AuPd catalysts were vent (5.6 g MeOH and 2.9 g H2O), purged three times with
reported to be more active and the AuPd particles were CO2 (3 MPa) and then lled with 5% H2/CO2 and 25% O2/
found with a coreshell structure, with Pd concentrated CO2 to give a hydrogen to oxygen ratio of 1:2, at a total
on the surface [12]. Zeolite HZSM-5 and zeolite Y were pressure of 3.7 MPa at 20 C. Stirring (1200 rpm) was
also used as the supports for Au and the catalysts can give started on reaching the desired temperature, and experi-
comparable rates of hydrogen peroxide formation to alu- ments were run for 30 min. H2O2 yield was determined
mina-supported Au catalysts prepared using a similar by titration of aliquots of the nal ltered solution with
impregnation method although they are not as stable acidied Ce(SO4)2 (7 10 3 mol/l). Ce(SO4)2 solutions
[13]. In this paper, the work on the use of zeolites as sup- were standardised against (NH4)2Fe(SO4)2 6H2O using
ports has been extended and four microporous materials, ferroin as indicator.
zeolites HZSM-5, Y, Beta and TS-1, have been used as sup- The in situ H2O2 oxidation was carried out in the same
ports to prepare catalysts using impregnation or deposition autoclave using a similar procedure, but organic substrate
precipitation. The catalysts were tested in the direct synthe- (cyclohexane or cyclohexene) was added at the start of
sis of hydrogen peroxide from H2 and O2 and in situ oxida- the reaction. The products of the reaction were analysed
tion of hydrocarbons using the H2O2 generated in this using a gas chromatograph (Varian Star 3400 CX) tted
system was also investigated. with a DB-5 column and a ame ionization detector (FID).
The catalytic activity for CO oxidation was determined
2. Experimental in a xed bed quartz micro-reactor, operated at atmo-
spheric pressure. The feed consisted of 1% CO in N2,
2.1. Catalyst preparation CO2 and H2 and 5% O2 in N2. The typical reaction condi-
tion is catalyst 50 mg, 1% CO mixed gas 10 ml/min, 5% O2/
Supported catalysts were prepared using impregnation N2 1 ml/min and temperature 80 C. Catalysts were tested
method using an aqueous solution of HAuCl4 3H2O (John- for a minimum of 300 min and analysis of the reaction
son Matthey). For the 2.5 wt% Au-supported catalyst the product was carried out using on-line gas chromatography.
detailed procedure was as follows. Aqueous solution (5 ml) Conversion was calculated on the basis of CO2 concentra-
of HAuCl4 3H2O (2 g dissolved in 100 ml water) was added tion in the euent and carbon balances were 100 2%.
to the support (1.9 g) and stirred at room temperature. Four
zeolite supports were used namely, zeolite HZSM-5 (Grace 3. Results and discussion
ZL5201, SiO2/Al2O3 = 45, total BET surface area 424 m2/
g), zeolite Y (Grace ZL5100, SiO2/Al2O3 = 6, total BET sur- 3.1. Eect of preparation method
face area 760 m2/g), zeolite Beta (SiO2/Al2O3 = 166, Cardi
University) and zeolite TS-1 (Cardi University). The paste The gold catalysts supported on zeolite HZSM-5 or Y
formed was dried at 90 C for 16 h and some samples were were prepared using impregnation (IM) or deposition pre-
also calcined at 400 C in static air for 3 h. Further catalysts cipitation (DP). The catalysts were tested in the synthesis of
containing dierent amounts of Au or other metal were also hydrogen peroxide and CO selective oxidation (Table 1).
prepared using a similar procedure.
Supported catalysts were also prepared using deposition
precipitation method. For the 4.3 wt% Au-supported cata-
lyst the detailed procedure was as follows. Aqueous solu- Table 1
Performance of catalysts prepared using impregnation (IM) or deposition
tion (7 ml) of HAuCl4 3H2O (2 g dissolved in 100 ml
precipitation (DP)
water) was added into 75.9 ml distilled water, then 2.1 g
Catalyst Preparation Productivitya CO Conv.b
urea and 1.53 g zeolite supports were added. The mixture
(mol-H2O2/ (%)
was heated to 80 C and kept at that temperature for 4 h h/Kgcat)
with stirring. Then the solid was ltered and washed using
4.3% Au/HZSM-5 IM, dried 4.73 1.7
distilled water. The paste formed was dried at 90 C for 4.3% Au/HZSM-5 IM, calcined 400 C 1.75 <1.0
16 h and some samples were also calcined at 400 C in sta- 4.3% Au/Y IM, dried 3.61 <1.0
tic air for 3 h. 4.3% Au/Y IM, calcined 400 C 2.99 <1.0
4.3% Au/HZSM-5 DP, dried 3.72 2.4
4.3% Au/HZSM-5 DP, calcined 400 C 3.38 5.1
2.2. Catalyst testing
4.3% Au/Y DP, dried 6.60 5.1
4.3% Au/Y DP, calcined 400 C 5.78 4.7
Hydrogen peroxide synthesis was performed using a a
Reaction conditions: methanol 5.6 g, water 2.9 g, catalyst 10 mg, 5%
Parr instruments stainless steel autoclave with a nominal H2/CO2 420 psi., 25% O2/CO2 150 psi., temperature 02 C, time 0.5 h.
volume of 50 ml and a maximum working pressure of b
Reaction conditions: catalyst 50 mg, 1% CO mixed gas 10 ml/min, 5%
14 MPa. The autoclave was equipped with an overhead O2/N2 1 ml/min, temperature 80 C.
 G. Li et al. / Catalysis Communications 8 (2007) 247250 249

Higher yields for hydrogen peroxide are found with the cat-
alysts prepared without calcination step, but these catalysts
are found to leach signicant amounts of Au in the course
of the rst reaction and cannot be successfully reused [13].
Catalysts calcined at 400 C are more stable and can be
reused eectively. The catalysts prepared using deposition
precipitation exhibited higher productivity in the synthesis
of hydrogen peroxide than the catalysts prepared using
impregnation, which may be due to a more homogeneous
distribution of the gold on the supports. However, all the
catalysts exhibited low activity in CO selective oxidation.
Characterisation of the catalysts calcined at 400 C using
powder X-ray diraction showed reections at 37.8, 43.9,
63.9 and 77.0 2h which can be assigned to metallic Au.
The intensity of these reections are more marked for the
samples prepared using impregnation than those for the
samples prepared using deposition precipitation, and it is
concluded that the particle size of gold in the samples pre-
pared using deposition precipitation is larger than 10 nm as
deduced from the Scherrer equation. It is reported that the
TOF of CO oxidation will increase sharply if the diameter
of the Au particles decrease to 4 nm [14] and as the parti-
cles in the present catalysts are signicantly larger in diam-
eter, this can help to explain the low activity we observe for
CO oxidation. The deposition precipitation using urea or
impregnation can transfer most gold in the preparation
mixture onto the supported catalyst. These catalysts have
higher gold concentrations and larger gold particle size,
so they exhibited higher activity in the synthesis of
hydrogen peroxide while lower activity in CO selective
oxidation.
3.2. Eect of supports
Four zeolites, HZSM-5, Y, Beta and TS-1, were used as
the supports to prepare the catalysts with same gold
amount. The gold catalysts were tested in the synthesis of
hydrogen peroxide and CO selective oxidation (Table 2).
It is interesting that the Beta supported-gold catalyst has
the highest activity in the synthesis of hydrogen peroxide.
It seems that the pore structure of the supports has some
eect to the performance of the catalyst.
3.3. Eect of dierent metal
The zeolite Y supported catalysts containing a range of
single metals were also prepared using impregnation. The
Pd catalyst exhibited the highest activity in the synthesis
of hydrogen peroxide (Fig. 1). The Au and Pt catalyst
has some catalytic activity, but much lower than that of
Pd catalyst. The Cu, Ag, Rh or Ru catalyst exhibited very
low catalytic activity or had no activity in synthesis of
H2O2. However, we consider that mixed metal systems
can provide enhanced activity [7,8] and this will be pursued
subsequently.
3.4. In situ H2O2 oxidation
An organic substrate, such as cyclohexane or cyclooc-
tene, was introduced into the system carrying out
H2O2 synthesis and in situ oxidation was investigated
(Table 3). Cyclohexane can be converted to cyclohexalco-
hol in this system, but the larger molecule cyclooctene was
not oxidized under the same conditions. The H2O2 pro-
ductivity of the catalyst containing gold only is low and
as noted earlier [7,8,1113] signicantly higher yields of
H2O2 can be produced if the catalyst also contains Pd,
but the more side reactions can then occur as the organic
substrate can be hydrogenated as well as oxidised. For
example, most of the cyclohexene converts to cyclohexane
in the presence of Pd and hydrogen. 1-Pentene or crotyl-
alcohol convert to several products in this system. In this
respect, it is important that the synthesis rate of H2O2
matches the oxidation rate of organic substrate in
in situ oxidation.

9
Pro d u ctivity (mo l-H2O2/h /Kg cat)

8
Table 2 7
Performance of catalysts prepared using dierent zeolites
6
Catalyst Preparation Productivitya CO Conv.b
(mol-H2O2/h/Kgcat) (%) 5

2.5% Au/HZSM-5 Dried 0.5 1.2 4

2.5% Au/HZSM-5 Calcined 400 C 0.69 <1.0 3
2.5% Au/Y Dried 0.95 7.7
2.5% Au/Y Calcined 400 C 1.02 <1.0 2
2.5 wt% Au/Beta Dried 3.99 1
2.5 wt% Au/Beta Calcined 400 C 3.58
2.5 wt% Au/TS-1 Dried 0.62 2.7 0
2.5 wt% Au/TS-1 Calcined 400 C 0.92 2.1 4.3Au 4.3Cu 4.3Ag 5 Pt 5Ru 5 Rh 0.3Pd
a
Reaction conditions: methanol 5.6 g, water 2.9 g, catalyst 10 mg, 5% Fig. 1. Performance of dierent metal supported on Y for the synthesis of
H2/CO2 420 psi., 25% O2/CO2 150 psi., temperature 02 C, time 0.5 h. H2O2. Reaction conditions: methanol 5.6 g, water 2.9 g, catalyst 10 mg,
b
Reaction conditions: catalyst 50 mg, 1% CO mixed gas 10 ml/min, 5% 5% H2/CO2 420 psi., 25% O2/CO2 150 psi., temperature 2 C, reaction
O2/N2 1 ml/min, temperature 80 C. time 0.5 h.
250 G. Li et al. / Catalysis Communications 8 (2007) 247250

Table 3
In situ oxidation of organic substrates
Catalyst Temperature Time Organic substrate (g) Products Residual H2O2
(C) (h) concentration (wt%)
2.5% Au 2.5% Pd/Y 40 2 0.26 Cyclohexane 2.4% Conversion to cyclohexalcohol 0.041
2.5% Au 2.5% Pd/Y 20 1 0.22 Cyclooctene No product 0.057
2.5% Au 2.5% Pd/Y 20 2 0.27 Cyclohexene 61.7% Conversion to cyclohexane 0.023
2.5% Au/TS-1 20 1 0.26 Cyclohexene No product 0.002
2.5% Au 2.5% Pd/TS-1 20 1 0.25 Cyclohexene 36.1% Conversion to cyclohexane 0.056
2.5% Au 2.5% Pd/TS-1 40 2 0.22 Cyclooctene No product 0.046
2.5% Au 2.5% Pd/TS-1 60 2 0.22 Cyclooctene No product 0.013
2.5% Au 2.5% Pd/TS-1 60 2 0.25 1-Pentene 21.6% Conversion, 1,2-epoxypentane selectivity 23.5% 0.045
2.5% Au 2.5% Pd/TS-1 60 4 0.25 1-Pentene 52.9% Conversion, 1,2-epoxypentane selectivity 10.5% 0.006
2.5% Au 2.5% Pd/TS-1 60 4 0.25 Crotylalcohol 81.9% Conversion, crotonaldehyde selectivity 31.5% 0.006
2.5% Au 2.5% Pd/TS-1 60 2 0.27 Crotylalcohol 58.9% Conversion, crotonaldehyde selectivity 28.9% 0.006
Reaction conditions: methanol 5.6 g, water 2.9 g, catalyst 10 mg, 5% H2/CO2 420 psi., 25% O2/CO2 150 psi., organic substrate added as specied.

4. Conclusions References

The zeolite supported-gold catalysts prepared with the
deposition precipitation using urea or impregnation exhibit
activity in the synthesis of hydrogen peroxide while demon-
strating lower activity in CO oxidation. Beta zeolite
appears to be an eective support for gold for this reaction.
The activity for the synthesis of H2O2 of dierent metal
supported on zeolite Y is as follows: Pd  Au, Pt > Cu,
Ag, Rh, Ru. In situ oxidation of organic substrates is fea-
sible with these catalysts and higher amounts of H2O2
can be synthesized to oxidize the organic substrate if the
catalyst contains Pd, but side reactions can become domi-
nant in this case.
Acknowledgements
The nancial support for G. Li of the Foundation of
National Excellent Doctoral Dissertation of PR China
(No. 200346), Program for New Century Excellent Talents
in University, National Natural Science Foundation of
China (No. 20406005), China Scholarship Council (CSC)
and Cardi University, is gratefully acknowledged.
[1] S.-E. Park, J.W. Yoo, W.J. Lee, J.-S. Chang, UK Park, C.W. Lee, US
Patent 5 972 305 (1999).
[2] B. Zhou, L.-K. Lee, US Patent 6 168 775 (2001).
[3] J.H. Lunsford, J. Catal. 216 (2003) 455.
[4] D.P. Dissanayake, J.H. Lunsford, J. Catal. 214 (2003) 113.
[5] R. Burch, P.R. Ellis, Appl. Catal. B: Environ. 42 (2003) 203.
[6] V.R. Choudhary, S.D. Sansare, A.G. Gaikwad, Catal. Lett. 84 (2002)
81.
[7] P. Landon, P.J. Collier, A.J. Papworth, C.J. Kiely, G.J. Hutchings,
Chem. Commun. (2002) 2058.
[8] P. Landon, P.J. Collier, D. Chadwick, A.J. Papworth, A. Burrows,
C.J. Kiely, G.J. Hutchings, Phys. Chem. Chem. Phys. 5 (2003)
1917.
[9] M. Okumura, Y. Kitagawa, K. Yamagcuhi, T. Akita, S. Tsubota, M.
Haruta, Chem. Lett. (2003) 822.
[10] T. Ishihara, Y. Ohura, S. Yoshida, Y. Hata, H. Nishiguchi, Y. Takita,
Appl. Catal. A: Gen. 291 (2005) 215.
[11] J.K. Edwards, B.E. Solsona, P. Landon, A.F. Carley, A. Herzing, M.
Watanabe, C.J. Kiely, G.J. Hutchings, J. Mater. Chem. 15 (2005)
4595.
[12] J.K. Edwards, B.E. Solsona, P. Landon, A.F. Carley, A. Herzing,
C.J. Kiely, G.J. Hutchings, J. Catal. 236 (2005) 69.
[13] G. Li, J.K. Edwards, A.F. Carley, G.J. Hutchings, Catal. Today 114
(2006) 369.
[14] M. Haruta, Gold Bull. 37 (2004) 27.