Low activation energy dehydrogenation of aqueous formic acid on platinum-ruthenium-bismuth oxide at near ambient temperature and pressure

Article abstract of DOI:10.1039/b916507j

Highly selective dehydrogenation of formic acid in water was observed at near ambient temperature on a metal/metal oxide catalyst composed of platinum ruthenium and bismuth with a low activation energy of 37.3 kJ mol^-1.

Full text of DOI:10.1039/b916507j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Low activation energy dehydrogenation of aqueous formic
acid on platinum–ruthenium–bismuth oxide at near ambient
temperature and pressurew
Siu-Wa Ting, Shaoan Cheng,z Kwok-Ying Tsang, Nicole van der Laak and Kwong-Yu Chan*
Received (in Cambridge, UK) 11th August 2009, Accepted 12th October 2009
First published as an Advance Article on the web 22nd October 2009
DOI: 10.1039/b916507j
3
7
Highly selective dehydrogenation of formic acid in water was
observed at near ambient temperature on a metal/metal oxide
catalyst composed of platinum ruthenium and bismuth with a low
adsorbed on Bi. Previous investigations of formic acid
decomposition have been mostly made with a single metal or
metal supported on a metal oxide which may not be
sufficiently selective or still possesses a high activation energy
barrier. Studies of bimetallic catalysts have not been reported
ꢀ1
activation energy of 37.3 kJ mol
.
3
2
Formic acid decomposition has been investigated over the
until very recently and there have also been no studies of
bismuth or bismuth oxide. We report here, for the first time, a
bismuth containing catalyst composited with platinum and
ruthenium. This heterogeneous catalyst decomposes formic
acid in liquid water at ambient temperature with high rates,
yielding exclusively hydrogen and carbon dioxide with a very
low activation energy.
1
–20
2,5–7
8–13
decades
metal supported on oxides.
without catalysts under hydrothermal conditions
using catalysts of metals,
metal oxides,
and
1
4–20
The reaction can also proceed
2
1–23
to
temperatures of 280–330 1C and a pressure of 275 bar. Recent
studies have been motivated by the interest of generating
2
4,25
hydrogen to feed fuel cells
and lowering the temperature
1
6–19
of the water–gas-shift process.
Most previous studies have
The catalyst can be synthesized (ESIw) with various
compositions of precursors containing Pt, Ru, and Bi with
bismuth in an oxidized form, and is generally denoted
been made at an elevated temperature and with gaseous
reactants and products. A gas phase reaction will require
heating above 100 1C, the normal boiling point of formic acid,
or pulling a vacuum, or introducing an inert carrier gas to
dilute formic acid below its saturated vapor pressure. These
constraints would add complexity to a hydrogen generation
device. A few recent studies have focused on a selective
dehydrogenation reaction in the liquid phase at near ambient
PtRuBiO
with ratios of the metals to be further specified.
catalyst at room temperature,
x
Upon contacting the PtRuBiO
x
aqueous formic acid shows steady gas evolution. The product
gas was continuously analyzed and contained equi-molar
hydrogen and carbon dioxide without detection of carbon
monoxide (Table S1 and Fig. S1w). A typical 7 h gas evolution
from an 80 mL 15% v/v formic acid in water immersed with
400 mg catalyst of 30% metal loaded carbon is shown in
Fig. 1. The rate of gas generation is initially high but decreased
gradually before reaching steady-state after 1 h. This apparent
2
4–32
temperature.
Selective dehydrogenation of aqueous
formic acid at near ambient temperature was observed
2
5,27–31
with soluble metal complexes of ruthenium
and
and the presence of
It is desirable to have an
2
4,26
27–31
rhodium.
An inert atmosphere
2
9–31
an amine may be required.
deactivation can be explained by product CO
catalyst active sites which were initailly all available for formic
acid decomposition. The forward rate of CO (ad) to CO (aq)
2
occupation of
insoluble catalyst for fixed bed flow-through operations and
that excess water from aqueous formic acid feed can be
removed without draining the catalyst. The recent report of
Pd–Au and Pd–Ag catalysts shows good decomposition rates
2
2
3
2
at near ambient temperature. Kinetically, the dehydrogena-
tion reaction is favored in the presence of a metal catalyst such
as platinum, which is thought to catalyze the C–H bond
cleavage. It is well known in fuel cell literature that ruthenium
3
3,34
acts as a CO tolerant co-catalyst to platinum,
whereas a
bimetallic PtBi catalyst was shown to have good activity for
3
5–37
electrooxidation of formic acid to CO
2
.
The enhanced
activity is suggested to be caused by a change of lattice spacing
5
3
in PtBi which reduces CO affinity. The other suggestion is
the promoted oxidation of adsorbed CO by OH preferentially
Department of Chemistry, The University of Hong Kong,
Pokfulam Road, Hong Kong SAR, P.R. China.
E-mail: hrsccky@hku.hk; Fax: +852 28571586
w Electronic supplementary information (ESI) available: Experimental
details and supporting information. See DOI: 10.1039/b916507j
z Present address of Shaoan Cheng: State Key Laboratory of Clean
Energy Utilization, Zhejiang University, Hangzhou, China 310027.
Fig. 1 Decomposition of formic acid on Pt–Ru–BiO
: 1 : 4) at 80 1C over 7 h. Inset: activity of initial 30 min for better
comparison with Pt–Bi /C and Ru–Bi /C.
x
/C (Pt/Ru/Bi B
2
2
O
3
2 3
O
This journal is ꢁc The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7333–7335 | 7333

View Article Online
2
to CO (gas) is gradually balanced by a backward rate at
steady-state.
The typically synthesized PtRuBiO catalyst is comprised of
x
nanoparticles with an average size of 5 nm supported on
Vulcan XC-72 carbon as shown in Fig. 2 and Fig. S2.w At a
fixed concentration and volume of aqueous formic acid, the
decomposition rate increased linearly with catalyst mass
(
Fig. S3w) and can be scaled to precious metal content as
.3 mol/(g of PtRu) over the 7 h period in Fig. 1. Over a much
1
longer time interval, gas evolution will decline unless formic
acid is replenished to compensate for its consumption. For
x
different compositions of PtRuBiO catalyst used, a varying
Fig. 3 An Arrhenius plot for the HCOOH decomposition on
degree of reactivity was found as shown by the initial as well as
steady-state rates in Table S2.w The initial rates on different
catalysts vary between 0.57 to 2.65 times of the initial rate
shown in Fig. 1. Metal dispersion was estimated using TEM
particle size distribution shown in Fig. S3.w Based on Pt
and Ru surface atoms, the turn over frequency (TOF) was
x
PtRuBiO /C.
and dehydration (II) are mildly endothermic as shown by their
standard enthalpies of reactions at 25 1C with
ꢀ
1
HCOOH(l) - CO
2
(g) + H
2
(g) D
r
H1 = 31.2 kJ mol , (I)
ꢀ1
estimated to be 312 h in the first hour of the decomposition
shown in Fig. 1. Bismuth oxide is not acting as a support since
no activity was found in PtRu supported on carbon or
titanium dioxide or alumina. Bismuth oxide plays an enabling
role for the ambient temperature reaction as control experi-
ments using a single metal of Pt and Ru, Bi, their oxides,
binary metal mixture and oxides all show no activity except
ꢀ1
and HCOOH(l) - CO(g) + H
2
O(l) D
r
H1 = 28.4 kJ mol
.
(
II)
Dehydration becomes more endothermic (an additional
ꢀ
1 39
44 kJ mol ) when water produced is gaseous. On the other
hand, dissolved CO (aq) in liquid phase can lower the
2
ꢀ1
enthalpy of dehydrogenation by 20.3 kJ mol . Experimental
19,22,22
results
PtBiOx and RuBiO . The activity of the last two binary
x
showed higher selectivity towards CO generation
2
mixtures with bismuth are, however, very low compared to
under hydrothermal conditions, suggesting a possible thermo-
dynamic enhancement. The possible lower activation energy
of dehydrogenation of formic acid in water has also been
the PtRuBiO , as shown in Fig. 1. The oxidation state of
x
bismuth was determined by XPS to be exclusively +3, whereas
appreciable amounts of Pt and Ru are in oxides or hydroxide
4
0
investigated theoretically.
(
Fig. S4w). The bismuth content and state in the used catalysts
was analyzed to be unchanged.
The reaction rate is sensitive to the concentration of formic
acid in water with a maximum rate at ca. 15% v/v or 3 M, as
shown in Fig. 4. This optimum concentration is consistent
Initial reaction rates determined over the range of 40–90 1C
yields a linear inverse temperature plot with an activation
2
7,28
with the concentration used by Laurenczy and co-workers.
ꢀ1
energy of 37.3 kJ mol as shown in Fig. 3. This value is lower
Concentrations of molecular formic acid and formate ion
change interactively with changing concentration of the formic
41
than all previously reported values except that on a mixture
1
of chromium oxide and potassium oxide. Most reported
2
acid/water mixture. Detailed potentiometric titration studies
activation energies for both dehydrogenation and dehydration
ꢀ1,12
1 20,24,26
show maximum formic acid dissociation in the range of
ꢀ
for formic acid are 4100 kJ mol
with a few reported
The apparent lower
25–65 wt% HCOOH in water with a max HCOO concentration
ꢀ
values between 40 to 90 kJ mol
.
of 0.07 M. The rate dependence in Fig. 4 can therefore be
interpreted as an effect of concentration of one or both of
activation energy may be attributed to thermodynamic and
kinetic factors in the aqueous environment using the specific
catalyst including the efficiency of PtRu for CO resistant C–H
ꢀ
HCOOH, HCOO and to a minor extent water at high
concentration of formic acid. Recent reports of aqueous
cleavage; and the lower enthalpy value of CO
2
dissolved in
formic acid decomposition have all used a mixture of formic
2
7,28,32
water or adsorbed on the basic Bi . Both dehydrogenation (I)
O
2 3
acid/sodium formate with a definite ratio.
In our experi-
ments using the PtRuBiO catalyst, the decomposition rate is
x
linearly dependent on formic acid concentration when the
formate ion is in excess (Fig. S6).w It is also linearly dependent
on formate ion concentration when formic acid is in excess
(
Fig. S7).w The reaction can therefore be bimolecular in
nature and depends on both HCOOH and HCOO
ꢀ
concentrations. Borowiak et al. have discussed the bimolecular
ꢀ
HCOOH/HCOO
mechanism giving higher selectivity
4
towards dehydrogenation. Through ab initio calculations
2
of activation complexes with a few water molecules present,
4
0
Akiya and Savage showed the role of water in lowering the
activation energy for the dehydrogenation path. Hence the
reaction mechanism may involve a combination of water,
molecular formic acid, and formate ion.
Fig. 2 Transmission electron microscopy image of the PtRuBiO
x
catalyst used for the reaction shown in Fig. 1.
7
334 | Chem. Commun., 2009, 7333–7335
This journal is ꢁc The Royal Society of Chemistry 2009

View Article Online
5
Y. K. Sun, J. J. Vajo, C. Y. Chan and W. H. Weinberg, J. Vac. Sci.
Technol., A, 1988, 6, 854.
6
7
D. H. S. Ying and R. J. Madix, J. Catal., 1980, 61, 48.
N. Aas, Y. Li and M. Bowker, J. Phys.: Condens. Matter, 1991, 3,
S281.
8
9
X. D. Peng and M. A. Barteau, Catal. Lett., 1991, 7, 395.
P. A. Dilara and J. M. Vohs, J. Phys. Chem., 1993, 97, 12919.
1
1
0 V. A. Gercher and D. F. Cox, Surf. Sci., 1994, 312, 106.
1 J. Stubenrauch, E. Brosha and J. M. Vohs, Catal. Today, 1996, 28,
4
31.
2 R. Larsson, M. H. Jamroz and M. A. Borowiak, J. Mol. Catal. A:
Chem., 1998, 129, 41.
3 G. Y. Popova, T. V. Andrushkevich, Y. A. Chesalov and
E. S. Stoyanov, Kinet. Catal., 2000, 41, 805.
4 A. Bandara, J. Kubota and A. Wada, J. Phys. Chem. B, 1997, 101,
1
1
1
Fig.
4 The initial rate of Pt/Ru/BiO on Vulcan-72 carbon
3
5 J. Rasko, T. Kecskes and J. Kiss, J. Catal., 2004, 224, 261.
6 T. Shido and Y. Iwasawa, J. Catal., 1993, 141, 71.
7 G. Jacobs, P. M. Patterson, U. M. Graham, A. C. Crawford and
B. H. Davis, Int. J. Hydrogen Energy, 2005, 30, 1265.
18 G. Jacobs, P. M. Patterson, U. M. Graham, A. C. Crawford,
61.
(Pt/Ru/Bi B 2 : 1 : 4) towards formic acid decomposition at 80 1C
1
1
1
with different concentrations of formic acid.
The high selectivity and high rate may be attributed to the
synergistic roles of platinum, ruthenium, and bismuth oxide in
cleaving C–H bonds, minimizing CO adsorption or oxidizing
A. Dozier and B. H. Davis, J. Catal., 2005, 235, 79.
19 Y. Yasaka, H. Yoshida, C. Wakai, N. Matubayasi and
M. Nakahar, J. Phys. Chem. A, 2006, 110, 11082.
20 M. Ojeda and E. Iglesia, Angew. Chem., Int. Ed., 2009, 48, 4800.
2
CO, and facilitating generation and removal of CO .
Chemisorption measurements showed significantly less CO
sorption on PtRuBiOx than PtRu on Vulcan carbon
21 A. Bjerre and E. Sorensen, Ind. Eng. Chem. Res., 1992, 31, 1574.
22 J. Yu and P. E. Savage, Ind. Eng. Chem. Res., 1998, 37, 2.
23 P. G. Maiella and T. B. Brill, J. Phys. Chem. A, 1998, 102, 5886.
24 S. Fukuzumi, T. Suenobu and S. Ogo, WO/2008/059630,
PCT/JP2007/060115.
(
see ESI Part 7w). The role of Ru in promoting Pt for efficient
33,34
C–H cleavage without CO poisoning has been discussed
3
5–37
and a similar role has been suggested for Bi.
Bismuth
2
5 G. Laurenczy, C. Fellay and P. Dyson, WO/2008/047312,
PCT/IB2007/054222.
6 S. Fukuzumi, T. Kobayashi and T. Suenobu, ChemSusChem, 2008,
oxide appears to offer a unique role in lowering the activation
energy and offering high selectivity to dehydrogenation.
2
Bismuth has different carbonates including (BiO)
and bismuth subcarbonate (BiO) (CO ) that can be formed
and play a possible role in
removing the product or forming the intermediate as
adsorbed CO
4 3
(OH)CO
1
, 827.
27 C. Fellay, P. J. Dyson and G. Laurenczy, Angew. Chem., Int. Ed.,
2008, 47, 3966.
2
3
3
8
2 2 3
from CO , water, and Bi O
2
8 C. Fellay, N. Yan, P. J. Dyson and G. Laurenczy, Chem.–Eur. J.,
009, 15, 3752.
2
2
.
29 B. Loges, A. Boddien, H. Junge and M. Beller, Angew. Chem., Int.
Ed., 2008, 47, 3962.
The high selectivity and high reaction rate of the PtRuBiOx
catalyst at ambient conditions show promise for a convenient
hydrogen generation device. Further increase in reaction rates
is possible with a continuous flow-through reactor packed
with the heterogeneous catalyst. Optimizations can be made in
fine tuning catalyst composition, particle size, properties of the
support, and operation pressure. The low activation energy
of the reaction and the unique role of bismuth oxide
suggest new perspectives in the mechanisms of formic acid
dehydrogenation in the aqueous phase.
3
3
3
3
0 H. Junge, A. Boddien, F. Capitta, B. Loges, J. R. Noyes,
S. Gladiali and M. Beller, Tetrahedron Lett., 2009, 50, 1603.
1 A. Boddien, B. Loges, H. Junge and M. Beller, ChemSusChem,
2008, 1, 751.
2 X. Zhou, Y. Huang, W. Xing, C. Liu, J. Liao and T. Lu, Chem.
Commun., 2008, 3540.
3 M. Watanabe, M. Uchia and S. Motto, J. Electroanal. Chem.,
1
34 H. Schmidt, P. Buchner, A. Datz, K. Dennerlein, S. Lang and
987, 229, 395.
M. Waidbas, J. Power Sources, 2002, 105, 243.
5 E. Casado-Rivera, Z. Gal, A. C. D. Angelo, C. Lind, F. J. DiSalvo
and H. D. Abruna, ChemPhysChem, 2003, 4, 193.
3
Financial support from The Hong Kong Research Grants
Council GRF HKU 700208P, ‘‘Initiative on Clean Energy and
Environment’’ of The University of Hong Kong Development
Fund, and the HKU Strategic Research Theme on
36 E. Casado-Rivera, D. Volpe, L. Alden, C. Lind, C. Downie,
T. Vazquez-Alvarez, A. C. D. Angelo, F. J. DiSalvo and
H. D. Abruna, J. Am. Chem. Soc., 2004, 126, 4043.
3
7 A. V. Tripkovic, K. Dj. Popovic, R. M. Stevanovic, R. Socha and
A. Kowal, Electrochem. Commun., 2006, 8, 1492.
‘
‘Clean Energy Research’’ are acknowledged.
38 P. Taylor, S. Sunder and V. J. Lopata, Can. J. Chem., 1984, 62,
863.
2
3
9 I. Martin, T. Skalicky, J. Langer, H. Abdoul-Carime, G. Karwasz,
E. Illenberger, M. Stano and S. Matejcik, Phys. Chem. Chem.
Phys., 2005, 7, 2212.
Notes and references
1
2
A. Mailhe and P. Sabatier, Compt. Rend. Acad. Sci., 1911, 152, 1212.
K. Hirota, K. Kuwata and Y. Nakai, Bull. Chem. Soc. Jpn., 1958,
40 N. Akiya and P. E. Savage, AIChE J., 1998, 44, 405.
41 J. S. Ferrer, E. Couallier, M. Rakib and G. Durand, Electrochim.
Acta, 2007, 52, 5773.
42 M. A. Borowiak, M. H. Jamro
Chem., 2000, 152, 121.
31, 861.
3
4
P. Mars, Proceedings of symposium on the mechanism of
heterogeneous catalysis, Amsterdam, The Netherlands, 1959.
P. Mars, J. J. F. Scholten and P. Zwietering, Adv. Catal., 1963, 14, 35.
´
z and R. Larsson, J. Mol. Catal. A:
This journal is ꢁc The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7333–7335 | 7335