Water dispersible acetate stabilized ruthenium(0) nanoclusters as catalyst for hydrogen generation from the hydrolysis of sodium borohyride

Article abstract of DOI:10.1016/j.molcata.2006.05.037

The development of new storage materials will facilitate the use of hydrogen as a major energy vector in near future. In the hydrogen economy, chemical hydrides such as NaBH₄, KBH₄, LiH, NaH have been tested as precursor materials for supplying hydrogen at ambient temperature. Among these chemical hydrides, sodium borohydride (NaBH₄) provides a safe and practical mean of producing hydrogen. Sodium borohydride is stable in basic solution; however, hydrolysis generates hydrogen gas in the presence of a suitable catalyst. All of the prior catalysts having been used for this hydrolysis reaction are heterogeneous and, therefore, have limited activity because of the surface area. Here, we report the employment of water dispersible metal(0) nanoclusters as catalyst for the hydrolysis of sodium borohydride. The water dispersible ruthenium(0) nanoclusters were prepared from the reduction of RuCl₃·3H₂O and stabilized by using acetate ion. They provide 5170 turnovers in the hydrolysis of sodium borohydride at 25.0 ± 0.1 °C before deactivation. The acetate stabilized ruthenium(0) nanoclusters of 2.8 ± 1.4 nm particle size were characterized by TEM, XPS, and FTIR spectroscopy. Kinetic study shows that the ruthenium(0) nanocluster-catalyzed hydrolysis of sodium borohydride is first order in catalyst concentration and zero order in substrate concentration. The acetate stabilized ruthenium(0) nanoclusters provide a low activation energy for the hydrolysis of sodium borohydride. Even in basic solution, the activation energy was found to be smaller than that obtained by using the carbon-supported ruthenium as catalyst.

Full text of DOI:10.1016/j.molcata.2006.05.037

Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
Water dispersible acetate stabilized ruthenium(0) nanoclusters
as catalyst for hydrogen generation from the
hydrolysis of sodium borohyride
∗
¨
Mehmet Zahmakıran, Saim Ozkar
Department of Chemistry, Middle East Technical University, 06531 Ankara, Turkey
Received 23 February 2006; received in revised form 3 May 2006; accepted 11 May 2006
Available online 21 June 2006
Abstract
The development of new storage materials will facilitate the use of hydrogen as a major energy vector in near future. In the hydrogen economy,
chemical hydrides such as NaBH , KBH , LiH, NaH have been tested as precursor materials for supplying hydrogen at ambient temperature.
Among these chemical hydrides, sodium borohydride (NaBH ) provides a safe and practical mean of producing hydrogen. Sodium borohydride
4
4
4
is stable in basic solution; however, hydrolysis generates hydrogen gas in the presence of a suitable catalyst. All of the prior catalysts having been
used for this hydrolysis reaction are heterogeneous and, therefore, have limited activity because of the surface area. Here, we report the employment
of water dispersible metal(0) nanoclusters as catalyst for the hydrolysis of sodium borohydride. The water dispersible ruthenium(0) nanoclusters
were prepared from the reduction of RuCl
3
·3H
2
O and stabilized by using acetate ion. They provide 5170 turnovers in the hydrolysis of sodium
◦
borohydride at 25.0 ± 0.1 C before deactivation. The acetate stabilized ruthenium(0) nanoclusters of 2.8 ± 1.4 nm particle size were characterized
by TEM, XPS, and FTIR spectroscopy. Kinetic study shows that the ruthenium(0) nanocluster-catalyzed hydrolysis of sodium borohydride is first
order in catalyst concentration and zero order in substrate concentration. The acetate stabilized ruthenium(0) nanoclusters provide a low activation
energy for the hydrolysis of sodium borohydride. Even in basic solution, the activation energy was found to be smaller than that obtained by using
the carbon-supported ruthenium as catalyst.
©
2006 Elsevier B.V. All rights reserved.
Keywords: Ruthenium; Water dispersible; Nanoclusters; Catalyst; Hydrolysis; Sodium borohydride; Acetate; TEM; NIH image; X-ray photoelectron spectroscopy;
FTIR
1
. Introduction
solar energy can not be fulfilled. The obvious solution to this
predicament is the storage of energy. In this context, hydrogen
appears to be the best energy carrier as it is clean to use, renew-
able, and widely available from diverse sources [2]. Although
hydrogen can be used in many ways for generating energy, the
most efficient one is its conversion to electrical energy following
combustion in a fuel cell [3]. Dramatic improvements, particu-
larly in the existing technologies for production, storage, and
usage of hydrogen, are still needed if hydrogen is to become a
major energy carrier in the future. The hydrogen-based energy
economy deals with the development of diversity of sources
for hydrogen production using efficient technologies, finding
viable methods for hydrogen storage, lowering the cost of fuel
cells and improving their on-board usage. Advanced storage
concepts include the use of metal hydrides and nanostructured
materials, for example, encapsulation in the porous materials.
Fossil fuels (i.e., petroleum, natural gas, and coal) will over-
whelmingly be used to meet the increasing energy demand
at least for the next few decades [1]. Because of increasing
concern about the environmental problems, including global
warming caused by the emission of gases from the combus-
tion of fossil fuels, there has been a growing interest for the
use of renewable energy sources such as solar and wind energy.
However, besides their relatively high cost, energy production
from renewable sources has the problem of discontinuity, for
example, requirement of uninterrupted sunshine in the usage of
∗
Corresponding author. Tel.: +90 312 2103212; fax: +90 312 2103200.
¨
E-mail address: sozkar@metu.edu.tr (S. Ozkar).
1
381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.05.037

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
9
6
Recently, sodium borohydride, NaBH4, has been suggested as
a new fuel source for supplying hydrogen under mild condi-
tions [4]. NaBH4 yields hydrogen gas and water-soluble sodium
metaborate, NaBO2, upon hydrolysis in the presence of certain
catalysts.
2.1. Preparation of acetate stabilized ruthenium(0)
nanoclusters
Following the procedure given in the literature [19], acetate
stabilized ruthenium(0) nanoclusters were prepared from the
reduction of 1.33 mM solution of ruthenium(III) chloride by
sodium borohydride (9.3 mM) in aqueous sodium acetate solu-
tion (80 mM) at room temperature. A molar ratio of NaBH4
to RuCl3 greater than 5 was used to ensure complete reduc-
tion of ruthenium(III) to its zerovalent state. Acetate stabilized-
ruthenium(0) nanoclusters, dark brown in color, were found to
be stable in aqueous media (no precipitation was observed after
a few days of storage). Aliquots of the ruthenium(0) nanoclus-
ters solution were used to catalyze the hydrolysis of sodium
borohydride. The nanoclusters could be isolated by centrifu-
gation followed by drying in vacuum. The isolated nanoclusters
were characterized by XPS and TEM. The isolated ruthenium(0)
nanoclusters are redispersed in water and were tested for their
catalytic activity. When redipersed, the ruthenium(0) nanoclus-
ters retain 70% of their initial catalytic activity in the hydrolysis
of sodium borohydride.
catalyst
NaBH4 + 2H2O−−−−−→NaBO2 + 4H2,
−
1
ꢀH = −300 kJ mol
(1)
The catalysts tested for the hydrolysis of sodium borohydride,
listed chronologically are: platinum or rhodium salts [5], cobalt
salt [6], iron, ruthenium, palladium, osmium, iridium, and plat-
inum salts [7], nickel, Raney nickel, and bulk cobalt [8], alloys
such as LaNi4.5T0.5 (T = Mn, Cr, Co, Fe, Cu) [9], bulk ruthenium
[
10], mixed metal/metal oxides such as Pt–LiCoO2 [11], nickel
boride [12] and filamentary nickel–cobalt [13]. With the excep-
tion of metal salts of platinum, ruthenium [5] and cobalt [6]
all the catalyst used for the hydrolysis of sodium borohydride
are heterogeneous. The limited surface area of heterogeneous
catalysts restricts their catalytic activity to a low value. As the
activity of catalyst is directly related to its surface area, using
water dispersible nanoclusters is a promising way to increase
the catalytic activity.
2.2. Self-hydrolysis of sodium borohydride
At the present time there is a considerable interest in the
study and use of metal nanoclusters because of their novel and
attractive physical and chemical properties [14]. For example,
transition metal nanoclusters have been studied as catalysts in
various organic and inorganic reactions [15] and as electrocat-
alysts in fuel cells [16]. Metal nanoclusters are more active
catalysts than the respective bulk metal counterparts, because a
large percentage of metal atoms are on the surface of nanoclus-
ters, and the surface atoms do not necessarily order themselves in
the same way as those in the bulk do [17]. For the above reasons,
we started an investigation of water-dispersible transition-metal
nanoclusters as catalyst for the hydrolysis of sodium borohy-
dride. We selected ruthenium, which is the most widely used
catalyst in the hydrolysis of sodium borohydride, though in the
bulk form, for the preparation of water dispersible metal nan-
oclusters and employ them herein as catalyst for the hydrolysis
of sodium borohydride [18]. The results of our study show that
acetate stabilized ruthenium(0) nanoclusters, prepared from the
reduction of ruthenium(III) chloride with sodium borohydride
in aqueous solution containing acetate anion, are highly active
catalyst for the hydrolysis of sodium borohydride at room tem-
perature.
In order to determine the rate of self hydrolysis of sodium
borohydride in the absence of catalyst, 284 mg sodium borohy-
dride was dissolved in 50 mL water and the solution was trans-
ferred with a 50 mL pipette into the reaction flask thermostated
◦
at 25.0 ± 0.1 C. The experiment was started by closing the reac-
tion flask and turning on the stirring at 1000 rpm simultaneously.
The volume of hydrogen gas generated was measured every 30 s.
The pH of the solution was recorded to be 9.6 at the beginning
of reaction and to increase to its final value of 10.8 when all
sodium borohydride was converted to sodium metaborate.
2
.3. Catalytic activity test of ruthenium(0) nanoclusters in
the hydrolysis of sodium borohydride
In a series of experiments, acetate stabilized-ruthenium(0)
nanoclusters were tested for their catalytic activity in the hydrol-
ysis of sodium borohydride. The catalytic activity of ruthe-
nium(0) nanoclusters in the hydrolysis of sodium borohydride
was determined by measuring the rate of hydrogen generation.
Before starting the catalytic activity test, a jacketed reaction
flask (75 mL) containing a Teflon-coated stir bar was placed
on a magnetic stirrer (Heidolph MR-301) and thermostated to
◦
2
5.0 ± 0.1 C by circulating water through its jacket from a con-
2
. Experimental
stant temperature bath. Then, a graduated glass tube (50 cm
in height and 5.0 cm in diameter) filled with water was con-
nected to the reaction flask to measure the volume of the
hydrogen gas to be evolved from the reaction. Next, 284 mg
(7.47 mmol) NaBH4 was dissolved in 40 mL water (correspond-
Ruthenium(III) chloride trihydrate, sodium acetate trihy-
drate (99%), sodium borohydride (98%) were purchased from
Aldrich. The water content of RuCl3·xH2O was determined by
TGA and found to be x = 3. Deionized water was distilled by
water purification system ( S¸ im s¸ ek SL-200, Turkey). All glass-
ware and Teflon coated magnetic stir bars were cleaned with
acetone, followed by copious rinsing with distilled water before
◦
ing to 30 mmol = 672 mL H2 at 25.0 ± 0.1 C and 0.91 atm
pressure). The solution was transferred with a 50 mL glass-
◦
pipette into the reaction flask thermostated at 25.0 ± 0.1 C.
Then, aliquots of acetate stabilized ruthenium(0) nanocluster
solutions was transferred into the reaction flask using a 10 mL
◦
drying in an oven at 100 C.

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
97
gastight syringe. The initial pH of the solution was recorded to
be same (pH 9.6) as in the self hydrolysis of sodium borohy-
dride. The volume of hydrogen gas evolved was measured by
recording the displacement of water level every 30 s at constant
atmospheric pressure of 693 Torr. The reaction was ceased when
we performed a catalytic activity test of Ru(0) nanoclusters for
the hydrolysis of sodium borohydride in aqueous solution con-
taining 10% NaOH (w/w), exactly in the same way as described
above. Hydrolysis of sodium borohydride (150 mM) in the pres-
ence of Ru(0) nanoclusters (0.32 mM) in 10% NaOH (w/w)
7
0% conversion was achieved.
solution was performed at four different temperatures, 25, 35,
◦
45, and 55 C.
2
.4. Kinetic study of Ru(0) nanocluster-catalyzed
hydrolysis of sodium borohydride
2.7. Catalytic lifetime of acetate stabilized-ruthenium(0)
nanoclusters
In order to establish the rate law for catalytic hydrolysis of
NaBH4 usingwaterdispersibleRu(0)nanoclusters, twodifferent
sets of experiments were performed in the same way as described
in Section 2.3. In the first set of experiments, the concentration
of NaBH4 was kept constant at 150 mM and the Ru(0) nan-
oclusters concentration was varied in the range of 0.08, 0.16,
The catalytic lifetime of acetate stabilized-ruthenium(0) nan-
oclusters in the hydrolysis of sodium borohydride was deter-
mined by measuring the total turnover number (TTO). Such a
lifetimeexperimentwasstartedwitha50 mLsolutioncontaining
0.32 mM Ru(0) nanoclusters and 450 mM NaBH4 (correspond-
ing to maximum possible total turnover number of 5400) at
0
.32, 0.48, 0.64, 0.8 and 1.12 mM. In the second set of exper-
iments, Ru(0) nanoclusters concentration was held constant at
.32 mM while the NaBH4 concentration was varied to get var-
◦
25.0 ± 0.1 C. The reaction was continued until no hydrogen
0
gas evolution was observed. Acetate stabilized ruthenium(0)
nanoclusters could provide 5170 turnovers for the hydrolysis
of sodium borohydride over 1.4 h before deactivation.
ious NaBH4/Ru(0) ratio in the range of 100, 200, and 500. In all
of these experiments, the initial pH of the solution was recorded
to be same (pH 9.6) as in the self hydrolysis of sodium borohy-
dride. Finally, we performed the catalytic hydrolysis of NaBH4
in the presence of Ru(0) nanoclusters at constant NaBH4 and
2.8. Mercury poisoning of acetate stabilized-ruthenium(0)
nanoclusters
Ru(0) concentrations at various temperatures in the range of
◦
2
(
5–45 C in order to obtain the activation energy (Ea), enthalpy
In a poisoning experiment 0.80 mM ruthenium(0) nanoclus-
ters used in the hydrolysis of 150 mM sodium borohydride
(54.8 mg) in 10 mL solution was poisoned by using 4.1, 3.3, 2.5,
1.65, 0.83, 0.41 and 0.20 mg mercury corresponding to a Hg/Ru
molar ratio of 0.125, 0.100, 0.075, 0.05, 0.025, 0.013, and 0.063,
respectively, and the rate of hydrogen generation measured at
#
#
ꢀH ) and entropy (ꢀS ). The rate constants were found to be
−
1
−1
s at 25,
k = 2.3, 3.8, 4.9, 6.1, 6.3 [mol NaBH4]·[mol Ru(0)]
◦
3
0, 35, 40, 45 C, respectively.
2
.5. Effect of acetate concentration on the catalytic activity
of Ru(0) nanoclusters
◦
25.0 ± 0.1 C.
In order to study the effect of acetate concentration on the cat-
alytic activity of Ru(0) nanoclusters in the hydrolysis of sodium
borohydride (150 mM), catalytic activity tests were performed
2.9. Isolation of acetate stabilized-ruthenium(0)
nanoclusters
◦
at 25.0 ± 0.1 C starting with various concentrations of sodium
The acetate stabilized ruthenium(0) nanoclusters were iso-
lated by centrifugation at 5000 rpm for 6 h and then drying
under vacuum. Then the catalytic activity of these isolated ruthe-
nium(0) nanoclusters was tested for the hydrolysis of sodium
borohydride.
acetate (0.5, 0.8, 1.0, 1.5, 2.0 and 10 M) in the preparation of
Ru(0) nanoclusters. In all the experiments the total volume of
solution was kept constant at 50 mL.
2
.6. Effect of sodium hydroxide on the Ru(0)
nanocluster-catalyzed hydrolysis of sodium borohydride
2.10. TEM sample preparation and TEM analyses
In order to study the hydrolysis of sodium borohydride in
strongly basic solution, the same catalytic hydrolysis of sodium
borohydride was performed in a solution containing 10% (w/w)
NaOH solution having an initial pH of 12.4. Here again, a con-
trol experiment was also performed to determine the rate of
the self hydrolysis of sodium borohydride in the absence of
ruthenium(0) nanoclusters; it was observed that there is no sig-
nificant amount of hydrogen production, 0.013 mL H2/s, i.e.,
The samples used for the TEM experiments were harvested
from the preparation of acetate stabilized-ruthenium(0) nan-
oclusters solution as described above: 5 mL aliquot of acetate
stabilized-ruthenium(0) nanoclusters solution was transferred
into a clean screw-capped glass vial with a disposable polyethy-
lenepipette. ThissamplewassenttotheUniversityofOregonfor
TEM investigation. There, one drop of the colloidal solution was
deposited on the chloroform cleaned, carbon coated Cu TEM
grid and the solvent was then evaporated. TEM analyses were
performed at the University of Oregon with expert assistance
of Dr. Eric Schabtach, using the sample preparation proce-
dure and Philips CM-12 TEM with a 70 ␮m lens operating at
0
.02% of the rate in a normal catalytic run using 0.08 mM Ru(0)
nanoclusters. A critical reexamination of the literature shows
that the catalytic activity of some heterogeneous catalysts in the
hydrolysis of sodium borohydride has been investigated in basic
solutions (<10% NaOH). Hence, for the purpose of comparison,
˚
100 kV and with a 2.0 A point-to-point resolution. Samples were

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
9
8
examined at magnification between 100 and 400k. One drop of
desired colloidal solution was deposited on the amorphous car-
bon membrane of the transmission electron microscope grid and
the solvent was then evaporated.
dride liberating hydrogen gas, Eq. (2).
Ru(0) nanoclusters
NaBH4 + 2H2O −− −−−−−−−−−−− −→ NaBO2 + 4H2
(2)
First of all, the rate of hydrogen generation from the self
hydrolysis of sodium borohydride was measured in a control
experiment starting with 150 mM sodium borohydride solution,
that is, under the exactly same conditions. The observed rate
is less than 0.8 mL H2/min (the total amount of hydrogen gas
produced over 120 min is 94 mL).
2
.11. Particle size measurements
Particle size analysis was performed using the public
domain NIH Image J 1.62 program [20], developed at the
U.S. National Institutes of Health and available on the Inter-
net at http://www.rsb.info.nih.gov/nih-image/. The image was
obtained as a TIF file directly from the TEM measurement.
Using Adobe Photoshop the contrast and brightness and channel
curves were adjusted so that particles stand out clearly from the
background. In NIH Image J 1.62, after having set the scale and
the threshold, the “analyze particles” feature was used to gen-
erate a table of particle areas and diameters (major and minor
axes). This table was then exported into Microsoft Excel XP
where histograms, statistical analysis and histogram plotting
were performed. For each particle, the diameter was calculated
from the area by assuming that the nanoclusters are circular. Size
distributions are quoted as the mean diameter ± the standard
deviation.
The acetate stabilized-ruthenium(0) nanoclusters were found
to be highly active catalyst for the hydrolysis of sodium borohy-
dride as shown in Fig. 1, which plots the volume of H2 generated
versus time during the catalytic hydrolysis of 150 mM NaBH4
solution in the presence of Ru(0) nanoclusters in different con-
◦
centrations at 25.0 ± 0.1 C and an initial pH of 9.6. It is seen
that acetate stabilized-ruthenium(0) nanoclusters have high cat-
alytic activity, in the range of 20–370 mL H2/min even at low
catalyst concentrations (0.08–1.12 mM Ru(0), respectively) and
room temperature. The hydrogen evolution starts immediately
without any induction period since preformed nanocluster cat-
alyst was used. The hydrogen generation rate was determined
from the linear portion of the plot for each experiment with dif-
ferent Ru(0) concentration. Fig. 2 shows the plot of hydrogen
generation rate versus Ru(0) concentration, both on logarithmic
scales. One obtains a straight line, the slope of which is found to
be 1.07–1.0 within experimental error. This result indicates that
the hydrolysis is first order with respect to the concentration of
acetate stabilized ruthenium(0) nanoclusters catalyst.
2
.12. XPS sample preparation and XPS analysis
Briefly, 10 mL of aliquots of acetate stabilized-ruthenium(0)
nanoclusters solutions were transferred via disposable pipette
into eight new 15 mm × 100 mm glass tubes separately and each
samples was simultaneously centrifuged for 3 h. The centrifuged
material was rinsed three times with a 30% (v/v) water/ethanol
mixture to remove water-soluble impurities. Next, the precipi-
tates were dried under vacuum for 2 h. All the nanoclusters sam-
ples dried were collected in a screw-capped glass-vial and sent
to the Colorado State University for XPS analysis. X-ray pho-
toelectron spectroscopy (XPS) was performed at the Colorado
State University using a Physical Electronics 5800 spectrometer
equipped with a hemispherical analyzer and using monochro-
matic Al K␣ radiation (1486.6 eV, the X-ray tube working at
The effect of NaBH substrate concentration on the hydro-
4
genation rate was also studied by performing a series of experi-
mentsstartingwithvaryinginitialconcentrationofNaBH while
4
keeping the catalyst concentration constant at 0.32 mM Ru(0).
Fig. 3 shows the plot of hydrogen generation rate versus NaBH4
concentration, both on logarithmic scales and indicates that the
catalytic hydrolysis is zero order with respect to substrate con-
centration. Taking both results together, the rate law for the
1
5 kV and 350 W) and pass energy of 23.5 eV.
2
.13. FTIR analyses
The sample prepared for the XPS analysis was also used
for the FTIR analysis. FTIR spectrum of the acetate stabilized-
ruthenium(0) nanoclusters was taken from KBr pellet on a Nico-
let 510 FTIR spectrophotometer using Omnic software.
3
. Results and discussion
3
.1. Use of water dispersible acetate stabilized
ruthenium(0) nanoclusters as catalyst for the hydrolysis of
sodium borohydride
Fig. 1. Plot of the volume of hydrogen (mL) vs. time (s) for the hydrolysis
of sodium borohydride catalyzed by ruthenium(0) nanoclusters with different
catalyst concentrations at 25.0 ± 0.1 C. [NaBH4]0 = 150 mM.
Water dispersible acetate stabilized-ruthenium(0) nanoclus-
ters were used as catalyst in the hydrolysis of sodium borohy-
◦

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
99
Fig. 2. Plot of the hydrogen generation rate vs. the catalyst concentration (both
in ln scale) in the hydrolysis of 150 mM NaBH4 catalyzed by ruthenium(0)
◦
nanoclusters at 25.0 ± 0.1 C.
catalytic hydrolysis of sodium borohydride can be given as
d[H2]
Rate =
= k[Ru]
(3)
dt
The hydrolysis of sodium borohydride was carried out at var-
◦
ious temperature in the range of 25–45 C starting with the initial
substrate concentration of 150 mM NaBH4 and an initial catalyst
concentration of 0.32 mM Ru(0). The values of the rate constant
k determined at five different temperatures are used to create the
Arrhenius and Eyring plots shown in Fig. 4. The Arrhenius acti-
vation energy was found to be Ea = 41 ± 2 kJ/mol for the acetate
stabilized ruthenium(0) nanoclusters catalyzed hydrolysis of
sodium borohydride. This value compares favorably with the
activation energy of 47 kJ/mol found at 5.6 M NaBH4 concen-
tration [21] and 56 kJ/mol found at 2 M NaBH4 concentration for
thehydrolysiscatalyzedbybulkruthenium[4a]. Itislessthanthe
activation energies found for the same hydrolysis using various
bulk metal catalyst: 75 kJ/mol for cobalt, 71 kJ/mol for nickel,
and 63 kJ/mol for Raney nickel [8]. The Eyring plot in Fig. 4b
provides the activation enthalpy and activation entropy values:
Fig. 4. (a) Arrhenius and (b) Eyring plots for the ruthenium(0) nanoclusters-
catalyzed hydrolysis of 150 mM NaBH4. [Ru] = 0.32 mM.
#
#
ꢀ
H = 38 ± 1 kJ/mol and ꢀS = −117 ± 6 J/K mol for the ruthe-
nium(0) nanoclusters catalyzed hydrolysis of 150 mM sodium
borohydride solution. The small value of activation enthalpy
and the large negative value of activation entropy are indicative
of an associative mechanism for the ruthenium(0) nanocluster-
catalyzed hydrolysis of sodium borohydride, in line with the
mechanism suggested for the hydrolysis of sodium borohy-
*
dride given in the literature [22] consisting of [BH ] formed as
5
transient.
Since most of the prior studies on the catalytic hydrolysis of
sodium borohydride have been carried out in basic medium,
mostly in the presence of 10% (bw) NaOH, for comparison
we also performed the hydrolysis of sodium borohydride in the
same medium by using the acetate stabilized ruthenium(0) nan-
oclusters as catalyst. The values of the rate constant k of the
catalytic hydrolysis of sodium borohydride solution containing
10% sodium hydroxide at varying temperatures were used to
obtain the Arrhenius plot shown in Fig. 5. The Arrhenius acti-
vation energy was found to be 43 ± 2 kJ/mol. This is slightly
greater than the activation energy in the absence of sodium
hydroxide (41 ± 2 kJ/mol), but still smaller than the literature
value of 47 kJ/mol found for bulk ruthenium at 5.6 M NaBH4
concentration [21]. The small increase in the activation energy
upon addition of sodium hydroxide indicates that the hydrolysis
reaction can be inhibited by the hydroxide ion in agreement with
Fig. 3. Plot of the hydrogen generation rate vs. the NaBH4 concentration (both
in ln scale) in the hydrolysis of NaBH4 catalyzed by ruthenium(0) nanoclusters
◦
(
[Ru] = 0.32 mM) at 25.0 ± 0.1 C.

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
1
00
stabilization against coagulation of the particles. The acetate ion
adsorbed on the nanoclusters surface is in dynamic equilibrium
with the acetate ion in the solution, so that some of the sur-
face metal atoms are still available providing active sites for the
catalysis. However, at higher stabilizer concentration, the sur-
face active sites are blocked by the acetate ions adsorbed on the
surface and the catalytic activity decreases with the increasing
concentration of the acetate ion. The acetate ion concentration
around 1.0 appears to be the optimum one.
Ruthenium(0) nanoclusters were isolated by centrifugation
followed by drying in vacuum. These isolated ruthenium(0)
nanoclusters were redispersed in water and tested for their cat-
alytic activity for sodium borohydride hydrolysis after 2 days of
storage. The isolated ruthenium(0) nanoclusters could be easily
dispersed in water and giving rise to a transparent dark brown
colloidal solution of ruthenium. After 2 days of storage, the
sample prepared by redispersing the isolated ruthenium(0) nan-
oclusters was found to be highly active catalyst in the hydrolysis
of sodium borohydride. The reaction rates indicate that ruthe-
nium(0) nanoclusters retain 70% of their catalytic activity in
the hydrolysis of sodium borohydride, when isolated and redis-
persed in water. The deactivation of ruthenium(0) nanoclusters
catalyst may be due to agglomeration or partial oxidation of
ruthenium. In fact, an XPS study shows the presence of ruthe-
nium oxide.
Fig. 5. Arrhenius plot for ruthenium(0) nanoclusters catalyzed hydrolysis of
sodium borohydride in the alkaline solution containing 10% NaOH (w/w),
[
5
Ru] = 0.32 mM, [NaBH4] = 150 mM, at different temperatures 25, 35, 45, and
5 C.
◦
the previous finding [4a]. According to the established mech-
anism [22], the hydrolysis of sodium borohydride is initiated
by the attack of borohydride anion on the hydronium ion. In
basic solution the reduction of proton concentration causes a
decrease in the rate of hydrolysis. However, the activation energy
obtained for the ruthenium(0) nanoclusters catalyzed hydrolysis
of sodium borohydride in basic medium is still lower than the
ones reported for the use of heterogeneous bulk ruthenium [4a].
As a result, Ru(0) nanoclusters are more active catalyst than bulk
ruthenium metal in the hydrolysis of sodium borohydride in the
absence or presence of sodium hydroxide.
The effect of the stabilizer concentration on the catalytic
activity of the ruthenium(0) nanoclusters was investigated by
performing the catalytic hydrolysis of sodium borohydride start-
ing with 0.16 mM ruthenium(III) ion and different acetate ion
concentration. Fig. 6 shows how crucial the effect of concentra-
tion of the stabilizer is on the catalytic activity of ruthenium(0)
nanoclusters. The highest catalytic activity was achieved for
ruthenium(0) nanoclusters stabilized by 1.0 M sodium acetate. It
is worth to note that the high activity of ruthenium(0) nanoclus-
ters is obtained only in a narrow range around 1.0 M acetate ion
concentration. The acetate ion is adsorbed on the nanoclusters
surface generating a double layer, which provides electrostatic
Fig. 7. (a) TEM image and (b) associated histogram for acetate-stabilized ruthe-
nium(0) nanoclusters prepared from the reduction of 1.33 mM ruthenium(III)
chloride trihydrate by 9.3 mM sodium borohydride in the presence of 1.0 M
acetate ion in aqueous solution.
Fig. 6. Turnover frequency for the hydrolysis of sodium borohydride (150 mM)
catalyzed by acetate stabilized Ru(0) nanoclusters with different acetate con-
centrations at 25 C. [Ru] = 0.16 mM.
◦

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
101
3
.2. Characterization of water-dispersible acetate
Fig. 9 shows the FTIR spectrum of the isolated ruthenium(0)
nanoclusters taken from KBr pellet shows the absorption bands
due to the acetate ion at 1580 and 1425 cm indicating the
existence of acetate ion in the nanocluster sample, most probably
on the surface of ruthenium(0) nanoclusters.
stabilized ruthenium(0) nanoclusters catalyst
−1
Fig. 7 gives the TEM image and histogram of ruthenium(0)
nanoclusters prepared from the reduction of ruthenium(III) chlo-
ride trihydrate by sodium borohydride in the presence of sodium
acetate. The histogram was constructed from the TEM image by
using NIH image program, whereby 235 non-touching particles
were counted. The particle size ranges from 1.0 to 6.0 nm with
a mean value of 2.8 nm and a standard deviation of 1.4 nm.
Fig. 8 shows the XPS spectrum of acetate stabilized ruthe-
nium(0) nanoclusters exhibits two signals at 280 and 462 eV for
3.3. Catalyst poisoning
Catalyst poisoning experiments are vital to show that the
acetate stabilized ruthenium(0) nanoclusters are the true (het-
erogeneous; metal particle) catalyst [25]. A series of mercury
poisoning experiments were carried out by adding mercury in
varying amount during the catalytic hydrolysis of sodium boro-
hydride and measuring the catalytic activity before and after
addition. Fig. 10 shows the results of eight independent experi-
ments as a plot of the relative rate of hydrogen generation versus
the relative mercury concentration as mol Hg/mol Ru. Experi-
mentalpointscanbefittedtoaline, theintersectionofwhichwith
the concentration axis provides the critical molar ratio of mer-
Ru 3d5 and Ru 3p3/2, respectively. These signals are readily
/2
assigned to the Ru(0). Additional peaks are observable at 284
and 466 eV, which can be assigned to higher oxidation states of
ruthenium such as Ru(IV) in RuO2. The formation of Ru oxides
in the XPS sampling procedure is a known phenomenon [23].
The signals beyond C 1s peak at 285 eV are attributable to the
Ru 3d3/2 of higher oxidation species [24,19].
Fig. 8. X-ray photoelectron spectrum of acetate stabilized ruthenium(0) nanoclusters: (a) in the region of 275–295 eV and (b) in the region of 458–478 eV, exhibiting
two prominent signals at 280 and 462 eV for Ru(0) 3d5/2 and Ru(0) 3p3/2, respectively. The peaks at 284 and 466 eV are due to a Ru(IV) species. The peaks higher
beyond the C 1s peak at 285 eV are attributable to the Ru 3d3/2 in higher oxidation states.

¨
1
02
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
Fig. 9. FTIR spectra of free acetate anion (top curve) and acetate stabilized Ru(0) nanoclusters (bottom curve).
cury to ruthenium [26]. Thus, the minimum amount of mercury
required for the complete poisoning of the ruthenium(0) nan-
oclusters catalyst was found to be 0.13 ± 0.05 mol of Hg/mol of
total ruthenium. This indicates that the acetate stabilized ruthe-
nium(0) nanoclusters are the true, heterogeneous catalyst. By
assuming a 1/1 Hg/Ru stoichiometry for the poisoning [25], one
estimates 13% of the total ruthenium atoms to be active in cataly-
sis. Notethatabout42%rutheniumatomsareonthenanoclusters
surface as estimated from the magic number nanoclusters size
sodium borohydride over 84 min before deactivation. This cor-
−
1
responds to a turnover frequency of 1.0 s
.
4. Conclusion
In summary, our study of the synthesis and characterization
of water soluble ruthenium(0) nanoclusters as catalyst for the
hydrolysis of sodium borohydride have led to the following con-
clusions and insights.
Water dispersible acetate stabilized ruthenium(0) nanoclus-
ters can be formed from the reduction of a commercially avail-
able precursor RuCl3·3H2O by sodium borohydride in the pres-
enceofacetateanion[19], whicharestableenoughtobeisolated,
but which retain high catalytic activity for the hydrolysis of
sodium borohydride in aqueous solution, Eq. (2).
A kinetic study shows that the catalytic hydrolysis of sodium
borohydride is first order in ruthenium concentration and zero
orderinsubstrateconcentration. Theapparentsaturationkinetics
in NaBH4 implies that the nanocluster surface is saturated by
borohydride anions.
[
14a].
3
.4. Catalyst lifetime
A catalyst lifetime experiment was performed starting with
0
.32 mM Ru(0) nanoclusters and 450 mM sodium borohydride
◦
in 50 mL aqueous solution at 25.0 ± 0.1 C, corresponding to
a maximum total turnover number of 5625. It was found that
the acetate stabilized ruthenium(0) nanoclusters provide 5170
turnovers of hydrogen gas generation from the hydrolysis of
Using ruthenium(0) nanoclusters as catalyst provides ≈10%
loweractivationenergyforthehydrolysisofsodiumborohydride
comparedtothebulkcounterpart, sothatthereactionoccurseven
at room temperature. The acetate stabilized ruthenium(0) nan-
oclusters are highly active catalyst with long lifetime providing
5
170 total turnovers in the hydrolysis of sodium borohydride at
◦
25.0 ± 0.1 C over 1.4 h before they are deactivated.
Overall, acetate-stabilized ruthenium(0) nanoclusters can
easily be formed by using a commercially available precursor
material and serve as highly active catalysts even at ambient tem-
peratures for the hydrolysis of sodium borohydride to generate
hydrogen gas.
Acknowledgments
Fig. 10. Plot of relative rate of hydrolysis of sodium borohydride vs. moles of
Hg/mol of total Ru(0) for the hydrolysis of sodium borohydride by 2.8 ± 1.4 nm
ruthenium(0) nanoclusters.
Partial support of this work by TUBITAK (BOREN Project
105M357), Turkish Academy of Sciences and METU-DPT-

¨
M. Zahmakıran, S. Ozkar / Journal of Molecular Catalysis A: Chemical 258 (2006) 95–103
103
OYP Program. The XPS and TEM images herein were obtained
with the assistance of the Department of Energy, Office of Basic
Energy Sciences, via DOE grant FG06-089ER13998 to Prof. Dr.
R.G. Finke.
[15] (a) L.N. Lewis, Catalysis by Di- and Polynuclear Metal Cluster Complexes,
VCH, Weinheim, 1988, p. 373;
(
(
b) L.N. Lewis, Chem. Rev. 93 (1993) 2693;
c) K.S. Weddle, J.D. Aiken, R.G. Finke, J. Am. Chem. Soc. 120 (1998)
5653;
(
d) J.D. Aiken, R.G. Finke, J. Mol. Catal. A., Chem. 145 (1999) 1;
References
(e) H. B o¨ nnemann, G. Braun, W. Brijoux, A.S. Tiilling, K. Seevogel, K.
Siepen, J. Organomet. Chem. 520 (1996) 143;
(
(
(
(
f) H. B o¨ nnemann, G. Braun, Chem. Eur. J. 3 (1997) 1200;
g) J.U. K o¨ hler, J.S. Bradley, Langmuir 14 (1998) 2730;
h) J. Schulz, A. Roucoux, H. Patin, Chem. Eur. J. 6 (2000) 618;
i) Y. Na, S. Park, S.B. Han, H. Han, S. Ko, S. Chang, J. Am. Chem. Soc.
[
1] Annual Energy Outlook 2005 With Projections To 2025, Energy
Information Administration, February 2005, http://www.eia.doe.gov/oiaf/
aeo/pdf/0383(2005).pdf.
[
2] Basic Research Needs For the Hydrogen Economy, Report of the Basic
Energy Sciences Workshop on Hydrogen Production, Storage and Use,
May 13–15, 2003, Office of Science, U.S. Department of Energy,
http://www.sc.doe.gov/bes/hydrogen.pdf.
1
26 (2004) 250.
16] (a) M. Zhao, R.M. Crooks, Adv. Mater. 11 (1999) 217;
b) M.M. Maye, Y. Lou, C.J. Zhong, Langmuir 16 (2000) 7520.
[
(
[
[
17] R. Pool, Science 248 (1990) 1186.
[
[
3] M.H. Sossina, Acta Mater. 51 (2003) 5981.
4] (a) S.C. Amendola, J.M. Janjua, N.C. Spencer, M.T. Kelly, P.J. Petillo, S.L.
Sharp-Goldman, M. Binder, Int. J. Hydr. Energy 25 (2000) 969;
¨
18] S. Ozkar, M. Zahmakıran, J. Alloys Compd. 404–406 (2005) 728.
Preliminary results of our study on the use of ruthenium(0) nanoclusters as
catalyst in the hydrolysis of sodium borohydride were presented in the 9th Inter-
national Symposium on Metal-Hydrogen Systems, Fundamentals, Applications,
Cracow, Poland, and published in its proceedings.
(
1
b) J.Y. Lee, H.H. Lee, J.H. Lee, D.M. Kim, J.H. Kim, J. Electrochem. Soc.
49 (2002) A603.
[
5] H.I. Schlesinger, H.C. Brown, A.B. Finholt, J.R. Gilbreath, H.R. Hockstra,
E.K. Hydo, J. Am. Chem. Soc. 75 (1953) 215.
6] A. Levy, J.B. Brown, C.J. Lyons, Ind. Eng. Chem. 52 (1960) 211.
7] H.C. Brown, C.A. Brown, J. Am. Chem. Soc. 84 (1962) 1493.
8] C.M. Kaufman, B. Sen, J. Chem. Soc. Dalton. Trans. 83 (1985) 307.
9] I.I. Korobov, N.G. Mozgina, L.N. Blinova, Kinet. Catal. 48 (1995) 380.
10] S.C. Amendola, P. Onnerud, M.T. Kelly, P.J. Petillo, S.L. Sharp-Goldman,
M. Binder, J. Power Sources 85 (2000) 186.
11] Y. Kojima, K.I. Suzuki, K. Fukumoto, M. Sasaki, T. Yamamoto, Y. Kawai,
H. Hayashi, Int. J. Hydr. Energy 27 (2002) 1029.
[
[
19] J. Yang, T.C. Deivaraj, H.P. Too, J.Y. Lee, Langmuir 20 (2004) 4241.
¨
20] J.E. Hutchison, G.H. Woehrle, S. Ozkar, R.G. Finke, Turkish J. Chem. 30
2006) 1.
[
[
[
[
(
[
[
21] S.C. Amendola, J.M. Janjua, N.C. Spencer, M.T. Kelly, P.J. Petillo, G.S.L.
Sharp, M. Binder, J. Power Sources 85 (2000) 186.
22] (a) R.E. Davis, C.G. Swain, J. Am. Chem. Soc. 82 (1960) 5950;
[
[
[
[
[
(
b) R.E. Mesmer, W.L. Jolly, Inorg. Chem. 1 (1962) 608.
In these studies on the acid catalyzed hydrolysis of sodium borohydride, a
*
mechanism has been suggested involving the formation of [BH5] in the tran-
−
+
*
sition state; BH4 + H → [BH5] → BH3 + H2. Two activated complexes have
been found to be reasonable, one having an attack on hydrogen, the other on
boron of the borohydride ion.
12] D. Hua, Y. Hanxi, A. Xinping, C. Chuansain, Int. J. Hydr. Energy 28 (2003)
1
095.
13] J.H. Kim, H. Lee, S.C. Han, H.S. Kim, M.S. Song, J.Y. Lee, Int. J. Hydr.
Energy 29 (2004) 263.
14] (a) G. Schmid, Chem Rev. 92 (1992) 1709;
[
[
[
[
23] M.W.F. Clune, Powder Diffraction File Alphabetical Index Inorganic
Phase, JCPPDS, Swarthnore, PA, 1980.
24] N. Chakroune, G. Viau, S. Ammar, L. Poul, D. Veatuier, M.M. Chemimi,
C. Manganey, F. Villain, F. Fievet, Langmuir 21 (2005) 6788.
25] R.G. Finke, B.J. Hornstein, J.D. Aiken, Inorg. Chem. 41 (2002)
(
b) G. Schmid (Ed.), Clusters and Colloids From Theory to Applications,
VCH, Weinheim, 1994;
c) K.J. Klabunde, G.C. Trivino, n Active Metals; Preparation Characteri-
zation, Applications, VCH, Weinheim, 1996;
d) N. Toschima, T. Yonezawa, New J. Chem. 22 (1998) 1179.
(
1625.
26] E.B. Maxted, Adv. Catal. 3 (1951) 129.
(