Hydrogen generation from solvolysis of sodium borohydride in ethylene glycol-water mixtures over a wide range of temperature

Article abstract of DOI:10.1039/c3ra43136c

A high-performance hydrogen generation system with fast kinetics and a wide range of operational temperature is highly desirable for promoting the implementation of hydrogen fuel cell technology. In the present study, we report a comparative study of the hydrogen generation properties from the reactions between sodium borohydride (NaBH₄) and ethylene glycol (EG), water (H₂O) or their mixture. Our study found that the glycolysis kinetics of NaBH₄ is faster than the hydrolysis kinetics at moderate temperatures, but gets sluggish at low temperatures. As a solution, the combined usage of EG-water mixture as solvent and cobalt chloride (CoCl₂) as a promoting additive enables the system to rapidly deliver H₂ at low temperatures. A series of control experiments have been conducted to evaluate the hydrogen generation property dependence on EG concentration, CoCl ₂ amount and ratio of EG-H₂O mixture to NaBH₄. The reaction byproducts were characterized by powder X-ray diffraction and Fourier transform infrared spectroscopy techniques. Our study demonstrated a high-performance hydrogen generation system with a wide range of operational temperature, which may lay the foundation for developing practical hydrogen source for mobile/portable applications.

Full text of DOI:10.1039/c3ra43136c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Hydrogen generation from solvolysis of sodium
borohydride in ethylene glycol–water mixtures over a
Cite this: RSC Adv., 2013, 3, 23810
wide range of temperature
Da-Wei Zhuang, Hong-Bin Dai* and Ping Wang*
A high-performance hydrogen generation system with fast kinetics and a wide range of operational
temperature is highly desirable for promoting the implementation of hydrogen fuel cell technology. In
the present study, we report a comparative study of the hydrogen generation properties from
the reactions between sodium borohydride (NaBH
mixture. Our study found that the glycolysis kinetics of NaBH
moderate temperatures, but gets sluggish at low temperatures. As a solution, the combined usage of
EG–water mixture as solvent and cobalt chloride (CoCl ) as a promoting additive enables the system to
rapidly deliver H at low temperatures. A series of control experiments have been conducted to evaluate
the hydrogen generation property dependence on EG concentration, CoCl amount and ratio of
EG–H O mixture to NaBH . The reaction byproducts were characterized by powder X-ray diffraction and
4
) and ethylene glycol (EG), water (H
2
O) or their
4
is faster than the hydrolysis kinetics at
2
2
2
Received 22nd June 2013
2
4
Accepted 10th October 2013
DOI: 10.1039/c3ra43136c
www.rsc.org/advances
Fourier transform infrared spectroscopy techniques. Our study demonstrated a high-performance
hydrogen generation system with a wide range of operational temperature, which may lay the
foundation for developing practical hydrogen source for mobile/portable applications.
regions or in cold weather. Second, the byproduct from the
1
. Introduction
alcoholysis reaction of NaBH can be readily removed from the
4
9
Lack of safe and efficient means for hydrogen storage is widely reactor, which is of clear benet for the apparatus design of
recognized as a major obstacle to the widespread implementa- practical hydrogen sources. In addition, according to a patent
1
tion of hydrogen fuel cell technology. Recently, on-demand disclosed by Ramachandran et al., the alcoholysis byproduct
14
hydrogen generation (HG) from chemical hydrides, in
conjunction with spent fuel regeneration, emerges as a prom-
ising approach for vehicular or portable hydrogen storage
can be regenerated back to NaBH
Stimulated by this situation, we conducted a comparative
study of the HG properties from the reactions between NaBH
4
in one step.
4
applications. Among the chemical hydrides of interest, sodium and EG, water or their mixture in a wide temperature range.
2
,3
borohydride (NaBH
4
) received the most extensive studies.
Here, the selection of EG or EG–H O mixture as reactant is
2
4,5
NaBH can react with water or primary alcohols to liberate H
4
2
.
mainly based on their low freezing points, e.g. a mixture of 60%
ꢁ
In the past decade, numerous reports have been published EG and 40% H O freezes at ꢀ45 C, which offers opportunity for
2
dealing with the HG properties of NaBH -based hydrolysis operating the NaBH -based system at low temperatures. Our
4
4
systems, whereas the studies of alcoholysis of NaBH were quite
sparse. This preference appears understandable from the
advantages of hydrolysis systems over alcoholysis systems in
4
study found that the HG kinetics of glycolysis reaction is
superior to that of the hydrolysis reaction. In particular, the
system using EG–water mixture as reactive solvent and cobalt
terms of hydrogen density and hydrogen cost. But according to chloride (CoCl ) as a promoting additive exhibited fast HG
2
the open literatures, NaBH
favorable attributes that are desirable for practical hydrogen make it promising for mobile/portable applications.
sources. First, NaBH can readily react with low-molecular-
weight primary alcohols, such as methanol, ethanol or ethylene
4
-based alcoholysis system does have kinetics and high fuel conversion at low temperatures, which
4
6
–14
glycol (EG) at low temperatures,
which is an important
2
. Experimental
requirement for vehicular and portable applications in cold
2.1. Chemicals
NaBH (96%), CoCl (99%), EG (99%) and hydrated sodium
4
2
Shenyang National Laboratory for Materials Science, Institute of Metal Research,
Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016, P. R. China.
E-mail: hbdai@imr.ac.cn; pingwang@imr.ac.cn; Fax: +86 24 2389 1320; Tel: +86 24
metaborate (NaB(OH) , 99%) were all purchased from Sino-
4
pharm. Deionized water was used in preparation of all the
2
397 1622
EG–water mixtures.
2
3810 | RSC Adv., 2013, 3, 23810–23815
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
.2. Hydrogen generation performance testing
RSC Advances
2
The HG performance testing was conducted in a 250 ml three-
ꢁ
neck ask. In the experiments conducted at above 0 C, the ask
was placed in a thermostat that was equipped with a water
circulating system to maintain the reaction temperature, typically
ꢁ
within ꢂ1 C. In a typical measurement run, the liquid reactant
2
with or without CoCl additive was pre-heated and held at the
designated temperature, and then NaBH powder was dropped
4
into the solution to initiate the solvolysis reaction. For conducting
ꢁ
the experiments at or below 0 C, we installed the experimental
setup in a refrigerator with a controllable temperature range of
0
ꢃ ꢀ20 ꢁC. The generated hydrogen amount was accurately
measured by water-displacement method using an electronic
balance with an accuracy of 0.01 g. The weight change was
automatically recorded by data acquisition soware. The reaction
temperature was monitored using a thermocouple inserted in the
fuel solution and recorded using an online recorder. Each
experiment was repeated at least two times to ensure the repro-
ducibility. The determined relative error was no more than 2%.
Fig. 1 HG kinetics curves of the glycolysis and hydrolysis reactions of NaBH
varied temperatures in the absence of catalyst. The system is composed of (a) 0.1
and 10 ml of EG; (b) 0.1 g of NaBH and 10 ml of water. (c) and (d) give
4
at
2
.3. Characterization of the byproducts
The byproducts from the solvolysis reactions were collected and g of NaBH
4
4
the Arrhenius plots of the temperature-dependent first-order rate constant,
characterized by powder X-ray diffraction (XRD, Rigaku D/MAX-
500, Cu Ka radiation) and Fourier transform infrared spectros-
which determined the apparent activation energies of the glycolysis and hydro-
2
4
lysis reactions of NaBH , respectively.
ꢀ
1
copy (FTIR Bruker TENSOR 27, 4 cm resolution). FTIR spectra
of the samples were collected using KBr-pellet method and the
obtained spectra were normalized using OPUS 6.5 soware.
3.2. Hydrogen generation from catalyzed reactions between
4
NaBH and ethylene glycol–water mixture
3
. Results and discussion
In an effort to improve the HG kinetics at low temperatures, we
employed a combination of antifreeze and catalysis strategies
to constitute a new system that is composed of solid NaBH
3
.1. Kinetic study of glycolysis and hydrolysis reactions of
NaBH
The solvolysis kinetics of NaBH
solvent choice. Fig. 1 presents a comparison of HG kinetics
between the glycolysis and hydrolysis reactions of NaBH at
4
4
,
4
is highly dependent on the
EG–H O mixture and CoCl additive. Here, the use of EG–H O
2
2
2
mixture is to lower the freezing point of the reactive solvent,
which lays foundation for developing practical hydrogen source
with a wide range of operational temperature. The selection of
4
varied temperatures. It was found that the HG kinetics of spon-
taneous glycolysis reaction is much faster than that of the
hydrolysis reaction. For example, NaBH can immediately react
with EG at 30 C and the system fullls a 100% fuel conversion
within 3 min. In contrast, it takes over 8 h for the aqueous
solution of NaBH
temperature. Consistently, the apparent activation energy of the
glycolysis reaction of NaBH
CoCl additive is based on its well-established promoting effect
2
4
16–23
on the hydrolysis reaction of NaBH
CoCl was pre-dissolved in the EG–H
that the transparent solution immediately turned into black upon
contacting with NaBH powder, indicative of in situ formation of
4
.
In the present study,
ꢁ
2
2
O mixture. It was observed
4
to fulll ꢃ61% fuel conversion at the same
4
ꢀ1
4
was determined to be 25.8 kJ mol
using the Arrhenius equation, which is much lower than that of
ꢀ1
the hydrolysis reaction of NaBH (95.9 kJ mol ).
4
As shown in Fig. 2, the glycolysis reaction of NaBH obeys
4

rst-order kinetics with respect to NaBH4 concentration, but
shows a lag time at the initial stage. A similar observation was
also reported by Lo et al. in their study of the NaBH /methanol
system. Presumably, the appearance of lag time is associated
with the solution of NaBH in alcohol solvents.
The glycolysis reaction of NaBH is kinetically rapid at
4
5
4
4
ambient temperature. But upon reducing the reaction tempera-
ꢁ
ture to 0 C (as seen in Fig. 1a), the kinetics of glycolysis reaction
becomes sluggish and the HG process is entangled with a long
Fig. 2 Fitting the experimental data of the system (0.1 g of NaBH
4
and 10 ml of
lag time. Similar problems were also encountered in the NaBH
4
/
ꢁ
EG at 30 C) using a first-order kinetics with respect to NaBH
4
concentration. Y
5,7,8,15
methanol system.
4
denotes the fractional conversion of NaBH .
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23810–23815 | 23811

View Article Online
RSC Advances
Paper
ꢁ
25
Co–B catalyst via the reaction between NaBH
following eqn (1).
4
and CoCl
2
107 C. Here, it should be noted that using the EG–H O
2
16
mixture with a proper ratio not only depresses the freezing
point, but also elevates the boiling point such that the oper-
ating range of the system is broadened on both ends of the
temperature scale.
CoCl + 2NaBH
2
4
+ 3H
2
O /
2 2
B + 2NaCl + 1.5HBO + 6.25H (1)
0.5Co
2
Next, we conducted another set of control experiments to
Property examination found that the catalyzed NaBH
4
/
evaluate the effect of CoCl
system. Interestingly, our study found that the required amount
of CoCl for rapid HG from the system is highly dependent on
2
amount on the HG properties of the
(
EG–H O) system showed considerable property dependence on
2
the composition of the system. In an effort to optimize the system
composition, we rst examined the effect of EG concentration on
the HG properties of the system. Fig. 3 presents the HG kinetics
2
the operational temperature. As seen in Fig. 4, the system
composed of 0.5 g of NaBH , 10 g of 50 wt% EG–50 wt% H O
4
2
curves of the systems with EG concentrations ranging from 0 to
ꢁ
mixture and 20 mg of CoCl can rapidly deliver H at 0 C. When
2
2
ꢁ
1
00 wt%. All the measurements were conducted at 0 C and no
ꢁ
the operational temperature was lowered to ꢀ10 and ꢀ20 C, the
required CoCl amount for achieving fast HG kinetics remark-
ably increased to 50 and 100 mg, respectively. Presumably, the
unusual requirement of a large amount of CoCl is an indication
of the low reactivity between NaBH and EG or water at low
attempt was made to control the reaction temperature. It was
found that the HG kinetics became increasingly favorable with
increasing EG concentration. This result is not unexpected in
view of the kinetics advantage of glycolysis over hydrolysis reac-
tion (Fig. 1). A similar trend was also reported by Fernandes
2
2
4
temperatures, even with the presence of Co–B catalyst. As a
solution, we can make use of the heat released from the reaction
between CoCl and NaBH to preheat the system to over the
15
9
et al. and Hannauer et al., respectively, in their studies of the
reaction kinetics between NaBH and methanol–H O mixture.
4
2
2
4
Fundamentally, the negative effect of water on the HG kinetics is
an indication of reaction competition between hydrolysis and
alcoholysis on the catalyst surface. In both reactions, the
adsorption of water or alcohol molecules on the active sites of
the catalyst is a critical step. As the H–O bond of water is more
polarized than that of primary alcohols, water molecules
should be more strongly adsorbed on the catalyst surface
threshold temperature, and thereby initiating the catalyzed
glycolysis and hydrolysis reactions of NaBH . As a support to this
speculation, we observed that addition of Co–B catalyst instead
4
of CoCl
EG–H
2
additive could not initiate the HG process of the NaBH
4
/
ꢁ
(
2
O) system at ꢀ10 C (Fig. 4). These results clearly indicate
that CoCl additive plays a dual role in the HG process at low
2
temperatures. One is the precursor for formation of Co–B cata-
lyst according to eqn (1), and the other is a reactive component
9
,24
relative to alcohol molecules.
As a consequence, the HG
process becomes increasingly dominated by the relatively
slow hydrolysis reaction with increasing the water content.
Although water exerts negative effect on the HG kinetics, the
for providing heat via the reaction with NaBH
NaBH + 2CH OHCH OH / NaB(OCH
NaBH + 4H O / NaB(OH) + 4H
2
4
.
4
2
2
2
)
4
+ 4H
2
(2)
(3)
use of EG–H O mixture can effectively lower the freezing point
2
of reactive solvent. In addition, properly increasing water
content is benecial for enhancing the effective hydrogen
density of the system. In an overall consideration of HG
kinetics, hydrogen density and freezing point of the solvent,
we selected the system using a mixture of 50 wt% EG–50 wt%
4
2
4
H
2
O for detailed study. According to the tabulated data, this
ꢁ
solvent has a freezing point of ꢀ34 C and a boiling point of
Fig. 3 Effect of EG concentration on the HG kinetics of the system composed of
Fig. 4 HG kinetics (top) and reaction temperature (bottom) of the systems
composed of 0.5 g of NaBH , 10 g of 50 wt% EG–50 wt% H O mixture and varied
amounts of CoCl . The measurements were conducted at an initial temperature
of 0, ꢀ10 and ꢀ20 ꢁC, respectively.
0
.5 g of NaBH
4
, 10 g of xEG–(100 ꢀ x)H
2
O mixture (x ¼ 0, 25, 50, 75 and 100 wt%,
4
2
respectively) and 20 mg of CoCl
2
. The measurements were conducted at an initial
2
ꢁ
temperature of 0 C.
2
3812 | RSC Adv., 2013, 3, 23810–23815
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
4 2
Finally, we examined the effect of NaBH : (EG + H O) mass
ratio on the HG kinetics and hydrogen density of the system. As
1
3
the glycolysis and hydrolysis reactions following eqn (2) and (3),
respectively, consume NaBH and the 50 wt% EG + 50 wt% H O
4
2
mixture in a mass ratio of 1 : 2.4, we selected a NaBH
mass ratio range of 1 : 2.4 ꢃ 1 : 20 in the set of control experi-
ments. For comparison purpose, the amounts of NaBH and CoCl
were xed at 0.5 g and 20 mg, respectively. As shown in Fig. 5, the
system with a NaBH : (EG + H O) mass ratio of 1 : 2.4 showed the
4 2
: (EG + H O)
4
2
4
2
worst HG performance among the examined systems, which ful-
ꢁ

lled a fuel conversion of only 45% at 0 C even aer prolongation
of the reaction duration to 10 min. In contrast, the systems with
NaBH : (EG + H O) mass ratios ranging from 1 : 4 to 1 : 20 were
4
2
observed to achieve a 100% fuel conversion. Clearly, the low fuel
conversion in the former system should be attributed to mass-
transfer limitation in a solvent-decient environment. But on the
Fig. 6 Photos taken in the reaction process and at the ending stage of the
system (a) composed of 0.5 g of NaBH
system (b) composed of 0.5 g of NaBH , 5 g of 50 wt% EG–50 wt% H
and 20 mg of CoCl . The illustrative experiments were conducted at room
4
, 5 g of water and 20 mg of CoCl
2
and
4
2
O mixture
2
other hand, overloading the EG–H O mixture will mitigate the
2
temperature rising of the system that arises from the proceeding temperature.
of exothermic solvolysis reactions, thus negatively impacting the
HG kinetics. This is actually the case for the systems with
NaBH
consideration of HG kinetics, fuel conversion and hydrogen
density, we determined the optimum NaBH : (EG + H O) mass
ratio to be 1 : 4. Thus-constituted system can deliver 4.2 wt% H
4 2
: (EG + H O) mass ratios higher than 1 : 10. In an overall
hydrolysis byproduct. Fig. 6 presents the photos of the two
systems that were taken in the reaction process and at the
ending stage. It was observed that the byproduct from the
hydrolysis reaction severely deposited on the tube wall, whereas
the byproduct(s) of the NaBH /(EG–H O) system largely settled
4 2
down at the bottom of the test tube. This is of clear benet for
the practical applications.
4
2
2
ꢁ
within 6 min at 0 C. This HG performance compares favorably
with the literature results of the NaBH /(methanol–water) system
4
ꢁ
ꢁ
(
3.4 wt% at 20 C (ref. 9) and 2.9 wt% at 45 C (ref. 15)).
The solvolysis byproduct(s) of the NaBH /(EG–H O) system
4
2
with an optimized composition was collected, ltered and dried
at 200 C under Ar atmosphere for 2 h. The obtained white
3.3. Analysis of the byproducts from solvolysis reactions
ꢁ
An added advantage of the NaBH /(EG–H O) system over the powder was characterized by XRD and FTIR. As shown in Fig. 7,
4
2
hydrolysis system is that the glycolysis byproduct does not tend the XRD pattern of the byproduct(s) showed a new set of
to adhere to the reactor wall. This may be associated with the diffraction peaks that should be assigned to the glycolysis
weaker polarity of the glycolysis byproduct relative to the byproduct NaB(OCH ) . Here, the absence of NaBH peaks
2
4
4
Fig. 5 HG kinetics (top), reaction temperature (middle) and hydrogen density
Fig. 7 XRD patterns of the byproducts from the reactions (a) between NaBH
and EG; (b) between NaBH and 50 wt% EG–50 wt% H O mixture; (c) between
NaB(OH) and EG. For comparison, the standard XRD pattern of the NaB(OH)
was included (JCPDF card 06–0122).
4
(
bottom) of the system composed of 0.5 g of NaBH
amounts of 50 wt% EG–50 wt% H
NaBH : (EG + H O) mass ratio ranges from 1 : 2.4 to 1 : 20.
4
, 20 mg of CoCl
2
and varied
4
2
ꢁ
2
O mixture at an initial temperature of 0 C. The
4
4
4
2
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23810–23815 | 23813

View Article Online
RSC Advances
Paper
4
In our preliminary study of regeneration of NaBH , we
mechanically milled the NaB(OCH ) –MgH mixture in a 1 : 2
2
4
2
26,27
molar ratio under argon atmosphere.
Unfortunately, XRD
and FTIR analyses of the post-milled samples showed no
evidence for the formation of NaBH . Further studies are still
required for developing advanced technologies for cost-effective
and energy-efficient regeneration of NaBH from spent fuel.
4
4
4. Conclusions
A multi-component system composed of solid NaBH , ethylene
4
glycol–water mixture and CoCl additive has been developed
2
and examined in terms of hydrogen generation properties. Our
study found that the new system shows notable advantages
and over the neat hydrolysis or glycolysis systems on hydrogen
generation kinetics and operational temperature. For
Fig. 8 FTIR spectra of the byproducts from the reactions (a) between NaBH
4
4 2
EG; (b) between NaBH and 50 wt% EG–50 wt% H O mixture.
example, the system of an optimal composition can liberate
ꢁ
4
.2 wt% H
2
within 6 min at 0 C. Properly increasing the CoCl
2
2 4
Table 1 IR band positions and their assignments of the byproduct (NaB(OCH ) )
amount enables the system to rapidly deliver H at tempera-
2
from the reaction between NaBH
4
and 50 wt% EG–50 wt% H
Band assignment
C–H stretching
2
O mixture
ture as low as ꢀ20 ꢁC. The favorable combination of high
ꢀ1
hydrogen density, rapid hydrogen generation kinetics and a
wide range of operational temperature makes the new system
promising for mobile and portable applications. An added
Band position (cm
)
2
1
872
479
C–H deformation (scissoring
mode)
4 2
advantage of the NaBH /(EG–H O) system is that the reaction
byproduct can be readily removed from the reactor, which is of
clear benet for the design of hydrogen source. In addition, for
the rst time, our study provided the XRD and FTIR results of
the glycolysis byproduct (NaB(OCH ) ), which may lay foun-
1
1
1
1
1
9
7
6
6
356
240
120
109
089
33
52
94
49
B–O stretching
B–O deformation
2
4
C–O stretching
B–O asymmetry stretching
C–H rocking (–CH
C–H rocking
dation for the basic research into the reaction mechanism of
glycolysis of NaBH
4
.
2 2
–CH –)
O–B–O stretching
Acknowledgements
The nancial supports for this research from the National Basic
Research Program of China (973 program, Grant no.
indicates its complete consumption in the solvolysis reactions,
which is consistent with the nearly 100% fuel conversion as
determined by property examination. But the disappearance of
2010CB631305), the National Outstanding Youth Science
Foundation of China (Grant no. 51125003) and the National
Natural Science Foundation of China (Grant no. 51071155) are
gratefully acknowledged.
the NaB(OH)
unexpected result. As the supplied EG–H
EG in excess of that required for the stoichiometric reaction
4
crystalline phase in the XRD analysis is an
2
O mixture contains
following eqn (2), we surmised that the generated hydrolysis Notes and references
byproduct subsequently reacted with EG following eqn (4),
1
2
L. Schlapbach and A. Z u¨ ttel, Nature, 2001, 414, 353.
S. S. Muir and X. D. Yao, Int. J. Hydrogen Energy, 2011, 36,
983.
B. H. Liu and Z. P. Li, J. Power Sources, 2009, 187, 527.
H. I. Schlesinger, H. C. Brown, A. E. Finholt, J. R. Gilbreath,
H. R. Hoekstra and E. K. Hyde, J. Am. Chem. Soc., 1953, 75,
yielding NaB(OCH
NaB(OH) in EG and magnetically stirred the solution for 2 h at
room temperature. XRD analysis of the collected and dried
2 4
) . To test this speculation, we dissolve
4
5
3
4
sample clearly showed the complete conversion of NaB(OH)
NaB(OCH , as seen in Fig. 7c.
4
to
2 4
)
2
15.
C. T. F. Lo, K. Karan and B. R. Davis, Ind. Eng. Chem. Res.,
007, 46, 5478.
NaB(OH)
4
+ 2CH
2
OHCH
2
OH / NaB(OCH
2
)
4
+ 4H
2
O
(4)
5
2
Fig. 8 presents the FTIR spectra of the solvolysis byproduct.
In good agreement with the XRD result, FTIR analysis did not
detect NaBH residue and NaBO byproduct. By reference to the
6 R. E. Davis and J. A. Gottbrath, J. Am. Chem. Soc., 1962, 84,
895.
7 C. T. F. Lo, K. Karan and B. R. Davis, Ind. Eng. Chem. Res.,
2009, 48, 5177.
4
2
9
literature results of the methanolysis byproduct, the detected
IR modes were assigned to B–O, C–H, C–O bands of
8 D. Y. Xu, Z. Lin, P. Dai and S. F. Ji, J. Nat. Gas Chem., 2012, 21,
NaB(OCH
2
)
4
, respectively, as tabulated in Table 1.
488.
2
3814 | RSC Adv., 2013, 3, 23810–23815
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
9
J. Hannauer, U. B. Demirci, G. Pastor, C. Geantet, 19 O. Akdim, U. B. Demirci, D. Muller and P. Miele, Int. J.
J. M. Herrmann and P. Miele, Energy Environ. Sci., 2010, 3, 1796. Hydrogen Energy, 2009, 34, 2631.
0 C. C. Su, M. C. Lu, S. L. Wang and Y. H. Huang, RSC Adv., 20 U. B. Demirci and P. Miele, Phys. Chem. Chem. Phys., 2010,
012, 2, 2073. 12, 14651.
1 K. Ramya, K. S. Dhathathreyan, J. Sreenivas, S. Kumar and 21 A. M. Ozerova, O. A. Bulavchenko, O. V. Komova,
1
1
2
S. Narasimhan, Int. J. Energy Res., 2013, DOI: 10.1002/
O. V. Netskina, V. I. Zaikovskii, G. V. Odegova and
er.3006.
V. I. Simagina, Kinet. Catal., 2012, 53, 511.
1
1
1
1
1
1
1
2 H. C. Brown, E. J. Mead and B. C. S. Rao, J. Am. Chem. Soc., 22 G. M. Arzac, T. C. Rojas and A. Fernandez, ChemCatChem,
955, 77, 6209. 2011, 8, 1305.
3 J. S. Zhang, T. S. Fisher, J. P. Gore, D. Hazra and 23 H. B. Dai, G. L. Ma, X. D. Kang and P. Wang, Catal. Today,
P. V. Ramachandran, Int. J. Hydrogen Energy, 2006, 31, 2292. 2011, 170, 50.
4 P. V. Ramachandran, D. Hazra, B. C. Raju, V. R. Mereddy and 24 K. Bourikas, C. Kordulis, J. Vakros and A. Lycourghiotis, Adv.
A. Bhattacharyya, US pat., 11/125,796, 2009. Colloid Interface Sci., 2004, 110, 97.
5 V. R. Fernandes, A. M. F. R. Pinto and C. M. Rangel, Int. J. 25 E. W. Flick, Industrial solvents handbook, Noyes: William
Hydrogen Energy, 2010, 35, 9862. Andrew Publishing, 5th edn, 1998.
6 A. Levy, J. B. Brown and C. J. Lyons, Ind. Eng. Chem., 1960, 52, 26 Z. P. Li, N. Morigazaki, B. H. Liu and S. Suda, J. Alloys
11. Compd., 2003, 349, 232.
7 C. M. Kaufman and B. Sen, J. Chem. Soc., Dalton Trans., 1985, 27 C. L. Hsueh, C. H. Liu, B. H. Chen, C. Y. Chen, Y. C. Kuo,
1
2
307.
K. J. Hwang and J. R. Ku, Int. J. Hydrogen Energy, 2009, 34,
8 B. H. Liu, Z. P. Li and S. Suda, J. Alloys Compd., 2009, 468, 493.
1717.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23810–23815 | 23815