Optimizing preparation of carbon supported cobalt catalyst for hydrogen generation from NaBH4 hydrolysis

Article abstract of DOI:10.1016/j.jallcom.2012.07.099

Modified carbon-supported cobalt catalysts have been synthesized by optimization of the preparation conditions including ball-milling pretreatment of the carbon and screening the solvents used for synthesis of the catalysts. Ball-milling treatment on carbon can not only reduce its particle size but also functionalize the surface with oxygen-containing groups, resulting in good affinity to Co²⁺ ions. The non-aqueous solvent N,N-dimethylformamide (DMF) is beneficial to improve the immersional wetting and adsorption of the cobalt salt solution on the surface of carbon, leading to uniform deposition of cobalt nanoparticles upon chemical reduction. The carbon-supported cobalt catalyst with the Co loading of 13.20 wt% obtained under optimum conditions exhibits excellent catalytic activity toward NaBH₄ hydrolysis and the average hydrogen generation rate at 27 °C is 10.29 LH₂min ^-1g(cobalt)-1 corresponding to a turnover frequency of 1473.4 mol H₂ (mol cobalt)^-1 h^-1, which is higher than the most reported data. Furthermore, the effects of the cobalt loading, catalyst dosage, concentration of NaOH and NaBH₄, and reaction temperature on the hydrogen generation have been investigated by hydrolysis of alkaline NaBH₄ solution, and the corresponding rate equation can be described as r=Aexp(-42190/RT)[cobalt]^0.16[NaOH]^0.23[NaBH ₄]^0.30.

Full text of DOI:10.1016/j.jallcom.2012.07.099

Journal of Alloys and Compounds 543 (2012) 159–166
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Optimizing preparation of carbon supported cobalt catalyst for hydrogen
generation from NaBH4 hydrolysis
Weiling Niu a, Dengbo Ren b, Yuanyuan Han a, Yanjun Wu a, Xinglong Gou a,
⇑
a Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, College of Chemistry and Chemical Engineering, China West Normal University, Nanchong 637000,
PR China
b Division of Academic Affairs, Sichuan University of Arts and Science, Dazhou 635000, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 April 2012
Received in revised form 12 July 2012
Accepted 14 July 2012
Available online 31 July 2012
Modified carbon-supported cobalt catalysts have been synthesized by optimization of the preparation
conditions including ball-milling pretreatment of the carbon and screening the solvents used for synthesis
of the catalysts. Ball-milling treatment on carbon can not only reduce its particle size but also functionalize
the surface with oxygen-containing groups, resulting in good affinity to Co2+ ions. The non-aqueous sol-
vent N,N-dimethylformamide (DMF) is beneficial to improve the immersional wetting and adsorption of
the cobalt salt solution on the surface of carbon, leading to uniform deposition of cobalt nanoparticles
upon chemical reduction. The carbon-supported cobalt catalyst with the Co loading of 13.20 wt% obtained
Keywords:
Hydrogen generation
Sodium borohydride
Hydrolysis
Cobalt catalyst
Ball-milling
under optimum conditions exhibits excellent catalytic activity toward NaBH4 hydrolysis and the average
hydrogen generation rate at 27 °C is 10.29 LH minꢀ1
g
corresponding to a turnover frequency of
ꢀ1
2
ðcobaltÞ
1473.4 mol H2 (mol cobalt)ꢀ1 hꢀ1, which is higher than the most reported data. Furthermore, the effects
of the cobalt loading, catalyst dosage, concentration of NaOH and NaBH4, and reaction temperature on
the hydrogen generation have been investigated by hydrolysis of alkaline NaBH4 solution, and the corre-
Kinetics
0:16
0:23
0:30
sponding rate equation can be described as r ¼ A expðꢀ42190=RTÞ½cobaltꢁ ½NaOHꢁ ½NaBH4ꢁ
.
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
with the advantages of low cost and high catalytic activity [15].
Generally, the carbon-supported Co catalysts are prepared by
impregnation of carbon powders into an aqueous CoCl2 solution
followed by chemical reduction. It is also found that the catalytic
activity of the catalysts is sensitive to their preparation conditions.
Kim et al. investigated the effects of different precursors (CoCl2 and
CoSO4), NaBH4/Co2+ mole ratios (0.67–3) and calcination tempera-
tures (130–450 °C) on the properties of the prepared Co–B catalyst,
Hydrogen is a clean and green fuel with little or nor harmful
emissions, and is now widely recognized as an attractive alterna-
tive to fossil fuels, capable of providing energy security and sus-
tainability, along with environmental and socioeconomic benefits
[1]. However, efficient hydrogen generation and storage still re-
main a great challenge for its large scale applications [2–5].
Recently, sodium borohydride (NaBH4) has gained increasing
attention as a promising source to supply hydrogen for proton ex-
change membrane fuel cells due to its combined advantages of
high hydrogen content (10.8 wt%), nontoxicity, and excellent sta-
bility of its alkaline solution [6–9]. Moreover, high purity of hydro-
gen can be controllably generated on demand by catalytic
hydrolysis of NaBH4 in the presence of appropriate catalyst [10].
Therefore, development of effective catalysts for NaBH4 hydrolysis
is of great scientific and technological importance.
and found that a maximum hydrogen generation rate (HGR) of
2.4 L minꢀ1
g
can be obtained only under the optimum condi-
ꢀ1
cat
tions with CoCl2 as the precursor, a NaBH4/Co2+ mole ratio of 1.5
and calcination at 250 °C [16]. Liu et al. found that the catalytic
properties of the Co–B catalysts were sensitive to the pH value of
the NaBH4 reduction solution and mixing manner of the precursors
[17]. To further improve the performance of cobalt-based catalysts
some measurements have been taken to control the cobalt disper-
sion and modify the surface chemistry of the carbon used. For
example, Chen et al. employed ethylenediamine as a complex
agent for Co2+ ions to improve the dispersion of Co–B particles
During the past decades a variety of catalysts including precious
metals and non-noble metals (usually functionalized with a sup-
port) have been developed for hydrogen generation from NaBH4
hydrolysis [11–14]. Among the investigated catalysts, carbon-sup-
ported cobalt metals or alloys are proven to be efficient catalysts
on the surface of the carbon, and achieved a maximum HGR of
2.07 LH minꢀ1
g
[18]. Shen et al. used oxygen-containing groups
ꢀ1
2
cat
functionalized multiwalled carbon nanotubes (MWCNTs) as the
support of Co–B catalysts, resulting in a HGR of 5.1 LH minꢀ1
g
ꢀ1
2
cat
[19]. Patel et al. prepared Co–B nanoparticles supported on carbon
film catalyst by pulsed laser deposition technique to improve the
⇑
Corresponding author. Tel.: +86 817 256 8081; fax: +86 817 256 8081.
E-mail address: gouxlr@126.com (X. Gou).
0925-8388/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jallcom.2012.07.099

160
W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
specific surface area of both the carbon support and Co–B nanopar-
2.3. Hydrogen generation measurements
ticles, and obtain a high HGR of 8.1 LH minꢀ1
g
in the hydrolysis
ꢀ1
cat
2
of NaBH4 [20]. Although great progresses have been made in the
field of hydrogen generation from NaBH4 hydrolysis catalyzed by
carbon-supported cobalt catalysts, facile methods for preparing
the needed catalysts with excellent activity are still highly desired.
In this paper, influences of some preparation parameters such as
mechanical milling pretreatment of the carbon support under aer-
obic/anaerobic conditions and the solvents (de-ionized water, abso-
lute ethanol, and DMF) used for preparation of the catalyst on its
Catalytic activity of the as-prepared catalysts for hydrogen gen-
eration from the hydrolysis of alkaline NaBH4 solution was evalu-
ated in a batch reactor. In a typical experiment, an appropriate
amount of the catalyst was first loaded into a 50 mL round-bottom
flask, which was immersed in a thermostatic water bath to keep
the temperature constant. Then, 10 mL of aqueous solution con-
taining NaBH4 and NaOH with a given concentration was quickly
added into the flask under stirring to initiate the hydrolysis reac-
tion. The volume of the hydrogen evolved was measured by a clas-
sic water-displacement method and was transformed to the value
in standard temperature and pressure. The hydrogen generation
catalytic activity toward hydrolysis of sodium borohydride have
been investigated, and an average HGR of 10.29 LH minꢀ1
corresponding to
g
,
ꢀ1
2
ðcobaltÞ
a turnover frequency (TOF) of 1473.4 mol
H2 (mol cobalt)ꢀ1 hꢀ1, was achieved at 27 °C by using the optimized
rate (HGR) was normalized by per unit weight of the active cobalt
catalyst and was defined as LH minꢀ1
g
.
ꢀ1
carbon-supported Co catalysts.
2
ðcobaltÞ
2. Experimental
3. Results and discussion
2.1. Catalyst preparation
3.1. Optimizing the preparation conditions of carbon-supported cobalt
catalyst
All chemicals including CoCl2ꢂ6H2O, NaBH4, NaOH, ethanol, and
dimethylformamide (DMF) are of analytical grade. Commercial
cylindrical activated carbon was used as a typical carbon support.
It was first manually ground into powder, and then partial of the
powder was further mechanically milled under aerobic or anaero-
bic conditions using a planetary high-energy ball mill (model QM-
3SP2, Nanjing NanDa Instrument Plant) with different weight ratio
of ball to carbon (10:1–20:1), rotation speeds (350–550 rpm), and
milling time (5–15 h) to investigate the effect of milling pretreat-
ment on the catalytic activity of the carbon-supported cobalt
catalysts.
The carbon-supported cobalt catalysts were prepared by a mod-
ified impregnation-chemical reduction method, in which DMF,
absolute ethanol as well as de-ionized water (DW) was respectively
used as the solvent for dissolving both cobalt salt and NaBH4 to
investigate the solvent effect. In a typical procedure, 0.5 g carbon
powder (hand-grinded or milled ones under different conditions)
was first dispersed into 10 mL CoCl2 solution (0.14 mol Lꢀ1) with
the help of ultrasonication, and then the mixture was stirred over-
night to reach adsorption equilibrium. Subsequently, freshly pre-
pared NaBH4 solution (10 mL, 0.56 mol Lꢀ1) with the same solvent
as the CoCl2-carbon dispersion was added dropwise under
continuous stirring. Finally, the resulting black precipitates were
collected by centrifugation, washed and dried in vacuum at 60 °C
for 24 h. By adjusting the initial ratio of carbon to Co2+, a series of
catalysts with different Co loading (13.20, 17.48, and 19.34 wt%)
were obtained.
NaBH4 hydrolysis catalyzed by a carbon-supported catalyst is a
heterogeneous reaction, and thus the HGR strongly depends on the
surface properties of the catalyst, which is directly related to the
preparation conditions. The purpose of this work is to investigate
the potential effects of both the pretreatment of carbon support
with ball milling and the solvents used for preparation of the com-
posite catalyst on the hydrogen generation activity of the prepared
catalyst.
3.1.1. Effect of ball-milling pretreatment of the carbon support
High-energy ball milling is reported to be an efficient mechano-
chemical technique to reduce the particle size and enhance the
surface properties of the studied materials [21–23]. Herein, milling
technique is employed to improve the surface activity of the car-
bon support, and the effects of milling parameters on the catalytic
activity of the supported catalyst were evaluated by the average
HGR for catalytic hydrolysis of 1 wt% NaBH4 solution containing
8 wt% NaOH at 27 °C in the presence of 20 mg the carbon-
supported cobalt nanoparticles with a Co loading of 13.20 wt%.
Fig. 1(a) shows the effect of the ball-to-carbon weight ratio on
the HGR for the carbon-supported cobalt catalysts. Obviously, the
ball-milled carbon supported cobalt catalysts always exhibit much
higher HGR than the one using the hand-ground carbon as the sup-
port. Moreover, the HGR improved markedly when the ball-to-car-
bon weight ratio increased from 10:1 to 15:1, and an average HGR
of 10.23 L minꢀ1 gꢀ1 was achieved at this point, which is almost 4
times higher than that for the hand-ground carbon supported cat-
alyst. It was also observed that further increase in the ball-to-car-
bon weight ratio had little effect on the HGR. Therefore, the
optimum ball-to-carbon weight ratio was selected as 15:1 in this
work.
2.2. Catalyst characterization
The chemical composition of the samples was characterized by
X-ray diffraction (XRD, Rigaku Ultima IV, Cu Ka radiation). The mor-
phology of the catalysts was observed by scanning electron micro-
scope (SEM, JEOL JSM-6510LV) coupled with an energy-dispersive
X-ray spectroscopy (EDS, Oxford instruments X-Max). The cobalt
loading of the supported catalyst as well as the elemental distribu-
tion on the carbon support was determined by EDS analysis, and an
average value of six different areas was reported for each sample
(the standard derivation was less than 0.3%). Transmission electron
microscope (TEM) images were taken using a FEI Tecnai G20 micro-
scope operated at 200 kV. The specific surface area of the sample
was measured by Brunauer–Emmett–Teller (BET) nitrogen adsorp-
tion–desorption method on a Belsorp-Mini II instrument (BEL
Japan, Inc.). Fourier transform infrared spectroscopy (FTIR) was
recorded on Thermo Sientific Nicolet 6700 FT-IR spectrometer.
From the inset of Fig. 1(b) we can see that the milling speed has
an important effect on the HGR. Especially, a sharp increase in HGR
is observed when the milling speed rises from 350 to 450 rpm.
Afterwards, the HGR increases only slightly when the milling speed
rises from 450 to 550 rpm. This trend is consistent with the order
of the BET specific surface area, which is 405.4, 429.5 and
430.7 m2 gꢀ1 for the milled carbon obtained at 350, 450 and
550 rpm, respectively. Therefore, the optimum milling speed was
chosen as 450 rpm to further investigate the influence of the mill-
ing time on HGR, and the result is presented in Fig. 1(b). With the
milling time changing from 5 to 10 h, an evident increase in the
HGR is observed from 9.69 to 10.29 LH minꢀ1 gꢀ1. Thereafter, a
2
very gentle change in HGR occurs with continuous increase of

W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
161
Fig. 2. FTIR spectra of the hand-ground carbon and the ball-milled ones obtained
under anaerobic and aerobic conditions.
resulting in good dispersion of the chemically reduced metal nano-
particles on its surface and high catalytic activity for hydrolysis of
NaBH4.
The above experiments suggest that ball-milling treatment on
carbon in air can improve its surface area and chemical composi-
tion, leading to an improvement of the catalytic hydrogen genera-
tion activity of cobalt nanoparticles supported on it in the
hydrolysis of sodium borohydride. The optimum milling conditions
were found to be 15:1 for the ball-to-carbon weight ratio, 450 rpm
for the milling speed, and 10 h for the milling time.
3.1.2. Effect of the solvents used for preparation of the catalyst
Carbon-supported cobalt catalyst is usually prepared by an
impregnation-chemical reduction method in the literatures, where
de-ionized water (DW) is often used as the solvent to dissolve co-
balt salt and also as the disperse media for carbon support. How-
ever, the hydrophobic characteristic of the carbon support might
restrict effective contact between the carbon surface and the aque-
ous Co2+ solution, and thus would affect the deposition and disper-
sion of the cobalt metals in the course of chemical reduction
process. We surmise that the catalytic performance of the car-
bon-supported cobalt catalysts can also be improved by selecting
an appropriate solvent, which not only dissolve the cobalt salt
but also infiltrate the surface and pore of the carbon support. Based
on this consideration, we prepared three kinds of carbon-sup-
ported cobalt catalysts by respectively using DW, absolute ethanol
and DMF as the solvent and disperse media. Fig. 3 displays the HGR
over the three kinds of catalysts under the same hydrolysis condi-
tions. As expected, the corresponding HGR at 27 °C increases from
Fig. 1. Effects of milling conditions on the hydrogen generation activity of the
carbon-supported cobalt catalysts (20 mg, the Co loading is 13.20 wt%) by catalytic
hydrolysis of 1 wt% NaBH4 solution containing 8 wt% NaOH at 27 °C. (a) Influence of
the ball-to-carbon weight ratio on the HGR, where the milling time and speed was
set as 5 h and 550 rpm, respectively. (b) Influence of milling time on the HGR at a
milling speed of 450 rpm. The inset is the influence of milling speed on the HGR,
where the milling time was set as 5 h.
the milling time. Accordingly, 10 h is sufficient for the ball milling
treatment of the carbon support.
FTIR spectrum can reveals the details about the surface chemi-
cal composition of a material. For the sake of comparison Fig. 2 pre-
sents the FTIR spectra of the hand-ground carbon, and the ball-
milled ones under both anaerobic and aerobic conditions. Overall,
all the studied carbon samples exhibits three pronounced bands at
1137, 1582, and 3450 cmꢀ1, respectively. The broad peak near
1137 cmꢀ1 is a superposition of a number of overlapping bands
of ether, epoxide, lactone and phenolic structures in different envi-
ronments [24]. The peak at 1582 cmꢀ1 is a characteristic peak for
carbon materials containing aromatic rings or ether-like bonds
[25]. The broad peak around 3450 cmꢀ1 can be assigned to O–H
stretching vibrations of surface hydroxylic groups and chemi-
sorbed water [25]. However, the intensity of the peaks related to
the oxygen-containing groups in the spectrum of the carbon milled
in air increases significantly, indicating that the high energy pro-
vided by ball milling exacerbates the oxidization of carbon in air,
and thus produces a large quantity of oxygen-containing groups
on its surface. In contrast, the oxidization of carbon is heavily
inhibited in the ball-milling process without air, or it is negligible
during the hand-ground treatment in air due to lack of enough en-
ergy. As a support for catalyst the oxygen-rich functional groups on
the ball-milled carbon exhibits enhanced affinity to metal ions,
2.41, 6.64 to 10.29 LH minꢀ1 gꢀ1, corresponding to a turnover fre-
2
quency of 345.06, 950.71 and 1473.4 mol H2 (mol Co)ꢀ1 hꢀ1
,
respectively. Therefore, DMF is the best solvent for preparation of
the carbon-supported cobalt catalysts.
3.2. Chemical composition and structure of the catalyst
Fig. 4 presents the XRD patterns of the unsupported cobalt par-
ticles (a), the carbon support after ball-milling treatment (b), and
the carbon-supported Co catalyst prepared under the optimum
conditions (c). Fig. 4(a) indicates that the unsupported cobalt par-
ticles are partially oxidized into cobalt oxides. Similar phenome-
non is often observed for the cobalt-based catalysts [26]. By

162
W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
depicted in Fig. 6 we can clearly see that numerous sphere-like co-
balt nanoparticles were uniformly deposited on the surface of the
carbon support, in good agreement with the result of EDS mapping.
The dimension of the ball-milled carbon is at micrometer scale,
while the average diameter of the cobalt nanoparticles is about
15 nm. This fine microstructure is favorable for catalysis applica-
tions, which will be the focus of the succeeding parts.
3.3. Catalytic performance of the carbon-supported cobalt catalyst
3.3.1. Effect of Co loading
The effect of Co loading on the catalytic activity was evaluated
by hydrolysis of 1 wt% NaBH4 solution containing 8 wt% NaOH at
27 °C in the presence of 20 mg the carbon-supported cobalt nano-
particles with different Co loading. Fig. 7 displays the accumulated
hydrogen volume as a function of time in the initial NaBH4 hydro-
lysis process. For all the tested catalysts, the volume of generated
hydrogen is almost proportional to the reaction time, indicating
the stable catalytic activity of the catalysts. The H2 yield, HGR
and TOF for the catalysts with different Co loadings are tabulated
in Table 1. It is clear that the catalytic activity of the catalyst in-
creases with the Co loading, because higher Co loading leads to
more active catalytic sites. This means that the hydrogen genera-
tion rate can be readily controlled by adjusting the Co loading.
Fig. 3. Effects of the solvents used for preparation of the carbon-supported cobalt
catalysts on their catalytic hydrogen generation activity by hydrolysis of 1 wt%
NaBH4 solution containing 8 wt% NaOH at 27 °C.
3.3.2. Effect of catalyst dosage
The effect of catalyst dosage on the hydrogen generation is
shown in Fig. 8, and the corresponding HGR normalized by the
weight of cobalt and TOF are listed in Table 2. As expected, the
HGR and TOF increase slightly with the amount of the cobalt cata-
lyst used. This means that the hydrogen generation can also be
controlled by tuning the dosage of the catalyst. From the inset of
Fig. 8 we can see that the plot of ln(HGR) vs ln(mcobalt) is a straight
line with a slope of 0.16. Therefore, the reaction order of the NaBH4
hydrolysis catalyzed by the carbon-supported cobalt nanoparticles
is 0.16 with respect to the concentration of cobalt. Possibly, the
more the carbon-supported catalyst used, the more water mole-
cules in the NaBH4 solution were adsorbed, indirectly leading to
the slight increase of NaBH4 concentration and more contact be-
tween NaBH4 and cobalt nanoparticles.
Hydrogen generation performance from NaBH4 catalytic Hydro-
lysis is not only dependent on the catalytic activity of the catalyst
used, but also on the hydrolysis conditions such as the concentra-
tion of both NaOH and NaBH4, and the hydrolysis temperature
[27]. Below we will discuss the effect of these outer conditions
on the hydrolysis of NaBH4 solution catalyzed by 20 mg of the
milled-carbon-supported cobalt catalysts with a Co loading of
13.20 wt%.
Fig. 4. XRD patterns of the unsupported cobalt particles (a), the carbon support (b),
and the carbon-supported Co catalysts with the Co loading of 13.20 wt% which was
synthesized under the optimum conditions (c).
comparing Fig. 4(c) with Fig. 4 (b) and (a) we can see that the dif-
fraction peaks of (002), (101), and (004) in Fig. 4(c) are ascribable
to the carbon support, and the rest peaks are very similar to those
in the XRD pattern of the unsupported cobalt particles, implying
that the carbon-supported cobalt catalyst has been successfully
prepared.
Fig. 5 shows the typical SEM image of the carbon-supported co-
balt catalyst prepared under the optimum conditions. It can be
seen that the catalyst is comprised of irregular particles. The corre-
sponding element composition and distribution are analyzed by
EDS mapping, as shown in Fig. 5(b)–(d). These results indicate that
the obtained catalyst consists of C, Co and O, which distribute very
uniformly on the surface of carbon. The signal of oxygen originates
from the oxidation products of both the carbon-support and the
deposited cobalt, in accord with the analyses of FTIR and XRD
aforementioned. The Co loading in the catalyst is determined to
be 13.20 wt%. It is also found that increase of the Co loading in
the range of 13.20–19.34 wt% does not change the particle-like
morphology of the products.
3.3.3. Effect of NaOH concentration
NaOH is usually added as a stabilizer into the NaBH4 solution to
prevent its self-hydrolysis. Fig. 9 demonstrates the real-time
hydrogen generation from NaBH4 hydrolysis by changing the ini-
tial NaOH concentration in the range of 1–15 wt%. With increasing
NaOH concentration from 1 to 10 wt%, the average HGR rises from
6.616 to 11.402 LH minꢀ1
g
. However, further increase in the
ꢀ1
2
cobalt
NaOH leads to a gradual decrease of the HGR due to the inhibition
function of OH-with high concentration towards the hydrolysis
reaction. Similar case was also reported for the colloidal carbon
supported Co nanoparticles [27] and the activated carbon-sup-
ported Co–B catalysts catalyzed NaBH4 hydrolysis reaction [28].
The quantitative relationship between the HGR and the NaOH
concentration in the range of 1–10 wt% is given in the inset of
Fig. 9. The plot of ln(HGR) vs ln(CNaOH) is a straight line with a slop
of 0.23, suggesting that the reaction order is 0.23 with respect to
the NaOH concentration. Herein, the increase of the NaOH
TEM observation can provide more information about the
microstructure of the catalyst. From the typical TEM images

W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
163
Fig. 5. A typical SEM image (a) of the carbon supported cobalt catalyst along with the corresponding EDS mapping of the constituent element C (b), Co (c), and O (d).
Fig. 6. Typical TEM images under different magnifications of the cobalt nanoparticles supported on the surface of the ball-milled carbon.
Table 1
The effect of Co loading on the hydrogen generation performance.
Co loading (wt%)
13.20
17.48
19.34
H2 Yield (%)
57
10.92
79
11.54
89
11.72
HGR ðLH gꢀ1 minꢀ1
Þ
2
cobalt
TOF (mol H2 (mol Co)ꢀ1 hꢀ1
)
1563.8
1651.9
1678.1
OHꢀ is beneficial to disperse the milled-carbon supported cobalt
catalyst which carried a large quantity of oxygen-containing func-
tional groups as discussed before, resulting in good contact be-
tween the catalyst and the NaBH4 solution.
3.3.4. Effect of NaBH4 concentration
Fig. 10 depicts the effect of the initial NaBH4 concentration on
the hydrogen generation performance. As can be seen from
Fig. 10 that the hydrogen generation rate slightly increases with
the NaBH4 concentration ranging from 0.25 to 1.00 wt%, while
gradually deceases with further increase of the NaBH4 concentra-
tion. Theoretically, higher NaBH4 concentration leads to higher
Fig. 7. Volume of the evolved H2 as a function of time obtained by catalytic
hydrolysis of 1 wt% NaBH4 solution containing 8 wt% NaOH and 20 mg carbon-
supported cobalt catalyst at 27 °C.
concentration in the range of 1–10 wt% has positive effect on the
hydrogen generation. Possibly, an appropriate concentration of

164
W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
Fig. 8. Hydrogen generation as a function of reaction time obtained by hydrolysis of
1 wt% NaBH4 solution containing 8 wt% NaOH and different amount of catalyst with
the same Co loading of 13.20 wt% at 27 °C. The inset is the plot of the hydrogen
Fig. 10. Real-time hydrogen generation at 27 °C from the NaBH4 solution of
different concentration, which contains 8 wt% NaOH and 20 mg the carbon-
supported cobalt catalyst with the Co loading of 13.20 wt%. The inset is the plot
generation rate (HGR) as
a function of cobalt concentration (mcobalt), both in
of ln(HGR) vs ln CNaBH
.
4
logarithmic scale.
Table 2
The effect of catalyst dosage on the hydrogen generation performance.
Catalyst dosage (mg)
20
40
60
HGR ðLH gꢀco1baltminꢀ1
Þ
10.29
11.33
12.25
2
TOF (mol H2 (mol Co)ꢀ1 hꢀ1
)
1473.4
1622.2
1753.9
Fig. 9. Real-time hydrogen generation from the hydrolysis of 1 wt% NaBH4 solution
containing different concentration of NaOH at 27 °C in the presence of 20 mg the
carbon-supported cobalt catalyst with the Co loading of 13.20 wt%. The inset is the
plot of ln(HGR) vs ln(CNaOH).
hydrogen density. However, the alkalinity and viscosity of the
hydrolysis system also increase with the NaBH4 concentration,
which prevents hydrogen generation and leads to a decrease of
the HGR [29]. Similar case was also observed by Xu et al. in the
NaBH4 hydrolysis system using the cobalt catalysts supported by
the modified activated carbon with concentrated HNO3 oxidization
[30].
Fig. 11. (a) Real-time hydrogen generation from the 1 wt% NaBH4 solution
containing 8 wt% NaOH and 20 mg the carbon-supported cobalt catalyst with a
Co loading of 13.20 wt% at different hydrolysis temperatures. (b) The Arrhenius plot
of ln (HGR) vs 1/T.
kinetics of NaBH4 hydrolysis catalyzed by Co–P–B is dependent
on the concentration of NaBH4 [31]. For example, a first order
kinetics was observed when the NaBH4 concentration changed in
the range of 0.005 to 0.05 M, while a near-zero kinetics was ob-
tained in the range of 0.075–0.25 M. On the other hand, Zou
The plot of ln(HGR) vs ln CNaBH by fitting the points shown in
4
the inset of Fig. 10 leads to a straight line with a slop of 0.30, imply-
ing that the reaction order is 0.30 with respect to the NaBH4 con-
centration in the range of 0.25–1 wt%. Patel et al. found that the

W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
165
Table 3
The effect of hydrolysis temperature on the hydrogen generation performance.
Hydrolysis temperature (oC)
25
30
35
40
45
HGR ðLH gꢀco1baltminꢀ1
Þ
7.63
11.17
13.83
18.56
22.50
2
TOF (mol H2 (mol Co)ꢀ1 hꢀ1
)
1092.5
1599.3
1980.2
2657.4
3221.5
Table 4
Comparison of the key parameters for the NaBH4 hydrolysis reaction catalyzed by different catalysts.
Activation energy (kJ molꢀ1
)
Hydrolysis temperature (oC)
Ref.
H2 generation rate ðLH minꢀ1
g
Þ
ꢀ1
met
Catalysts
2
Ru nanoclusters
28.51
60.4 3.1
57.8
3.65
–
2.3
5.1
8.1
0.07
0.22
0.94
5.0
4.8
3.0
2.7
0.132
0.4
1.27
0.46
2.5
2.91
9.36
6.09
–
25
25
25
30
25
30
30
30
25
30
25
25
25
35
25
25
30
25
25
25
25
25
30
27
[13]
[14]
[18]
[19]
[20]
[29]
[29]
[33]
[35]
[36]
[37]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
This work
Zeolite framework/Ni
Vulcan XC-72/Co–B
MWCNT/Co–B
Carbon film/Co–B
Active carbon/Co
40.40
ꢃ31
45.64
32.63
38.14
c
-Al2O3/Co
AMPS hydrogels/Co
hydroxyapatite/Co
Co–Cu–B
LiCoO2/Ru
LiCoO2/Pt
IR-120/Ru
Ni–Ru
Attapulgite clay/Co–B
Pd–TiO2–P/Co–Ni
Co–Mn–B
53
2
49.6
68.5
70.4
49.72
52.73
56.32
57.0
52.1
57
Ni–Fe–B
PVP-stabilized Co(0)
Intrazeolite/Co
PEG-stabilized Fe(0)
Zeolite/Ru nanoclusters
Ni foam/Co–W–B
Modified carbon/Co
37
34
2
2
37.0 2.0
34.9 2.0
29
14.7
15
10.29
42.19
energy (kJ molꢀ1), R is the gas constant (8.314 J molꢀ1 Kꢀ1), and T is
the hydrolysis temperature (K). The Arrhenius plot of ln(HGR) vs
ln(1/T) is shown in Fig. 11(b). The slope of the straight line gives
the activation energy of 42.19 kJ molꢀ1, which is lower than most
of the reported data as summarized in Table 4.
On the basis of the above discussion about the influence factors
of the hydrogen generation, the final rate equation for the carbon-
supported cobalt nanoparticles catalyzed NaBH4 hydrolysis reac-
tion is described as:
et al. reported the reaction order with regard to NaBH4 is ꢀ0.25
and 0.27, respectively, for the powder-like and spherical Ru/C cat-
alysts [32]. Sahiner et al. found that the hydrolysis reaction of
NaBH4 with p(AMPS)-Co composite catalyst system is zero order
with respect to initial concentration of NaBH4 [33]. Shang et al. re-
ported that the hydrolysis NaBH4 over ruthenium catalyst is zero
order to NaBH4 concentration when water is sufficient [34].
Accordingly, the kinetics of NaBH4 hydrolysis is not only depen-
dent on its concentration range, but also on the type of the cata-
lysts used for hydrogen generation. Further investigations should
be performed to better understand the effects of the NaBH4 con-
centration on the mechanisms involved in the hydrolysis process.
0:16
0:23
0:30
r ¼ A expðꢀ42190=RTÞ½cobaltꢁ ½NaOHꢁ ½NaBH4ꢁ
ð2Þ
3.3.5. Effect of hydrolysis temperature
4. Conclusion
Fig. 11(a) presents the effect of hydrolysis temperature on the
hydrogen generation performance of the NaBH4 hydrolysis reac-
tion. As can be seen from the slope of the linear portion and the
hydrolysis duration time that hydrogen generation rate increase
with the reaction temperature. The corresponding average HGR
as well as the corresponding TOF at each temperature is listed in
Table 3, indicating that the hydrolysis rate can also be controlled
by adjusting the reaction temperature.
The average HGR data in Table 3 are further used to determine
the activation energy for the NaBH4 hydrolysis catalyzed by the
carbon-supported cobalt nanoparticles according to the Arrhenius
equation:
The catalytic hydrogen generation activity of the carbon-sup-
ported cobalt catalysts toward NaBH4 hydrolysis reaction was sig-
nificantly improved by ball-milling pretreatment of the carbon
support and screening the solvents used for preparing the cata-
lysts. The high performance of the modified carbon-supported co-
balt catalyst is supported by its low activation energy
(42.19 kJ molꢀ1
)
and high average hydrogen generation rate
ð10:29 LH minꢀ1
g
Þ.
ꢀ1
2
cobalt
Acknowledgements
This work was financially supported by the National Natural
Science Foundation of China (51071131), Program for New Century
Excellent Talents in University (NCET-10-0890), and the Scientific
Research Foundation for the Returned Overseas Chinese Scholars
([2008]488).
Ea
RT
lnðHGRÞ ¼ ln A ꢀ
ð1Þ
where HGR is the average hydrogen generation rate
ðLH minꢀ1
g
Þ, A is the pre-exponential factor, Ea is the activation
ꢀ1
2
cobalt

166
W. Niu et al. / Journal of Alloys and Compounds 543 (2012) 159–166
[23] C. Liu, Y. Kuo, B. Chen, C. Hsueh, K. Hwang, J. Ku, F. Tsau, M. Jeng, Int. J.
References
Hydrogen Energy 35 (2010) 4027–4040.
[24] A.W. Heinen, J.A. Peters, H. Bekkum, Appl. Catal. A 194–195 (2000) 193–202.
[25] J.P. Chen, S. Wu, Langmuir 20 (2004) 2233–2242.
[1] M.Z. Jacobson, W.G. Colella, D.M. Golden, Science 308 (2005) 1901–1905.
[2] P. Chen, M. Zhu, Mater. Today 11 (2008) 36–43.
[3] I.P. Jain, P. Jain, A. Jain, J. Alloys compd. 503 (2010) 303–339.
[4] J. Yang, A. Sudik, C. Wolverton, D.J. Siegel, Chem. Soc. Rev. 39 (2010) 656–675.
[5] C. Liang, Y.F. Liu, H. Fu, Y. Ding, M. Gao, H.G. Pan, J. Alloys compd. 509 (2011)
7844–7853.
[6] D.M.F. Santos, C.A.C. Sequeira, Renew. Sust. Energ. Rev. 15 (2011) 3980–4001.
[7] B. Sakintuna, F. Lamari-Darkrim, M. Hirscher, Int. J. Hydrogen Energy 32 (2007)
1140–1211.
[8] S. Orimo, Y. Nakamori, J.R. Eliseo, A. Züttel, C.M. Jensen, Chem. Rev. 107 (2007)
1140–1211.
[9] B.H. Liu, Z.P. Li, J. Power Source 187 (2009) 527–534.
[10] U.B. Demirci, O. Akdim, J. Andrieux, J. Hannauer, R. Chamoun, P. Miele, Fuel
Cells 10 (2010) 335–350.
[11] S.S. Muir, X. Yao, Int. J. Hydrogen Energy 36 (2011) 5983–5997.
[12] B.H. Liu, Z.P. Li, S. Suda, J. Alloys Compd. 415 (2006) 288–293.
[13] S. Özkar, M. Zahmakıran, J. Alloys Compd. 404–406 (2005) 728–731.
[14] M. Zahmakıran, T. Ayvalı, S. Akbayrak, S. Çalısßkan, D. Çelik, S. Özkar, Catal.
Today 170 (2011) 76–84.
[15] D. Xu, P. Dai, X. Liu, C. Cao, Q. Guo, J. Power Sources 182 (2008) 616–620.
[16] S.U. Jeong, E.A. Cho, S.W. Nam, I.H. Oh, U.H. Jung, S.H. Kim, Int. J. Hydrogen
Energy 32 (2007) 1749–1754.
[17] B.H. Liu, Q. Li, Int. J. Hydrogen Energy 33 (2008) 7385–7391.
[18] J. Zhao, H. Ma, J. Chen, Int. J. Hydrogen Energy 32 (2007) 4711–4716.
[19] Y. Huang, Y. Wang, R. Zhao, P. Shen, Z. Wei, Int. J. Hydrogen Energy 33 (2008)
7110–7115.
[26] U.B. Demirci, P. Miele, Phys. Chem. Chem. Phys. 12 (2010) 14651–14665.
[27] J. Zhu, R. Li, W. Niu, Y. Wu, X.L. Gou, J. Power Source 211 (2012) 33–39.
[28] D. Xu, H. Wang, Q. Guo, S. Ji, Fuel Process. Technol. 92 (2011) 1606–1610.
[29] W. Ye, H. Zhang, D. Xu, L. Ma, B. Yi, J. Power Sources 164 (2007) 544–548.
[30] D. Xu, P. Dai, Q. Guo, X. Yue, Int. J. Hydrogen Energy 33 (2008) 7371–7377.
[31] N. Patel, R. Fernandes, A. Miotello, J. Power Source 188 (2009) 411–420.
[32] Y.C. Zou, M. Nie, Y.M. Huang, J.Q. Wang, H.L. Liu, Int. J. Hydrogen Energy 36
(2011) 12343–12351.
[33] N. Sahiner, O. Ozay, E. Inger, N. Aktas, Appl. Catal. B 102 (2011) 201–206.
[34] Y. Shang, R. Chen, G. Jiang, Int. J. Hydrogen Energy 33 (2008) 6719–6726.
[35] M. Rakap, S. Özkar, Catal. Today 183 (2012) 17–25.
[36] X. Ding, X. Yuan, C. Jia, Z. Ma, Int. J. Hydrogen Energy 35 (2010) 11077–11084.
[37] Z. Liu, B. Guo, S.H. Chan, E.H. Tang, L. Hong, J. Power Source 176 (2008) 306–
311.
[38] C. Hsueh, C. Chen, J. Ku, S. Tsai, Y. Hsu, F. Tsau, M. Jeng, J. Power Sources 177
(2008) 485–492.
[39] C. Liu, B. Chen, C. Hsueh, J. Ku, M. Jeng, F. Tsau, Int. J. Hydrogen Energy 34
(2009) 2153–2163.
[40] H. Tian, Q. Guo, D. Xu, J. Power Sources 195 (2010) 2136–2142.
[41] M. Rakap, E.E. Kalu, S. Özkar, J. Alloys Compd. 509 (2011) 7016–7021.
[42] X. Yuan, C. Jia, X. Ding, Z. Ma, Int. J. Hydrogen Energy 37 (2012) 995–1001.
[43] M. Nie, Y.C. Zou, Y.M. Huang, J.Q. Wang, Int. J. Hydrogen Energy 37 (2012)
1568–1576.
[44] Ö. Metin, S. Özkar, Energy Fuel 23 (2009) 3517–3526.
[45] M. Rakap, S. Özkar, Appl. Catal. B 91 (2009) 21–29.
[46] M. Dinç, Ö. Metin, S. Özkar, Catal. Today 183 (2012) 10–16.
[47] M. Zahmakiran, S. Özkar, Langmuir 25 (2009) 2667–2678.
[48] H.B. Dai, Y. Liang, P. Wang, X.D. Yao, T. Rufford, M. Lu, H.M. Cheng, Int. J.
Hydrogen Energy 33 (2008) 4405–4412.
[20] N. Patel, R. Fernandes, N. Bazzanella, A. Miotello, Catal. Today 170 (2011) 20–
26.
[21] L. Gao, C. Chen, L. Chen, Q. Wang, C. Wang, Y. An, J. Alloys Compd. 424 (2006)
338–341.
[22] S. Liu, L. Sun, Y. Zhang, F. Xu, J. Zhang, H. Chu, M. Fan, T. Zhang, X. Song, J.P.
Grolier, Int. J. Hydrogen Energy 34 (2009) 8079–8085.