Surface modification of Ni/Al2O3 with Pt: Highly efficient catalysts for H2 generation via selective decomposition of hydrous hydrazine

Article abstract of DOI:10.1016/j.jcat.2012.10.012

Hydrous hydrazine, such as hydrazine monohydrate (N₂H ₄·H₂O), is a promising hydrogen carrier material due to its high content of hydrogen (8.0 wt%). The decomposition of hydrous hydrazine to H₂ with a high selectivity and a high activity under mild conditions is the key to its potential usage as a hydrogen carrier material. Platinum-modified Ni/Al₂O₃ catalysts (NiPt _x/Al₂O₃) were prepared starting from Ni-Al hydrotalcite and tested in the decomposition of hydrous hydrazine. Compared with Ni/Al₂O₃, the TOF was enhanced sevenfold over NiPt _0.057/Al₂O₃; meanwhile, the selectivity to H₂ was increased to 98%. Characterization results by means of HAADF-STEM, XRD, and EXAFS revealed the presence of surface Pt-Ni alloy in this Pt-promoted catalyst. The formation of Pt-Ni alloy could significantly weaken the interaction between adspecies produced (including H₂ and NH _x) and surface Ni atoms, which is confirmed by microcalorimetry and TPD results. The weakening effect could account for the greatly enhanced reaction rate, as well as H₂ selectivity on NiPt_x/Al ₂O₃ catalysts.

Full text of DOI:10.1016/j.jcat.2012.10.012

Journal of Catalysis 298 (2013) 1–9
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Surface modification of Ni/Al2O3 with Pt: Highly efficient catalysts for H2
generation via selective decomposition of hydrous hydrazine
Lei He a,b, Yanqiang Huang a, Aiqin Wang a, Yu Liu a, Xiaoyan Liu a, Xiaowei Chen c, Juan José Delgado c,
Xiaodong Wang a, Tao Zhang a,
⇑
a State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, China
b University of Chinese Academy of Sciences, 19A Yuquan Road, Beijing 100039, China
c Departamento de Ciencia de los Materiales e Ingeniería Metalúrgica y Química Inorgánica Facultad de Ciencias, Universidad de Cádiz, Puerto Real, Cadiz, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Hydrous hydrazine, such as hydrazine monohydrate (N2H4ꢀH2O), is a promising hydrogen carrier material
due to its high content of hydrogen (8.0 wt%). The decomposition of hydrous hydrazine to H2 with a high
selectivity and a high activity under mild conditions is the key to its potential usage as a hydrogen carrier
material. Platinum-modified Ni/Al2O3 catalysts (NiPtx/Al2O3) were prepared starting from Ni–Al
hydrotalcite and tested in the decomposition of hydrous hydrazine. Compared with Ni/Al2O3, the TOF
was enhanced sevenfold over NiPt0.057/Al2O3; meanwhile, the selectivity to H2 was increased to 98%.
Characterization results by means of HAADF-STEM, XRD, and EXAFS revealed the presence of surface
Pt–Ni alloy in this Pt-promoted catalyst. The formation of Pt–Ni alloy could significantly weaken the
interaction between adspecies produced (including H2 and NHx) and surface Ni atoms, which is con-
firmed by microcalorimetry and TPD results. The weakening effect could account for the greatly enhanced
reaction rate, as well as H2 selectivity on NiPtx/Al2O3 catalysts.
Received 27 July 2012
Revised 5 October 2012
Accepted 6 October 2012
Keywords:
Hydrous hydrazine
Hydrogen storage
Pt–Ni alloy
EXAFS
TPD
Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction
and Rh–Ni bimetallic nanoparticles exhibited selectivity close to
100% for hydrogen generation at room temperature [10–12]. How-
Chemical hydrogen storage appears to be one of the most prom-
ising approaches for safe and efficient hydrogen storage [1]. Search-
ing for effective hydrogen storage materials to release hydrogen
conveniently under mild conditions is attracting considerable
research interest from both academia and industry [2–4]. Recently,
hydrous hydrazine, such as hydrazine monohydrate (N2H4ꢀH2O),
with a high content of hydrogen (8.0 wt%), was discovered as a
promising hydrogen carrier material, with the advantage that
the only by-product of complete decomposition is nitrogen
(H2NNH2 ? N2(g) + 2H2(g)) [5,6]. However, to efficiently make use
of the hydrogen storage properties of hydrous hydrazine, its incom-
plete and undesired decomposition (3H2NNH2 ? 4NH3(g) + N2(g))
must be avoided [7,8]. This is because the NH3 as a by-product could
not only complicate the separation process, but also poison the
Nafion membrane and the fuel-cell catalysts. Thus, finding highly
selective and active catalysts for quick hydrogen release is of great
importance for practical applications.
ever, the addition of surfactants during the preparation of nanopar-
ticles makes it difficult to separate them from the reactants and
also significantly reduces the reaction rate.
To improve the kinetic properties of hydrous hydrazine decom-
position, several studies have contributed to the development of
efficient heterogeneous catalysts. Wang et al. [15] designed a sup-
ported catalyst by depositing Rh–Ni nanoparticles on graphene,
which exhibited much higher activity than the bare nanoparticle
catalysts. Moreover, Shao and coworkers [16,17] discovered that
Ba0.5Sr0.5Co0.8Fe0.2O3-d (BSCF) could catalyze hydrous hydrazine
decomposition under mild conditions. Unfortunately, for all the
supported catalyst systems tested so far, the presence of alkaline
additives (NaOH or KOH) was necessary for high selectivity and
activity. This promotion effect could be explained in two ways
[14,15]. On one hand, the addition of alkalis decreased the undesir-
able N2Hþ in aqueous solution ðN2H4 þ H2O ¼ N2Hþ þ OHꢁÞ. On
5
5
the other hand, it facilitated the privileged N–H bond cleavage
Many attempts have been put forward to achieve this aim. Xu
and coworkers synthesized various kinds of monometallic as well
as bimetallic nanoparticles [9–14]. Among them, Pt–Ni, Ir–Ni,
ðN2H4 ¼ N2Hꢂ þ HꢂÞ during the decomposition of N2H4 to N2 and
3
H2.
The hydrotalcite-like compounds are
a class of inorganic
layered materials represented by the general formula
h
i
MII MIIIðOHÞ2 ðAnꢁÞ ꢀ zH2O. The bidimensional structure is com-
posed of positively charged brucite-like layers of divalent and
ð1ꢁxÞ
x
x
⇑
Corresponding author. Fax: +86 411 84691570.
E-mail address: taozhang@dicp.ac.cn (T. Zhang).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.10.012

2
L. He et al. / Journal of Catalysis 298 (2013) 1–9
trivalent metals whose excess of positive charge is compensated
for by anions present in the interlayer region. Upon heating, this
layered compound forms a homogeneous mixture of oxides with
small crystal size. Part of this oxide mixture can be reduced to form
small and thermally stable metal particles. In particular, hydrotal-
cite and/or its calcination product is very relevant as a solid base,
which is increasingly regarded as a good alternative to these corro-
sive dissolved catalysts (e.g., NaOH, KOH): easy separation from
the reactant, recycling possibility, decreased corrosion of the reac-
tor, and so forth. Recently, we have reported that a basic Ni-based
catalyst, derived from Ni–Al hydrotalcite precursor, could catalyze
the selective decomposition of hydrous hydrazine with H2 selectiv-
ity up to 93% [18]. This unique catalysis is due to the cooperation of
Ni nanoparticles and strong basic sites derived from the alumina
support. However, this type of catalyst is not efficient enough for
practical applications and deactivates severely. In this work, we
deposited different amounts of Pt onto Ni-based hydrotalcite
was added as a precipitation agent. The suspension, thermostated
at 90 °C, was vigorously stirred for 8 h. After being washed with
water, the catalyst was dried at 80 °C for 10 h and then reduced
at 300 °C in pure hydrogen for 1 h to get Pt/Al2O3.
2.2. Hydrous hydrazine decomposition test
The test was carried out in a three-necked round-bottom flask
containing 4 ml of water with ꢃ40.0 mg of catalyst (the precursor
before reduction was 90.0 mg). The reaction temperature was con-
trolled by a water bath. Under magnetic stirring, the reaction was
initiated by injecting 0.5 ml N2H4ꢀH2O solution (3.2 mol Lꢁ1) into
the reactor. Gaseous product for the reaction was passed through
a trap containing 0.055 mol Lꢁ1 hydrochloric acid to ensure the
absorption of ammonia. The gas volume was monitored by an air-
tight gas buret connected to one neck of the reactor. Turnover fre-
quencies (TOFs, hꢁ1) for different catalysts were calculated by
assuming that all the Ni atoms took part in the reaction as
described by Xu et al. [12] and with the conversion of N2H4 at 50%.
The selectivity toward hydrogen generation (x) could be evalu-
ated on the basis of the following equation:
precursor to produce
a series of Pt-modified Ni catalysts
(NiPtx/Al2O3), which showed strongly enhanced and stable activity
for hydrous hydrazine decomposition at ambient temperatures.
Compared with other reported catalysts for this reaction, this novel
NiPtx/Al2O3 catalyst has the following advantages: (1) highly cata-
lytic activity even under ambient temperature; (2) avoiding the
usage of alkaline additives; and (3) highly efficient utilization of
noble metals.
3N2H4 ! 4ð1 ꢁ xÞNH3 þ 6xH2 þ ð1 þ 2xÞN2:
As NH3, if there were any, was absorbed by water and HCl
previously, the gas volume measured at the end of the reaction
contained only N2 and H2, from which the molar ratio of
n(N2 + H2)/n(N2H4) (k) was obtained. Therefore, selectivity (xvol
could be calculated as follows:
)
2. Experimental methods
ꢂ
ꢀ
ꢁꢃ
3k ꢁ 1
nðN2 þ H2Þ
nðN2H4Þ
1
3
2.1. Catalyst preparation
xvol
¼
k ¼
6 k 6 3
:
8
The Ni–Al hydrotalcite was used as a precursor for preparing
Ni/Al2O3 catalyst and NiPtx/Al2O3 catalysts. It was prepared by a
co-precipitation method as described elsewhere [18–20]. In brief,
an aqueous solution (50 mL) containing 10.89 g of Ni(NO3)2ꢀ6H2O
and 4.68 g of Al(NO3)3ꢀ9H2O (the total concentration of Ni2+ and
Al3+ was 1 mol Lꢁ1 with a Ni/Al molar ratio of 3:1) was added drop-
wise into a 50-mL continually stirred mixed solution containing
4.24 g of Na2CO3 (0.8 mol Lꢁ1) and 2.40 g of NaOH (1.2 mol Lꢁ1).
Afterward, a green suspension was obtained, and the pH value
was adjusted to 10 by introducing 3 mol Lꢁ1 NaOH aqueous solu-
tion. Then, the suspension was kept at 65 °C for 18 h to get a pre-
cipitate with hydrotalcite-like structure. After filtration, repeated
washing with deionized water and drying at 80 °C, the Ni–Al
hydrotalcite was synthesized. Finally, after reduction at 500 °C
(according to the H2 TPR results) in pure hydrogen for 1 h, Ni/
Al2O3 catalyst was obtained from the Ni–Al hydrotalcite precursor.
In order to keep metallic nickel from oxidizing, the catalyst was
protected in water before the hydrous hydrazine decomposition
test.
As the amount of HCl in the acid trap did not change after the
decomposition test, all of the produced NH3 was dissolved in
water. The amount of NH3 (d) in the supernatant was titrated using
calibrated HCl (0.055 mol Lꢁ1), and selectivity (xtitr) could be calcu-
lated as follows:
xtitr ¼ 1 ꢁ d=4:
To trace the selectivity of different catalysts in the entire
decomposition process, analysis of the gas released during the
reaction was performed with an online gas chromatograph (GC)
equipped with a 13X column and a thermal conductivity detector.
The peak areas of the H2 and N2 were calibrated for a standard gas
mixture with a molar ratio H2:N2 = 2:1.
For the stability test, the catalyst was separated by a magnet
after each cycle.
2.3. Catalyst characterization
Thermo IRIS Intrepid II inductively coupled plasma (ICP) was
used to determine the exact content of Pt, Ni, and Al in the
catalysts (Table 1). In addition, the content of Pt, Ni, and Al after
reaction was also checked to determine the amounts of these
species leached into the solution.
NiPtx/Al2O3 catalysts were synthesized by the deposition–pre-
cipitation (DP) method using urea as the precipitant. For NiPt0.057
/
Al2O3 catalyst, for example, the Ni–Al hydrotalcite precursor
(0.92 g) was first dispersed into 150 mL of deionized water with
the aid of ultrasound. Then, the H2PtCl6 solution (5 mL, containing
80.0 mg Pt) and 0.90 g of urea were added into the suspension
under magnetic stirring. Afterward, the temperature was increased
to 90 °C and kept for 8 h. Urea decomposition leads to a gradual
rise in pH to 9. The as-obtained deposit was filtered, washed with
water, dried at 80 °C, and reduced at 300 °C for 1 h in pure hydro-
gen to get NiPt0.057/Al2O3. Catalysts with other Pt contents were
synthesized as described above.
Table 1
Catalyst composition determined by inductively coupled plasma (ICP)
spectrophotometry.
Catalysts
Pt (wt%)
Ni (wt%)
Al (wt%)
Pt/Ni (molar ratio)
Ni/Al2O3
0
38.3
40.1
38.8
36.6
39.9
36.1
6.0
6.1
6.2
5.7
6.2
5.6
0
NiPt0.014/Al2O3
NiPt0.027/Al2O3
NiPt0.044/Al2O3
NiPt0.057/Al2O3
NiPt0.074/Al2O3
1.8
3.6
5.4
7.6
9.0
0.014
0.027
0.044
0.057
0.074
Pt/Al2O3 catalyst was also prepared using the DP method, as
mentioned above. H2PtCl6 solution (5 mL, containing 80.0 mg Pt)
was added to a 150-mL suspension containing dispersed 0.92 g
c
-Al2O3 (BET surface area 243.5 m2 gꢁ1). Afterward, 0.90 g of urea

L. He et al. / Journal of Catalysis 298 (2013) 1–9
3
Temperature-programmed reduction (TPR) experiments were
carried out on a Micromeritics AutoChem II 2920 automated
catalyst characterization system. First, about 100 mg of unreduced
precursor (Ni–Al hydrotalcite with or without Pt) was loaded into a
U-shaped quartz reactor and pretreated with air at 150 °C for 1 h to
remove adsorbed water. Then, after cooling to room temperature,
the flowing gas was switched to 10 vol% H2/Ar, and the catalyst
was heated to 900 °C at a ramping rate of 10 °C minꢁ1
X-ray diffraction (XRD) patterns were recorded with a PANalyt-
ical X’Pert-Pro powder X-ray diffractometer, using Cu K mono-
chromatized radiation (k = 0.1541 nm) at a scan speed of 5° minꢁ1
.
a
.
HAADF-STEM images were recorded on a JEOL2010F instrument
using an electron probe (diameter 0.5 nm) at a diffraction camera
length of 10 cm. The reduced catalysts were suspended on a copper
grid before the analysis.
The extended X-ray absorption fine structure (EXAFS) spectra of
the Pt L3-edge were measured at the BL-9C station of the Photon
Factory at the BL14W1 beamline of the Shanghai Synchrotron
Radiation Facility (SSRF) in a fluorescence mode. The samples were
first reduced at the same temperature as for the decomposition
tests and protected under inert atmosphere to prevent oxidation.
The IFFEFIT 1.2.11 data analysis package (Athena, Artemis, Atoms,
and FEFF6) was used for the data analysis and fitting. The raw data
were reduced by aligning the scans to the foil standard and deglit-
ching when necessary to diminish experimental error. The AUTOBK
algorithm in Athena was used to remove the isolated-atom
background function from the EXAFS data, and then, the data were
Fourier transformed into R-space. Local structural information was
obtained by using Artemis to fit each data set with theoretical stan-
dards generated by FEFF6 in R-space. The theoretical Pt–Pt photo-
electron amplitudes and phases were calculated for the bulk Pt fcc
structure. Pt–Ni contributions were calculated by replacing Pt with
Ni in the first nearest-neighbor shell.
Microcalorimetric measurements of H2 at 40 °C were performed
by using a BT 2.15 heat-flux calorimeter. The calorimeter was con-
nected to a gas handling and a volumetric system employing MKS
698A baratron capacitance manometers for precision pressure
measurement ( 1.33 ꢄ 10ꢁ2 Pa). The maximum apparent leak rate
of the volumetric system was 10ꢁ4 Pa minꢁ1 with a system volume
of approximately 80 cm3. The ultimate dynamic vacuum of the sys-
tem was ca. 10ꢁ5 Pa.
Temperature-programmed desorption (TPD) experiments were
carried out on a Micromeritics AutoChem II 2920 automated cata-
lyst characterization system. The samples for either H2 TPD or NH3
TPD tests were first reduced in situ under the same conditions as
for N2H4 decomposition. Then, the freshly reduced catalysts were
flushed with Ar for 60 min at a temperature 10 °C higher than for
reduction to remove the extra hydrogen adsorbed on the surface.
Afterward, for H2 TPD tests, when the temperature decreased to
40 °C, pure H2 was injected until saturation. For NH3 TPD tests,
NH3 was injected until saturation when the temperature decreased
to 30 °C. Finally, after flushing with Ar until the baseline was
stable, the TPD process was carried out until 800 °C at a rate of
10 °C minꢁ1. During the TPD process, the desorbed products were
monitored by both a thermal conductivity detector (TCD) and a
quadrupole mass spectrometer (MS).
Fig. 1. Time profiles for the hydrous hydrazine decomposition tests at 30 °C over (a)
NiPt0.074/Al2O3 catalyst; (b) NiPt0.057/Al2O3 catalyst; (c) NiPt0.044/Al2O3 catalyst; (d)
NiPt0.027/Al2O3 catalyst; (e) NiPt0.014/Al2O3 catalyst; (f) Ni/Al2O3 catalyst; (g) Pt/
Al2O3 catalyst. Insert: the comparison of TOFs versus Pt/Ni (molar ratio) for NiPtx/
Al2O3 catalysts.
for this reaction. Over Ni/Al2O3 catalyst, the reaction finished in
70.0 min. However, after being doped with a small amount of plat-
inum, hydrous hydrazine completely converted within 48.0 min
over NiPt0.014/Al2O3 catalyst. When the Pt/Ni (molar ratio) in-
creased to 2.7%, 4.4%, and 5.7%, the reaction time reduced to 26.0,
21.0, and 11.5 min, respectively (Table 2). Turnover frequencies
(TOFs, hꢁ1) have been calculated when the conversion was 50% over
different catalysts containing the same amount of Ni. The TOF over
NiPt0.057/Al2O3 was more than seven times higher than that over Ni/
Al2O3. Obviously, the addition of Pt to Ni/Al2O3 significantly pro-
moted the rate of hydrous hydrazine decomposition. In other
words, the activity was much higher on Pt-modified catalysts. Also,
the additive effect of Pt was strongly dependent on the amount of
Pt. An optimum amount of 5.7% Pt was observed, and further
increasing the loading of Pt led to a slight decrease in reaction rate.
Interestingly, in addition to the promotion of the decomposition
rate, H2 selectivity was also slightly higher after the introduction
of a small amount of Pt (more than 97%), which was further con-
firmed by GC analysis (Supporting information, Fig. S1). We traced
the components of gaseous production over the entire decomposi-
tion procedure on NiPt0.057/Al2O3 catalyst. The molar ratio of n(H2)/
n(N2) was above 1.99 and unchanged during the whole procedure,
which substantiates the high selectivity to H2 over the Pt-modified
catalysts. Furthermore, the H2 selectivity was calculated according
to the amount of NH3 determined by acid–base titration, which
coincided well with the volumetric observation (Table 2).
In the temperature range 30–80 °C, hydrous hydrazine
decomposition performances over Ni/Al2O3 and NiPt0.057/Al2O3
catalysts were also compared (Fig. 2). The apparent activation
energies (Ea) were estimated by linear fittings for the plots of lnk
versus the reciprocal absolute reaction temperature (1/T).
The Ea values were 34.0 1.8 kJ molꢁ1 for NiPt0.057/Al2O3 and
49.3 3.2 kJ molꢁ1 for Ni/Al2O3. The significant difference in Ea
suggests a facile decomposition of hydrous hydrazine in the pres-
ence of NiPt0.057/Al2O3 catalyst. It is noted that NiPt0.057/Al2O3 cat-
alyst exhibited high H2 selectivity (P97%) over the entire
investigated temperature range (Table 3). By contrast, Ni/Al2O3
catalyst showed 93% H2 selectivity at 30–50 °C, but the selectivity
decreased to 82% when the temperature increased further to 80 °C
(Table 3). As hydrous hydrazine decomposition is an exothermic
process, the wider temperature range for operation with a sustain-
able high selectivity gives us more confidence in the application of
NiPtx/Al2O3 catalysts.
3. Results and discussion
3.1. Catalytic performance in hydrous hydrazine decomposition
The hydrous hydrazine decomposition plots over Ni/Al2O3, Pt/
Al2O3, and NiPtx/Al2O3 catalysts at 30 °C are shown in Fig. 1. It is
clear that Ni/Al2O3 and NiPtx/Al2O3 catalysts could catalyze 100%
conversion of hydrous hydrazine, while Pt/Al2O3 catalyst was inert

4
L. He et al. / Journal of Catalysis 298 (2013) 1–9
Table 2
Comparison of selectivity and activity for hydrous hydrazine decomposition tests over different catalysts.
a
Catalysts
Reaction time (min)
Gas volume selectivity xvol (%)
Titration selectivity xtitr (%)
TOF (hꢁ1
)
Ni/Al2O3
70.0
48.0
26.0
21.0
11.5
15.5
86.0
93
95
97
98
99
95
96
94
97
97
98
98
98
97
2.2
3.9
7.8
NiPt0.014/Al2O3
NiPt0.027/Al2O3
NiPt0.044/Al2O3
NiPt0.057/Al2O3
NiPt0.074/Al2O3
9.9
16.5
11.4
1.8
b
K–Ni/Al2O3
a
The TOF was calculated when the conversion was 50%.
The mass percentage of K for K–Ni/Al2O3 was 0.23%.
b
investigated the structure and the surface adsorption properties
of the serial Pt-modified Ni/Al2O3 catalysts.
3.2.1. H2 TPR
H2 TPR profiles of the precursors of NiPtx/Al2O3 and Ni/Al2O3
catalysts are presented in Fig. 4. During the reduction in the Ni/
Al2O3 precursor, there was a broad hydrogen consumption peak
centered at 490 °C. This peak is assigned to the reduction of highly
dispersed NiO, which interacts strongly with the alumina support
[18]. It is noted that the ratio of reduced Ni was only 93% based
on Ni content according to the hydrogen consumption calculation,
which indicates that 7% of Ni2+ ions were located deep in the Al2O3
lattice as a solid solution and became hard-to reduce Ni2+ species.
On the other hand, the main reduction peak for NiPtx/Al2O3 cata-
lysts appeared below 300 °C, which was much lower than for Ni/
Al2O3 catalysts. The results strongly elucidates that the addition
of trace Pt significantly promoted the reduction of Ni. A similar ten-
dency was also observed for the previous noble-metal-modified Ni
catalysts [21]. This promotion effect was reported due to easy H2
dissociation on Pt, followed by spillover of hydrogen to Ni. This
H2 spillover effect resulted in the simultaneous reduction of Pt
and Ni at lower temperature. Moreover, the temperature of Ni
reduction peaks for NiPtx/Al2O3 decreased from 296 °C to 259 °C
as the content of Pt increased from 1.7% to 5.7%. Obviously, the
reducibility of Ni is notably promoted as the Pt content increased.
In addition, only a small shoulder peak at higher temperature
(412 °C) was observed for NiPt0.017/Al2O3. This small peak corre-
sponds to the reduction in the Ni2+ species, which are not influ-
enced by Pt because of the low Pt content and high Ni content.
These TPR results are also evidence of choosing reduction temper-
atures for different catalysts.
Fig. 2. A comparison of Arrhenius plots for (a) NiPt0.057/Al2O3 catalyst; (b) Ni/Al2O3
catalyst.
To further investigate the stability of Pt-modified catalysts, we
performed six runs of decomposition tests over the same NiPt0.057
/
Al2O3 catalyst at 30 °C (Fig. 3). After each cycle, the catalyst was
separated by a magnet and reused. Although the reaction rate
decreased a little after the first three cycles, the complete decom-
position of hydrous hydrazine could still occur within 18 min after
six cycles with a conversion rate of 7.8 hꢁ1. There is little deactiva-
tion in the last four cycles. Meanwhile, the selectivity to H2 gener-
ation remained above 98% over the Pt-modified catalysts. In
addition, ICP analysis of the supernatant after each cycle demon-
strated that neither Pt nor Ni was detected, indicating no leaching
of active components during the reaction.
3.2.2. XRD, HAADF-STEM, and EXAFS
XRD patterns gave a preliminary evidence for different crystal
structures of Ni/Al2O3 and NiPtx/Al2O3 catalysts (Fig. 5A). In the
case of Ni/Al2O3, the major diffraction peaks with 2h values of
44.5° and 51.7° can be indexed to the (111) plane and the (200)
plane of fcc Ni (PDF #01-087-0712), which are the most intensive
peaks for metallic Ni. It is important to notice that for all the
3.2. Catalyst characterization
It is obvious that the addition of Pt significantly promoted the
H2 generation rate as well as the selectivity in the decomposition
experiment. In order to explore the promoting effect of Pt, we
Table 3
Comparison of hydrous hydrazine decomposition tests over NiPt0.057/Al2O3 and Ni/Al2O3 catalysts at different reaction temperatures.
Reaction temperature (°C)
NiPt0.057/Al2O3
Ni/Al2O3
a
Reaction time (min)
Selectivity (%)
TOF (hꢁ1
)
Reaction time (min)
Selectivity (%)
TOF (hꢁ1
)
30
40
50
60
70
80
11.5
6.5
4.0
3.3
2.0
1.5
97
>99
99
98
97
16.5
28.2
56.1
72.0
93.9
137.7
70.0
50.0
20.0
14.0
7.0
93
93
93
91
87
82
2.2
2.8
7.8
11.6
23.3
34.6
97
5.5
a
The TOF was calculated when the conversion was 50% for both catalysts.

L. He et al. / Journal of Catalysis 298 (2013) 1–9
5
To confirm the existence of Pt–Ni alloy, Fig. 5B–D show the
high-angle annular dark-field scanning transmission electron
microscopy (HAADF-STEM) images of NiPt0.057/Al2O3 catalyst. The
HAADF result clearly shows the presence of bright spots that cor-
respond to supported metal nanoparticles on the alumina. The
HAADF-STEM technique allows us to locate the convergent elec-
tron beam (0.5 nm) at any position of the sample. In combination
with X-ray energy dispersive spectroscopy (X-EDS), it can be used
to obtain valuable chemical information of the samples at the
subnanometric scale. Four points were randomly chosen in the
area shown in Fig. 5C. The corresponding spectra show that all four
particles contain both Pt and Ni elements (Fig. 5D), although the
composition is not homogeneous. The EDX line scanning result
was also obtained by moving the electron probe along the line
indicated in Fig. S2 (Supporting information). It is evident that
the intensity of Ni (red) varied simultaneously with that of Pt
(blue), which further confirms the formation of Pt–Ni alloy.
Extended X-ray absorption fine structure (EXAFS) measure-
ments were conducted to further substantiate the formation of
Pt–Ni alloy. The normalized Pt L3-edge X-ray absorption near-edge
structure (XANES) spectra of NiPt0.014/Al2O3, NiPt0.057/Al2O3, as
well as the reference spectra of Pt foil and PtO2, are shown in
Fig. 6A. The intensities in the near-edge spectra reflect the oxida-
tion state of Pt in different samples. In both NiPt0.014/Al2O3 and
NiPt0.057/Al2O3 samples, the spectra were similar to that of Pt foil,
suggesting that all of Pt species are fully reduced [23]. As specu-
lated above, Pt–Ni alloy was formed after reduction, which may
lead to electron transfer between Ni and Pt atoms and therefore
influences the extended-edge spectra. Fig. 6B shows the Fourier
transform of Pt L3-edge EXAFS oscillations for the corresponding
samples. The k-range used in the transform was chosen in order
to exclude the regions of noise at high k. Obviously, the first
nearest-neighbor distance for NiPtx/Al2O3 catalysts was shorter
than the Pt–Pt bond, but longer than the Pt–O bond, implying
the existence of a Pt–Ni bond [19,24–27]. Curve-fitting results of
the catalysts are listed in Table 4. For Pt foil, the coordination num-
ber (CN) was 12. However, the sums of the CNs of Pt–Pt and Pt–Ni
bonds for NiPt0.014/Al2O3 and NiPt0.057/Al2O3 were 9.0 and 9.6,
respectively, which were smaller than the CN for Pt foil, indicating
that Pt was highly dispersed on the support. For NiPt0.014/Al2O3, the
CN of Pt–Ni took up 73% of the total CN of Pt, while this ratio in-
creased to 82% for NiPt0.057/Al2O3. The results produced evidence
in support of the existence of Pt–Ni alloy, as there were Pt–Ni
bonds for both low and high Pt content catalysts.
Fig. 3. The molar ratio plots of n(N2 + H2)/n(N2H4) versus time for the recycling test
of hydrous hydrazine decomposition over NiPt0.057/Al2O3 catalyst at 30 °C. The
catalyst was separated by a magnet after each cycle.
Fig. 4. H2 TPR results for the precursors of Pt–Ni serial catalysts (a) NiPt0.017/Al2O3
catalyst; (b) NiPt0.027/Al2O3 catalyst; (c) NiPt0.044/Al2O3 catalyst; (d) NiPt0.057/Al2O3
catalyst; (e) Ni/Al2O3 catalyst. TPR conditions: 10% H2/Ar, 10 K minꢁ1. The intensity
of H2 consumption is normalized by the catalysts’ Ni content.
Pt-modified NiPtx/Al2O3 catalysts, two main diffraction peaks are
still related to Ni0. Peaks related to Pt0 were not detected for the
NiPtx/Al2O3 catalysts, indicating that Pt atoms are highly dispersed
among the catalysts with very small particle size, which could not
be detected by XRD. Notably, as the content of Pt increased from
1.7% to 5.7%, the peaks for Ni (111) shifted to 44.2°, 44.1°, 43.9°,
and 43.5°, respectively. This shift in the 2h values could be attrib-
uted to the alloying of the metals involved [22]. The partial
replacement of Ni atoms by Pt atoms having a bigger atomic radius
contributed to the shift of 2h to lower angles. In accord with
Vegard’s law, we made a calculation of the qualitative composition
of Pt–Ni alloy in different NiPtx/Al2O3 catalysts by using the peak
with highest intensity. The alloys are Ni97Pt3 for NiPt0.014/Al2O3,
Ni94Pt6 for NiPt0.027/Al2O3, and Ni88Pt12 for NiPt0.057/Al2O3. The
average size of Ni0 particles, calculated by the Scherrer equation,
was ꢃ4.4 nm for Ni/Al2O3, but decreased to 4.0–2.6 nm, depending
on the amount of Pt introduced. Moreover, there was a very weak
diffraction peak observed at ꢃ63°, which was attributed to nickel
aluminum spinel, that is, the unreduced part of nickel. Interest-
ingly, the intensity of this diffraction peak decreased as the Pt con-
tent increased, indicating that the introduction of Pt significantly
promoted the reduction of Ni species in the NiPtx/Al2O3 catalysts.
3.2.3. Microcalorimetry and H2 TPD
The results presented in this paper clearly show that NiPtx/Al2O3
catalysts exhibit higher decomposition activity than the corre-
sponding monometallic catalysts. The increased activity is directly
related to Pt–Ni bimetallic bond formation. Further experiments
were conducted to explain the enhanced catalytic activity of Pt-
modified catalysts. Microcalorimetric experiments of H2 adsorption
were used to probe the change of H2 adsorption properties on NiPtx/
Al2O3 catalysts. This microcalorimetric measurement can provide
not only the adsorbed amount of the probe molecules (e.g., H2,
CO), but also the energetic distribution of the surface sites as a func-
tion of the coverage. Fig. 7 displays the differential heat of H2
adsorption at 40 °C as a function of H2 uptake over Ni/Al2O3,
NiPt0.027/Al2O3, and NiPt0.057/Al2O3 catalysts. The saturated amount
of H2 adsorption was 729
ger than those of NiPt0.027/Al2O3 catalyst (419
NiPt0.057/Al2O3 catalyst (241
mol gꢁ1). Simultaneously, the corre-
l
mol gꢁ1 for Ni/Al2O3, which is much lar-
mol gꢁ1) and
l
l
sponding initial heats of H2 adsorbed were 87.0, 79.8, and
69.6 kJ molꢁ1, respectively, indicating that the adsorption strength
was much weaker on Pt–Ni alloy surfaces. From the percentage of
active sites for heat distribution histograms (Fig. 7, right) obtained

6
L. He et al. / Journal of Catalysis 298 (2013) 1–9
Fig. 5. (A) XRD patterns for (a) Ni/Al2O3 catalyst; (b) NiPt0.014/Al2O3 catalyst; (c) NiPt0.027/Al2O3 catalyst; (d) NiPt0.057/Al2O3 catalyst. (B) STEM-HAADF image of NiPt0.057/Al2O3
catalyst. (C) STEM-EDX point spectrum of NiPt0.057/Al2O3 catalyst. (D) Corresponding EDX analysis results for the four randomly chosen particles in C.
Fig. 6. (A) The normalized intensity of Pt L3-edge XANES spectra for (a) Pt foil; (b) NiPt0.014/Al2O3 catalyst; (c) NiPt0.057/Al2O3 catalyst; (d) PtO2. (B) Corresponding Fourier
transform k3-weighted EXAFS spectra (without phase correction) for (a) Pt foil; (b) NiPt0.014/Al2O3 catalyst; (c) NiPt0.057/Al2O3 catalyst; (d) PtO2.
D .
k = 3.7–12.6 Åꢁ1
a.u. = arbitrary units.
by a method described by Cortright and Dumesic [28], it could be
seen that a total of 59.3% of active sites on Ni/Al2O3 catalyst had a
differential heat of H2 adsorption above 80 kJ molꢁ1. In contrast,
49.9% of active sites on NiPt0.057/Al2O3 surface had a differential

L. He et al. / Journal of Catalysis 298 (2013) 1–9
7
Table 4
Curve fitting results of Pt L3-edge EXAFS of NiPt0.014/Al2O3 and NiPt0.057/Al2O3
catalysts.
Samples
Pt foil
Shells CNa
Pt–Pt 12
R (Å)b
D
r2 (Å2)c E0 (eV)d R factor
2.76
0.0041
0.0063
0.0063
0.0065
0.0065
5.50
4.62
4.62
4.39
4.39
0.0044
0.0065
NiPt0.014/Al2O3 Pt–Ni 6.6 1.3 2.54
Pt–Pt 2.4 0.4 2.71
NiPt0.057/Al2O3 Pt–Ni 7.9 1.5 2.54
0.0058
Pt–Pt
1.7 0.3 2.67
a
b
c
Coordination number.
Distance between absorber and backscatter atoms.
Change in the Debye–Waller factor value relative to the Debye–Waller factor of
the reference compound.
d
Inner potential correction to account for the difference in the inner potential
between the sample and the reference compound.
heat between 60 and 70 kJ molꢁ1, indicating much weaker H2
adsorption. Obviously, the presence of Pt in Ni/Al2O3 significantly
decreased the strength of the Ni–H bond.
Fig. 8. Mass spectra of H2 TPD for (a) Ni/Al2O3 catalyst; (b) NiPt0.014/Al2O3 catalyst;
(c) NiPt0.027/Al2O3 catalyst; (d) NiPt0.057/Al2O3 catalyst. The samples were first
reduced in situ at the same condition as for hydrous hydrazine decomposition tests.
To further confirm the unique properties of Pt-modified surface,
TPD studies were performed to investigate the H2 desorption from
the series of NiPtx/Al2O3 catalysts (Fig. 8). The temperature of
desorption is related to the metal–H bond strength [29]. There is
only one peak at ꢃ200 °C for the Ni/Al2O3 catalyst, which is attrib-
uted to H2 desorption from Ni particles. This result is consistent
with the studies by other authors with thermal desorption of
N2H4 on Ni(110), on which H2 desorption appears in the
27–245 °C range [30]. It is apparent that the temperature of H2
desorption is significantly decreased with increasing amounts of
Pt in NiPtx/Al2O3 catalysts, indicating a weaker metal–H bond on
surface Pt–Ni alloy [31–33]. The more the Pt atoms locate on the
surface, the more easily the H2 desorption occurs. Considering that
the hydrazine decomposition was tested at 30 °C in this paper,
hydrogen desorption may be one of the rate-determining steps in
hydrazine decomposition in this case. Consequently, when Pt was
deposited on Ni/Al2O3 to form an alloy catalyst, the produced
hydrogen became easier to release from the Pt–Ni alloy surface.
The much lower energy needed for product desorption may effec-
tively enhanced the reactivity of hydrous hydrazine decomposi-
tion. To further examine the influence of metal–H bond strength
on activity, we introduced small amounts of alkali (such as potas-
sium) to Ni/Al2O3 to strengthen the Ni–H bond [34–36]. The results
show that the heat of H2 adsorption on Ni/Al2O3 was raised by
doping with trace potassium (Supporting information, Fig. S3). As
expected, the K-doped Ni/Al2O3 exhibited decreased reactivity in
hydrous hydrazine decomposition (Table 2). It could be expected
that the decrease in activity was due to the strengthening of
Ni–H bonds caused by doping with potassium.
3.2.4. NH3 TPD
It has been reported that alloying of Pt brings about a significant
change in the surface structure and affects the adsorption proper-
ties (weakening the adsorption strength between the metal atoms
and the adsorbed species) [37,38]. Thus, it is reasonable to assume
that the alloying of Pt and Ni could also tune the metal–N bond
strength [39,40]. Thereafter, NH3 TPD was carried out to investi-
gate the strength of adsorbed N species on NiPtx/Al2O3 catalysts.
The mass spectra of desorbed NH3 and N2 are compared in Fig. 9.
The NH3 desorption peak was at 97 °C for all Ni/Al2O3, NiPt0.027
/
Al2O3, and NiPt0.057/Al2O3 catalysts. Notably, there was a N2
desorption peak recorded above 350 °C originated from an ad-
sorbed ammonia intermediate on all the samples, indicating their
high activity for breaking N–H bonds. However, the amount of N2
desorption on Ni/Al2O3 was significantly lowered by the addition
of Pt atoms. This may be related to the relatively weaker Ni–N
bond due to the formation of alloy [41]. These N species were
Fig. 7. Differential heat of H2 adsorption on (a) Ni/Al2O3 catalyst; (b) NiPt0.027/Al2O3 catalyst; (c) NiPt0.057/Al2O3 catalyst, at 40 °C in static system (left); and the respective
heat distribution histograms (right).

8
L. He et al. / Journal of Catalysis 298 (2013) 1–9
Fig. 9. Mass spectra of NH3 TPD-MS for (A) NH3 desorption signal and (B) N2 desorption signal for (a) Ni/Al2O3 catalyst; (b) NiPt0.027/Al2O3 catalyst; (c) NiPt0.057/Al2O3 catalyst.
difficult to desorb under mild condition and caused catalyst deac-
tivation. The reaction rate was decreased after three cycles, which
is probably due to the accumulation of strongly bonded N-ada-
toms. The weaker adsorption of N species on the Pt–Ni alloy sur-
face reduced the deactivation and enhanced the stability for the
catalyst.
The existence of Pt–Ni alloy weakens both Ni–H and Ni–N
bonds during the hydrous hydrazine decomposition process. Such
a change lowers the energy barrier of molecular H2 desorption
from the surface of catalysts, which may be the source of the pro-
moting conversion rate over NiPtx/Al2O3 catalysts. On the other
hand, weakened Ni–N bond reduces the potential N self-poisoning
problem on Ni catalysts and enhances the stability of the catalysts.
It has been reported that the prevailing cleavage of N–H bonds in-
stead of N–N bonds over Ni catalysts may be the reason for high H2
selectivity [42–45]. In our previous work, we have known that Ni/
Al2O3 itself shows a high selectivity for this reaction. The alloy of Ni
with Pt may further adjust the prevailing break of N–H bond,
which leads to an enhanced H2 selectivity from 93% to P97%.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2012.10.012.
References
[1] U.B. Demirci, P. Miele, Energy Environ. Sci. 4 (2011) 3334.
[2] L. Schlapbach, A. Züttel, Nature 414 (2001) 353.
[3] U. Eberle, M. Felderhoff, F. Schüth, Angew. Chem. Int. Ed. 48 (2009) 6608.
[4] J. Yang, A. Sudik, C. Wolverton, D.J. Siegel, Chem. Soc. Rev. 39 (2010) 656.
[5] E.W. Schmidt, In Hydrazine and its Derivatives: Preparation, Properties,
Applications, second ed., Wiley, New York, 2001.
[6] H.L. Jiang, S.K. Singh, J.M. Yan, X.B. Zhang, Q. Xu, ChemSusChem 3 (2010)
5412.
[7] S. Mary, C. Kappenstein, S. Balcon, S. Rossignol, E. Gengembre, Appl. Catal. A
182 (1999) 317.
[8] R. Eloirdi, S. Rossignol, C. Kappenstein, D. Duprez, N. Pillet, J. Propul. Power 19
(2003) 213.
[9] S.K. Singh, X.B. Zhang, Q. Xu, J. Am. Chem. Soc. 131 (2009) 9894.
[10] S.K. Singh, Q. Xu, J. Am. Chem. Soc. 131 (2009) 18032.
[11] S.K. Singh, Q. Xu, Chem. Commun. 46 (2010) 6545.
[12] S.K. Singh, Q. Xu, Inorg. Chem. 49 (2010) 6148.
[13] S.K. Singh, Z.H. Lu, Q. Xu, Eur. J. Inorg. Chem. 14 (2011) 2232.
[14] S.K. Singh, A.K. Singh, K. Aranishi, Q. Xu, J. Am. Chem. Soc. 133 (2011) 19638.
[15] J. Wang, X. Zhang, Z. Wang, L. Wang, Y. Zhang, Energy Environ. Sci. 5 (2012)
6885.
4. Conclusion
[16] J. Song, R. Ran, Z. Shao, Int. J. Hydrogen Energy 35 (2010) 7919.
[17] B. Zhao, J. Song, R. Ran, Z. Shao, Int. J. Hydrogen Energy 37 (2012) 1133.
[18] L. He, Y. Huang, A. Wang, X. Wang, X. Chen, J.J. Delgado, T. Zhang, Angew.
Chem. Int. Ed. 51 (2012) 6191.
Taking Ni–Al hydrotalcite as a precursor, doping a small amount
of Pt, even 1.4%, could efficiently promote the reaction rate and H2
selectivity for the decomposition of hydrous hydrazine at room
temperature. According to the HAADF-STEM, XRD, and EXAFS re-
sults, Pt–Ni alloy is formed on the near surface of Pt-modified
NiPtx/Al2O3 catalysts during the reduction process. The formation
of surface Pt–Ni alloy affects the nature of chemisorption bond be-
tween the adspecies and the Ni surface atoms. Such a change is
manifested as a weakening of the surface bonding of H and NHx
adspecies to Ni, which associates with the high reaction rate and
enhanced H2 selectivity for hydrous hydrazine decomposition.
Although this is a common feature by alloying Ni with Pt, the more
detailed experimental and theoretical studies are needed to further
understand the origin. This kind of NiPtx/Al2O3 catalysts gives con-
fidence in the practical application of hydrous hydrazine as a
hydrogen carrier material.
[19] K. Takehira, J. Nat. Gas. Chem. 18 (2009) 237.
[20] J. Feng, Y. Lin, D.G. Evans, X. Duan, D. Li, J. Catal. 266 (2009) 351.
[21] A. Tanksale, J.N. Beltramini, J.A. Dumesic, G.Q. Lu, J. Catal. 258 (2008) 366.
[22] N.H.H. Abu Bakara, M.M. Bettahara, M. Abu Bakarb, S. Monteverdia, J. Ismailb,
M. Alnotc, J. Catal. 265 (2009) 63.
[23] S. Qi, B.A. Cheney, R. Zheng, W.W. Lonergan, W. Yu, J.G. Chen, Appl. Catal. A 393
(2011) 44.
[24] B. Li, S. Kado, Y. Mukainakano, T. Miyazawa, T. Miyao, S. Naito, K. Okumura, K.
Kunimori, K. Tomishige, J. Catal. 245 (2007) 144.
[25] S. Lu, W.W. Lonergan, J.P. Bosco, S. Wang, Y. Zhu, Y. Xie, J.G. Chen, J. Catal. 259
(2008) 260.
[26] W.W. Lonergan, D.G. Vlachos, J.G. Chen, J. Catal. 271 (2010) 239.
[27] W.W. Lonergan, T. Wang, D.G. Vlachos, J.G. Chen, Appl. Catal. A 408 (2011) 87.
[28] R.D. Cortright, J.A. Dumesic, J. Catal. 148 (1994) 771.
[29] J. Greeley, M. Mavrikakis, J. Phys. Chem. B 109 (2005) 3460.
[30] S. Huang, T. Rufael, J. Gland, Surf. Sci. Lett. 290 (1993) L67.
[31] M. García-Diéguez, E. Finocchio, M. Larrubia, L.J. Alemany, G. Busca, J. Catal.
274 (2010) 11.
[32] J.G. Chen, C.A. Menning, M.B. Zellner, Surf. Sci. Rep. 63 (2008) 201.
[33] S. Velu, S.K. Gangwal, Solid State Ion. 177 (2006) 803.
[34] N. Al-Sarraf, J.T. Stuckless, D.A. King, Nature 360 (1992) 243.
[35] Y. Zhai, D. Pierre, R. Si, W. Deng, P. Ferrin, A.U. Nilekar, G. Peng, J.A. Herron, D.C.
Bell, H. Saltsburg, M. Mavrikakis, M. Flytzani-Stephanopoulos, Science 329
(2010) 1633.
[36] J.M. Lee, Y.J. Min, K.B. Lee, S.G. Jeon, J.G. Na, H.J. Ryu, Langmuir 26 (2010)
18788.
[37] V.R. Stamenkovic, B.S. Mun, K.J.J. Mayrhofer, P.N. Ross, N.M. Markovic, J. Am.
Chem. Soc. 128 (2006) 8813.
Acknowledgments
This work was supported by the National Natural Science Foun-
dation of China (21076211, 21103173, and 21176235). Xiaowei
Chen is grateful for ‘‘Ramón y Cajar’’ program from Spanish Minis-
try of Science and Innovation. We also acknowledge the Shanghai
Synchrotron Radiation Facility (SSRF) for the EXAFS experiment.

L. He et al. / Journal of Catalysis 298 (2013) 1–9
9
[38] J.R. Kitchin, J.K. Nørskov, M.A. Barteau, J.G. Chen, Phys. Rev. Lett. 93 (2004)
156801.
[39] D.A. Hansgen, D.G. Vlachos, J.G. Chen, Nat. Chem. 2 (2010) 484.
[40] J. Hannaner, O. Akdim, U.B. Demirci, C. Geantet, J. Herrmann, P. Miele, Q. Xu,
Energy Environ. Sci. 4 (2011) 3355.
[42] H.L. McKay, S.J. Jenkins, D.J. Wales, J. Phys. Chem. C 115 (2011) 17812.
[43] P. Zhang, Y. Wang, Y. Huang, T. Zhang, G. Wu, J. Li, Catal. Today 165 (2011)
80.
[44] M. Zheng, R. Cheng, X. Chen, N. Li, L. Li, X. Wang, T. Zhang, Int. J. Hydrogen
Energy 30 (2005) 1081.
[41] Y.K. Al-Haydari, J.M. Saleh, M.H. Matloob, J. Phys. Chem. 89 (1985) 3286.
[45] S. Dahl, A. Logadottir, C.J.H. Jacobsen, J.K. Nørskov, Appl. Catal. A 222 (2001) 19.