Hollow nickel-coated silica microspheres containing rhodium nanoparticles for highly selective production of hydrogen from hydrous hydrazine

Article abstract of DOI:10.1039/c4ta03550j

The synthesis of hollow nickel-coated silica microspheres containing rhodium nanoparticles (NPs) (Rh/Ni@SiO₂) via thermal hydrolysis of urea using core-shell silica microspheres as templates is described. This dissolution-and-deposition method

Full text of DOI:10.1039/c4ta03550j

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Hollow nickel-coated silica microspheres
containing rhodium nanoparticles for highly
selective production of hydrogen from hydrous
hydrazine†
Cite this: J. Mater. Chem. A, 2014, 2,
18929
*
Jung Bo Yoo, Han Sol Kim, Seung Hee Kang, Byeongno Lee and Nam Hwi Hur
The synthesis of hollow nickel-coated silica microspheres containing rhodium nanoparticles (NPs)
(Rh/Ni@SiO2) via thermal hydrolysis of urea using core–shell silica microspheres as templates is
described. This dissolution-and-deposition method using urea as a precipitating agent provided uniform
hollow microspheres composed of amorphous Ni(OH)2 and silica (SiO2) layers along with small amounts
of Rh species even without etching; these hollow microspheres transformed to crystalline Rh/Ni@SiO2
microspheres after annealing at 750 ꢀC under a reducing atmosphere. The formation of a hollow
structure is dependent on the concentration of urea and unique dissolution behavior of the core–shell
silica. The bimetallic Rh/Ni@SiO2 microsphere with a low Rh content (6.35 wt%) is a highly active catalyst
for complete dissociation of hydrous hydrazine into hydrogen and nitrogen. Complete release of
hydrogen from hydrous hydrazine was accomplished at 25 ꢀC with a H2 selectivity of 99.4% and turnover
number of 66. The used Rh/Ni@SiO2 catalyst, which was recovered by a magnet, was reused in
subsequent reactions with virtually identical activity.
Received 11th July 2014
Accepted 16th September 2014
DOI: 10.1039/c4ta03550j
www.rsc.org/MaterialsA
catalytically dissociate to yield hydrogen. Numerous boron–
nitrogen compounds have been evaluated as potential
Introduction
hydrogen-storage materials; they are suitable for producing
hydrogen under relatively mild conditions but yield undesired
products that cannot be regenerated by hydrogenation under
practical conditions.9–15
Combustion of fossil fuels produces large amounts of carbon
dioxide (CO2), which is a main contributor to climate change. In
contrast, hydrogen (H2) is free of carbon and its combustion
produces water as the only waste product. Accordingly,
hydrogen plays a major role in the generation of clean energy
and reduction of CO2 emissions.1–3 However, a major drawback
is that hydrogen is very difficult to store and transport because it
exists as a ammable gas under ambient conditions. An alter-
native approach is to develop hydrogen storage materials that
satisfy several technical requirements, including sufficiently
high volumetric and gravimetric capacities, facile release of
hydrogen at a reasonably low temperature, and efficient
regeneration at a practical temperature.4 Thus far, a wide range
of hydrogen storage materials, such as nano materials,5 metal
hydrides,6 chemical hydrides,7 and liquid organic compounds,8
has been explored to facilitate the use of hydrogen as a fuel. In
particular, chemical hydrides composed of boron, nitrogen, and
hydrogen atoms have drawn signicant interest because they
are light atoms with high gravimetric hydrogen capacities. In
addition, hydridic B–H and protic N–H bonds can thermally or
On the other hand, liquid organic hydrogen-storage mate-
rials possess signicant advantages over solid hydrides,8
particularly in on-board applications, because they can be
handled like fuel in a vehicle. Aer releasing hydrogen, the
dehydrogenated materials can be regenerated by hydrogenation
or stored in another tank. However, most of these reversible
organic compounds have low gravimetric capacities and require
signicant energy to dissociate the C–H and N–H bonds.
Anhydrous hydrazine (H2NNH2) is a promising hydrogen
storage material because it has a high H2 content (12.5 wt%)
and is in the liquid state at room temperature. Moreover, it can
ideally be dissociated into hydrogen and nitrogen.16 However,
direct use of anhydrous hydrazine is restricted because it is
extremely toxic, highly reactive, and even explosive. Hydrous
hydrazine, which is prepared by diluting hydrazine with water,
is thus used as an alternative to mitigate the toxicity and reac-
tivity of anhydrous hydrazine.17–20 Complete decomposition of
hydrazine yields only hydrogen and nitrogen according to the
following equation: H2NNH2 / N2 (g) + 2H2 (g). Hydrazine can
also incompletely decompose into ammonia and nitrogen, as
follows: 3H2NNH2 / 4NH3 (g) + N2 (g). The dissociation
Department of Chemistry, Sogang University, Seoul 121-742, Korea. E-mail: nhhur@
sogang.ac.kr
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ta03550j
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18929–18937 | 18929

View Article Online
Journal of Materials Chemistry A
Paper
pathways depend signicantly on the catalyst and reaction
Herein, we report hollow nickel-coated microspheres con-
conditions. A wide range of catalysts based mostly on metal taining Rh NPs (Rh/Ni@SiO2) that were prepared by thermal
nanoparticles (NPs) have been developed for complete decom- hydrolysis of urea using core–shell silica microspheres as
position of hydrous hydrazine.21–28 Rh NPs have proven to be templates followed by annealing at 750 ꢀC under owing 5% H2
very active for the production of H2 from hydrous hydrazine: in an Ar atmosphere. The Rh/Ni@SiO2 microspheres maintain a
their H2 selectivity is close to 43%.29 Monometallic catalysts spherical shape, have large cavities in the core, and contain less
containing Co, Ru, or Ir NPs are also active for the decompo- than 6.35 wt% Rh. At room temperature, the Rh/Ni@SiO2
sition reaction but their selectivity is very low (<7%). Most catalyst generates H2 with over 99% selectivity within 1.5 h.
Ni-based bimetallic catalysts with precious metals show Moreover, the used Rh/Ni@SiO2 catalyst recovered by a magnet
enhanced selectivity towards H2 production;30–35 their activities produces H2 at the same selectivity; these results generate
not only depend on the metal composition but also on the signicant promise for hydrous hydrazine as a potential
compositional ratio. Thus far, an Rh-rich alloy (i.e., Rh4Ni) hydrogen-storage material.
shows the highest selectivity (ꢁ100%) for the production of H2
from hydrous hydrazine at room-temperature.35 However,
complete replacement of Rh by abundant metals or reduction of
the Rh content by alloying with inexpensive metals is necessary
Experimental section
Chemicals
for this technology for producing H2 from hydrazine to be
industrially viable.36–38
All chemicals were purchased from commercial suppliers
and used without further purication unless otherwise stated.
Tetraethyl orthosilicate (TEOS), nickel acetylacetonate
(C10H14NiO4), and rhodium chloride hydrate (RhCl3$xH2O)
were purchased from Tokyo Chemical Industry Co. (Tokyo,
Japan). Octadecyl trimethoxysilane (C18-TMS), hydrazine
hydrate (H2NNH2$H2O, 98%), benzyl aldehyde, phenol, sodium
hypochlorite (NaOCl), sodium hydroxide (NaOH), sodium
nitroferricyanide dihydrate (Na2[Fe(CN)5NO]$2H2O), and urea
were purchased from Sigma-Aldrich (St Louis, USA). Ethanol,
deionized water, sulfuric acid, and ammonia solution (25–28%)
were obtained from Samchun Chemical Reagent Co. (Seoul,
Korea).
Thus far, most of the catalysts developed for complete
decomposition of hydrous hydrazine are NPs that are typi-
cally composed of noble metals and transition metals.
Bimetallic NPs have a high surface area, which is very
advantageous, particularly for catalytic activity. The synthesis
of bimetallic NPs typically requires high temperatures, which
lead to agglomeration of the NPs. To prevent aggregation,
NPs can be covered with capping agents; however, these oen
hamper the catalytic reaction. Another drawback is that
bimetallic NPs are difficult to separate and reuse in a
subsequent reaction mainly because of their extremely small
sizes. From this perspective, hollow microspheres containing
active-metal NPs on their outer surfaces are signicant
challenges both synthetically and catalytically. An important
benet of hollow microspheres is that they can be used as
recyclable supports with high surface areas.
Synthesis of core–shell silica microspheres (core–shell SiO2)
The core–shell silica microsphere templates were prepared
using a slightly modied Stober procedure.53 In brief, 75 mL of
¨
ethanol, 10 mL of deionized water, and 3 mL of ammonia
solution were added to a round-bottom ask, and the solution
was stirred for 10 min. Using a syringe, TEOS (6 mL) was
injected into the solution. Aer vigorous stirring for 2 h, a
mixture of TEOS (5 mL) and C18-TMS (2 mL) was added, and the
solution was stirred for another 2 h. A white precipitate was
formed and separated from the solution by centrifugation. To
remove all organic residues in the precipitated powders, the
powders were sintered in air at 550 ꢀC for 6 h, yielding 2.0 g of
silica powders. Silica has a spherical shape and a core–shell
structure. The core diameter is approximately 300 nm and the
shell thickness ranges from 20 to 30 nm.42
Hollow materials containing metal NPs are typically
prepared using sacricial templates.39–41 This process involves
coating the desired materials onto the templates and then
etching the templates away to form the hollow structure. In
previous work, we prepared hollow materials through selective
etching of the templates using a concentrated alkaline solu-
tion.42 This methodology is simple but partially degrades the
active shell components. The reports on the syntheses of hollow
materials without etching have thus far been limited. In the
present work, we use core–shell silica (SiO2) microspheres as
templates and employ urea as an alkaline medium, thereby
forming hollow microspheres that contain metal species
without etching. This dissolution-and-deposition method
proceeds through thermal hydrolysis of urea. Urea is a simple
organic compound that has two NH2 groups connected by a
Synthesis of hollow nickel-coated silica microspheres
(Ni@SiO2)
carbonyl (CO) group; it is neither acidic nor alkaline in water The prepared core–shell SiO2 microspheres (0.495 g) were
under ambient conditions. Upon heating, however, it produces dispersed in 250 mL of deionized water in a 500 mL round-
ammonia in water, which causes the solution to become alka- bottom ask. Nickel acetylacetonate (0.4 g, 1.6 ꢂ 10ꢃ3 mol) and
line. A urea solution at high temperature can dissolve silica in a urea (4.0 g, 6.7 ꢂ 10ꢃ2 mol) were dissolved into the solution,
controlled manner even though it is only slightly soluble.43–52 which was then stirred at 80 ꢀC for 2 h. The white microspheres
The weak alkalinity of the urea solution leads to the uniform gradually became green colloidal spheres, which were separated
deposition of metal precursors and gradual dissolution of the from the solution by centrifugation (3000 rpm), and dried in an
ꢀ
silica core.
oven at 100 C. The dried microspheres (Ni(OH)2@SiO2) were
18930 | J. Mater. Chem. A, 2014, 2, 18929–18937
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
annealed at 750 ꢀC for 20 h under a reducing atmosphere
Methods
(Ar/H2 ¼ 95 : 5) to yield the desired nickel-coated microspheres.
Powder X-ray diffraction (XRD) patterns were recorded using a
Rigaku DMAX 2500 diffractometer (Cu Ka; Rigaku, Japan)
operating at 40 kV and 150 mA. Transmission electron
microscopy (TEM) was performed using a JEOL JEM-2100F
microscope (JEOL, Japan). Specimens for TEM examinations
were prepared by dispersing nely ground powders of the
samples in anhydrous ethanol and then allowing a drop of the
suspension to evaporate on a 400 mesh carbon-coated grid.
Energy-dispersive X-ray (EDX) spectroscopy was performed
using the same microscope as used for TEM. Line-scan analyses
were performed in scanning transmission electron microscope-
mode using a real-time interactive imaging system with a high-
angle annular dark-eld detector. Adsorption and desorption
measurements were carried out at 77 K using an ASAP 2420
instrument (Micromeritics, Norcross, USA) with nitrogen as the
adsorptive gas. The Brunauer–Emmett–Teller (BET) surface
areas were calculated using the following: P/P0 ¼ 0.05–0.3 from
the adsorption curve using the BET equation. The pore-size
distributions were obtained from the desorption curve using
the density functional theory method. Prior to each sorption
Synthesis of hollow nickel-coated silica microspheres
containing rhodium (Rh/Ni@SiO2)
The procedure for the synthesis of Rh/Ni@SiO2 is virtually
identical to that of Ni@SiO2 except that rhodium chloride
hydrate is added as a source of Rh. The core–shell SiO2 micro-
spheres (0.495 g) were dispersed in 250 mL of deionized water in
a 500 mL round-bottom ask. Rhodium chloride hydrate (0.04
g, 1.9 ꢂ 10ꢃ4 mol), nickel acetylacetonate (0.4 g, 1.6 ꢂ 10ꢃ3
mol), and urea (4.0 g, 6.7 ꢂ 10ꢃ2 mol) were dissolved into the
solution, which was then stirred at 80 ꢀC for 2 h. The white
microspheres gradually became green colloidal spheres, which
were separated from the solution by centrifugation (3000 rpm),
and dried in an oven at 100 ꢀC. The dried microspheres
(Rh/Ni(OH)2@SiO2-1T) were annealed at 750 ꢀC for 20 h under a
reducing atmosphere (Ar/H2 ¼ 95 : 5). The rhodium-deposited
nickel-coated silica microspheres are denoted as Rh/Ni@SiO2-
1T. Prior to annealing, the dried microspheres were coated once
more in a similar manner. The resulting microspheres were
ꢀ
measurement, the sample was out-gassed at 300 C for 24 h in
ꢀ
annealed at 750 C for 20 h under a Ar/H2 (95 : 5) atmosphere.
vacuo to completely remove the impurities. To investigate the
elemental compositions, X-ray photoelectron spectroscopy
(XPS; Theta probe AR-XPS System, Thermo Fisher Scientic, UK)
analysis using a monochromated Al Ka X-ray source (hv ¼
1486.6 eV) was performed at the Korea Basic Science Institute
(KBSI) in Busan. Field emission scanning electron microscopy
(FE-SEM) was performed using a Hitachi S-4700 microscope at
KBSI in Jeonju. The contents of Ni and Rh in Rh/Ni@SiO2 were
analyzed using inductively coupled plasma atomic emission
spectroscopy (ICP-AES, JY Ultima2C) at KBSI in Seoul.
These annealed microspheres are denoted as Rh/Ni@SiO2-2T.
Similarly, a Rh/Ni@SiO2-3T sample was prepared by repeating
the procedure three times.
Hydrogen production from hydrous hydrazine
The evolution of hydrogen from hydrazine was examined using
a two-necked round-bottom ask at room temperature; one
neck was connected to a gas burette and the other neck was
used to inject a solution of hydrazine hydrate. The catalyst
(Rh/Ni@SiO2, 50 mg) was loaded into the ask and dispersed in
5.0 mL of deionized water containing 1 mL of 0.5 M NaOH
under an Ar atmosphere. Hydrazine hydrate (0.1032 g, 2.06
mmol) was then added to the catalyst via a syringe. Gas evolu-
tion was observed immediately. The gases released from the
solution rst passed through a trap containing sulfuric acid
(0.1 M); this trap absorbed ammonia gas, which might be
generated by the reaction. The amounts of passed gases were
volumetrically measured using a gas burette. To determine the
amounts of NH3 and N2H4 le in the ask aer the reaction, the
indophenol-blue method was employed. Two solutions were
prepared to perform this test. One solution comprised of 0.5 g of
phenol and 2.83 mg of Na2[Fe(CN)5NO] $2H2O in 50 mL of
deionized water and the other comprised of 0.251 g of NaOH
and 0.15 mL of NaOCl in 50 mL of deionized water. The pH of
the solution that remained in the round-bottom ask was
adjusted to 7.0 using a 0.1 M H2SO4 solution. Then, 1 mL of the
pH-adjusted solution and 5 mL of each of the two solutions was
Results and discussion
Our strategy to prepare hollow microspheres containing nickel
hydroxide and rhodium species is illustrated in Fig. 1; this
gure schematically represents the process for the synthesis of
three different materials starting from core–shell SiO2 micro-
spheres: (a) yolk–shell microspheres with large yolks and thin
shells, (b) yolk–shell microspheres with small yolks and thick
shells, and (c) hollow microspheres with very thick shells. Our
Fig. 1 Schematic representation of the preparation of hollow micro-
spheres via the dissolution-and-deposition pathway using core–shell
mixed, and the resultant solution was stirred thoroughly for 1 h. SiO2 microspheres as templates: (a) Rh/Ni(OH)2@SiO2-1T, (b) Rh/
Ni(OH)2@SiO2-2T, and (c) Rh/Ni(OH)2@SiO2-3T. The cores and shells
of the parent SiO2 template are depicted in dark and light blue,
respectively. Rh NPs are represented as red dots. The shells of (a), (b),
and (c) are depicted in yellow and blue, which represent Ni(OH)2 and
SiO2, respectively. The synthesis of Rh/Ni(OH)2@SiO2-3T involves
three dissolution-and-deposition steps.
The absorbance of the solution was measured and the total
concentration was determined using
a calibration curve
obtained from a standard solution.54,55 In a similar manner, the
amounts of NH3 and N2H4 in the H2SO4 trap were also
determined.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18929–18937 | 18931

View Article Online
Journal of Materials Chemistry A
Paper
method relies on the thermal hydrolysis of urea to produce a
hollow structure. The urea solution containing Rh and Ni
precursors readily inltrates into the silica template via meso-
ꢀ
pores in the shell. At 80 C, urea dissociates into ammonium
cations (NH4 ) and hydroxide anions (OHꢃ) according to the
+
+
following equation: NH2CONH2 + 3H2O / CO2 + 2NH4
+
2OHꢃ. The hydroxide anions generated from thermal hydrolysis
of urea drive the ionization and dissolution processes on the
SiO2 surface. Deposition of Ni2+ and Rh3+ species appears to
occur through electrostatic interactions on the ionized surface
of SiO2. Metal species are presumably deposited on the silica
shell in their hydroxide forms. Simultaneously, the silica cores
gradually dissolve and form the inside of the silica shell.
Repetition of this dissolution-and-deposition process eventually
leads to the formation of hollow microspheres, which are
composed of Ni(OH)2, SiO2, and Rh species. An important
benet of this method is that bimetallic hollow materials can
be prepared simply by controlling the dissolution-and-deposi-
tion step.
Mavredaki et al. demonstrated that hydroxide ions promote
the self-condensation of silicic acid in the pH range of 5–10 to
yield dissolved silicate species.56 Furthermore, this self-
condensation process becomes more dominant when metal-
hydroxide moieties are present on the SiO2 surface. Tomiyama
et al. also reported that nickel hydroxide particles on silica were
prepared through deposition-and-precipitation processes.57
These reported methods are straightforward; however, it is
difficult to use them to achieve the desired shape and func-
tionality. Our approach offers simple and convenient synthesis
of hollow structures containing active metal species on the
silica surface in a controlled manner. It is worth noting that the
Rh/Ni(OH)2@SiO2 microspheres cannot be obtained using core
SiO2 microspheres with non-porous shells as templates; these
microsphere templates yielded SiO2 microspheres with irreg-
ular deposition of the metal species (Fig. S1†). These results
clearly suggest that the porous shell also plays a crucial role in
the dissolution-and-deposition processes. The cavity size and
surface area can be tuned by adjusting the concentration of urea
and heating temperature.
Typical TEM and SEM images of core–shell SiO2, Rh/
Ni(OH)2@SiO2-1T, Rh/Ni(OH)2@SiO2-2T, and Rh/Ni(OH)2@
SiO2-3T are shown in Fig. 2 and demonstrate the distinctive
changes in surface morphology aer thermal hydrolysis of urea.
The core–shell silica templates (Fig. 2(a)) had average diameters
of 300 nm with shell thicknesses ranging from 20 to 30 nm.
Fig. 2(b) shows a TEM image of Rh/Ni(OH)2@SiO2-1T, which is
the rst dissolution-and-deposition product; this image clearly
illustrates the formation of void space between the shell and
core, which suggests that thermal hydrolysis of urea resulted in
the slow dissolution of silica. Interestingly, the average diam-
eter of Rh/Ni(OH)2@SiO2-1T is slightly larger at ꢁ340 nm. The
void space increased in the second dissolution-and-deposition
product, Rh/Ni(OH)2@SiO2-2T, which had a clear yolk–shell
structure. The TEM images shown in Fig. 2(c) reveal that the
diameter of the yolk (core) was ꢁ120 nm and the shell thickness
drastically increased to about 60–80 nm; this suggests that the
core silica gradually dissolved while the shell expanded through
Fig. 2 Representative TEM images of (a) core–shell SiO2, (b) Rh/
Ni(OH)2@SiO2-1T, (c) Rh/Ni(OH)2@SiO2-2T, and (d) Rh/Ni(OH)2@SiO2-
3T are displayed in the top panels. The bottom panels show SEM
images of the corresponding samples, i.e., (e) core–shell SiO2, (f) Rh/
Ni(OH)2@SiO2-1T, (g) Rh/Ni(OH)2@SiO2-2T, and (h) Rh/Ni(OH)2@SiO2-
3T. The insets show higher magnifications of the TEM and SEM images.
the deposition of dissolved silica as well as the Ni(OH)2 coating.
As illustrated in Fig. 2(d), hollow microspheres with shell
thicknesses of ꢁ120 nm formed aer three repeated reactions;
this clearly demonstrates that the core is completely dissolved.
The overall diameters of Rh/Ni(OH)2@SiO2-3T ranged from 370
to 390 nm and were larger than that of the core–shell silica
template. The TEM images suggest that dissolved silica accu-
mulated inside the silica shell while the metal components were
coated on the outer shell. This is in good agreement with the
increment in diameter of the hollow microspheres.
Representative SEM images of the same samples are shown
in the bottom panel of Fig. 2. The overall diameters of the
samples were virtually identical to those estimated from the
TEM images. The SEM images illustrate that the surface had a
unique ower-like morphology. The ower-like surface
becomes more distinctive with increasing number of dissolu-
tion-and-deposition steps. It has been reported that Ni(OH)2
grows in ower-like architectures in strongly alkaline solution.
It is typically crystalline and shows distinctive XRD patterns
corresponding to a- or b-phase Ni(OH)2. Interestingly, the XRD
patterns of the as-prepared samples did not show any distinc-
tive peaks (Fig. S2†), which suggests that all the phases in Rh/
Ni(OH)2@SiO2 were amorphous. Thus far, an amorphous
Ni(OH)2 phase with a ower-like morphology has not been
reported. This amorphous nature suggests that the dissolved
silica might interfere with the formation of a long-range
ordered crystalline Ni(OH)2 phase. Instead, Ni(OH)2 petals grew
along with dissolved silica species on the shell, which resulted
in the formation of disordered amorphous Ni(OH)2. To inves-
tigate the elemental distribution and conrm the hollow
structure, EDX line scan elemental proles of the three Rh/
Ni(OH)2@SiO2 samples were obtained as given in Fig. 3. The
results conrm the presence of Ni, Si, and Rh, although the
signal intensity of Rh is very low due to the minute amount of
Rh in the sample. The EDX scan data clearly show that the
intensity of Ni increases with increasing dissolution-and-depo-
sition time while the intensity of Si decreases. In particular, the
intensity of Si in the core is drastically reduced, conrming that
the sample had a hollow structure. A notable feature in the line-
18932 | J. Mater. Chem. A, 2014, 2, 18929–18937
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Fig.
3 EDX elemental line scans of (a) Rh/Ni(OH)2@SiO2-1T, (b)
Fig. 4 The top panel shows TEM images of (a) Rh/Ni@SiO2-1T, (b)
Rh/Ni@SiO2-2T, and (c) Rh/Ni@SiO2-3T after heating the as-prepared
samples at 750 ꢀC under flowing 5% H2 in an Ar atmosphere. The
corresponding SEM images are illustrated in (d) Rh/Ni@SiO2-1T, (e)
Rh/Ni@SiO2-2T, and (f) Rh/Ni@SiO2-3T. The insets show higher
magnifications.
Rh/Ni(OH)2@SiO2-2T, and (c) Rh/Ni(OH)2@SiO2-3T. The top panels
show the TEM images that include the positions of the line scan (yellow
line). The bottom panel shows the EDS signal intensities of Ni, Si, and
Rh across the diameter of the corresponding microsphere.
scan curves of Rh/Ni(OH)2@SiO2-3T is that the intensity of Ni is
stronger than that of Si, which suggests that Ni(OH)2 was identical shell thickness. However, annealing caused the ower-
predominantly coated on the outer shell while SiO2 was mostly like surface morphology to disintegrate and large quasi-spher-
deposited on the inner shell. The EDX line-scan data are ical agglomerates to grow along with the coated Ni layers. Thus,
consistent with the TEM and SEM images shown in Fig. 2. The annealing transformed the ower-like Ni(OH)2 phases into large
presence of Rh was not clear from the line-scan data but was metallic Ni granules and layers. This transformation process is
conrmed by ICP-AES and EDX elemental analyses; this quite similar to the melt-grown agglomeration procedure. The
ꢀ
suggests that Rh species were embedded in the shell as tiny Ni(OH)2 phase deposited on the outer shell melts at ꢁ230 C
NPs. As evident from Table S1,† the two techniques yielded and the melted Ni particles initially coat on the surface. Some of
slightly different results because of their different measurement the Ni phases agglomerate to form granules on the surface. In
methods. We used the ICP-AES analysis data to evaluate the contrast, the extent of coarsening of the Rh species was limited,
catalytic activity. With increasing dissolution-and-deposition which is mainly because of their high melt temperature. The
time, the Rh content increased and was roughly proportional to SEM images show that large Ni granules with spherical shapes
the nominal composition. The maximum amount of Rh in protruded from the Ni surface (Fig. 4).
Rh/Ni(OH)2@SiO2-3T was 6.35 wt%. In addition, the presence of
Rh and the hollow structure were also evidenced by elemental characterized by XRD. As illustrated in Fig. 5, XRD patterns of
mapping images shown in Fig. S7.† the three Rh/Ni@SiO2 samples show two distinctive peaks at
The structures of the bimetallic phases in the shell were
To obtain crystalline bimetallic Rh/Ni phases, the Rh/ 44.2 and 52.4ꢀ, which are typical of the Ni phase (space group:
Ni(OH)2@SiO2 samples were annealed at 750 ꢀC for 20 h under Fm3m) with a face-centered cubic structure (JCPDS no. 70-
owing 5% H2 in an Ar atmosphere, which converted them into 1849).60–63 The two peaks were assigned to the (111) and (200)
hollow Rh/Ni@SiO2 microspheres containing Ni and Rh NPs. As reections. A very broad peak near 20ꢀ was assigned to SiO2
illustrated in the TEM images (Fig. 4), the parent hollow (JCPDS no. 29-0085). No diffraction peaks corresponding to Rh
structures were maintained even aer heating at high temper- appeared even in Rh/Ni@SiO2-3T, which indicates that the
atures. Usually, annealing of hollow spheres composed of metal amount of Rh in the sample was too small to be detected by
components at high temperatures leads to collapse or shrinkage XRD.64,65 Another plausible scenario is that Rh alloyed with Ni to
of the parent hollow structure.58,59 However, in the case of Rh/ yield a Ni-rich bimetallic phase. The XRD data provide indirect
Ni(OH)2@SiO2, the hollow shape was retained throughout the evidence of the formation of bimetallic NPs. Namely, a part of
heat treatment because of the presence of silica layers in the the Rh metal was incorporated in the Ni metal to form alloy NPs
inner shell, which appear to play a crucial role in maintaining while the remainder was present in the shell as sub-nanometer
the hollow structure. In addition, the silica layers act as sup- Rh NPs. This conjecture is also supported by the XRD patterns
porting materials for the formation of the Ni layers. During the of Rh@SiO2, which is shown in Fig. S6.†
heating process, small amounts of the Rh species in the shells
The oxidation states and chemical environments of Rh, Ni,
were deposited onto the abundant Ni surface to form bimetallic Si, and O in Rh/Ni@SiO2-1T, Rh/Ni@SiO2-2T, and Rh/Ni@SiO2-
NPs. Relative to Rh/Ni(OH)2@SiO2-3T, the diameter of Rh/ 3T were further characterized by XPS. Fig. S3† shows detailed
Ni@SiO2-3T remained almost constant at 370 nm and the shell XPS spectra of the important regions (i.e., Rh 3d, Ni 2p, Si 2p,
thickness ranged from 100 to 120 nm. Accordingly, the and O 1s) for each element. All species of interest are displayed
annealed sample retained the hollow structure with a virtually in the XPS spectra. The Rh 3d signals are observed at 306 and
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18929–18937 | 18933

View Article Online
Journal of Materials Chemistry A
Paper
illustrate the selective production of H2 and N2. To determine
the efficiency of H2 production using the bimetallic catalyst, we
reacted hydrous hydrazine (0.404 M) with the catalyst at 25 ꢀC in
a round-bottom ask and measured the production of H2
volumetrically using a gas burette. To ensure that the catalyst
surface was basic and to prevent the production of unwanted
NH3 gas, NaOH (0.5 M) was added to the hydrazine solution.
When Rh/Ni@SiO2-1T was used as a catalyst, gases evolved for
approximately 3.3 h and the amount of gases released over the
course of the reaction was 2.77 equivalents (equiv.), as illus-
trated in Fig. 6. This corresponds to a H2 selectivity of 91.3%.
The turnover number (TON) of Rh/Ni@SiO2-1T was about 143,
which was calculated on the basis of the Rh content. Under the
same reaction conditions, the second catalyst (Rh/Ni@SiO2-2T)
yielded 2.78 equiv. of gases in 1.6 h and had a 91.4% H2 selec-
tivity (TON ¼ 90). The catalytic activity of Rh/Ni@SiO2-3T
towards the same reaction was also tested. Almost complete
conversion of hydrazine was achieved in 1.5 h, yielding 2.99
equiv. of gases (TON ¼ 66). The selectivity of Rh/Ni@SiO2-3T
toward H2 production was remarkably high (99.4%). These
results suggest that increasing the Rh content in the Rh/
Ni@SiO2 catalyst increases the H2 selectivity as well as the yield.
The optimal amount of Rh in the Ni-coated hollow catalyst was
ꢁ6.36 wt%, which is remarkably low compared with the values
Fig. 5 Powder XRD patterns of (a) Rh/Ni@SiO2-1T, (b) Rh/Ni@SiO2-2T,
and (c) Rh/Ni@SiO2-3T. Theoretical XRD patterns of Ni and Rh are also
displayed for reference.
310 eV, which were assigned to Rh 3d5/2 and Rh 3d3/2, respec- reported previously.36 In a recent paper by Singh and Xu, it was
tively. The two Rh 3d signals in Rh/Ni@SiO2-3T are substantially demonstrated that the H2 selectivity strongly depends on the
more intense than those of Rh/Ni@SiO2-1T, suggesting that the Rh/Ni ratio in RhxNiy alloys, and the Rh4Ni catalyst with 87.5
Rh/Ni@SiO2-3T sample contained more Rh. Assignments of the wt% Rh reaches a maximum H2 selectivity close to 100%. Thus,
Ni 2p peaks are complicated due to the presence of several Ni Rh/Ni@SiO2-3T could be the catalyst with the lowest Rh content
species in different chemical environments. Four peaks are for the selective decomposition of hydrous hydrazine. Its high
displayed at ꢁ857.2, 861.1, 874.8 and 879.7 eV. On the basis of selectivity at low Rh loadings is unprecedented in RhNi alloys.
previous reports on Ni metal, the two peaks at 857.2 and 874.8
For comparison, monometallic catalysts that contain only Ni
eV were assigned to Ni 2p3/2 and Ni 2p1/2, respectively.66–73 The or Rh, such as Ni@SiO2 and Rh@SiO2, were used as catalysts
two broad remaining peaks are presumably associated with NiO under the same conditions. As illustrated in Fig. 7, no gas
or could be ascribed to the satellite peak. The presence of products were obtained when Ni@SiO2 was used as a catalyst.
characteristic Ni 2p signals in the Ni 2p XPS spectra indicates Even over Rh@SiO2 catalysts, a low yield and medium selectivity
that metallic Ni was coated on the outer shell and NiO formed of 1.3 equiv. and 42%, respectively, were observed. Moreover,
presumably at the boundary between Ni and SiO2. The binding the release of gases terminated aer about 6.3 h. These results
energies of Si 2p and O 1s are very close to the values reported
for a range of silicate species.74–76
Prior to testing the catalytic activities of Rh/Ni@SiO2-1T,
Rh/Ni@SiO2-2T, and Rh/Ni@SiO2-3T, we measured their N2
adsorption and desorption isotherms to estimate their surface
areas and pore sizes. Type-IV isotherm curves with distinctive
hysteresis loops were observed for all three samples (Fig. S4†),
which suggests the presence of mesopores in the shells.77–81 The
specic surface areas of the samples were very high and quite
similar, ranging from 172.63 to 180.21 m2 gꢃ1. However, the
pore-size distribution curves show that the pore sizes increased
substantially from 4.7 nm in Rh/Ni@SiO2-1T to 5.6 nm in Rh/
Ni@SiO2-3T. In addition, Rh/Ni@SiO2-3T had a considerable
number of pores that are 10 nm or larger. This suggests that a
greater variety of pores sizes were produced with increasing
number of dissolution-and-deposition steps.
Because of the high specic surface area of the bimetallic
Rh/Ni@SiO2 samples, they were expected to show great promise
Fig. 6 Production of H2 and N2 from hydrous hydrazine as a function
of time in the presence of NaOH (0.5 M) at 25 ꢀC using Rh/Ni@SiO2-1T
(black circles), Rh/Ni@SiO2-2T (blue triangles), and Rh/Ni@SiO2-3T
catalytically. Hydrous hydrazine was selected as a substrate to (red squares) catalysts.
18934 | J. Mater. Chem. A, 2014, 2, 18929–18937
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
For the second cycle, the reaction provided a virtually identical
yield of gases (2.99 equiv.). Even for the third cycle, the yield of
gases released was 2.99 equiv. although the reaction took
slightly longer (2.0 h) than the rst run. These results clearly
demonstrate that the bimetallic NPs supported on hollow
Ni-coated microspheres are highly resistant to deactivation in
alkaline solution and retain excellent catalytic activity for the
complete decomposition of hydrous hydrazine. Thus, hollow
Rh/Ni@SiO2-3T microspheres can be considered to be one of
the most promising catalysts for hydrogen production from
hydrous hydrazine in commercial applications.
Conclusions
Fig. 7 Production of H2 and N2 from hydrous hydrazine as a function
of time in the presence of NaOH (0.5 M) at 25 ꢀC using Ni@SiO2 (black
triangles), Rh@SiO2 (blue circles), and Rh/Ni@SiO2-3T (red squares)
catalysts.
In conclusion, we developed a simple and efficient methodology
for the synthesis of hollow Ni-coated microspheres
(Rh/Ni@SiO2) containing Rh NPs using core–shell silica
microspheres as templates. The Ni-coated hollow microsphere
with a low loading of Rh showed excellent catalytic performance
for complete conversion of NH2NH2 into H2 and N2 at room
temperature with over 99% H2 selectivity. Our study highlights a
method for the exclusive production of H2 from NH2NH2 using
a low-cost and recyclable catalyst under industrially realistic
conditions, which opens up the possibility of accelerating the
clearly demonstrate that the bimetallic alloy is crucial for high
yield and selectivity. The RhNi bimetallic species anchored on
the Ni surface appear to promote the adsorption of hydrous
hydrazine so that complete decomposition of hydrous hydra-
zine into H2 and N2 predominantly occurs. Thus, the bimetallic
hollow catalysts (Rh/Ni@SiO2) clearly show a synergetic effect
and superior catalytic activity compared to those of the mono-
metallic congeners. Table S2† presents an overview of recently
reported catalysts that efficiently produced hydrogen from
practical application of H2NNH2 as
material.
a hydrogen-storage
hydrous hydrazine. Compared with other Rh alloy catalysts Acknowledgements
(entries 3 and 9), the Rh/Ni@SiO2 catalyst shows excellent H2
selectivity even at 25 C despite the low Rh content.
This work was supported by the Korea CCS R&D Center Grant
funded by the Korea government (Ministry of Science, ICT &
Future Planning, 2013M1A8A1035853). BL thanks the Research
Fellow Program (2012R1A12043256).
ꢀ
Because of the presence of magnetic Ni layers in the
Rh/Ni@SiO2 microsphere, the catalyst was separated simply by
using a magnet aer completion of the reaction (Fig. S5†). A
remarkable advantage of these catalysts is that they can be
reused without loss of catalytic activity. The Rh/Ni@SiO2-3T
catalyst was recycled three times; the results are shown in Fig. 8.
Notes and references
1 Y. Xia, Z. Yang and Y. J. Zhu, J. Mater. Chem. A, 2013, 1, 9365.
2 K. Shimura and H. Yoshida, Energy Environ. Sci., 2011, 4,
2467.
¨
3 L. Schlapbach and A. Zuttel, Nature, 2001, 414, 353.
4 N. Armaroli and V. Balzani, ChemSusChem, 2011, 4, 21.
5 J. B. O. Santos, G. P. Valenca and J. A. J. Rodrigues, J. Catal.,
2002, 210, 1.
6 B. Sakintuna, F. Lamari-Darkrim and M. Hirscher, Int.
J. Hydrogen Energy, 2007, 32, 1121.
7 G. Moussa, R. Moury, U. Demirci, T. Sener and P. Miele, Int.
J. Energy Res., 2013, 37, 825.
8 D. Teichmann, W. Arlt, P. Wasserscheid and R. Freymann,
Energy Environ. Sci., 2011, 4, 2767.
9 L. Chong, J. Zou, X. Zeng and W. Ding, J. Mater. Chem. A,
2013, 1, 3983.
Fig. 8 Recycle test of the Rh/Ni@SiO2-3T catalyst: production of H2
and N2 from hydrous hydrazine as a function of time in the presence of
NaOH (0.5 M) at 25 ꢀC using a Rh/Ni@SiO2-3T catalyst. The red
squares, blue triangles, and black circles indicate the first, second, and
third cycles, respectively. The used catalysts were separated by using a
magnet after the reaction.
10 S. S. Muir and X. Yao, Int. J. Hydrogen Energy, 2011, 36, 5983.
11 J. Zhang, T. S. Fisher, J. P. Gore, D. Hazra and
P. V. Ramachandran, Int. J. Hydrogen Energy, 2006, 31, 2292.
12 H. Dai, Y. Liang, L. Ma and P. Wang, J. Phys. Chem. C, 2008,
112, 15886.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18929–18937 | 18935

View Article Online
Journal of Materials Chemistry A
Paper
13 H. I. Schlesinger, H. C. Brown, A. E. Finholt, J. R. Gilbreath, 44 C. Chen, L. K. Yee, H. Gong, Y. Zhang and R. Xu, Nanoscale,
H. R. Hoekstra and E. K. Hyde, J. Am. Chem. Soc., 1953, 75,
215.
14 B. H. Liu and Z. P. Li, J. Power Sources, 2009, 187, 527.
2013, 5, 4314.
45 W. Di, S. K. P. Velu, A. Lascialfari, C. Liu, N. Pinna, P. Arosio,
Y. Sakka and W. Qin, J. Mater. Chem., 2012, 22, 20641.
15 M. J. F. Ferreira, L. Gales, V. R. Fernandes, C. M. Rangel and 46 Z. Xu, Y. Gao, T. Liu, L. Wang, S. Bian and J. Lin, J. Mater.
A. M. F. R. Pinto, Int. J. Hydrogen Energy, 2010, 35, 9869. Chem., 2012, 22, 21695.
16 S. J. Cho, J. Lee, Y. S. Lee and D. P. Kim, Catal. Lett., 2006, 47 Z. H. Zhang, L. Chen, Q. H. Gan, Q. Tian, C. L. Chen and
109, 181.
Z. L. Huang, Phys. Status Solidi A, 2013, 210, 378.
48 H. Qin, X. Zhang, H. Liu, Y. Sang and J. Wang,
CrystEngComm, 2013, 15, 5076.
17 S. K. Singh and Q. Xu, Catal. Sci. Technol., 2013, 3, 1889.
18 H. Jiang, S. K. Singh, J. Yan, X. Zhang and Q. Xu,
ChemSusChem, 2010, 3, 541.
49 L. Su, C. Ma, T. Hou and W. Han, RSC Adv., 2013, 3, 19807.
19 D. G. Tong, W. Chu, P. Wu, G. F. Gu and L. Zhang, J. Mater. 50 P. Burattin, M. Che and C. Louis, J. Phys. Chem. B, 1997, 101,
Chem. A, 2013, 1, 358. 7060.
20 E. W. Schmidt, Hydrazine and its Derivatives: Preparation, 51 M. Khoudiakov, M. C. Gupta and S. Deevi, Appl. Catal., A,
Properties, Applications, John Wiley and Sons, Inc., 2nd
edn, 2001, vol. 1–2.
2005, 291, 151.
52 J. A. Schwarz, Chem. Rev., 1995, 95, 477.
¨
21 L. He, Y. Huang, X. Y. Liu, L. Li, A. Wang, X. Wang, C. Mou 53 W. Stober and A. Fink, J. Colloid Interface Sci., 1968, 26, 62.
and T. Zhang, Appl. Catal., B, 2014, 147, 779.
22 K. V. Manukyan, A. Cross, S. Rouvimov, J. Miller,
54 T. T. Ngo, A. P. H. Phan, C. F. Yam and H. M. Lenhoff, Anal.
Chem., 1982, 54, 46.
´
A. S. Mukasyan and E. E. Wolf, Appl. Catal., A, 2014, 476, 47. 55 P. Moreno, E. Sanchez, A. Pons and A. Palou, Anal. Chem.,
23 T. Umegaki, J. Yan, X. Zhang, H. Shioyama, N. Kuriyama and
Q. Xu, Int. J. Hydrogen Energy, 2009, 34, 2303.
24 A. K. Singh and Q. Xu, Int. J. Hydrogen Energy, 2014, 39, 9128.
1986, 58, 585.
56 E. Mavredaki, E. Neofotistou and K. D. Demadis, Ind. Eng.
Chem. Res., 2005, 44, 7019.
25 L. He, Y. Huang, A. Wang, X. Wang, X. Chen, J. J. Delgado 57 S. Tomiyama, R. Takahashi, S. Sato, T. Sodesawa and
and T. Zhang, Angew. Chem., Int. Ed., 2012, 51, 6191. S. Yoshida, Appl. Catal., A, 2003, 241, 349.
26 L. He, Y. Huang, A. Wang, X. Wang and T. Zhang, AIChE J., 58 R. Nakamura and H. Nakajima, J. Phys.: Conf. Ser., 2009, 165,
2013, 59, 4297. 012072.
27 L. He, Y. Huang, A. Wang, Y. Liu, X. Liu, X. Chen, 59 R. Nakamura, D. Tokozakura, J. Lee, H. Mori and
J. J. Delgado, X. Wang and T. Zhang, J. Catal., 2013, 298, 1.
28 M. Yadav and Q. Xu, Energy. Environ. Sci., 2012, 5, 9698.
29 S. K. Singh, X. Zhang and Q. Xu, J. Am. Chem. Soc., 2009, 131,
9894.
30 S. K. Singh and Q. Xu, Chem. Commun., 2010, 46, 6545.
31 S. K. Singh, Z. Lu and Q. Xu, Eur. J. Inorg. Chem., 2011, 2011,
2232.
32 J. Zhang, Q. Kang, Z. Yang, H. Dai, D. Zhuang and P. Wang, J.
Mater. Chem. A, 2013, 1, 11623.
33 S. K. Singh, Y. Iizuka and Q. Xu, Int. J. Hydrogen Energy, 2011,
36, 11794.
H. Nakajima, Acta Mater., 2008, 56, 5276.
60 J. Gao, C. Jia, M. Zhang, F. Gu, G. Xu and F. Su, Catal. Sci.
Technol., 2013, 3, 2009.
61 A. Jangam and S. Kawi, ACS. Catal., 2014, 4, 289.
62 S. Wang, X. Guo, H. Yang, J. Dai, R. Zhu, J. Gong, L. Peng and
W. Ding, Appl. Surf. Sci., 2014, 288, 530.
63 Y. Luo, J. Zhang, Y. Shen, S. Jiang and G. Liu, J. Mater. Sci.
Technol., 2007, 5, 587.
64 H. Jiang, T. Umegaki, T. Akita, X. Zhang, M. Haruta and
Q. Xu, Chem.–Eur. J., 2010, 16, 3132.
65 K. S. Han and O. H. Han, Bull. Korean Chem. Soc., 2006, 26,
1121.
66 Y. Hou and S. Gao, J. Alloys Compd., 2004, 365, 112.
34 S. K. Singh, A. K. Singh, K. Aranishi and Q. Xu, J. Am. Chem.
Soc., 2011, 133, 19638.
35 D. G. Tong, D. M. Tang, W. Chu, G. F. Gu and P. Wu, J. Mater. 67 F. Liu, J. Y. Lee and W. Zhou, J. Phys. Chem. B, 2004, 108,
Chem. A, 2013, 1, 6425.
17959.
36 S. K. Singh and Q. Xu, J. Am. Chem. Soc., 2009, 131, 18032.
37 A. K. Singh, M. Yadav, K. Aranishi and Q. Xu, Int. J. Hydrogen
Energy, 2012, 37, 18915.
38 J. Wang, X. Zhang, Z. Wang, L. Wang and Y. Zhang, Energy.
Environ. Sci., 2012, 5, 6885.
39 C. Yan and D. Xue, J. Phys. Chem. B, 2006, 110, 7102.
40 X. W. Lou, L. A. Archer and Z. Yang, Adv. Mater., 2008, 20,
3987.
68 A. P. Grosvenor, M. C. Biesinger, R. St. C. Smart and
N. S. McIntyre, Surf. Sci., 2006, 600, 1771.
69 L. Xia, X. Hu, X. Kang, H. Zhao, M. Sun and X. Cihen,
Colloids Surf., A, 2010, 367, 96.
70 H. B. Li, M. H. Yu, F. X. Wang, P. Liu, Y. Liang, J. Xiao,
C. X. Wang, Y. X. Tong and G. W. Yang, Nat. Commun.,
2013, 4, 1894.
71 L. Mi, Q. Ding, W. Chen, L. Zhao, H. Hou, C. Liu, C. Shen and
Z. Zheng, Dalton Trans., 2013, 42, 5724.
41 X. Gong, L. Wang and W. Wen, Chem. Commun., 2009, 4690.
42 J. B. Yoo, H. J. Yoo, B. W. Lim, K. H. Lee, M. H. Kim, D. Kang 72 Z. Xu, J. Yu, G. Liu, B. Cheng, P. Zhou and X. Li, Dalton
and N. H. Hur, ChemSusChem, 2012, 5, 2334. Trans., 2013, 42, 10190.
43 E. Hemmer, N. Venkatachalam, H. Hyodo, A. Hattori, 73 J. Yang, X. Duan, Q. Qin and W. Zheng, J. Mater. Chem. A,
Y. Ebina, H. Kishimoto and K. Soga, Nanoscale, 2013, 5,
2013, 1, 7880.
11339.
18936 | J. Mater. Chem. A, 2014, 2, 18929–18937
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
74 G. He, G. W. Meng, L. D. Zhang and M. Liu, Appl. Phys. Lett., 78 J. Li, J. Lin, X. Xu, X. Zhang, Y. Xue, J. Mi, Z. Mo, Y. Fan,
2007, 91, 232910.
L. Hu, X. Yang, J. Zhang, F. Meng, S. Yuan and C. Tang,
Nanotechnology, 2013, 24, 155603.
79 R. Jia, J. Chen, J. Zhao, J. Zheng, C. Song, L. Li and Z. Zhu, J.
Mater. Chem., 2010, 20, 10829.
75 S. N. Arafat, S. Dutta, M. Perring, M. Mitchell, P. J. A. Kenis
and N. B. Bowden, Chem. Commun., 2005, 3198.
76 T. D. Shen, I. Shmagin, C. C. Koch, R. M. Kolbas, Y. Fahmy,
L. Bergman, R. J. Nemanich, M. T. McClure, Z. Sitar and 80 Y. Liang, D. Wu and R. Fu, Sci. Rep., 2013, 3, 1119.
M. X. Quan, Phys. Rev. B: Condens. Matter Mater. Phys., 81 S. Feng, W. Li, Q. Shi, Y. Li, J. Chen, Y. Ling, A. M. Asiri and
1997, 55, 7615.
D. Zhao, Chem. Commun., 2014, 50, 329.
77 H. Lin, S. Wong, C. Mou and C. Tang, J. Phys. Chem. B, 2000,
104, 8967.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18929–18937 | 18937