Synthesis of novel nanotubular mesoporous nickel phosphates with high performance in epoxidation

Article abstract of DOI:10.1039/b805843a

Two novel mesoporous nickel phosphates (NiPO-1 and NiPO-2) with nanotubular structures were developed in the presence of cationic surfactant and different bases by a sol-gel method. The mesostructures of NiPO-1 and NiPO-2 are composed of nanotubes of different lengths, which are arranged by a less-ordered lamellar symmetry as synthesized. ICP elemental analysis revealed that they are different compounds. The Ni/P molar ratios of NiPO-1 and NiPO-2 are 1.57 and 1.20, respectively. Both materials possess a relatively high BET surface area (205-292 m² g^-1) and good thermal stability up to 673 K. It is found that these novel materials exhibited high activity (>50%) and high selectivity to epoxide (95.6% and 99.0%) in the epoxidation of cyclododecene with H₂O₂ as oxidant.

Full text of DOI:10.1039/b805843a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Synthesis of novel nanotubular mesoporous nickel phosphates with high
performance in epoxidation†
a
a
a
Jing Yu, Anjie Wang, Juan Tan, Xiang Li,a Jeroen Anton van Bokhovenb and Yongkong Hua
*
*
Received 7th April 2008, Accepted 27th May 2008
First published as an Advance Article on the web 1st July 2008
DOI: 10.1039/b805843a
Two novel mesoporous nickel phosphates (NiPO-1 and NiPO-2) with nanotubular structures were
developed in the presence of cationic surfactant and different bases by a sol–gel method. The
mesostructures of NiPO-1 and NiPO-2 are composed of nanotubes of different lengths, which are
arranged by a less-ordered lamellar symmetry as synthesized. ICP elemental analysis revealed that they
are different compounds. The Ni/P molar ratios of NiPO-1 and NiPO-2 are 1.57 and 1.20, respectively.
Both materials possess a relatively high BET surface area (205–292 m2 gꢀ1) and good thermal stability
up to 673 K. It is found that these novel materials exhibited high activity (>50%) and high selectivity to
epoxide (95.6% and 99.0%) in the epoxidation of cyclododecene with H2O2 as oxidant.
catalytic reactions of bulky cyclododecene, due to diffusion
Introduction
limitation. Naturally, nickel phosphates with mesopores were
Since the first reports of the M41S family of ordered mesoporous
expected to present interesting and novel catalytic properties in
silica by Mobil researchers,1 considerable efforts have been made
this kind of reaction for bulky molecules. However, the synthesis
to extend the range of the elemental composition of mesoporous
of mesoporous nickel phosphates is difficult. Up till now, only
materials. The synthesis of mesoporous materials which contain
one related material has been reported—an agglomerate of fine
transition metals has drawn much attention due to their potential
primary nickel phosphate particles with some mesoporosity
applications in acid and redox catalysis, or in photocatalytic
(BET surface area 50–130 m2 gꢀ1 and pore size around 2–4 nm).13
processes.2 Besides the transition metal-substituted mesoporous
One of the main difficulties in the synthesis of mesoporous nickel
silicates,3 a variety of mesostructured transition metal (Ti, Zr,
phosphates lies in the differences between NiO6 and PO4 in bond
Nb, W, and Sn) oxides4 and transition metal (Zr,5 Ti,6 Fe,7 Sn,8
lengths and angles.14 Moreover, nickel and phosphorus species
and Nb9) phosphates have been reported in the past several
readily combine to form dense precipitates.11
years.
We report, for the first time, the synthesis of two mesoporous
From the viewpoint of heterogeneous catalysis, transition
nickel phosphates (denoted NiPO-1 and NiPO-2) using a sol–
metal-containing mesoporous materials favorably combine the
gel method in the presence of cationic template cetyltrimethyl-
redox catalytic properties of metal species with easy diffusion of
ammonium bromide (CTAB). Different bases, aqueous ammonia
reactant molecules. One of the most commonly concerned
and tetramethylammonium hydroxide (TMAOH) were used to
reactions is catalytic liquid-phase oxidation with H2O2 or
control the pH and assist the assembly of surfactant and
organic peroxide as oxidant. However, significant leaching of
inorganic species. The obtained mesoporous nickel phosphate
active metal components from the material into the liquid phase
materials were characterized by means of XRD, TEM, ICP, N2
remains a serious problem which limits the possible applications
adsorption, FT-IR and UV-Vis. They were evaluated in the
of these materials. The leaching is supposed to be generally
catalytic epoxidation of cyclohexene and cyclododecene with
a result of solvolysis of metal–oxygen bonds attached to the
H2O2 aqueous solution under mild conditions.
structures by polar molecules.3,10 Consequently, higher stability
of transition metals has been one of the main aims in the
synthesis of transition metal-containing mesoporous materials.
Recently, Guillou et al.11 have reported a zeolitic nickel
Experimental
Synthesis
phosphate, VSB-5, which exhibited high catalytic activities and
no leaching of nickel species was observed in epoxidation of
An aqueous solution of CTAB (25 wt.%) was mixed with phos-
cyclic olefins (cyclohexene and cyclooctene).12 However, the
phoric acid at 333 K to form an acidified surfactant solution,
narrow pore channels of this material led to low conversion in
which was aged for 1 h. Then the pH of the mixture was adjusted
to 8–10 with aqueous ammonia (25 wt.%) or TMAOH (25 wt.%).
Ni(NO3)2ꢁ6H2O was dissolved in 10 mL de-ionized water, and
aState Key Laboratory of Fine Chemicals, Dalian University of
Technology, 158 Zhongshan Road, Dalian, 116012, P.R. China. E-mail:
ajwang@dlut.edu.cn; tanjuan@dlut.edu.cn; Fax: +86 411 88993693; Tel:
+86 411 88993693
bInstitute of Chemical and Bioengineering, ETH Zurich, Wolfgang-Pauli-
Strasse 10, 8093 Zurich, Switzerland
added to the above solution to form a starting gel of molar
composition 1.0 P : 0.7 Ni : (0.5–1.0) CTAB : (2.1–2.4)
NH3$H2O/(1.7–2.1) TMAOH : (150–250) H2O. The gel was
heated and kept at 373 K for 24 h. The resulting solid products
were collected through filtration, washed first with water and
subsequently with ethanol, and dried at 373 K for 12 h. The
† Electronic supplementary information (ESI) available: XRD patterns
of mesoporous NiPO materials. See DOI: 10.1039/b805843a
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 3601–3607 | 3601

View Article Online
samples synthesized in the presence of aqueous ammonia and
TMAOH were nominated as NiPO-1 and NiPO-2, respectively.
To remove the surfactants, the samples were extracted with
a 0.15 mol Lꢀ1 CH3COONa–EtOH solution (100 mL solution for
1 g sample) at room temperature for 3–8 h, followed by washing
and drying at 343 K for 12 h.
Characterization
Powder X-ray diffraction (XRD) patterns were recorded on
a Rigaku D/max-2400 diffractometer with Cu-Ka radiation.
Transmission electron micrographs (TEM) were obtained with
a Tecnai G220 S-Twin microscope operating at an accelerating
voltage of 200 kV. Nitrogen adsorption and desorption experi-
ments were carried out at 77 K on a Micromeritics ASAP 2010
analyzer. Prior to the measurements, the samples were outgassed
at 413 K for over 4 h. The specific surface areas were calculated
using the BET (Brunauer, Emmett, and Teller) equation. Pore
size distributions were calculated by the BJH (Barrett, Joyner,
and Halender) method using the desorption branches of the
isotherms. Bulk chemical analyses for the as-synthesized samples
were determined by an inductively coupled plasma analyzer
(ICP, Therm Jarrell Ash IRIS/AP) after completed solubilization
of the samples in HCl. Fourier transform infrared spectra
(FT-IR) were obtained on a Bruker IFS 120 FT-IR spectrometer
using ca. 0.5 mm KBr pellets containing 2.5 wt.% sample.
UV-Vis diffuse reflectance spectra were recorded on a Shimadzu
UV-2501 PC spectrophotometer, using BaSO4 as a reference.
UV-Vis absorption spectrum of Ni(NO3)2$6H2O solution was
determined on an HP8453-Shimadzu UV3100 spectrophoto-
meter equipped with a quartz flat cell.
Fig. 1 XRD patterns of as-synthesized NiPO-1 (a) and NiPO-2 (b).
broadening of the three-line XRD patterns suggest that the two
materials have less-ordered lamellar symmetry.15 We found that
no diffraction peaks appeared in the high angle XRD patterns
(not shown) for both materials, which means that their frame-
works are amorphous in nature.
Fig. 2 shows the TEM images of NiPO-1 and NiPO-2. The
appearances of the two NiPOs in TEM views are unexpectedly
inconsistent with their lamellar symmetry revealed by the XRD
patterns. NiPO-1 (Fig. 2a, b) appears as nanotubes piled
together, and the tube walls can be observed as darker edges.
Most nanotubes are 200 to 400 nm in length, 4 to 5 nm in
diameter and randomly piled, while some are gathered into
bundles and extend roughly in the same directions (as shown by
the circle in Fig. 2a). For NiPO-2, disordered mesopores and
randomly piled short nanotubes were observed (Fig. 2c, d). In
some regions where the piling is not compact (as shown by the
circle in Fig. 2c), the short nanotubes present a worm-like
appearance, with a diameter of 4–5 nm and a length/diameter
ratio of 3–6. We noticed that the average diameter of the nano-
tubes observed under TEM is close to the d100 spacing measured
by the XRD analysis. To our knowledge, this unique nano-
tubular structure feature was not previously observed among
mesostructured inorganic/surfactant composite materials.
The examination of the thermal stability of the samples may
help to explain the contradiction between the XRD results and
the TEM views. After removal of the surfactant by extraction,
the first intense diffractions of the materials increased signifi-
cantly (as shown in Fig. 3 a1 and a2). The d100 spacing of NiPO-1
increased to 5.1 nm and that of NiPO-2 to 4.9 nm, while the
second and third diffractions of both samples were well retained.
The thermal stabilities of mesoporous NiPO-1 and NiPO-2
materials were examined by heating the extracted samples at
different temperatures (573 and 673 K) for 4 h in air or in N2. The
presence of mesophase after heat treatment was confirmed by
XRD characterization (Fig. 3). After heating to 573 K in N2,
only one primary low angle peak at a smaller d-spacing (4.7 nm)
appears in the XRD pattern of NiPO-1, while the second and
third peaks almost disappeared, which indicates that the sample
has experienced structural rearrangement of the framework to
a less-ordered structure during heat treatment (Fig. 3 a1).
Increasing temperature to 673 K for 4 h in N2 led to much
lowered intensity of the primary peak, which indicates partial
Catalytic activity measurement
The catalytic performances of NiPO-1 and NiPO-2 were evalu-
ated in the epoxidation reaction of cycloalkenes. The reactions
were performed in a three-necked flask fitted with a reflux
condenser and stirrer. NiPO-1 or NiPO-2 was de-surfactanted by
extraction and heat treatment at 573 K for 4 h before use. In
a typical reaction run, the reactor was loaded with 0.1 g of
NiPO-1 or NiPO-2, 1 mL of cycloalkene (cyclohexene or cyclo-
dodecene), 3.6 mL of acetonitrile, and 30 wt.% aqueous solution
of H2O2 (H2O2/cycloalkene ¼ 3). The epoxidation was carried
out at 333 K and atmospheric pressure under stirring. The
products were analyzed using an Agilent-6890+ gas chromato-
graph equipped with
a flame ionization detector (FID).
Identification of the products was performed using a gas
chromatograph–mass spectrometer with an electron ionization
detector (Hewlett Packard 6890 GC/5973 MSD).
Results and discussion
Structures of mesoporous NiPO-1 and NiPO-2
The XRD patterns of the as-synthesized NiPO-1 and NiPO-2 are
shown in Fig. 1. For both samples, the presence of three broad
reflections at low angles suggests the formation of meso-
structures. These reflections can be indexed to (100), (200), and
(300) diffraction with d100 ¼ 4.3–4.6 nm and 4.2–4.5 nm for
NiPO-1 and NiPO-2 respectively. The repeat distances and
3602 | J. Mater. Chem., 2008, 18, 3601–3607
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Fig. 2 TEM images of the as-synthesized NiPO-1 (a),(b) and NiPO-2 (c),(d).
Scheme 1 Schematic depiction of interaction of inorganic species with
surfactant micelles to form NiPO-1 and NiPO-2.
Fig. 3 XRD patterns of extracted NiPO-1 (a) and NiPO-2 (b) materials
followed by heat treatment at 573 or 673 K for 4 h in N2 or in air,
respectively.
Within a layer, the nanotubes develop roughly along one
dimension, but not strictly parallel. Along the other dimension,
the nanotubes array side by side to extend to a layer. Then the
layers array to form a lamellar arrangement. In the structures of
NiPO-1 and NiPO-2, the arrangement of nanotubes should be
similar because of their similar XRD patterns, though they
possess nanotubes of different lengths.
collapse of the mesostructure. Heating in air led to an increase in
d-spacing (5.4 nm and 5.2 nm) and decrease in intensity of the
peaks (Fig. 3 d1, e1). The thermal stability of NiPO-2 showed
a similar trend to that of NiPO-1 (Fig. 3 a2–e2). Their
mesostructures could be retained up to 673 K.
With respect to the absence of ordering when the samples were
observed in the TEM views, we propose that the lamellar
arrangement of nanotubes, and also the arrays in the same layer,
especially for NiPO-2, was destroyed by the high-energy electron
beam irradiation as in heating at elevated temperature. However,
the stability of the nanotubes of both NiPOs is sufficient to
survive heating or electron beam irradiation, as evidenced in the
On the basis of the above results, we propose that the lamellar
symmetry of the mesostructures in the as-synthesized samples
originates from the arrangement of nanotubes, as sketched in
Scheme 1. The lamellar arrangement is constructed by rows of
nanotubes. One layer is composed of one row of nanotubes.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 3601–3607 | 3603

View Article Online
TEM images. Therefore, this good thermal stability of NiPO-1
and NiPO-2 may be ascribed to the character of the tubular
materials.15–17 The poor combination of nanotubes might be
owing to the weak bonding effect between nanotubes, which is
inferred as van der Waals forces or hydrogen bonds.
stability studies have confirmed that the layered arrangement
was lost upon heating, while the nanotubes survived. Therefore,
it can be deduced that the remaining pore distribution is of the
nanotubular channels and the loss is of the spaces formed by
nanotube packing. In addition, the shapes of the hysteresis loops
(between type H1 and H2 for NiPO-1 and type H2 for NiPO-2)
did not significantly change after heat treatment at 573 K, when
compared with those of the extracted samples. It is clear that
NiPO-1 possesses channel-like pores in the nanotubes, which is in
agreement with the TEM observation. As for the type H2
hysteresis loop for NiPO-2, we guess it might stem from
ink-bottle-type pores shaped by smaller mouths of the shorter
nanotubes, though this can not be clearly observed in the TEM
views.
The mesoporosities of the solvent-extracted and heat-treated
nickel phosphates were characterized by means of N2 adsorption
at 77 K. As shown in Fig. 4, the isotherms of all the samples are
of type IV, which is characteristic of mesoporous materials.
The nitrogen-adsorption isotherms of extracted NiPO-1 and
NiPO-2 both show two distinct capillary condensation steps—
the first one at p/p0 ¼ 0.05–0.35 and the second one with
hystereses at p/p0 > 0.40, which suggests that both NiPOs have
bimodal pore structures in the mesopore region. For extracted
NiPO-1, the hysteresis loop appears in the relative pressure range
of 0.40–0.90 and represents a transitional type between H2 and
H1.18 Extracted NiPO-2 shows a larger hysteresis loop of type H2
in the p/p0 range of 0.40–0.85.19
The BET surface area and pore volume of extracted NiPO-1
are 292 m2 gꢀ1 and 0.36 cm3 gꢀ1. These values decreased to 205 m2
gꢀ1 and 0.24 cm3 gꢀ1 after heat treatment at 573 K. Extracted
NiPO-2 has a BET surface area of 210 m2 gꢀ1 and a pore volume
of 0.20 cm3 gꢀ1, which increased to 259 m2 gꢀ1 and 0.32 cm3 gꢀ1
after heating to 573 K. The surface areas of the above meso-
porous nickel phosphates are lower than those of mesoporous
silica MCM-41 or SBA-15, but they are close to the typical values
of non-siliceous mesoporous solids.5,8,9 It is noticeable that the
changing trends of porosities after heating are different for
NiPO-1 and NiPO-2. The reason may be complicated when
considering the lost pores generated by nanotube packing, the
As shown in the insets of Fig. 4, both extracted NiPO-1 and
NiPO-2 showed a bimodal pore size distribution (PSD). There
are two distinct peaks at 2.7 nm and 3.7 nm in the PSD of
extracted NiPO-1, and at 2.0 and 3.9 nm in that of extracted
NiPO-2. After heating to 573 K, the characteristics of the
bimodal pore structure disappeared from the isotherms and the
PSD curves of NiPO-1 and NiPO-2 show only one peak centered
around 3.7 nm. The previous results from TEM and thermal
Fig. 4 Nitrogen adsorption (-)–desorption (,) isotherms of NiPO-1 after solvent extraction (a) and heat treatment at 573 K for 4 h in N2 (b), and of
NiPO-2 after solvent extraction (c) and heating at 573 K for 4 h in N2 (d). BJH pore size distribution curves are shown in the insets.
3604 | J. Mater. Chem., 2008, 18, 3601–3607
This journal is ª The Royal Society of Chemistry 2008

View Article Online
reduction of remaining surfactant or the decomposition of
nanotubes.
to synthesize the mesostructured nickel phosphates depends on
the type of base. In the case of using aqueous ammonia to control
the pH, the pure phase of NiPO-1 falls in NH4+/P molar ratio
2.1–2.4 and Ni/P ratio 0.5–0.7. Outside this range, crystalline
particles of (NH4)Ni(PO4)$H2O or Ni3(PO4)2$8H2O are
obtained.20 When the base is TMAOH, the optimal range to
synthesize NiPO-2 is 1.8 to 2.1 for the TMA+/P ratio and ca.
0.7 for the Ni/P ratio. A higher or lower amount of TMAOH
leads to the formation of dense Ni3(PO4)2$8H2O crystals and the
reduction of the desirable mesophases.
ICP elemental analysis was employed to determine the
chemical compositions of the mesoporous nickel phosphates.
The elemental compositions of the two samples revealed that
they are different compounds. The Ni/P molar ratios of NiPO-1
and NiPO-2 samples are 1.57 and 1.20, respectively. As the molar
Ni/P ratio calculated from the theoretical composition of
nickel(II) phosphate is 1.5, NiPO-1 is more likely composed of
nickel(II) phosphate (Ni3(PO4)2). It should be noted that the Ni/P
ratio of NiPO-2 deviates far from the composition of Ni3(PO4)2,
which suggests that there might be abundant hydroxyls bonded
to PO4 tetrahedra in the framework of NiPO-2.
We propose that the formation of mesoporous NiPOs take
place as follows: (i) acidification of the CTAB micelle solution by
H3PO4 helps the phosphates to be uniformly dispersed on the
interface of surfactant micelles and form phosphate–surfactant
+
ion pairs;17,21 (ii) when the base is added to control the pH, NH4
Synthesis of mesoporous NiPO-1 and NiPO-2
or TMA+ ions react with the phosphate–surfactant ion pair and
The synthesis of mesoporous nickel phosphates was based on
a self-assembly between a cationic surfactant CTAB and two
precursors (phosphate and nickel species) in a particular pH
range. As shown in Fig. 5, the phases of pure mesoporous NiPOs
lie in a narrow region. The range of base/P or Ni/P ratio required
modify the charge equilibrium; (iii) nickel species coordinate with
3ꢀ
PO4 to form a nickel phosphate precursor around surfactant
micelles; (iv) the precursor condenses to
mesostructure at elevated temperatures.
a nanotubular
The nanotube units in the NiPO structures indicate that
cylindrical micelles of different lengths might form in the
presence of aqueous ammonia and TMAOH at a relatively high
surfactant concentration (ꢂ20%). Longer cylindrical micelles are
favorable in the presence of aqueous ammonia due to the lower
viscosity of the system, while short rod-like micelles are preferred
in TMAOH system.22 When the inorganic components condense
on the surface of these micelles, mesostructured materials with
nanotube units form accordingly.
Besides aqueous ammonia and TMAOH, NaOH was also
used to adjust the pH of the starting gel. The product shows an
XRD pattern with a very low intensity diffraction in the low-
angle range and some diffractions of crystalline nickel phosphate
Ni3(PO4)2$8H2O in the high-angle range. It is assumed that
the bases with different radii exert different influences on the
electrostatic interactions between nickel phosphate species and
cationic surfactant micelles to assemble S+Iꢀ/NH4+/TMA+/Na+
ion pairs.17 NH4+ or TMA+ possess much larger ionic radii than
Na+. They induce relatively weak interactions between nickel
phosphate species and surfactant micelles which is favorable for
the formation of pure mesostructured nickel phosphates.
Spectroscopic characterization
As shown in Fig. 6, the FT-IR spectra of the samples show that
NiPO-1 and NiPO-2 have very similar absorption bands.
Broad bands in the hydroxyl region with a maximum at
3420 cmꢀ1 are assigned to the O–H stretching vibration of the
residual water, exchangeable OHꢀ, and defective OH groups with
considerable hydrogen bonding.23 Due to the presence of the long
chain organic surfactant, strong absorption bands at 2800–
3000 cmꢀ1, 1380–1480 cmꢀ1, and 720 cmꢀ1 were observed in
as-synthesized NiPO-1 and NiPO-2 samples. The intensity of these
absorptions is significantly reduced after extraction, indicating
that the majority of the surfactants were removed, but small
amounts of surfactant still remain after heating to 573 K for 4 h.
For the as-synthesized and extracted NiPOs, the two bands in
the range of 1000–1100 cmꢀ1 correspond to the tetrahedral PO4
stretching vibrations, and the bands around 569 and 528 cmꢀ1 are
Fig. 5 Ternary phase diagrams of the structures obtained in the
synthesis of (a) NiPO-1 and (b) NiPO-2. The molar compositions of the
mixtures are x/y/z/1/200 Ni(NO3)2$6H2O/NH3$H2O/H3PO4/CTAB/H2O
or Ni(NO3)2$6H2O/TMAOH/H3PO4/CTAB/H2O.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 3601–3607 | 3605

View Article Online
Fig.
6 FT-IR spectra of NiPO-1 (a) and NiPO-2 (b) samples:
as-synthesized, extracted, and heat-treated at 573 K for 4 h in N2.
Fig. 7 UV-Vis diffuse reflectance spectra of NiPO-1 (a) and NiPO-2 (b)
samples: as-synthesized, extracted, and heat-treated at 573 K for 4 h in N2.
assigned to Ni–O and P–O vibrations.2 After heat treatment at
573 K, the closely adjacent absorptions around 1000–1100 cmꢀ1
and around 530–570 cmꢀ1 became joint bands at 1081 cmꢀ1 and
563 cmꢀ1, respectively. Moreover, it is noticeable that absorp-
tions at 850–910 cmꢀ1 for as-synthesized and extracted samples
were not observed in the heated samples, but a well-resolved
band appears around 920 cmꢀ1 instead, ascribed to P–O–P
vibration.24 The appearance of Ni–O bonds at 445 cmꢀ1 was
accompanied by a strong band at 569 cmꢀ1, which can be
assigned to the Ni species embedded in the OH groups.25 The
latter band disappeared after heat treatment due to the reduction
of the OH groups which surrounded the Ni atoms.25 All these
results show that some changes in the local structures have
occurred in the heat-treated NiPO-1 and NiPO-2 samples.
The coordination of the Ni cations by lattice oxygens in the
as-synthesized, extracted, and heat-treated NiPOs was studied by
UV-Vis diffuse reflectance. The results are shown in Fig. 7. Ni(II)
ions are present as octahedral [Ni(H2O)6]2+ complexes in
hydrated nickel nitrate, and have main absorption bands at 394,
657 and 720 nm. In comparison, the maxima of the absorbance
bands of all the mesoporous nickel phosphate samples shifted to
longer wavelengths. This indicates that ligand exchanges
occurred in the coordination spheres of the Ni cations upon sol–
gel process and heat treatment.11,26 As shown in Fig. 7, the
UV-Vis bands of the as-synthesized and extracted NiPOs
are located at 407–410, 480–495, 670–678, and 748–757 nm,
suggesting that the nickel ions are present in distorted
octahedra.27 The nickel ions in the framework of the mesoporous
nickel phosphates coordinated to six oxygen ions, some from
PO4 unit and some from hydroxyl or hydrated cations, leading to
different nickel oxygen bond lengths and producing a distorted
octahedral coordination.28 Nevertheless, after the samples are
heated at 573 K, most of the six-coordinated nickel cations are
transformed into four-coordinated cations. Although the main
absorption bands of heat-treated NiPO-1 and NiPO-2 around
430 and 550 nm are different from those of known tetrahedral
complexes,29 they lie in the ranges assigned to distorted tetra-
hedral complexes.30 Some nickel cations are still present in
distorted octahedral coordination, as evidenced by the appear-
ance of bands in the ranges of 670 to 707 and 757 to 770 nm.27
The presence of four-coordinated nickel species can be ascribed
to the elimination of the OH or H2O groups, which coordinated
with nickel cations, by heat treatment at high temperatures.
Catalytic performance in epoxidation
The catalytic performances of NiPO-1 and NiPO-2 were
evaluated in epoxidation of cyclohexene and cyclododecene
under mild conditions. The reaction results are summarized in
Table 1.
Under the given reaction conditions, both NiPO-1 and NiPO-2
showed high activities in the epoxidation of cyclohexene (over
3606 | J. Mater. Chem., 2008, 18, 3601–3607
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Table 1 Epoxidation of cyclohexene and cyclododecene over NiPO-1
and NiPO-2a
suggests that these materials show potential for use in liquid-
phase oxidation catalysis.
Selectivity (mol%)
Acknowledgements
Catalyst Substrate
Conversion (%)b Epoxide Diol Othersc
The authors acknowledge the financial support from the NSFC
(20333030, 20503003), the Education Ministry of China
(20030141026), NCET, 111 Project and the CNPC Innovation
Foundation.
NiPO-1 Cyclohexene
86.6
Cyclododecene 52.7
44.9
95.6
40.3
99.0
24.9 30.2
0.0 4.4
21.1 38.6
0.0 1.0
NiPO-2 Cyclohexene 87.2
Cyclododecene 56.4
a
Reaction conditions: substrate 1 mL, H2O2/substrate (molar ratio) ¼ 3,
acetonitrile 3.6 mL, 0.1 g catalyst, 333 K, the reaction time was 6 h
(cyclohexene) or 10 h (cyclododecene).b Conversion was mol% of
cyclododecene consumed during the reaction.c Others were allylic
oxidation products (cycloalken-1-ol and -1-one) and ring cleavage
products of cycloalkenes.
References
1 C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and
J. S. Beck, Nature, 1992, 359, 710.
2 M. P. Kapoor, S. Inagaki and H. Yoshida, J. Phys. Chem. B, 2005,
109, 9231.
3 M. Ziolek, Catal. Today, 2004, 90, 145.
4 P. D. Yang, D. Y. Zhao, D. I. Margolese, B. F. Chmelka and
G. D. Stucky, Nature, 1998, 396, 152.
5 W. H. J. Hogarth, J. C. D. da Costa, J. Drennan and G. Q. Lu,
J. Mater. Chem., 2005, 15, 754.
6 D. J. Jones, G. Aptel, M. Brandhorst, M. Jacquin, J. Jimenez-
Jimenez, A. Jimenez-Lopez, P. Maireles-Torres, I. Piwonski,
E. Rodriguez-Castellon, J. Zajac and J. Roziere, J. Mater. Chem.,
2000, 10, 1957.
7 U. R. Pillai and E. Sahle-Demessie, Chem. Commun., 2004, 826.
8 C. Serre, A. Auroux, A. Gervasini, M. Hervieu and G. Ferey,
Angew. Chem., Int. Ed., 2002, 41, 1594.
9 N. K. Mal and M. Fujiwara, Chem. Commun., 2002, 2702.
10 I. W. C. E. Arends and R. A. Sheldon, Appl. Catal., A, 2001, 212, 175.
11 N. Guillou, Q. Gao, P. M. Forster, J. S. Chang, M. Nogues,
S. E. Park, G. Ferey and A. K. Cheetham, Angew. Chem., Int. Ed.,
2001, 40, 2831.
12 S. H. Jhung, J. H. Lee, A. K. Cheetham, G. Ferey and J. S. Chang,
J. Catal., 2006, 239, 97.
13 K. Kandori, H. Nakashma and T. Ishikawa, J. Colloid Interface Sci.,
1993, 160, 499.
14 F. Sanz, C. Parada, J. M. Rojo and C. Ruiz-Valero, Chem. Mater.,
1999, 11, 2673.
15 J. M. Kim, S. Jun and R. Ryoo, J. Phys. Chem. B, 1999, 103, 6200.
16 S. A. Bagshaw, E. Prouzet and T. J. Pinnavaia, Science, 1995, 269,
1242.
17 Z. H. Luan, D. Y. Zhao, H. Y. He, J. Klinowski and L. Kevan,
J. Phys. Chem. B, 1998, 102, 1250.
18 T. Z. Ren, Z. Y. Yuan and B. L. Su, Langmuir, 2004, 20, 1531.
19 P. Schmidt-Winkel, W. W. Lukens, P. D. Yang, D. I. Margolese,
J. S. Lettow, J. Y. Ying and G. D. Stucky, Chem. Mater., 2000, 12,
686.
85%). As a further hydrolysis product of cycloalkene oxide, diol
is considered as an epoxidation product as well.11,31 The total
selectivity to cyclohexene oxide and 1,2-cyclohexanediol was
69.8% for NiPO-1, and 61.4% for NiPO-2. In the epoxidation of
cyclododecene, NiPO-1 and NiPO-2 exhibited high conversions
of 52.7% and 56.4%, respectively. Cyclododecene oxide was the
main product from the oxidation of cyclododecene, while no diol
was detected. The epoxide selectivities of NiPO-1 and NiPO-2
are remarkably high, reaching 95.6% and 99.0%, respectively.
The active species of NiPO-1 and NiPO-2 for the epoxidation
might be analogical to those of VSB-5, which has been assigned
to the Ni(II) sites in the framework.12 According to the UV-Vis
spectra of these materials, most of the nickel atoms are in
tetrahedral coordination in heat-treated NiPO-1 and NiPO-2,
which is in favor of the exposure of nickel active sites. Besides,
there was no destruction of the structures of mesoporous NiPOs
after reaction, as confirmed by XRD characterization. No nickel
was detected in the liquid mixture after reaction by ICP analysis.
Further studies are in progress to shed light on the catalysis
mechanism and the effect of nickel species in the mesoporous
NiPOs.
Conclusions
20 S. H. Jhung, J. W. Yoon, Y. K. Hwang and J. S. Chang, Microporous
Mesoporous Mater., 2006, 89, 9.
21 B. Tian, X. Liu, B. Tu, C. Yu, J. Fan, L. Wang, S. Xie, G. D. Stucky
and D. Y. Zhao, Nat. Mater., 2003, 2, 159.
22 Z. Lin, Langmuir, 1996, 12, 1729.
23 A. Bhaumik and S. Inagaki, J. Am. Chem. Soc., 2001, 123, 691.
24 A. G. Shikerkar, J. A. E. Desa, P. S. R. Krishna and R. Chitra,
J. Non-Cryst. Solids, 2000, 270, 234.
25 C. Faure, C. Delmas and M. Fouassier, J. Power Sources, 1991, 35,
279.
26 R. Garcia, I. J. Shannon, A. M. Z. Slawin, W. Z. Zhou, P. A. Cox and
P. A. Wright, Microporous Mesoporous Mater., 2003, 58, 91.
27 C. Lepetit and M. Che, J. Phys. Chem., 1996, 100, 3137.
28 Y. Sendoda, Y. Ono and T. Keii, J. Catal., 1975, 39, 357.
29 H. A. Weakliem, J. Chem. Phys., 1962, 36, 2117.
30 K. Klier and M. Ralek, J. Phys. Chem. Solids, 1968, 29, 951.
Two novel nanotubular mesoporous nickel phosphate materials
(NiPO-1 and NiPO-2) have been synthesized by using a cationic
surfactant as template. The base, aqueous ammonia or
TMAOH, determines the materials with distinct textural
characteristics. The XRD patterns and TEM views indicate that
the NiPOs have different length nanotubular units with an
irregular lamellar symmetry. This unique structure is responsible
for the good thermal stability and relatively high BET surface
areas. The evolution of coordination of the nickel cations in these
two NiPOs presents a similar state according to UV-Vis data. It
is noticeable that this kind of material exhibited high catalytic
activity and remarkably high selectivity in the epoxidation of
bulky cyclododecene with H2O2 as an oxidant. The good stability
of active metal and mesoporous structures in the reactions
´
31 J. M. Fraile, J. I. Garcıa, J. A. Mayoral and E. Vispe, Appl. Catal., A,
2003, 245, 363.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 3601–3607 | 3607