Microporous niobium phosphates and catalytic properties prepared by a supramolecular templating mechanism

Article abstract of DOI:10.1039/b300323j

Microporous hexagonal niobium phosphate synthesized using neutral surfactants of molecular length, C₆ to C₁₀ hydrocarbons, by a supramolecular templating mechanism (S⁰I₀) possesses strong hydrophilic character, which leads to high selectivity for catechol formation (95.3%) in the presence of protic solvent (MeOH) in the hydroxylation of phenol using aqueous H₂O₂.

Full text of DOI:10.1039/b300323j

Microporous niobium phosphates and catalytic properties prepared by a
supramolecular templating mechanism
Nawal Kishor Mal,a Asim Bhaumik,b Prashant Kumarc and Masahiro Fujiwara*a
a National Institute of Advanced Industrial Science and Technology (AIST) Kansai, 1-8-31 Midorigaoka,
Ikeda, Osaka 563-8577, Japan. E-mail: m-fujiwara@aist.go.jp
b Department of Materials Science, Indian Association for the Cultivation of Science, Jadavpur, Kolkata -
32, India
c
Chemical Technology and Heterogenous Catalysis, University of Technology, Aachen Worringerweg 1
52074, Aachen, Germany
Received (in Cambridge, UK) 10th January 2003, Accepted 25th February 2003
First published as an Advance Article on the web 10th March 2003
Microporous hexagonal niobium phosphate synthesized
using neutral surfactants of molecular length, C6 to C10
hydrocarbons, by a supramolecular templating mechanism
(S0I0) possesses strong hydrophilic character, which leads to
high selectivity for catechol formation (95.3%) in the
presence of protic solvent (MeOH) in the hydroxylation of
phenol using aqueous H2O2.
100 °C and calcined at 450 °C for 6 h. Elemental analyses of the
samples were carried out using ICP (Shimadzu ICPV-1017).
The samples were charaterized using XRD (Cu-Ka radiation, l
= 0.15406 nm, Shimadzu XRD-6000), TEM (Hitachi H-
9000NA, 300 kV), and N2 sorption at 2196 °C using a Bellsorp
28 instrument. The sorption measurement was carried out
gravimetrically in an electrobalance (Chan, USA) at 25 °C with
a fixed p/p0 ratio of 0.4 each of triisopropylbenzene and water
as adsorbents after equilibration for 3 h. 31P spectra were
recorded on a Bruker DSX500 machine at 202.42 MHz with a
spinning rate of 8 kHz, pulse time (P1) of 3.0 µs and a repetition
time (D1) of 30 s. Chemical shift are reported relative to 85 wt%
H3PO4. The catalytic activity of the niobium phosphate sample
was tested in the hydroxylation of phenol, batchwise in a round
bottomed flask in the presence of aqueous H2O2 (30 wt%) as
oxidant using different solvents. The products were analyzed
using GC (GC-17A, Shimadzu) using a capillary column and
confirmed by standard product and GC-MS.
The XRD patterns and TEM image of niobium phosphate are
shown in Fig. 1. The XRD spectrum of the as-synthesized
sample shows one sharp reflection at low angle and two broad
reflections at higher angles; assuming a hexagonal structure like
MCM-41, these reflection can be indexed as (100), (110) and
(200), respectively. The calcined sample shows a single
reflection which can be indexed as (100) indicating the local
disorder in the structure.12 The interplanar spacings, d100, of the
as-synthesized and calcined samples are 2.91 and 2.68 nm,
respectively. The TEM image of niobium phosphate confirms
its disordered hexagonal structure due to disordered packing of
cylindrical pores12,13 or wormholes, as reported by Pinnavaia et
al.14 for the synthesis of Ti-HMS using hexadecylamine as
surfactant (Fig. 1; right). Note that the pore size of the sample is
within the micropore range. In Fig. 2, N2 adsorption–desorption
isotherms and the solid-state 31P NMR spectrum of calcined
microporous niobium phosphate are shown. The microporosity
of niobium phosphate is further illustrated by the type I N2
adsorption–desorption isotherms. The pore size of the sample
calculated using the MP method is 1.25 nm (Fig. 2; left, inset),
which is slightly bigger than that of one-dimensional micro-
porous VPI-5 (pore diameter = 1.21 nm).2 This was confirmed
when a comparison of the adsorption capacities of niobium
Porous inorganic materials synthesized with high surface areas
and narrow pore size distributions are highly useful as active
and/or selective catalytic materials.1 Organic templates, as
discrete entities around which the framework crystallizes, have
been used to construct microporous zeolitic frameworks of sizes
ranging from small pore (0.4 nm) to extra large pore (1.2 nm,
VPI-5),2 but our understanding of their function in the process
remains rather limited and incongruent.3 In contrast, a more
generalized approach for the formation of mesoporous struc-
tures with a supramolecular templating mechanism has been
developed after the discovery of MCM-41 silicates by Mobil
scientists in 1992.4 Such a mechanism has been used for the
synthesis of materials less than mesopore in size ( < 2 nm) such
as super-microporous tin5 and niobium6 phosphates using long
chain alkyl (C16) groups, and microporous transition metal
(niobium) oxide using a short chain alkylamine (C6) as a
surfactant.7,8 Recently, we have reported the synthesis of super-
microporous distorted hexagonal niobium phosphate with high
anion exchange capacity (6.3 mmol g21) and its catalytic
activity using hexadecyl amine as a surfactant (S0I0).6 Super-
microporous materials are those porous materials whose pore
size is bigger than that of extra-large-pore zeolitic materials and
smaller than that of mesoporous materials. The dimensions
range from 1.4 to 2.0 nm.6,9,10 However, the synthesis of
microporous metal phosphates has not yet been reported.
Therefore, the synthesis of microporous metal phosphates with
tailored pore size should be useful in shape- and size-selective
molecular sieving applications. Herein we describe, for the first
time, the synthesis of a microporous metal (Nb) phosphate using
short chain alkylamines (C6 and C10) as surfactants, which has
interesting catalytic activity in the hydroxylation of phenol.
We mixed the metal source and phosphoric acid prior to
addition of surfactant, otherwise no reproducible phosphorus/
metal ratio in the final solids could be obtained.6,11 In a typical
synthesis, 3.46 g of H3PO4 (30 mmol, 85% aq., Wako Chem.)
was added to 4.09 g of NbCl5 (15 mmol, Aldrich) in 70 g of H2O
under stirring for 10 min. Aqueous ammonia was added until
the pH stabilized around 4.90. The precipitate was filtered and
washed with distilled water several times to remove excess
chloride ions. H2O (20 g) and 2.36 g of decylamine (15 mmol)
were added to the precipitate in a plastic beaker and stirred for
30 min. Finally, the required amount of H3PO4 (1.15 g, 10
mmol) was added, with the pH maintained at 3.88 under stirring
for 30 min. The gel was heated in a Teflon lined stainless steel
autoclave at 75 °C under autogenous pressure for 1 day. The
final product was filtered, washed with distilled water, dried at
Fig. 1 XRD profiles (left) of as-synthesized (a) and calcined (b), and TEM
image (right) of calcined niobium phosphate.
872
CHEM. COMMUN., 2003, 872–873
This journal is © The Royal Society of Chemistry 2003

Table 2 Catalytic activity in the oxidation of phenola
H2O2
Product distribution (mol%)
Conversion efficiency
Solvent (mol%)
(mol%)
CAT
HQ
PBQ
CAT/HQ
CH3CN 13.1
MeOH 15.3
40.2
46.0
62.6
95.3
35.0
4.5
2.4
0.2
1.79
21.18
a Reaction conditions: catalyst = 0.10 g, phenol = 1 g, solvent = 5 g, temp.
= 80 °C, phenol/H2O2 = 3 (molar ratio), reaction time = 15 h, conversion
= mol% of phenol consumed during the reaction, H2O2 efficiency = mol%
of H2O2 consumed in the formation of catechol (CAT), hydroquinone (HQ)
and para-benzoquinone (PBQ).
Fig. 2 N2 adsorption–desorption isotherms (inset: pore size distribution
curve from MP method) (left) and 31P MAS NMR of calcined microporous
niobium phosphate (right) (* spinning bands).
15.3 and 46%, respectively, in methanol. However, the catechol
selectivities in aprotic (acetonitrile) and protic solvents (metha-
nol) are quite different: 62.6 and 95.3%, respectively. The
catechol to hydroquinone ratios in acetonitrile and methanol are
1.79 and 21.18%, resepectively. This unusual high catechol
selectivity in protic solvent may be probably due to the
formation of hydrogen bonds between the OH group of phenol
and the solvent (MeOH) in a transition state, thereby inducing
the ortho position of phenol to form catechol (Fig. 3), since the
strong hydrophilic character of niobium phosphate (H2O
adsorption = 42 wt%) has the tendency to coordinate with
protic solvents and expand their coordination.
phosphate (0.122 cm3 g21) and VPI-5 (0.117 cm3 g21)2 for
triisopropylbenzene (diameter = 0.85 nm) was carried out,
suggesting the pore size of niobium phosphate is probably
slightly bigger than that of VPI-5 (Table 1). The H2O adsorption
capacities of niobium phosphate and VPI-5 are 0.42 and 0.35
cm3 g21, respectively, which might also be due to interparticle
adsorption. The higher hydrophilicity of microporous niobium
phosphate is due to the larger metal size. The BET specific
surface area of niobium phosphate is 283 m2 g21, which is
smaller than that of a sample prepared using hexadecylamine
(C16) as surfactant (482 m2 g21).6 The 31P MAS NMR spectrum
of calcined niobium phosphate gives a resonance at 223.1 ppm
with respect to H3PO4 (85% aq.), confirming the tetrahedral
coordination of phosphorus (Fig. 2; right). As the connectivity
increases, an upfield shift is observed from 25.3 to 210.6 ppm
for H2PO4, to 218.1 ppm for HPO4 and finally to 219 to 232.5
ppm for PO4.15,16 Niobium ions (Nb5+) in niobium phosphate
are tetra- or penta-coordinated and not hexa-coordinated.6 The
Fig. 3 Mechanism for the formation of catechol in protic solvent (MeOH)
(R = CH3 or H) over microporous niobium phosphate.
P/Nb molar ratio in the niobium phosphate is 1.13. Nb(
V
)
oxidation state is present in microporous niobium phosphate,
which is most stable under our hydrothermal reaction conditions
(acidic medium). Nb5+ ions are in tetrahedral coordination state;
in such a case, the P/Nb molar ratio is expected to be 1. In
addition, the double positive charge is due to framework Nb5+
and P5+ ions balanced by the extraframework excess of
phosphorus (HPO422) and Cl2 anions. The composition of the
microporous niobium phosphate is [Nb1.0(PO4)1.0](H-
PO4)0.13Cl0.74. Effects of pH of the gel, temperature and alkyl
chain length of the surfactant on mesophase formation are given
in Table 1. A mesophase was obtained for decylamine as
surfactant in the gel pH range 3 to 6.5 and temperature 65 to 120
°C, whereas above 130 °C a lamellar phase was produced.
However, when n-hexylamine was used as surfactant in a
narrow gel pH range 3.6 to 4.2 and temperature 65 to 80 °C, the
resulting materials were mesophases. TG analysis of niobium
phosphate shows that the weight loss between 150 and 450 °C
due to removal of surfactant and condensation of hydroxy
groups is 36.3%, which is 27.7% less than that of a sample
prepared using hexadecylamine (C16) as surfactant (50.2%).
The catalytic activity of niobium phosphate in the hydroxyla-
tion of phenol is presented in Table 2. The phenol conversion
(mol% phenol consumed) and H2O2 efficiency (mol% H2O2
consumed) are 13.1 and 40.2%, respectively, in acetonitrile and
In conclusion, a microporous transition metal (niobium)
phosphate has been synthesized by a supramolecular templating
mechanism for the first time, which exhibits an unusual high
selectivity for catechol formation (95.3%) in protic solvents
(MeOH) in the hydroxylation of phenol.
Notes and references
1 (a) J. M. Thomas, Angew. Chem., Int. Ed. Engl., 1988, 27, 1673–1691;
(b) W. F. Holderich and H. van Bekkum, Stud. Surf. Sci. Catal., 1991,
58, 631–726.
2 M. E. Davis, J. Am. Chem. Soc., 1989, 111, 3919–3924.
3 M. E. Davis and R. F. Lobo, Chem. Mater., 1992, 4, 756–768.
4 C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck,
Nature, 1992, 359, 710–714.
5 C. Serre, A. Auroux, A. Gervasini, M. Hervieu and G. Férey, Angew.
Chem., Int. Ed., 2002, 41, 1594–1597.
6 N. K. Mal and M. Fujiwara, Chem. Commun., 2002, 2702–2703.
7 T. Sun and J. Y. Ying, Nature, 1997, 389, 704–706.
8 T. Sun and J. Y. Ying, Angew. Chem., Int. Ed., 1998, 37, 664–667.
9 G. E. Bastardo, R. Mokaya and W. Jones, Chem. Commun., 2001,
1016–1017.
10 S. A. Bagshaw and A. R. Hayman, Adv. Mater., 2001, 13,
1011–1013.
11 C. Serre, M. Hervieu, C. Magnier, F. Taulelle and G. Férey, Chem.
Mater., 2002, 14, 180–188.
Table 1 Effect of pH, surfactant and temperature on mesophase for-
mationa
12 S. A. Bagshaw, E. Prouzet and T. J. Pinnavaia, Science, 1995, 269,
1242–1244.
13 P. Behrens, Angew. Chem., Int. Ed. Engl., 1996, 35, 515–518.
14 T. J. Pinnavaia and W. Zhang, Stud. Surf. Sci. Catal., 1998, 117,
23–35.
pH
Surfactant
Temperature/°C
Phase
3–6.5
Decylamine
Decylamine
Hexylamine
Hexylamine
65–120
> 130
65–80
> 90
m.p.
l.p.
m.p.
l.p.
15 B. Chakraborty, A. C. Pulikottil, S. Das and B. Viswanathan, Chem.
Commun., 1997, 911–912.
16 D. J. Jones, G. Aptel, M. Brandhorst, M. Jacquin, J. J. Jiménez, A. J.
López, P. M. Torres, I. Piwonski, E. R. Castellón, J. Zajac and J.
Roziére, J. Mater. Chem., 2000, 10, 1957–1963 and references
therein.
3–6.5
3.6–4.2
3–6.5
a m.p. = mesophase, l.p. = lamellar phase.
CHEM. COMMUN., 2003, 872–873
873