Dispersion of VOPO4?2H2O on different supports through exfoliation and their catalytic activity for the selective oxidation of cyclohexane

Article abstract of DOI:10.1016/j.catcom.2010.08.012

The VOPO₄?2H₂O phase has been exfoliated in 2-propanol and dispersed on different supports such as alumina, KIT-6, CeO ₂ and ZrO₂ at 50% loadings and at different VPO/support ratios in the case of alumina. It has been found that the nature and amount of the support has a profound influence on the crystallinity and morphology of the dispersed VPO phase which in turn, has a marked effect on its catalytic activity for the oxidation of the cyclohexane.

Full text of DOI:10.1016/j.catcom.2010.08.012

Catalysis Communications 12 (2010) 110–115
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Dispersion of VOPO4·2H2O on different supports through exfoliation and their
catalytic activity for the selective oxidation of cyclohexane
⁎
Parijat Borah, Arjun Ramesh, Arunabha Datta
Indian Institute of Petroleum, Dehradun, Uttarakhand, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 May 2010
Received in revised form 28 July 2010
Accepted 9 August 2010
Available online 15 August 2010
The VOPO4·2H2O phase has been exfoliated in 2-propanol and dispersed on different supports such as
alumina, KIT-6, CeO2 and ZrO2 at 50% loadings and at different VPO/support ratios in the case of alumina. It
has been found that the nature and amount of the support has a profound influence on the crystallinity and
morphology of the dispersed VPO phase which in turn, has a marked effect on its catalytic activity for the
oxidation of the cyclohexane.
© 2010 Elsevier B.V. All rights reserved.
Keywords:
VOHPO4.2H2O
Exfoliation
Dispersion
Supports
Cyclohexane oxidation
1. Introduction
for the oxidation of cyclohexane, although with low conversions [8]. It
has been observed that although vanadium containing compounds such
Vanadium phosphorous oxides (VPO) are a fascinating class of
compounds both because of their importance as the only commercial
catalyst for the activation of an alkane such as butane for its selective
oxidation to maleic anhydride [1] and also because of its structural
diversity arising from the different ways in which vanadium octahedra
and phosphate tetrahedra can be linked together to give different
layered solids. The structural diversity of the VPO system gains
importance in light of the fact that the catalytic activity of VPO phases
has been found to be sensitive to its structural and morphological
characteristics [2,3]. Consequently considerable research has been
devoted to the synthesis of new phases and the modification of existing
ones. In this context we have demonstrated a novel approach involving
exfoliation of a layered VPO phase in suitable solvents followed by their
surfactant assisted organization into a mesostructured phase [4]. The
exfoliation process prompted us to demonstrate a new approach to
prepare a dispersed VPO phase by exfoliating the VOPO4·2H2O in 2-
propanol followed by its dispersion on alumina [5] and dispersed phase
was evaluated as a catalyst for the oxidation of cyclohexane.
as mixed metal oxides V2O5-TiO2 [9], amorphous microporous
vanadium oxides mixed with silica [10], V-MCM-41, VMCM-48
[11,12] etc. have been used as catalysts for the oxidation of cyclohexane,
the exfoliated VOPO4·2H2O phase dispersed on alumina gave a very
high selectivity for cyclohexanol at high conversion (83.3%) of
cyclohexane [5].
In the present work we demonstrate that the dispersion of the
VOPO4·2H2O phase, after exfoliation, on different supports and with
different VPO/support ratios has a profound effect on the physical
properties of the dispersed VPO phase which in turn has a marked effect
on the conversions and product selectivity when these phases are used
as catalysts for the oxidation of cyclohexane.
2. Experimental
2.1. Synthesis
2.1.1. Preparation of VOPO4·2H2O
The use of VPO phases as catalysts for liquid phase oxidation
reactions is very limited. However it has also been demonstrated that
the vanadyl pyrophosphate phase is a good catalyst for the liquid phase
oxidation of cycloalkanes but not for linear alkanes [6,7]. Apart from the
vanadyl pyrophosphate phase, metalloporphyrins supported on VPO
phases like MnP/VOPO4 and FeP/VOPO4 have been reported as catalysts
A previously reported method was followed for the preparation of
VOPO4·2H2O [13]. The yellow solid was identified as VOPO4·2H2O by
powder X-ray diffraction [14] and FTIR studies [4]. This original
VOPO4·2H2O is denoted as VPO-O.
2.1.2. Dispersion of exfoliated VOPO4·2H2O on supports
A suspension of powdered VPO-O (1 g) in isopropyl alcohol
(50 ml) was placed in a flask (100 ml) equipped with a condenser and
heated at 323 K for 30 min with stirring which led to a homogeneous
yellow colored liquid. To this yellow colored solution, 1 g of different
⁎
Corresponding author. Tel.: +91 135 2660263.
E-mail address: adatta@iip.res.in (A. Datta).
1566-7367/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.08.012

P. Borah et al. / Catalysis Communications 12 (2010) 110–115
111
Fig. 1. (a) XRD patterns of VPO-O and the dispersed phases VPO-A, VPO-C, VPO-Z and
VPO-K. (b) XRD patterns of the dispersed phases VPO-A, VPO-A1, VPO-A2 and VPO-A3.
supports was added and the solvent evaporated by constant stirring.
The solid samples obtained using the supports Al2O3, CeO2, ZrO2 and
KIT-6 were coded as VPO-A, VPO-C, VPO-Z and VPO-K respectively. In
the case of alumina support however, in addition to VPO-A which had
50% loading, other samples with VPO loadings of 25%, 35% and 75%
were also prepared by taking the requisite amounts of VPO and Al2O3
and were coded as VPO-A1, VPO-A2 and VPO-A3 respectively.
Fig. 2. (a) FTIR spectrum of VPO-O and the dispersed phases VPO-A, VPO-C, VPO-Z and
VPO-K. The inset shows the O-H stretching region of different phases. (b) FTIR
spectrum of the dispersed phases VPO-A, VPO-A1, VPO-A2 and VPO-A3.
diluted in KBr (1%). The scanning electron micrographs and EDS data
were obtained using a FEI Quanta FEG 200 scanning electron microscope.
BET surface areas were obtained using a MICROMERITICS-ASAP 2010 unit
after activation at 423 K and 10−2 bar pressure in N2 atmosphere for 2 h.
2.2. Characterization
2.3. Catalytic test
The X-ray powder diffractograms (XRD) were recorded at room
temperature on a Bruker diffractometer using CuKα radiation
(λ=1.5406 Å) in the 2θ range 2° to 60°. FTIR spectra were recorded on
a Perkin-Elmer 1760X FTIR spectrometer with the sample powder
The catalytic tests were carried out in a two necked round bottom
flask. In a typical experiment 1.1 cm3 (0.01 mol) of cyclohexane
(Merck GR) was added to 10 cm3 of acetonitrile (Merck, GR)

112
P. Borah et al. / Catalysis Communications 12 (2010) 110–115
containing the VPO catalyst. For all the reactions the molar ratio of
cyclohexane to vanadium was 500:1. H2O2 (50% Rankem) was added
slowly over a period of 6 h and the resulting reaction mixture was
stirred at 250 rpm at 333 K for 24 h. The reactants and products were
analyzed by GC-MS and quantified by GC after determining the
response factors for the reactants and products individually.
addition there are other strong lines at d values of 3.72 Å, 3.1 Å, 2.29 Å
and 4.77 Å corresponding to the (002), (200), (103) and (101) planes
respectively. XRD patterns of VPO-A, VPO-K, VPO-C and VPO-Z also
show reflections at 7.33 Å, 7.3 Å, 7.24 Å and 7.16 Å which correspond
to the (001) plane of VOPO4·2H2O although the d values are lower
than the 7.4 Å observed for VPO-O. Moreover the peaks are broader
with low crystallinity compared to those of VPO-O indicating a
smaller crystallite size. At the same time, VPO-Z and VPO-C in
addition to the characteristic lines of VPO also show the reflections
due to ZrO2 and CeO2 although they are much weaker.
3. Results and discussion
3.1. XRD
The XRD patterns (Fig. 1b) of the VPO catalysts supported on
alumina with different VPO loadings show that on increasing the VPO
loading on the support, the crystallinity and crystallite size of the
The XRD pattern (Fig. 1a) of VOPO4·2H2O (VPO-O) shows an
intense line at a d=7.4 Å corresponding to the (001) plane. In
Fig. 3. (a) SEM micrographs of VPO-O and the dispersed phases VPO-A, VPO-C, VPO-Z and VPO-K. (b) SEM micrographs of the dispersed phases VPO-A, VPO-A1, VPO-A2 and VPO-A3.

P. Borah et al. / Catalysis Communications 12 (2010) 110–115
113
Fig. 3 (continued).
dispersed VPO phase improve. Thus VPO-A1 shows lines of alumina
only due to the low amount of VPO loading on alumina whereas the
XRD pattern of VPO-A2 shows the characteristics lines of VPO-O at d
values of 7.35 Å and 3.1 Å (although with much reduced intensity)
along with the alumina lines. The XRD pattern of VPO-A (as already
discussed) and VPO-A3 shows strong lines of VOPO4·2H2O only even
though the peaks are broader and less intense compared to those of
VPO-O.
spectra of VPO-A2, VPO-A and VPO-A3 are similar to that of the VPO-O
phase. However, interestingly in the case of VPO-A and VPO-K in the O-
H stretching region (Fig. 2a inset), a predominant band at 3384 cm−1
and 3443 cm−1 respectively are observed which is due to the O-H
stretch of the water molecule bound to vanadium. In contrast, the
spectra of both VPO-C and VPO-Z show two distinct bands of almost
equal intensity at 3421 cm−1 and 3564 cm−1 and at 3444 cm−1 and
3564 cm−1 respectively due to the O-H stretch of bound and interlayer
H2O. The higher O-H stretching frequency of the H2O molecule bound to
vanadium in the dispersed phases suggests a weaker interaction of the
H2O molecule with the vanadium as compared to VPO-O indicating a re-
organization of charge on the vanadium on dispersion on different
supports.
3.2. FTIR
FTIR spectrum (Fig. 2a) of VOPO4·2H2O (VPO-O) shows the
characteristic P-O and V-O stretching frequencies in the region
of 900–1200 cm−1 with bands centered at 947, 1038, 1088 and 1164
cm−1. The 1038 cm−1 band can be assigned to the stretching vibration
of V+5=O group [15] whereas the 947 cm−1 and 1088 cm−1 bands are
due to the asymmetric stretching of PO4 and the band at 1164 cm−1 due
to the antisymmetric stretching of PO4 group. In the hydroxyl stretching
region, two bands are observed at 3564 cm−1 and 3349 cm−1 which
can be assigned to interlayer free structural water and to the water
molecule coordinatively bound to the vanadium in the lattice opposite
to the vanadyl oxygen respectively. The FTIR spectra of VPO-A,VPO-K,
VPO-C and VPO-Z are also essentially similar to the VPO-O phase
indicating once again that the original VPO-O phase is obtained on
dispersion on the different supports. The characteristic vibrations of the
support are not observed asthey are much weaker than those of the VPO
phase. In the case of the spectrum of VPO-A1 (Fig. 2b) however, due
to the low loading of VPO, only the predominant band of Al2O3 at
1072 cm−1 is observed along with weak P-O and O-H bands of VPO. The
3.3. SEM
SEM micrographs (Fig. 3a) show that VPO-O has the typical
platelet structure of VOPO4·2H2O with the platelet size of ~6–10 μm.
In the case of VPO-A an agglomerated platelet structure with smaller
particles of ~1.5–3 μm is observed with some of the platelets having
agglomerated to a flower-petal like morphology. The VPO-K phase
consists of aggregates of thin plates with smaller crystallite size of ~3–
5 μm. It is interesting to note that both in the case of VPO-A and VPO-
K the EDS analysis of the agglomerated platelets showed the presence
of both the VPO and the support phases indicating that in these two
cases a composite morphology of the dispersed VPO phase and the
support was obtained. On the other hand in the case of VPO-C, large ill
defined platelets are observed on which smaller particles are
sprinkled. EDS analysis showed the large platelets to be the VPO

114
P. Borah et al. / Catalysis Communications 12 (2010) 110–115
Table 1
Variation of cyclohexane conversion and selectivity of products with 50% VPO dispersed on different supports as catalysts.
Catalyst
Support
Surface area
(m2/g)
Total pore volume
(ml/g)
Cyclohexane conversion
(%)
Selectivity
(%)
Cyclohexanol/cyclohexanone
Cyclohexanol
CHHPa
Cyclohexanone
VPO-O
VPO-A
VPO-K
VPO-C
VPO-Z
-
5.0
101.0
374.7
42.0
-
53.8
83.3
72.6
93.9
97.0
58.0
42.8
35.1
25.7
51.2
40.4
56.6
31.2
61.1
24.7
1.6
0.6
33.7
13.2
24.1
36.3
71.3
1.04
1.9
Al2O3
KIT-6
CeO2
ZrO2
0.173
0.462
0.222
0.104
21.4
2.1
Reaction conditions: H2O2: cyclohexane mole ratio=5:1, cyclohexane/V mole ratio=500:1, temperature 333 K, 24 h.aCyclohexyl hydrogen peroxide.
phase whereas the smaller particles are those of CeO2. A similar kind
of morphology is observed in the case of VPO-Z also with large
platelets of the VPO phase having much smaller particles of ZrO2
dispersed on them as evident from EDS data.
A similar variation of the cyclohexane conversion and selectivity of
products is observed (Table 2) when samples with different VPO
loadings on alumina are used as the catalyst. It is seen that the highest
conversion of 83.3% is observed in the case of VPO-A although the
surface area of the sample is lower that of VPO-A1 which shows a
conversion of only 23% indicating that the surface area is not a critical
factor. There is also a large variation of the cyclohexanol/cyclohexanol
ratio for the samples with the different VPO loadings, with VPO-A
displaying the highest ratio. The low cyclohexane conversions in the
case of VPO-A1 and VPO-A2 in spite of the higher surface areas can
probably be attributed to the fact that the dispersed VPO phase has a
very ill defined morphology and poor crystallinity as evident from the
SEM and XRD data. On the other hand although both VPO-A and VPO-
A3 have a very well defined agglomerated platelet structure of the
dispersed VPO phase the higher cyclohexane conversion in the case of
VPO-A can be attributed to the higher surface area.
In the case of the VPO phases dispersed on alumina with different
loadings (Fig. 3b), VPO-A3 which has the highest loading of 75% VPO,
displays a well defined rose petal agglomerated morphology. When
the VPO loading is reduced to 50% (VPO-A) the agglomerated platelet
structure is still retained although it is not as well defined as in the
case of VPO-A3. As the VPO loading is further reduced to 35% (VPO-
A2) the spherical particles of the alumina support begin to appear and
there is a distinct co-existence of the alumina particles and the ill
defined agglomerated platelet structure of the dispersed VPO phase.
Finally in the case of VPO-A1 the spherical particles of the alumina
support are predominantly observed with small platelets of the VPO
phase dispersed on them. The EDS analysis however showed in all the
cases the particles contained both the VPO and support phases.
It was observed during the experiments that the catalysts had gone
completely into solution and the total leaching of vanadium was
confirmed by ICP analysis. However, the recyclability of the catalysts
was checked by an experiment wherein fresh cyclohexane and H2O2
were added to the reaction mixture obtained after 24 h and the
reaction was then carried out for a further period of 24 h under
exactly identical conditions. It was observed in case of VPO-A that
although the conversions had reduced from 83.3% to 72.3% the
selectivity for cyclohexanol was in fact higher (52.4%) although the
cyclohexanol/cyclohexanone ratio decreased. Similarly in the case of
VPO-Z the total conversion of cyclohexane reduced to 89.3% from 97%
on recycling whereas in the case of VPO-O the conversion decreased
from 53.8 to 48%. These results clearly indicate that the catalytic
activity of the VOPO4·2H2O phase is dependent on whether it is used
as such or dispersed on a support and also on the nature of the support
used in spite of the vanadium being leached into the solution.
Otherwise all the different VPO solids should have given similar
activity once in solution. This very interesting variation in activity is
attributable to the different crystallinity and morphology of the VPO
phase obtained on different supports as well as the interaction of the
VPO phase with the support which probably affect the nature of the
catalytically active vanadium species obtained in solution in the
different cases. We are in fact planning a detailed investigation into
the nature of the vanadium species obtained in solution starting from
3.4. Catalytic activity
The catalytic activity of the different supported VPO phases for
cyclohexane conversion (Table 1) shows that the undispersed VPO-O
phase gives cyclohexanol as the predominant product with a
selectivity of 58% and only 1.6% for cyclohexanone at a cyclohexane
conversion of 53.8%. On dispersion on alumina however (VPO-A)
there is a marked increase in the cyclohexane conversion while the
higher selectivity for cyclohexanol is still retained. The higher
cyclohexane conversion can therefore be attributed to the higher
surface area (101 m2/g) of VPO-A. In the case of the VPO dispersed on
KIT-6 (VPO-K), the cyclohexane conversion of 72.6% though quite
substantial is the lowest amongst all the samples in spite of having the
highest surface area of 374.7 m2/g. Moreover the selectivity profile
changes quite significantly with cyclohexanol and cyclohexanone
being obtained in almost equal quantities indicating that the nature of
the support has a profound influence on the conversion and product
selectivity. It is very interesting that in the case of VPO-C and VPO-Z in
spite of the relatively low surface areas the cyclohexane conversion is
very high with values of 93.9% and 97.0% respectively and there is
preferential conversion to cyclohexanol though not to the same
extent as that obtained with VPO-A.
Table 2
Variation of cyclohexane conversion and selectivity of products using dispersed phases with different VPO/alumina ratios as catalysts.
Catalyst
VPO loading (%)
Surface area (m2/g)
Cyclohexane conversion (%)
Selectivity (%)
Cyclohexanol
Cyclohexanol/cyclohexanone
CHHPa
Cyclohexanone
VPO-A1
VPO-A2
VPO-A
25
35
50
75
139.8
108.4
101.0
56.4
23.0
40.0
83.3
59.0
16.3
35.7
42.8
49.2
54.3
54.8
56.6
36.0
29.4
9.5
0.6
0.55
3.8
71.3
3.3
VPO-A3
15.0
Reaction conditions: H2O2/cyclohexane mole ratio=5:1, cyclohexane/V mole ratio=500:1, temperature 333 K, 24 h.
aCyclohexyl hydrogen peroxide.

P. Borah et al. / Catalysis Communications 12 (2010) 110–115
115
[3] P.A. Agaskar, L. DeCaul, R.K. Grasselli, Catal. Lett. 23 (1994) 339–351.
the various dispersed VPO phases and the results will be communi-
cated later when complete.
[4] S. Dasgupta, M. Agarwal, A. Datta, Micropor. Mesopor. Mater. 67 (2004) 229–234.
[5] P. Borah, A. Datta, Appl. Catal. A: Gen. 376 (2010) 19–24.
[6] Unnikrishnan R. Pillai, Endalkachew Sahle-Demessie, New J. Chem. 27 (2003) 525.
[7] R. Unnikrishnan, Pillai, Endalkachew Sahle-Demessie, Chem. Commun. (2002)
2142–2143.
4. Conclusions
[8] E.C. Zampronio, M.C.A.F. Gotardo, M.D. Assis, H.P. Oliveira, Catal. Lett. 104 (2005)
53–56.
[9] A. Bellifa, D. Lahcene, Y.N. Tchenar, A. Choukchou-Braham, R. Bachir, S. Bedrane, C.
Kappenstein, Appl. Catal. A: Gen. 305 (2006) 1–6.
[10] R.S. da Cruz, M.M. Dauch, U. Schuchardt, R. Kumar, Stud. Surf. Sci. Catal. 130
(2000) 1037.
[11] P. Selvam, S.E. Dapurkar, J. Catal. 229 (2005) 64–71.
[12] P. Selvam, S.E. Dapurkar, Appl. Catal. A: Gen. 276 (2004) 257–265.
[13] G.Z. Ladwig, Z. Anorg. Allg. Chem. 338 (1965) 266–278.
[14] A. Datta, M. Agarwal, S. Dasgupta, J. Mol. Catal. A: Chem. 198 (2003) 205–214.
[15] A.R. Antonio, R.L. Barbour, P.R. Bulm, Inorg. Chem. 26 (1987) 1235–1243.
It is evident that the nature and amount of the support on which
the exfoliated VPO-O is dispersed has a profound influence on the
crystallinity, morphology and catalytic activity of the dispersed VPO
phase. The present work therefore provides a novel method for
dispersing the VOPO4·2H2O phase on different supports and tailoring
the catalytic activity of these phases for the selective oxidation of
cyclohexane through the nature of the support and the VPO/support
ratio.
References
[1] R.L.Bergmen, N.W.Frisch, US Pat. 3,293,268 (1996) assigned to Princeton chemical
research.
[2] E. Bordes, Catal. Today 16 (1993) 27–38.