Effect of group 13 compounds and bulky organic oxygenates as structural promoters for the selective oxidation of n-butane with vanadium phosphorus oxide catalysts

Article abstract of DOI:10.1039/b105473m

A detailed study of n-butane oxidation over vanadium phosphate catalysts derived from in situ activation of VOHPO₄·0.5H₂O modified with group 13 compounds or 2-norbonanemethanol is described and discussed. Two methods of preparation of VOHPO₄·0.5H₂O were studied: (a) the VPO method, in which V₂O₅ is reacted with H₃PO₄ in isobutanol, and (b) the VPD method, in which VOPO₄·2H₂O is reacted with isobutanol. Group 13 additives (B, Al, Ga, In) were found to change the morphology of the VPO catalyst precursor from platelet-like to the rosette morphology usually associated with the VPD type of preparation. Although all act as structural modifiers, the In- and Ga-doped materials give improved catalyst performance for the synthesis of maleic anhydride from the selective oxidation of n-butane. The Al, Ga and In additives were added as acetylacetonates and it was found that addition of 1% acetylacetonate also enhances the catalyst performance, but does not act as a structural promoter. In view of this, the addition of 2-adamantanol and 2-norbonane methanol was investigated for the VPD-type preparation method. The addition of 2-norbonanemethanol was found to enhance the catalyst surface area of the activated catalyst, but the catalyst performance, with respect to both n-butane conversion and maleic anhydride selectivity, was not improved and this is considered to be due to the alcohol being strongly adsorbed on the catalyst surface.

Full text of DOI:10.1039/b105473m

View Article Online / Journal Homepage / Table of Contents for this issue
Effect of group 13 compounds and bulky organic oxygenates as
structural promoters for the selective oxidation of n-butane with
vanadium phosphorus oxide catalysts
Javier Cabello Sanchez, J. Antonio Lopez-Sanchez, Richard P. K. Wells, Colin Rhodes
and Graham J. Hutchings*
Department of Chemistry, Cardiff University, P.O. Box 912, Cardiff, UK CF10 3TB
Received (in London, UK) 10th July 2001, Accepted 14th September 2001
First published as an Advance Article on the web
A detailed study of n-butane oxidation over vanadium phosphate catalysts derived from in situ activation of
VOHPO Á 0.5H O modified with group 13 compounds or 2-norbonanemethanol is described and discussed.
4
2
Two methods of preparation of VOHPO
with H PO in isobutanol, and (b) the VPD method, in which VOPO
4
Á 0.5H
2 2 5
O were studied: (a) the VPO method, in which V O is reacted
3
4
4
2
Á 2H O is reacted with isobutanol. Group
1
3 additives (B, Al, Ga, In) were found to change the morphology of the VPO catalyst precursor from platelet-
like to the rosette morphology usually associated with the VPD type of preparation. Although all act as
structural modifiers, the In- and Ga-doped materials give improved catalyst performance for the synthesis of
maleic anhydride from the selective oxidation of n-butane. The Al, Ga and In additives were added as
acetylacetonates and it was found that addition of 1% acetylacetonate also enhances the catalyst performance,
but does not act as a structural promoter. In view of this, the addition of 2-adamantanol and 2-norbonane
methanol was investigated for the VPD-type preparation method. The addition of 2-norbonanemethanol
was found to enhance the catalyst surface area of the activated catalyst, but the catalyst performance, with
respect to both n-butane conversion and maleic anhydride selectivity, was not improved and this is considered
to be due to the alcohol being strongly adsorbed on the catalyst surface.
an electronic promoter by forming solid solutions of the type
[(VO)1 M ] P O , where M represents a promoter cation,
Introduction
19
� x x 2
2 7
4
þ
Vanadium phosphate compounds are used commercially as
catalysts for the selective oxidation of n-butane for the pro-
duction of maleic anhydride. At present, vanadium phosphates
remain the only example of a heterogeneous catalyst employed
commercially for the partial oxidation of an alkane. It is,
therefore, not surprising that these catalysts have been exten-
in which the promoter cation substitutes for V . This may
form defects in the structure which can act as new active sites
or, alternatively, the metal cations may affect the redox
properties of the surface. Recently, we have reported detailed
studies on the structural and electronic effects of Co and Fe as
20,21
catalyst promoters.
However, it is interesting to note that
1
–8
sively studied.
The current catalyst is prepared by in situ
Á 0.5H O in a butane–air mixture under
which form a complex mixture of
group 13 compounds have, to date, been neglected as addi-
tives, although it is known that Ga O has been found to be
very active for the activation of carbon–hydrogen bonds in
activation of VOHPO
4
–7
2
2
3
5
reaction conditions
VO) and VOPO
been made recently concerning the nature of active sites for the
6
22
(
2
P
2
O
7
4
phases. Considerable progress has
alkanes. In this paper, we investigate the effect of group 13
compounds as catalyst promoters. In particular, the interplay
between structural and redox effects is examined in detail. In
addition, the effect of using bulky alcohols as structural
promoters is also explored.
pure vanadium phosphate catalysts. In situ Raman spectro-
þ
4
5 þ
scopy suggests that specific combinations of V
and V
phases are a necessary requirement for the catalyst to exhibit
high activity and high selectivity for the formation of maleic
5
8
5 þ
anhydride. In addition, Coulston et al. have shown that V
is essential for the initial activation of n-butane.
Although considerable progress has been achieved with pure
vanadium phosphate, it should be noted that the catalysts used
industrially contain a broad range of structural and electronic
promoters. In particular, a wide range of cations have been
added to vanadium phosphate catalysts and beneficial effects
Experimental
Catalyst preparation
Preparation of VOHPO
prepared using the VPO and VPD preparation methods.
the VPO method, V (11.8 g, Aldrich) was refluxed with
PO (16.49 g, 85%, Aldrich) in isobutanol (250 ml,
Aldrich) for 16 h. The light blue solid, VOHPO O, was
Á 0.5H
recovered by vacuum filtration, washed with isobutanol (200
4
.
0.5H O. VOHPO Á 0.5H O was
2
4
2
9
10
6,23
have been reported with: Co, Mo; Co, Cd, Ni; Zn, Bi, Cu,
Li;
In
11–13
14
15
16
17
V; Zr; La and Nb. Most of this literature has
O
2 5
4
been reviewed and the promoters are considered to improve
catalyst performance via two mechanisms. Firstly, the pro-
moter can increase the surface area of the catalyst and,
therefore, acts as a structural promoter. This is achieved by the
metal cation acting as a phosphorus scavenger ensuring that
inactive phases, such as the low surface area VO(H PO )
H
3
4
4
2
�
1
ml) and ethanol (250 ml) and refluxed with water (9 ml g )
ꢀ
for 3 h, filtered hot and dried in air (110 C, 16 h). In the VPD
method, V O (11.8 g, Aldrich) was refluxed with H P 4 (115
2 5 3 O
g, 85%, Aldrich) in water (300 ml) for 8 h. The yellow solid,
2
4 2
1
8
phase, are not formed. Secondly, the metal cation can act as
1
528
New J. Chem., 2001, 25, 1528–1536
DOI: 10.1039/b105473m
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2001

View Article Online
VOPO
with cold water and acetone and dried in air (110 C, 16 h).
4
Á 2H
2
O, was recovered by vacuum filtration, washed
ꢀ
generator with a Cu-Ka source fitted with an Inel CPS 120
hemispherical detector. BET (Brunauer–Emmett–Teller) sur-
face area measurements using nitrogen adsorption were carried
out using a Micromeritics ASAP 2000 instrument. Raman
spectra were obtained using a Renishaw Ramanscope Spec-
The resulting dried VOPO
isobutanol (40 ml) for 21 h. The blue solid, VOHPO
was recovered by vacuum filtration, washed with isobutanol
4
Á 2H
2
O (2 g) was refluxed in
4
Á 0.5H O,
2
�
1
þ
(200 ml) and ethanol (250 ml), refluxed in water (10 ml g
ꢀ
for 2 h, filtered hot and dried (110 C, l6 h).
)
trograph fitted with a green Ar laser (l ¼ 514.532 nm).
Scanning electron microscopy (SEM) was performed using a
Hitachi S-2460 N SEM, operating at an accelerating voltage
of 25 kV. Inductively coupled plasma mass spectroscopy
.
4 2
Preparation of VOHPO 0.5H O containing group 13
(ICPMS) was performed using a Perkin Elmer Elan 5000 ICP
mass spectrometer.
compounds. The VPO method, described above, was used
and the appropriate group 13 compounds were added,
together with the V
following compounds were used: aluminium acetylacetonate
0.41 g, Aldrich), gallium acetylacetonate (0.51 g, Aldrich),
indium acetylacetonate (0.54 g, Aldrich), tributylborate (0.31 g,
Aldrich), acetylacetone (0.129 g, Aldrich), tributylaluminate
2 5
O at the start of the preparation. The
(
Results and discussion
Vanadium phosphate catalysts modified with group 13
compounds
(
0.31 g, Aldrich). The materials were analysed using ICPMS
and the promoter=V molar ratios for the acetylacetonate
compounds were found to be Al=V ¼ 0.014, Ga=V ¼ 0.014,
In=V ¼ 0.012 and, for the tributyl compounds Al=V ¼ 0.01 and
B=V ¼ 0.009.
Catalyst precursor characterisation. Catalyst precursors
containing ca. 1 mol% group 13 compounds were prepared
using the modified VPO method and the laser Raman
spectra and PXRD patterns are shown in Fig. 1 and 2,
respectively. The laser Raman spectra (Fig. 1) do not show
much detail due to the fluorescence caused by the retention
of the alcohol within the catalyst precursor crystallites.
.
4 2
Preparation of VOHPO 0.5H O with mixed alcohols. The
VPD method, described above, was used and the appropriate
bulky alcohol (2-norbonanemethanol, 2-adamantanol) was
�
1
added with the isobutanol at the stage when VOPO
was refluxed in the alcohol.
4
2
Á 2H O
However, the main band at a Raman shift of 984 cm
minor bands at Raman shifts of 1151, 1110 and 340 cm
with
�
1
24,25
can all be assigned to VOHPO4 Á 0.5H
2
O
(1151 and 1110
� 1
�
1
� 1
cm , V–O–P; 984 cm , P–O stretch; 340 cm , coupled
V–O and P–O bonding mode). All the reflections in the
Catalyst testing and characterisation
The oxidation of n-butane was carried out using a micro-
reactor and a standard volume of catalyst (0.7 ml). n-Butane
and air were fed into the reactor via calibrated mass flow
controllers to give a feedstock composition of 1.8% n-butane
in air. The products were fed via heated lines to an on-line gas
chromatograph for product analysis. The reactor comprised a
stainless steel tube with the catalyst held in place by plugs of
quartz wool. A thermocouple was located in the centre of the
PXRD patterns (Fig. 2) can be indexed to VOHPO Á
4
26
0.5H O. No reflections are observed that can be indexed to
2
separate phases of group 13 compounds. The PXRD
patterns show an interesting difference when the group 13
additive is present (Fig. 2) compared to the unpromoted
VPO material. For the unpromoted VOHPO Á 0.5H O, the
4
2
main reflection corresponds to the d-spacing of 0.57 nm and
can be indexed to the (001) plane. However, addition of
group 13 additives, in all cases leads to the reflections at
d ¼ 0.29 nm, indexed to the (220) plane, being the most
intense reflection in the pattern, as is typically observed with
ꢀ
catalyst bed and temperature control was typically ‹1 C.
Carbon mass balances of 5 95% were typically observed.
VOHPO Á 0.5H O precursors were activated in situ (1.8% n-
4
2
�
1
butane in air, 2500 h ) at 400 C for 72 h.
ꢀ
23
the VPD method of preparation. To determine if the
A number of physical techniques were employed to char-
acterise the catalyst microstructure. Powder X-ray diffraction
organic ligands, i.e. the butoxy or acetylacetonate, could
cause this restructuring, a separate experiment was carried
out in which 1% acetylacetone was added to the VPO
(
PXRD) was performed using an Enraf Nonius FR590 X-ray
Fig. 1 Laser Raman spectra of group 13-modified precursors.
New J. Chem., 2001, 25, 1528–1536
1529

View Article Online
Fig. 2 Powder XRD patterns of group 13-modified precursors.
preparation method. The PXRD pattern, shown in Fig. 2, is
identical to that for VOHPO O prepared analogously
conversion of n-butane gradually increased [Fig. 4(a)] and the
selectivity to maleic anhydride also increased [Fig. 4(b)].
The results are summarised in Table 1. The surface area of the
activated catalysts was found to increase markedly for all the
catalysts tested, with the highest surface area being observed
4
Á 0.5H
2
in the absence of acetylacetone. Hence, it can be concluded
that the changes observed in the PXRD pattern can be
ascribed to the presence of the group 13 element. A similar
PXRD pattern was observed for the aluminium-doped
material prepared using either aluminium acetylacetonate or
tributylaluminate, which is a further indication that it is the
group 13 element that is responsible for the differences
observed. It is interesting to note that the PXRD pattern of
the group 13-doped precursors is very similar to that of
2
� 1
with the B-doped material (31 m
exhibited intrinsic activities (mol maleic anhydride m
and specific activities (mol maleic anhydride g
g
). All the catalysts
2
�
� 1
h )
) in
�
1
� 1
h
similar ranges to that of the undoped catalysts. However, the
specific activity of the In-doped material and the intrinsic
activity of the Ga-doped material are both higher than the
undoped catalyst or the catalyst prepared with 1%
acetylacetone as additive. The slight promotional effect
observed with In with respect to activity, and also the small
promotional effect on selectivity observed with B, are much
less marked than the effects observed previously with Co and
VOHPO
4
Á 0.5H
2
O prepared using the VPD method (Fig. 2).
SEM micrographs for these precursors are shown in Fig. 3.
The materials containing group 13 additives all display thin
platelets of VOHPO Á 0.5H O and often these are associated
4
2
with each other in the form of small spheres, similar to the
4
rosette morphology of undoped VOHPO Á 0.5H O prepared
Mo additives.
The PXRD patterns for the activated catalysts are shown in
Fig. 5 and the materials all have similar patterns that can be
4
2
by the VPD method. This is particularly marked for the
aluminium-doped material. The SEM of the VPD-prepared
material is shown for comparison and the morphology
observed is consistent with the PXRD patterns in which the
26
.
indexed to poorly crystalline (VO)
2
P
2
O
7
There does not
appear to be any marked difference between the XRD patterns
of the undoped and doped activated catalysts. The laser
Raman spectra of the activated catalysts do, however, show
very clear differences (Fig. 6). The B-, Al-, Ga- and In-
doped activated catalysts all show the two main Raman shifts
26
main reflection can be indexed to the (220) plane. In
contrast, the SEM micrographs for VOHPO
4
Á 0.5H
2
O
prepared using the VPO method, in the presence or absence
of 1% acetylacetone, have a more plate-like morphology
�
1
24,25
[Fig. 3(e) and (f)] which is in agreement with the PXRD
pattern, for which the main reflection can be indexed to the
at 926 and 1189 cm that can be assigned to (VO) P O .
2 2 7
The two undoped activated catalysts all show additional
5 þ
2
001) phane. Surface area measurements (Table 1) show
6
(
bands, and these can be assigned to V
phases being
It should be noted that the broad band observed
2
4,25
that the addition of 1% acetylacetone does not affect the
precursor surface area, but the addition of the group 13
additives does increase the surface area in all cases. The
surface area of the B-doped material is markedly higher than
the other additives, but this was prepared using
tributylborate as the boron source as boron acetylacetonate
is unavailable. The Al-doped material similarly prepared
present.
at 933 cm
�
1
for the undoped material is considered to be a
�
1
combination of two bands at 927 and 939 cm , which are
resolved in the Raman spectrum for the activated undoped
catalyst containing 1% acetylacetonate. The bands at 1089,
�
1
25
994, 939 and 470 cm
the bands at 1089, 1072, 1016 and 939 cm
can be assigned to a -VOPO
and
can be assigned
II
4
�
1
2
25
using tributylaluminate also has a higher surface area (12 m
�
to g-VOPO4 . Thus, the presence of the group 13 additive, in
combination with the organic ligand, appears to increase the
level of reduction of the catalyst surface. However, this does
not appear to affect the intrinsic activity of the catalyst
significantly. Such a finding has been previously noted for
1
2
� 1
g
), in contrast to 9 m
g for the Al-doped material
prepared using aluminium acetylacetonate. This indicates
that the surface area of the precursor can be sensitive to the
nature of the organic ligand added with the promoter.
26
catalysts prepared using the VPO and VPD methodologies
since, although the catalyst comprised different amounts of the
(VO) and VOPO phases, they exhibited the same
n-Butane oxidation and post-reaction characterisation. The
group 13-doped VOHPO Á 0.5H O precursors were used as
2
P
2
O
7
4
4
2
27
catalyst precursors for the oxidation of n-butane. The catalysts
were activated in situ in the reactor with flowing n-butane
intrinsic activity. This has been assigned to the activated
catalysts having the same surface composition, which is
considered to be amorphous in nature.
ꢀ
(
1.8%) in air at 400 C for 72 h. During this time, the
1
530
New J. Chem., 2001, 25, 1528–1536

View Article Online
Fig. 3 SEM micrographs of modified precursors: (a) Al-doped, prepared using aluminium acetylacetonate; (b) Ga-doped; (c) In-doped; (d) B-
doped; (e) 1% acetylacetonate added to preparation; (f) standard unpromoted VPO method with isobutanol; (g) standard unpromoted VPD
method with isobutanol.
New J. Chem., 2001, 25, 1528–1536
1531

View Article Online
a
Table 1 n-Butane oxidation over catalysts derived from group 13-doped vanadium phosphates
2
Surface Area=m g
� 1
Selectivity (%)
Catalyst
additive
Activated
catalyst
n-Butane
conversion (%)
Intrinsic activity=
10
Specific activity=
b
10
b
� 5
b
mol MA m
� 2 � 1
� 4
� 1 � 1
mol MA g h
Precursor
MA
CO
CO
2
h
None
1
7
7
14
9
10
9
25
21
31
21
16
25
64
55
55
58
54
67
56
59
62
52
53
57
17
16
14
15
16
17
27
22
22
33
31
22
1.79
1.96
1.42
1.85
2.22
1.90
4.48
4.18
4.40
3.89
3.55
4.85
%a acetylacetone
B
Al
Ga
In
d
d
d
a
ꢀ
� 1
b
c
d
1
.8% n-butane in air, 400 C, 2500 h , 72 h. MA ¼ maleic anhydride. Tributylborate. Acetylacetonate.
ꢀ
� 1
Fig. 4 Oxidation of n-butane to maleic anhydride showing the effect of catalyst activation (400 C, 18% n-butane, 2500 h ) on (a) n-butane
conversion and (b) maleic anhydride selectivity.
2
6
With respect to the intrinsic activity of the group 13-doped
activated catalysts, it is apparent that, with the exception of the
Ga-doped catalyst, the activities are very similar for the
undoped and Al- and In-doped materials. This indicates that
these additives are acting primarily as structural promoters
rather than as electronic promoters. This is consistent with
effects observed in the PXRD patterns and SEM images for
the precursor materials. The inclusion of a group 13 element,
in combination with the organic ligand, in the VPO method of
preparation appears to result in a material that would
normally be expected for the VPD methodology. However, it
is notable that: (i) the inclusion of 1% acetylacetone in the
preparation leads to an increase in the intrinsic activity of the
doped material by ca. 10% and (ii) the use of a tributylalu-
minate rather than aluminium acetylacetonate increases the
surface area of the VOHPO
4
Á 0.5H
2
O precursor from 9 to 12
2
� 1
m g . Both of these effects indicate that structural effects
may be expected to arise from the addition of bulkier oxyge-
nates to the preparation method and this is explored in the
next section.
1
532
New J. Chem., 2001, 25, 1528–1536

View Article Online
Fig. 5 Powder XRD patterns of group 13-modified activated catalysts.
Fig. 6 Laser Raman spectra of group 13-modified activated catalysts.
Vanadium phosphate catalysts modified with 2-adamantanol and
2-norbonanemethanol
isobutanol) and the PXRD patterns are shown in Fig. 7. All the
reflections can be indexed to VOHPO O, although, in the
preparation with the highest concentration of 2-adamantanol, it
is clear that some crystalline 2-adamantanol is present. The laser
Raman spectra showed the standard bands that can be assigned
4
Á 0.5H
2
Catalyst precursor characterisation. Previous studies using
-octanol in place of isobutanol in the VPD method of
1
preparation of VOHPO
chain alcohol consistently gave higher surface area catalyst
2
4,25
4
Á 0.5H
2
O have shown that the longer
to VOHPO Á 0.5H O.
The addition of 2-adamantanol did
4
2
not markedly affect the surface area of the precursor (Table 2),
although the addition of a small amount appears to decrease the
surface area. In the powder X-ray diffraction pattern, it is
notable that the precursors prepared using 2-adamantanol all
give a pattern for which the d ¼ 0.57 nm reflection is the most
intense, and this is assigned to the (001) plane of VOH-
2
3
precursors and activated catalysts. This effect is probably
caused by the alcohol being occluded within the layer
structure of the hemihydrate. Catalyst precursors were
prepared using the VPD method in which 2-adamantanol (1)
and 2-norbonanemethanol (2) were added to the isobutanol.
PO
4
Á 0.5H
2
O. These PXRD patterns differ from that of the
standard material prepared by the VPD method in the absence
of 2-adamantanol, for which the reflection at d ¼ 0.29 nm is the
most intense. The PXRD patterns for the precursors prepared
using 2-adamantanol are similar to those for the standard VPO
catalyst (Fig. 2) and also to those prepared using the VPD
method with alcohols other than primary alcohols as reducing
agents.
reducing agent in these reactions and, consequently, this affects
the morphology of the VOHPO O. In view of this, fur-
23,26
2-Adamantanol. A series of catalyst precursors were prepared
This indicates that the 2-adamantanol is acting as the
using the VPD method in which 2-adamantanol was added to
the isobutanol during the alcohol reduction step of the pre-
paration. Three different concentrations of 2-adamantanol were
studied (0.01, 0.06 and 0.2 mol 2-adamantanol per mol
Á 0.5H
4
2
ther studies were conducted using a bulky primary alcohol,
2-norbonanemethanol.
New J. Chem., 2001, 25, 1528–1536
1533

View Article Online
Fig. 7 Powder XRD patterns of 2-adamantanol precursors.
Table 2 Surface area of catalyst precursors
with the PXRD pattern of VOHPO Á 0.5H O prepared in the
23
expected, since both alcohols are primary alcohols. The
4
2
absence of 2-norbonanemethanol (Fig. 8). This is to be
2
Surface area=m g
� 1
Catalyst precursor
laser Raman spectra of the precursors prepared with 2-nor-
bonanemethanol show no differences to the standard
undoped catalyst precursor and all the bands can be assigned
to VOHPO Á 0.5H O. The most significant effect observed
Undoped VPD
18
14
20
18
14
12
27
20
9
b
0
0
0
3
1
1
3
5
.01
b
.06
.20
b
4
2
c
when 2-norbonanemethanol is included in the preparation
concerns the surface area of the precursor (Table 2). The
addition of a small amount of 2-norbonanemethanol (3, 10
vol%) leads to a marked decrease in the surface area, but
higher concentrations up to 33 vol% lead to an increase
in surface area. This indicates that the two alcohols used have
a complex effect on the porosity of the catalyst precursor.
c
0
c
6
c
3
c
0
a
b
Standard preparation procedure. Mol 2-adamantanol=mol iso-
c
butanol. % 2-norbonanemethanol.
n-Butane oxidation and post-reaction characterisation of
catalysts prepared using 2-norbonanemethanol. Representative
2
-Norbonanemethanol. A series of catalysts were prepared
VOHPO
4
Á 0.5H
O precursors prepared with the addition of
2
using the VPD method in which 2-norbonanemethanol was
added to the isobutanol. Five different concentrations of 2-
norbonanemethanol were studied (3, 10, 16, 33 and 50 vol%
2
are shown in Fig. 8. All the reflections can be indexed to
2-norbonanemethanol (0 vol%, standard VPD catalyst; 10
vol%, reduction in precursor surface area; 16 vol%, increase
in precursor surface area) were used as catalyst precursors
for the oxidation of n-butane. The catalysts were activated
in situ in flowing n-butane (1.8%) in air at 400 C for 12 h.
During this time, the n-butane conversion and maleic
anhydride selectivity increased and stabilised. The results
-norbonanemethanol in isobutanol) and the PXRD patterns
ꢀ
VOHPO Á 0.5H O and the most intense reflection is for
4
2
d ¼ 0.29 nm [assigned to the (220) plane], which is consistent
Fig. 8 Powder XRD patterns of 2-norbonanemethanol precursors.
1
534
New J. Chem., 2001, 25, 1528–1536

View Article Online
from the activated catalysts are given in Table 3. The surface
areas of the activated catalysts were found to be significantly
higher than those of the precursors, and the two catalysts
prepared using 2-norbonanemethanol gave much higher
surface areas than the standard catalyst. The PXRD patterns
and the laser Raman spectra of the activated catalysts are
shown in Fig. 9 and 10 respectively. All reflections in the
bands in the Raman spectra can similarly be assigned to
(VO) P O .
2
4,25
It is apparent that these characterisation
methods do not show any major differences in the activated
catalysts, which can be classified as poorly crystalline
2
2
7
2 2 7
(VO) P O . However, the catalytic performance of the 2-
norbonanemethanol catalysts is, unfortunately, poorer than
the standard catalyst. In particular, both the n-butane
conversion and selectivity to maleic anhydride decrease with
2 2 7
PXRD patterns can be assigned to (VO) P O and all the
Table 3 n-Butane oxidation over catalysts derived from group 13-doped vanadium phosphates
2
Surface Area=m g
� 1
Selectivity (%)
Catalyst=
% 2NBM
Activated
Precursor catalyst
Intrinsic activity=
10
Specific activity=
b
a
b
� 5
b
mol MA m
� 2 � 1
� 4
� 1 � 1
n-Butane conversion (%) MA
CO CO
2
h
10
mol MA g
h
0
1
1
18
12
27
22
27
35
65
60
45
58
56
55
23
27
22
19
17
23
2.39
1.70
1.01
5.26
4.59
3.54
0
6
a
.8% n-butane in air, 400 C, 2500 h� 1, 72 h, 2NBM ¼ 2-norbonanemethanol. MA ¼ maleic anhydride.
ꢀ
b
1
Fig. 9 Powder XRD patterns of 2-norbonanemethanol-activated catalysts.
Fig. 10 Laser Raman spectra of 2-norbonanemethanol-activated catalysts.
New J. Chem., 2001, 25, 1528–1536
1535

View Article Online
increasing 2-norbonanemethanol concentration, even though
the surface area of the catalysts increases. Consequently,
3
G. Centi, F. Trifiro, G. Busca, J. Ebner and J. Gleaves, Faraday
Discuss., 1989, 87, 215.
G. J. Hutchings, Appl. Catal. A, 1992, 721.
G. J. Hutchings, A. Desmartin-Chomel, R. Olier and J. C. Volta,
Nature, 1994, 368, 41.
C. J. Kiely, A. Burrows, G. J. Hutchings, K. E. Bere, J. C. Volta,
A. Tuel and M. Abon, Faraday Discuss., 1996, 105, 103.
G. Centi, Catal. Today, 1994, 16, 1.
G. W. Coulston, S. R. Bare, H. Kung, K. Birkeland, G. J. Bethke,
R. Harlow, N. Herron and P. L. Lee, Science, 1997, 275, 191.
G. J. Hutchings and R. Higgins, J. Catal., 1996, 162, 153.
R. Lemel and J. Vekemans, US Pat., 3987063, 1976.
E. M. Boghosian, US Pat., 3862146, 1974.
�
1
� 1
� 1
4
5
both the specific activity (mol maleic anhydride g
and intrinsic activity (mol maleic anhydride m
h
h
)
)
�
2
decrease when 2-norbonanemethanol is added to isobutanol.
The results of this study confirm that the use of bulky
6
23
reducing agents, as previously demonstrated with 1-octanol,
7
8
can significantly enhance the surface area of activated vana-
dium phosphate catalysts. However, although they can, in
principle, act as structural promoters, it is apparent that very
bulky alcohols do not enhance catalyst performance. This may
be due to the bulky alcohol being strongly adsorbed on the
catalyst surface from where it is not readily desorbed during
the activation process.
9
0
1
1
1
12 C. A. Udovich and R. J. Vertolacini, US Pat., 4328126, 1982.
13 E. C. Kilberger, S. R. Dolhyj and N. J. Bremer, US Pat., 4219484,
1
980.
N. J. Bremer, E. C. Milberger and S. R. Dolhyj, US Pat., 4002650,
977.
1
4
1
1
1
1
1
5
6
7
8
A. J. Jurewicz and L. B. Young, US Pat., 3932305, 1976.
R. Higgins and G. J. Hutchings, US Pat., 4222945, 1980.
R. Higgins and G. J. Hutchings, US Pat., 4147661, 1979.
J. K. Bartley, C. Rhodes, C. J. Kiely, A. F. Carley and G. J.
Hutchings, Phys. Chem. Chem. Phys., 2000, 2, 4999.
M. Otake, US Pat., 4337173, 1982.
´
S. Sajip, J. K. Bartley, A. Burrows, M.-T. Sananes-Schulz, A.
Tuel, J. C. Volta, C. J. Kiely and G. J. Hutchings, New J. Chem.,
2001, 25, 125.
S. Sajip, J. K. Bartley, A. Burrows, C. Rhodes, J. C. Volta, C. J.
Kiely and G. J. Hutchings, Phys. Chem. Chem. Phys., 2001, 32143.
J. S. J. Hargreaves, G. J. Hutchings, R. W. Joyner and S. H.
Taylor, Chem. Commun., 1996, 523.
M. T. Sananes, I. J. Ellison, S. Sajip, A. Burrows, C. J. Kiely, J. C.
Volta and G. J. Hutchings, J. Chem. Soc., Faraday Trans., 1996,
92137.
V. V. Guliants, J. B. Benzinger, S. Sundaresan, L. E. Wachs, J. M.
Jehng and J. E. Roberts, Catal. Today, 1996, 28, 275.
F. B. Abdelouahab, R. Olier, N. Guilhaume, F. Lefebvre and J. C.
Volta, J. Catal., 1992, 134, 151.
C. J. Kiely, A. Burrows, S. Sajip, G. J. Hutchings, M. T. Sananes,
A. Tuel and J. C. Volta, J. Catal., 1996, 162, 31.
G. J. Hutchings, J. K. Bartley, J. M. Webster, J. A. Lopez-San-
chez, D. J. Gilbert, C. J. Kiely, A. F. Carley, S. M. Howdle,
S. Sajip, S. Caldarelli, C. Rhodes, J. C. Volta and M. Poliakoff,
J. Catal., 2001, 197, 232.
Conclusions
Group 13 additives (B, Al, Ga, In) are found to change the
morphology of the VPO catalyst precursor (prepared by the
reaction of V O with H PO in isobutanol) from plate-like to
2 5 3 4
rosette morphology. The rosette morphology is normally
associated with the VPD catalyst precursor (prepared by the
reaction of VOPO
additives act as structural modifiers, the In- and Ga-doped
materials also enhance the catalyst performance for the selec-
tive oxidation of n-butane.
The addition of 2-norbonanemethanol in the VPD precursor
preparation was found to affect the surface area of the mate-
rial, but the catalyst performance was not enhanced. This is
considered to be due to the alcohol being strongly adsorbed on
the catalyst surface.
1
2
9
0
4
2
Á 2H O with isobutanol). Although, all the
2
2
2
1
2
3
24
25
26
27
References
1
2
Forum on Vanadium Pyrophosphate Catalysts, ed. G. Centi, Catal.
Today, 1994, 16.
E. Borders, Catal. Today, 1987, 1499.
1
536
New J. Chem., 2001, 25, 1528–1536