12-Tungstophosphoric acid/zirconia - A highly active stable solid acid - Comparison with a tungstated zirconia catalyst

Article abstract of DOI:10.1039/b200722c

A highly active and stable zirconia supported 12-tung-stophosphoric acid catalyst is found to be 2-3 times more active in benzylation and acylation reactions than a tungstated zirconia catalyst.

Full text of DOI:10.1039/b200722c

1
2-Tungstophosphoric acid/zirconia—a highly active stable solid
acid—comparison with a tungstated zirconia catalyst†
a
a
a
b
c
Biju M. Devassy, S. B. Halligudi,* S. G. Hegde, A. B. Halgeri and F. Lefebvre
a
Inorganic Chemistry & Catalysis Division, National Chemical Laboratory, Pune -411 008, India.
E-mail: halligudi@cata.ncl.res.in
Indian Petrochemicals Corp, Ltd, Baroda, India
Laboratoire de Chemie Organométallique de Surface, CNRS-CPE, Villeurbanne Cedex, France
b
c
Received (in Cambridge, UK) 27th January 2002, Accepted 3rd April 2002
First published as an Advance Article on the web 16th April 2002
A highly active and stable zirconia supported 12-tung-
stophosphoric acid catalyst is found to be 2–3 times more
active in benzylation and acylation reactions than a tung-
stated zirconia catalyst.
phthalene (2-MN) with acetic anhydride (Ac
the probe reactions.
2
O) were used as
Catalysts were prepared by wet impregnation of zirconia in a
methanolic solution of 12-TPA. The zirconia support was
prepared by the hydrolysis of an aqueous solution of zirconyl
chloride with a 10 M ammonia solution and the precipitate was
washed free of chloride and dried at 393 K for 24 h. A series of
catalysts were prepared by suspending a known amount of dried
zirconia in a methanolic solution of 12-TPA (4 ml methanol for
1 g of dried zirconia). This mixture was stirred in a rotary
evaporator for 8–10 h followed by evaporation to dryness. The
resulting samples were dried at 393 K for 24 h, powdered and
calcined at 1023 K in air for 4 h. Catalysts with different TPA
loadings (5–30 wt% of dried zirconia) were prepared by
changing its concentration. Similarly, catalysts with optimum
TPA loading (15 wt%) were prepared using different solvents
as well as without solvent (neat–physical mixture). Tungstated
zirconia (12.6 or 15 wt% W) was also prepared using tungstic
acid dissolved in ammonia as the precursor. The properties of
the catalysts with different TPA loadings prepared in methanol
and that of tungstated zirconia calcined at 1023 K are presented
in Table 1.
x
Metal oxides promoted by sulfur compounds; especially SO -
1
promoted zirconia containing iron and manganese oxides as
promoters have recently attracted much attention owing to their
catalytic performances in reactions such as hydrocarbon
2
isomerization, alkylation, acylation, etc. Their poor stability,
tendency to form volatile sulfur compounds during catalysis and
3
regeneration by oxidation limit their applicability. Tungstated
zirconia, WO
x
/ZrO
2
(WZ) is found to be an alternative stable
4,5
solid acid catalyst,
but has lower activity than sulfated
zirconia.6 Although Fe and Mn oxides promote catalytic
activity of sulfated zirconia they do not promote tungstated
7
zirconia. Thus there is a clear need for the development of a
highly active and stable solid acid catalyst.
In the present work, a series of catalysts with different
2-tungstophosphoric acid [H PW12O40·21H O (12-TPA)]
3 2
1
loadings were prepared by the wet impregnation method using
zirconia as the support. Tungstated zirconia catalysts were
prepared for comparison with 12-TPA/ZrO
2
catalysts. A direct
31P MAS NMR spectra of 5 and 10 wt% TPA/ZrO
2
catalysts
correlation between surface structure and catalytic activity is
reported.
exhibit a peak at 212 ppm.‡ All other catalysts with different
TPA loadings showed an additional peak at 230 ppm. The peak
at 212 ppm (P–OH) is due to the presence of phosphorus in the
The state of 12-TPA on the catalysts was elucidated by 31P
MAS NMR spectroscopy (Bruker DSX-300 spectrometer) and
crystallographic identification of the samples were performed
using X-ray powder diffraction with Cu-Ka radiation (Rigaku
Model D/MAXIII VC, Japan, l = 1.5418 Å). Acidity and acid
strength distribution of tungstated zirconia and catalysts with
different TPA loadings were measured by TPD of ammonia.
Brönsted and Lewis acidity of catalysts with different TPA
loadings were monitored by pyridine adsorption FT-IR spec-
8
Keggin unit and the peak at 230 ppm is due to the presence of
phosphorus in the decomposition product.9
The acidity of the catalysts with different TPA loadings and
that of tungstated zirconia were measured by TPD of ammo-
nia.§ TPD of NH
sites i.e. acidity (NH
reaches its maximum for a 15 wt% TPA/ZrO
3
shows that the surface concentration of acid
2
2
3
nm ) increases with TPA loading and
catalyst (Table
2
1). The maximum concentration of acid sites was observed at a
surface density6 of 7.2 W nm , where a TPA monolayer
completely covered the surface. Although the P–OH intensity is
2
1
22
troscopy (NICOLET MODEL 60SXB, 4 cm resolution and
averaged over 500 scans). To assay the acidity and catalytic
activities of the solid acid catalysts, benzylation of phenol with
100% in 5 and 10 wt% TPA/ZrO
2
and 80% in 15 wt% TPA/
benzyl alcohol (PhCH
2
OH) and acylation of 2-methoxyna-
ZrO (Table 1), the actual amount of TPA in its Keggin form is
2
at a maximum for the latter catalyst and hence it shows the
highest acidity. The acidity decreases when the TPA loading
exceeds monolayer coverage (15 wt% TPA loading). Decrease
in acidity is due to the decomposition of TPA into its constituent
†
Electronic supplementary information (ESI) available: W and P analysis
31
in the reaction mixtures. Fig. S1: P MAS NMR. Fig. S2: FT-IR pyridine
adsorption spectra. Fig. S3: conversion and acidity vs. TPA loading. See
http://www.rsc.org/suppdata/cc/b2/b200722c/
3
oxides as indicated by the appearance of crystalline WO by
Table 1 Physicochemical properties of the catalysts
Surface
area/m g
Surface
density/W nm
Volume (%) of
Acidity/
NH
nm22
P–OH
intensity (%)
2
21
22
No.
Catalyst (wt%)
5 wt% TPA/ZrO
10 wt% TPA/ZrO
15 wt% TPA/ZrO
20 wt% TPA/ZrO
30 wt% TPA/ZrO
12.6 wt% WZ
tetragonal ZrO
2
3
1
2
3
2
40.2
46.3
53.2
52.3
44.9
76.0
44.6
3.2
5.5
7.2
9.8
14.7
6.2
21.9
75.6
84.2
94.6
100
3.44
4.68
5.21
2.65
2.41
2.62
2.16
100
100
80
54
32
2
2
2
2
a
4
a
5
6
7
100
100
—
—
a
15 wt% WZ
12.6
a
XRD indicates the presence of crystalline WO
3
1
074
CHEM. COMMUN., 2002, 1074–1075
This journal is © The Royal Society of Chemistry 2002

Fig. 1 Conversion and B/L ratio vs. TPA loading (catalyst prepared in
methanol). Reaction conditions: benzylation: phenol = 4.2 g, PhCH OH =
.8 g, catalyst = 0.08 g, temp. = 363 K, time = 1 h; acylation: 2-MN = 2
g, Ac O = 3.86 g, catalyst = 0.2 g, temp. = 393 K, time = 12 h.
2
Fig. 2 Conversion vs. P–OH intensity of 15 wt% TPA/ZrO catalyst
2
prepared in different solvents. Order of solvents: 0.5 M aqueous ammonia,
neat, diethyl ether, 1,4-dioxane, acetic acid, ethyl acetate, acetonitrile, THF,
acetone, water, methanol, DMF. Reaction conditions as in Fig. 1.
0
2
XRD and phosphorus( ) oxide (a decrease in P–OH intensity)
V
31
conversions increase with P–OH intensity and is maximum for
catalysts prepared in DMF or methanol as solvent. To confirm
reusability, the above catalyst was recycled a few (5–6) times,
with no loss in catalytic activity.
In summary, for the first time, we have prepared a highly
active and stable zirconia supported 12-tungstophosphoric acid
catalyst, which is more active than tungstated zirconia cata-
lyst.
by P MAS NMR spectroscopy (Table 1) and the accumulation
of these oxides on the surface of active monolayer. However, a
2
2
tungstated zirconia with similar surface density (6.2 W nm
as that of 15 wt% TPA/ZrO
acidity.
)
2
2
2
(7.2 W nm ) has a lower
Brönsted and Lewis acidity of catalysts with different TPA
loadings prepared in methanol were measured by pyridine
adsorption in-situ FT-IR spectroscopy.¶ FT-IR spectra of these
2
1
One of the authors (B. M. D.) acknowledges CSIR, New
Delhi for the award of Senior Research Fellowship.
catalysts showed Brönsted (B) acidity at 1540 cm and Lewis
2
1
(
L) acidity at 1440 cm . The B/L ratio increases with TPA
loading up to 15 wt% (Fig. 1), and decreased with further
loading. Thus the sample containing 15 wt% TPA is the most
acidic in terms of the number of acidic sites and strength and
corresponds to monolayer coverage of TPA on the surface.
Benzylation and acylation reactions were carried out in a 50
ml glass batch reactor. The reaction mixture was heated to the
desired temperature and stirred for required time. The resulting
solution was diluted with 5 ml ortho-dichlorobenzene (for
acylation) and 5 ml methanol (for benzylation) and after the
catalyst separation it was analyzed by GC (SE-52 packed
column coupled with FID). Products of the reactions were
identified using authentic samples and GC–MS analysis.
Benzylation of phenol results in the formation of ortho- and
para-benzyl phenol, benzyl phenyl ether and benzyl ether
Notes and references
P MAS NMR spectra were recorded at 121.5 MHz with high power
decoupling using a Bruker 4 mm probehead. The spinning rate was 10 kHz
and the delay between two pulses was varied between 1 and 30 s to ensure
that a complete relaxation of the P nuclei occurred. The chemical shifts are
3 4
given relative to external 85% H PO .
‡ 31
31
§ NH -TPD was carried out after 0.1 g of the catalyst sample was
3
dehydrated at 773 K in dry air for 1 h and purged with helium for 0.5 h. The
temperature was decreased to 398 K under a flow of helium and then 0.5 ml
NH
3
pulses were supplied to the samples until no more uptake of ammonia
was desorbed in He flow by increasing the temperature
was observed. NH
to 813 K at 10 K min
detector.
3
21
3
measuring NH desorption using a TCD
(
< 2%) as the products. Acylation of 2-MN gave 1-acetyl-2-MN
as the only acylated product. In the absence of catalyst no
reaction was observed and catalyst removal after partial reaction
results in complete stoppage of the reaction which indicates the
heterogeneous catalytic nature of the reaction. The dissolution
of phosphorus or tungsten species from the catalyst into solution
during reaction was monitored at the end of the reaction using
inductively coupled plasma-optical emission spectroscopy
¶
Pyridine adsorption in-situ FT-IR spectroscopy was performed by pre-
treating a self-supporting wafer (10 mg) of the sample at 673 K for 1 h under
26
a vacuum (10 mbar) followed by cooling to 373 K. Pyridine vapor was
introduced into the cell followed by evacuation at 573 K for 30 min and the
IR spectrum recorded. The Brönsted/Lewis acid site density ratio (B/L ratio)
was determined from integration of IR bands corresponding to Brönsted
21
21
acid sites (1540 cm ) and Lewis acid sites (1440 cm ).
(
ICP-OES) after catalyst filtration. This technique indicated the
1
2
K. Arata., Appl. Catal. A: General, 1996, 146, 3.
G. D. Yadav and J. J. Nair, Microporous Mesoporous Mater., 1999, 33,
absence of phosphorus or tungsten in solution after reaction.
The conversion of 2-MN and benzyl alcohol with 12.6 wt%
WZ are 17.9 and 38%, while with 15 wt% WZ the correspond-
ing values are 7.1 and 19.6%, respectively. The change in
conversions with TPA loading and B/L ratio is shown in Fig. 1.
It is seen that conversions increases with TPA loading and is
1
.
3
R. Sreenivasan, R. A. Keogh, D. R. Milburn and B. H. Davis, J. Catal.,
995, 153, 123.
1
4 D. G. Barton, S. L. Soled, G. D. Meitzner, G. A. Fuentes and E. Iglesia,
J. Catal., 1999, 181, 57.
5 J. G. Santiesteban, J. C. Vartuli, S. Han, R. D. Bastain and C. D. Chang,
J. Catal., 1997, 168, 431.
maximum for the 15 wt% TPA/ZrO
2
catalyst in both reactions.
However, conversions decrease with further TPA loading. The
change in conversions with TPA loading is in line with the B/L
ratio and the acidity of the catalysts (Table 1). Tungstated
zirconia with similar tungsten surface density as that of
6
K. Shimizu, T. N. Venkatraman and W. Song, Appl. Catal. A: General,
002, 225, 33.
M. Scheithauer, R. E. Jentoft, B. C. Gates and H. Knözinger, J. Catal.,
000, 191, 271.
2
7
2
1
5 wt% TPA/ZrO
versions.
Conversions for 15 wt% TPA/ZrO
different solvents are shown in Fig. 2. This shows that the
2
prepared in methanol showed lower con-
8
9
M. Misono, Chem. Commun., 2001, 1141.
E. Lopez-Salinas, J. G. Hernandez-Cortez, I. Schifter, E. Torres-Garcia,
J. Navarrete, A. Gutierrez-Carrillo, T. Lopez, P. P. Lottici and D. Bersani,
Appl. Catal. A: General, 2000, 193, 215.
2
catalyst prepared in
CHEM. COMMUN., 2002, 1074–1075
1075