SURFACE STUDIES AND NATURE OF ACTIVE SITES OF SUPPORTED HETEROPOLYACIDS AS CATALYSTS IN METHANOL DEHYDRATION

Article abstract of DOI:10.1135/cccc19940820

Different amounts of molybdo- and tungstophosphoric acids were supported on α-Al2O3 to get information about their surface and catalytic properties.The surface study revealed that surface area, total pore volume and the mean pore radius decreased as the acid content increased.X-Ray diffraction analysis showed that there is no interaction between the acid and α-Al2O3.Using a continuous flow system, methanol transformation was carried out under atmospheric pressure.Some experiments were made to determine the nature of active centers and reaction mechanism.

Full text of DOI:10.1135/cccc19940820

8
20
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
SURFACE STUDIES AND NATURE OF ACTIVE SITES
OF SUPPORTED HETEROPOLYACIDS AS CATALYSTS
IN METHANOL DEHYDRATION
Amin ANWAR, Ali ABDEL-GHAFFAR*, Sameh ABOUL-FOTOUH
and Ebeid FIKRY
Department of Chemistry,
Faculty of Education, Roxy, Cairo, Egypt
Received December 8, 1992
Accepted June 11, 1993
Different amounts of molybdo- and tungstophosphoric acids were supported on α-Al O to get infor-
2
3
mation about their surface and catalytic properties. The surface study revealed that surface area, total
pore volume and the mean pore radius decreased as the acid content increased. X-Ray diffraction
analysis showed that there is no interaction between the acid and α-Al O . Using a continuous flow
2
3
system, methanol transformation was carried out under atmospheric pressure. Some experiments were
made to determine the nature of active centers and reaction mechanism.
It was shown by amine titration that heteropolyacids in the solid state have a large
1
2
number of acid sites . Moffat et al. concluded from chemisorption of ammonia that the
acid strength increases with calcination temperature whereas that of the salt decreases
in the order: H > Zn > Al > B > Mg > Ca > Na. Concerning the kind of heteropoly
3
compounds supported on SiO -200, the acidity was found to decrease in the order:
2
H PW O /SiO -200 > H PMo O /SiO -200 > H PO /SiO -200.
3
12 40
2
3
12 40
2
3
4
2
4
– 8
Misono and others
measured water content and surface area of 12-tungsto- and
1
2-molybdophosphoric acid after heat treatment in dry helium stream and after evacu-
ation, respectively. They found that the dehydration of molybdophosphoric acid pro-
2
ceeds at temperature as low as 25 °C. Its surface area is about 13 – 14 m /g at 100 °C
and decreases at the higher temperature. On the other hand, dehydration of tungsto-
phosphoric acid proceeds at 100 °C, and the decrease in surface area with increasing
9
temperature is smaller. Hayashi and Moffat found that the surface area of the am-
2
monium salt of tungstophosphoric acid is about 113 m /g, significantly higher than that
2
of the acid (3 – 6 m /g). With the salt, its surface area increases slightly on heating to
*
The author to whom correspondence should be addressed.
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

Heteropolyacids Supported on α-Al2O3
821
2
3
00 °C (163 m /g). On further heating, the surface decreases monotonously to approxi-
2
mately 5 m /g at 600 °C.
From X-ray structural analysis Bradly et al.10 have deduced that the lattice constant
of the tungstophosphoric acid, H PW O . 19 H O, is 23.3 Å. The acid dehydrates to
3
12 40
2
1
1
H PW O . 6 H O with the lattice constant 12.5 Å. They assumed that dehydration
3
12 40
2
causes shrinking of the unit cell because adjacent heteropolyanions come much closer
when water molecules between the anions are removed. From these results it was con-
1
2
cluded that the water of crystallization is not zeolitic. Hayashi et al. assumed that the
active sites responsible for the conversion of methanol are Brønsted acid sites.
This work was oriented to get information about the nature of active centers for
methanol transformation over heteropolyacids supported by α-Al O .
2
3
EXPERIMENTAL
Two series of supported 12-tungsto- and 12-molybdophosphoric acids on α-alumina were used. Four
samples from each series were prepared by impregnating 5.0 g of α-alumina in an aqueous solution
of the acid whose volume was just sufficient to produce 0.5, 1.0, 3.0 and 5.0 wt.%. The solvent,
twice distilled water, was evaporated on a hot surface (70 °C) for 3 h and the sample was dried at
1
10 °C for 24 h. Molecular formula and acid content are shown in Table I.
The adsorption of nitrogen at −196 °C was determined by means of conventional volumetric ap-
1
3
paratus . X-Ray diffraction analysis was made for various catalysts on a Philips Diffractometer
(
PW-1390) under 40 kV and 30 mA with Ni filter and CuKα radiation.
TABLE I
Chemical formulae and acid content for the acid samples used
Acid content/g Al O
2
3
Symbol
Chemical formula
mg
mmol
wt.%
HPMo
H3PMo12O40 . x H2O
500
5
0.2700
0.0027
0.0054
0.0160
0.0270
0.1800
0.0017
0.0034
0.0100
0.0180
100.0
0.5
0
1
3
5
.5 HPMo/α-Al2O3
.0 HPMo/α-Al2O3
.0 HPMo/α-Al2O3
.0 HPMo/α-Al2O3
0.5 H3PMo12O40 . x H2O/α-Al2O3
1.0 H3PMo12O40 . x H2O/α-Al2O3
3.0 H3PMo12O40 . x H2O/α-Al2O3
5.0 H3PMo12O40 . x H2O/α-Al2O3
H3PW12O40 . x H2O
10
30
50
500
5
1.0
3.0
5.0
HPW
100.0
0.5
0
1
3
5
.5 HPW/α-Al2O3
.0 HPW/α-Al2O3
.0 HPW/α-Al2O3
.0 HPW/α-Al O
0.5 H3PW12O40 . x H2O/α-Al2O3
1.0 H3PW12O40 . x H2O/α-Al2O3
3.0 H3PW12O40 . x H2O/α-Al2O3
5.0 H PW O . x H O/α-Al O
3
10
30
50
1.0
3.0
5.0
2
3
3
12 40
2
2
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

8
22
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
The catalytic activity of the samples was examined using methanol dehydration at 300 °C in a
continuous flow system. The product was analyzed by GC on a Varian 3700 chromatograph con-
nected to Varian 9176 recorder, using 4 m column packed with 10% squalane on Chromosorb.
RESULTS AND DISCUSSION
Surface Characterization of Supported Heteropolyacids
Adsorption–desorption isotherms of nitrogen at liquid nitrogen (−196 °C) were
measured for supported and nonsupported catalysts pretreated at 350 °C. It was ob-
served that some of the obtained isotherms showed characteristic behavior of the
1
4
type III isotherms while the others were type II isotherms . All the isotherms were
0
found to be irreversible in all cases, exhibiting small hysteresis beyond P/P = 0.5.
Specific surface area SBET, BET-constant and the monolayer capacity V were deter-
m
mined for both types II and III as indicated in Table II in parentheses. Surface area
_
4
derived from V1 − t plot (S ), mean pore radius calculated by VP0.95/SBET . 10 ( r ) and
t
0
total pore volume at P/P = 0.95 V
are summarized in Table II.
P0.95
2
The SBET values of all the samples are low (of the order 14 m /g or less). These small
values may be attributed to the fact that heteropolyanions of Keggin structure are
packed units into which other molecules cannot penetrate. Furthermore, non-polar ga-
seous molecules are unable to enter the interstitial sites of HPA. These findings are in
TABLE II
Some surface characteristics of heteropolyacids supported on α-Al O pretreated at 350 °C
2
3
a,b
b,c
c
a
Catalyst
BETconst
Vm
SBET
St
VP0.95
r, Å
α-Al2O3
10.0
2.4
3.4 (II)
2.6 (III)
0.8 (II)
2.6 (II)
2.5 (III)
2.0 (III)
1.8 (II)
0.9 (II)
2.5 (III)
1.9 (III)
1.8 (III)
1.5 (II)
14.8 (II)
11.3 (III)
3.6 (II)
14.2
11.5
4.2
0.031
0.029
0.010
1.019
0.020
0.020
0.017
0.015
0.017
0.023
0.026
0.019
21.3
25.2
28.8
17.3
20.3
31.4
22.0
40.0
15.8
28.5
33.9
30.0
H2O/α-Al2O3
HPMo
34.0
13.0
3.5
0
1
3
5
.5 HPMo/α-Al2O3
.0 HPMo/α-Al2O3
.0 HPMo/α-Al2O3
.0 HPMo/α-Al2O3
11.3 (II)
10.9 (III)
8.7 (III)
7.8 (II)
11.5
10.5
8.8
1.6
6.0
7.3
HPW
9.0
3.9 (II)
4.2
0
1
3
5
.5 HPW/α-Al2O3
.0 HPW/α-Al2O3
.0 HPW/α-Al2O3
.0 HPW/α-Al O
2.5
10.9 (III)
8.3 (III)
7.8 (III)
6.5 (II)
10.0
8.0
4.0
5.0
8.5
16.0
7.0
2
3
a
3
b
c
2
In cm /g; type of the isotherm in parentheses; in m /g.
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

Heteropolyacids Supported on α-Al2O3
823
agreement with those obtained for the BET surface area as calculated from the nitrogen
adsorption.
_
Effect of Acid Content on SBET, VP0.95 and the Mean Pore Radius r
The relation between the acid content and SBET (Fig. 1) shows that the surface area of
α-alumina decreases gradually with increasing loading of HPA. These results may be
explained by assuming that the impregnation with heteropolyacid results in blocking of
the pores of α-alumina, leading to a decrease of the accessibility of the pore system
towards nitrogen molecules and resulting thus in an apparent decrease of the surface
area.
2
In fact, 3.0 wt.% HPMo is enough to cover an area of 14.4 m /g, based on that a
1
5
2
Keggin unit covers 144 Å . It is also seen from Fig. 1 that the decrease in surface area
is more remarkable in the case of HPW samples supported on α-alumina. Indeed it is
1
5
known that the size of tungsten cation for the Keggin anion is greater than that of
molybdenum, and this difference may be responsible for the observed decrease of sur-
face area.
The data obtained for the effect of α-alumina impregnation with HPA on the total
pore volume and pore radius are given in Table II and shown in Figs 2 and 3. The
_
results indicate a decrease in both VP0.95 and r with the samples of 0.5 wt. % acid
content. This may be due to the blocking of some micropores and might indicate an
increased accessibility of the pore system towards nitrogen molecules. The observed
_
increase in VP0.95 and r up to 3.0 wt % acid content may be due to the strong acidity of
HPA which attacks the alumina substrate and causes the widening of some mesopores.
_
The increase in r values reflects a limited ability of nitrogen molecules to measure the
FIG. 1
Dependence of specific surface area (S
in
BET
2
m /g) on acid content (HPA). ❍ HPMo/α-
Al O , ❍ HPW/α-Al O , ∆ α-Al O , ❍
2
3
2
3
2
3
H O/α-Al O
2
2
3
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

8
24
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
narrower part of the total pore system (only mesopores). From Figs 2 and 3 it is seen
_
that with 5.0 wt.% acid content, there is a small decrease in both VP0.95 and r. This may
be due to multilayer formation and deposition of the acid, which results in narrowing of
the pores. The VP0.95 of the free acid samples is smaller than that of all the supported
samples with high mean radius. This can be ascribed to the low surface area of these
acids as a result of few micropores as well as to prevaling mesopores when compared
to α-alumina. The increase in r observed with HPW samples in comparison with HPMo
samples may be due to the larger size of tungsten atom relative to molybdenum one in
FIG. 2
Dependence of total pore volume (V ) on acid content (HPA). For designation of points see Fig. 1
P
FIG. 3
_
Dependence of mean pore radius (r in Å) on acid content (HPA). For designation of points see Fig. 1
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

Heteropolyacids Supported on α-Al2O3
825
1
5
the Keggin unit . From the data given in Table II it follows that the pore parameters of
α-alumina impregnated in solution of both HPMo and HPW are changed due to the
presence of water solvent and/or the acid.
One can thus conclude that (i) impregnation of α-alumina in pure water and/or with
_
HPMo and HPW acids results in a decrease of SBET and pore volume, (ii) the r value for
_
the impregnated samples is strongly dependent on the mean radius (r) of the free acid.
Thus, the mean pore radius of the α-Al O with supported HPW is generally higher
2
3
than that of α-Al O with supported HPMo.
2
3
Pore Structure Analysis
In order to determine the porosity of investigated samples, the t-method developed by
1
6
deBoer et al. was applied. The V1 − t plots obtained for these samples are shown in
Figs 4 – 6. The values of SBET and S for all the samples are nearly the same. The result
t
1
6
17
justifies the use of the methods by deBoer et al. and by Mikhail et al. . In general,
the V1 − t plots of α-alumina impregnated with molybdophosphoric acid (Fig. 5) do not
show any characteristic deviation from the initial straight line. Only at very high
relative pressure (in the capillary condensation region) there occurs upward deviation.
This means that the total pore system of these catalysts consists of limited size meso-
FIG. 4
3
VL (in cm /g) vs t (in Å) plots for a α-Al O , b H O/α-Al O , c HPW and d HPMo samples
2
3
2
2
3
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

8
26
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
pores. A somewhat different trend is observed with the HPW-impregnated catalysts
Fig. 6). An upward deviation appears around a t-value of 6 Å in all the cases. Such a
(
behavior is characteristic of mesoporous solids exhibiting multilayer building inside the
pores. However, the densely packed alumina surface with 0.5 wt.% HPW exhibits the
small downward bending at the high t-value, indicating a limited number of the layers,
i.e. limited pore dimensions.
Comparison of the texture parameters of the two series (shape of adsorption iso-
_
therms, C-values, SBET, r and trend of the V
plots) shows that the porous structure of
1 − t
both series is nearly the same. The HPW series is characterized by the relatively wider
pores, the lower adsorbate–adsorbent interactions and the slightly lower surface areas.
X-Ray Diffraction Analysis
The diffraction patterns of α-Al O , HPMo, HPW and their 0.5 and 5.0 wt.% supported
2
3
1
8
samples when compared with A.S.T.M. cards show that the heteropoly compounds
can be assigned the structures H PMo O . 24 H O and H PW O . 24 H O. Also
3
12 40
2
3
12 40
2
1
8
the structure of α-Al O is the same as given by A.S.T.M. cards .
2
3
FIG. 5
3
VL (in cm /g) vs t (in Å) plots for supported HPMo samples. Acid content: a 0.5 wt.%, b 1 wt.%,
c 3 wt.%, d 5 wt.%
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

Heteropolyacids Supported on α-Al2O3
827
The diffractograms of 0.5 and 5.0 wt.% HPA/α-Al O differ only in line intensities,
2
3
depending on the amount of the acid content. For the supported samples, the diffraction
lines indicate that the supported acids are located on the surface of α-Al O as separate
2
3
phases (new diffraction lines were not observed). However, the impregnation process
resulted in a slight increase in the intensity of XRD patterns. This may be explained by
that the doped acids might stimulate crystallization of α-alumina. Thus, impregnation
of α-Al O with heteropolyacids is mainly associated with an increase in the amor-
2
3
1
9
phous character . It is interesting to note that on supporting 0.5 wt.% HPA the well
crystalline α-Al O was formed in a greater extent than in the case of 5.0 wt.% acid.
2
3
From the results obtained by X-ray analysis it follows that there are no specific inter-
actions between α-Al O and the free acids which would result from the supporting
2
3
process.
Nature of Active Centers and Reaction Pathway
. In methanol transformation over this type of solid acid catalysts (100 – 300 °C) it
1
2
1
was found that :
FIG. 6
3
VL (in cm /g) vs t (in Å) plots for supported HPW samples. Acid content: a 0.5 wt.%, b 1 wt.%,
c 3 wt.%, d 5 wt.%
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

8
28
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
–
At lower reaction temperatures methanol undergoes dehydration to give dimethyl
ether (Eq. (A)) and traces of methane by a parallel reaction.
2
CH OH
CH OCH + H O
(A)
3
3
3
2
–
At higher temperatures, methanol and dimethyl ether participate22 in methane for-
mation via a consecutive reaction (Eq. (B)),
−H2O
2
CH OH
CH OCH
3
hydrocarbons.
(B)
3
3
By measuring the catalyst life time it was found that some catalysts require an “in-
duction period” to form new active sites involved in the methanol conversion. After
this period, the surface becomes saturated with the reactive species, which accounts for
the constant rate of dimethyl ether formation. The proposed species are weakly ad-
sorbed, as evident from the sharp decrease in the reaction rate when the alcohol-feed to
the reactor had been stopped. The catalyst activity depends on the electronegativity of
the central metal atom (Lewis acidity), which is evident from the increase in the reac-
tion rate in the case of the more electronegative tungsten atom.
2
. At 250 °C “pretreatment temperature”, the rate of dimethyl ether formation over
HPW increases on increasing the temperature from 250 – 500 °C, whereas the rate of
methane formation starts to decrease at 300 – 500 °C (Fig. 7). On increasing the pre-
treatment temperature, Brønsted acid sites are supposed to be converted to Lewis sites
2
3
by loosing the structural water (Eq. (C)).
>
300 °C
Brønsted
Lewis + H O
(C)
2
This result was explained by that Lewis acid sites play an important role in dimethyl
ether formation, whereas Brønsted acid sites are required for methane formation.
3
. Changes in pretreatment atmosphere (Fig. 8) were assumed to affect the number of
1
2
acidic sites . The effect of pretreatment environment on the catalyst activity was also
studied. It was found that argon seems to be the most beneficial environment for the
pretreatment which removes some of the more acidic Brønsted sites and leads to a
beneficial shift in the acid site distribution (high dispersion of Brønsted sites). Further,
hydrogen may increase the number of weakly bound acid sites which are attached to the
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

Heteropolyacids Supported on α-Al2O3
829
catalyst at the positions not normally occupied by hydrogen. Such sites are expected to
be readily removed at 350 °C, and not easily replaced. Hence, the rate of methanol
transformation to dimethyl ether and methane is greater in argon atmosphere than in
hydrogen. It was also found that the pretreatment in the argon saturated with water
vapour decreases the activity of HPMo to about 50% of its original value, as shown in
Fig. 8. This may be due to H O molecules which become strongly adsorbed on the most
2
active catalyst sites because of their higher absorptivity. This results in a decrease of
the affinity of HPMo for alcohol molecules. Similar results were obtained by Ohtsuka
2
4
et al. for the dehydration of tert-butyl alcohol over heteropoly compounds.
. Effect of the water produced during methanol dehydration on the rate of both
4
dimethyl ether and methane formation was evaluated in the following way: Acetic acid
was feeded with methanol in the amount 1 mole % after about 3 h of the reaction
(
Fig. 9). It was observed that the rate of dimethyl ether formation increases whereas the
rate of methane formation decreases sharply to about 8.5% of the value observed just
FIG. 7
FIG. 8
Effect of pretreatment temperature tr (°C) on
the rate r (mol/h kg) of dimethyl ether (1) and
methane (2) formation at 250 °C over HPW
Effect of pretreatment enviroment on the rate r
(mol/h kg) of dimethyl ether (a) and methane
(b) formation at 300 °C. ❍ Argon, ❍ hy-
drogen, ❍ water–argon atmosphere
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

8
30
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
before injection of acetic acid. This indicates that the release of H O of dehydration by
2
CH COOH shifts the reaction towards dimethyl ether formation. With continuous
3
removal of water from the reaction product, the rate of diffusion of methanol and of
2
5
dimethyl ether into the pseudo-liquid phase increases and, consequently, also the rate
of methane formation increases according to Eqs (A) and (D) – (F).
CH3COOH
CH OH
CH OCH + H O + CH
4
(D)
(E)
(F)
3
3
3
2
CH COOH + H O
CH OH + HCOOH
3
3
2
HCOOH
CO + H O
CH4
2
FIG. 9
FIG. 10
Effect of injection of acetic acid (at point A) on
the rate r (mol/h kg) of dimethyl ether (❍) and
methane (❍) formation at steady state condi-
tions over HPMo (300 °C, 10.6 kPa)
Effect of injection of pyridine (at point P) on
dimethyl ether (❍) and methane (❍) formation.
For experimental conditions see Fig. 9
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

Heteropolyacids Supported on α-Al2O3
831
Role of Brønsted Acid Sites in Reaction Mechanism
(
+)
To investigate the role of “H ” in the catalytic process, pyridine was injected with
methanol in 1 mole % (pyridine/alcohol) after 3 h from the beginning of the run at
3
8
00 °C (Fig. 10). It was found that the rate of methane formation decreased sharply to
% of the value observed just before pyridine injection, whereas the rate of dimethyl
ether formation increased to 18% within 15 min. Such a behavior may be due to the
adsorption of pyridine on Brønsted acid sites to form the strongly adsorbed pyridinium
2
6
ion . This result indicates that methane formation proceeds on Brønsted acid sites.
After that the rate of dimethyl ether formation decreases whereas the rate of methane
(
+)
formation increases. This may be explained by formation of new proton acid sites, H ,
as a result of the high electronegativity of pyridine, which assists methane formation on
the expense of the ether. The catalyst starts to be deactivated by increasing pyridine
concentration on the surface.
On the basis of these results and previous studies27,28 one can conclude that in meth-
(
+)
anol conversion the first step includes protonation of CH OH to form CH OH which
3
3
2
reacts with another physisorbed methanol molecule to form dimethyl ether, regenera-
ting proton. Methane could be formed from the ether which acts as a hydrogen transfer
(
+)
agent. The CH OCH species formed in this case may react further, i.e. it can
3
2
isomerize or produce the lighter products, H , CO, CO (Scheme 1).
2
2
SCHEME 1
Collect. Czech. Chem. Commun. (Vol. 59) (1994)

8
32
Anwar, Abdel-Ghaffar, Aboul-Fotouh, Fikry:
REFERENCES
. Otake M., Onoda T.: Shokubai (Catalysis) 17, 13 (1975).
1
2
3
4
. Moffat J. B., Hayashi J. H.: J. Catal. 19, 395 (1985).
. Ohtsuka R., Morioka Y., Kobayashi J.: Bull. Chem. Soc. Jpn. 63, 2071 (1990).
. Misono M., Mizuno N., Katamura K., Kasai A., Konishi Y., Sakata K., Okuhara T., Yoneda Y.:
Bull. Chem. Soc. Jpn. 55, 400 (1982).
5. Mizuno N., Katamura K., Yoneda Y., Misano M.: J. Catal. 83, 384 (1983).
6. Okuhara T., Hashimoto T., Hibi T., Misono M.: J. Catal. 93, 224 (1985).
7. Tonomura S., Aoshima A.: Shokubai (Catalysis) 27, 389 (1985).
8. McCarron E. M., Harlow R. L.: J. Am. Chem. Soc. 105, 6179 (1985).
9. Hayashi H., Moffat J. B.: J. Catal. 83, 182 (1983).
1
1
0. Brodley A. J., Hiling J. W.: Proc. R. Soc. London, A 157, 113 (1936).
1. Brown G. M., Neo-Spirlet M. R., Busing W. R., Levy H. A.: Acta Crystallogr., B 33, 1038
(
1977).
1
1
2. Moffat J. B., Hayashi J. H.: J. Catal. 77, 473 (1982).
3. Jouner L. G. in: Scientific and Industrial Glass Blowing (W. R. Ban and V. J. Ankrn, Eds), p.
2
57. Instrument Publishing, Pittsburgh 1949.
1
1
1
4. Brunauer S., Emmet P. H.: J. Am. Chem. Soc. 57, 1754 (1935).
5. Moffat J. B.: J. Mol. Catal. 52, 169 (1989).
6. Lippens B. C., Linsen B. C., DeBoer J. H.: J. Catal. 3, 32 (1964); Lippens B. C., DeBoer J. H.:
J. Catal. 4, 319 (1965).
1
1
1
2
7. Mikhail R. Sh., Guindy N. M., Hanafi S.: J. Chem. A.R.E. 1973, 53.
8. Smith J. V.: X-Ray Powder Data File and Index, p. 13. A.S.T.M., Philadelphia 1961.
9. El-Shobaky G. A., Fagal G. A., Amin N. H.: Thermochim. Acta 141, 205 (1989).
0. Cotton F. A., Wilkinson G.: Advanced Inorganic Chemistry, p. 103. Interscience, New York
1
967.
1. Fikry M., Abdel-Ghaffar A., Anwar A., Aboul-Fotouh S.: Collect. Czech. Chem. Commun. 58,
079 (1993).
2
2
2
2
2. Qinwei Z., Deng J.: J. Catal. 116, 298 (1989).
3. Matsuda T., Sato M., Kanno T., Miura H., Sugiyama K.: J. Chem. Soc., Faraday Trans. 1 77,
3
107 (1981).
24. Ohtsuka R., Monoka Y., Kobayashi J.: Bull. Chem. Soc. Jpn. 62, 3195 (1989).
25. Okuhara T., Kasai A., Hayakawa N., Yoneda Y., Misono M.: J. Catal. 83, 121 (1983).
26. Furuta M., Sakata K., Misono M., Yoneda Y.: Chem. Lett. 1979, 31.
27. Salvador P., Klading W.: J. Chem. Soc., Faraday Trans. 1 73, 1153 (1977).
28. Zataroski W., Kryzyzanouski S.: Acta Phys. Chem. 24, 347 (1978).
Revised by J. Hetflejs.
Collect. Czech. Chem. Commun. (Vol. 59) (1994)