Effect of the Sulfate Sulfur Content and the Preparation Procedure on the Catalytic Properties of Sulfated Zirconium Oxide

Article abstract of DOI:10.1023/A:1019887320752

The effects of the composition and the procedure of preparing sulfated zirconium oxides on their catalytic properties in the reactions of 1-butene isomerization to 2-butenes and isobutanol dehydration were studied. The activity depended on nature of parent zirconium oxide, the sulfate sulfur content, and the temperature of calcination. Catalysts prepared from a crystalline oxide showed the highest activity. Their activity depended on the sulfate sulfur content. The calcination of catalysts at temperatures higher than 400°C led to a detectable loss of sulfate. The catalytic behavior of the sulfated oxide in both the reactions was indicative of the absence of superstrong proton sites. The specific activity of catalysts based on zirconium oxide in the model reactions was higher than the activity of hydroxide-based catalysts.

Full text of DOI:10.1023/A:1019887320752

Kinetics and Catalysis, Vol. 43, No. 4, 2002, pp. 550–554. Translated from Kinetika i Kataliz, Vol. 43, No. 4, 2002, pp. 595–600.
Original Russian Text Copyright © 2002 by Shmachkova, Kotsarenko.
Effect of the Sulfate Sulfur Content and the Preparation
Procedure on the Catalytic Properties
of Sulfated Zirconium Oxide
V. P. Shmachkova and N. S. Kotsarenko
Boreskov Institute of Catalysis, Siberian Division, Russian Academy of Sciences, Novosibirsk, 630090 Russia
Received August 22, 2001
Abstract—The effects of the composition and the procedure of preparing sulfated zirconium oxides on their
catalytic properties in the reactions of 1-butene isomerization to 2-butenes and isobutanol dehydration were
studied. The activity was found to depend on the nature of parent zirconium oxide, the sulfate sulfur content,
and the temperature of calcination. Catalysts prepared from a crystalline oxide exhibited the highest activity.
Their activity depends on the sulfate sulfur content. The calcination of catalysts at temperatures higher than
400°C resulted in a detectable loss of sulfate. The activity of sulfated oxides prepared from an amorphous oxide
was noticeably lower; it depended only slightly on the temperature of calcination and sulfate sulfur content. The
catalytic behavior of the sulfated oxide in both of the reactions is indicative of the absence of superstrong proton
sites. Based on the results, assumptions on the nature of formed surface compounds were made.
INTRODUCTION
surfaces. The latter reaction mainly occurs at proton
sites; to a lesser degree, it also occurs with the partici-
pation of a pair of aprotic and basic sites, and the iso-
meric composition of butenes formed in this case is dif-
ferent. The rates of both reactions depend on the
strength and concentration of sites. It was found that the
activity with respect to the number of proton sites
increases proportionally to the strength of sites [10]. In
binary silicate catalysts [11], the activity and strength
of a proton site is almost independent of the ratio
between components, whereas is depends on the nature
of the metal.
The SOH group is the carrier of proton sites in sul-
fate catalysts; in crystalline sulfates, its activity
depends only slightly on the nature of the metal [12]. In
crystalline aluminum sulfate and sulfated aluminum
oxides, the activity of proton sites is almost indepen-
dent of the concentration of these groups [13]. Taking
into account these facts, we intended to obtain data on
the proton-donor properties of sulfated zirconium
oxides by studying their activity in the above reactions.
A change in the total activity will reflect a change in the
concentration of proton sites depending on the compo-
sition of catalysts and on various factors of their prepa-
ration.
The ability of sulfated zirconium oxides to catalyze
the skeletal isomerization of alkanes at low temperatures
generated considerable interest in them, and several
reviews on this subject matter were published [1–5].
However, although a great body of data was obtained,
many problems remain unsolved. This is especially true
in regard to the nature of the active components of solid
superacids and to the mechanisms of their action. The
occurrence of proton and aprotic acid sites on the sur-
face of these catalysts was found [6–8]; however, the
superacidity of these systems was not established by
direct measurements. Because the reaction mechanism
of skeletal isomerization is unknown, many problems
associated with the formation of catalyst acid sites in
the course of preparation are still unclarified. To eluci-
date the roles of various factors in the course of prepa-
ration of these catalysts and in the formation of these
unique properties, it is of interest to study their behav-
ior in typical catalytic reactions of the acid type, which
occur by a known mechanism, and whose rates depend
on the strength of sites.
The aim of this work was to examine the effects of
the concentration of sulfate ions and the nature of the
parent zirconium oxide on the catalytic properties of
sulfated zirconia.
EXPERIMENTAL
The activity of catalysts was studied in the model
reactions of double bond migration in 1-butene and of
Zirconium hydroxide and zirconium oxide were
isobutanol dehydration. It was found previously [9] that used as starting compounds for the preparation of sul-
the former reaction occurs at the proton sites of catalyst fated zirconia. Zirconium hydroxide was prepared from
0023-1584/02/4304-0550$27.00 © 2002 MAIK “Nauka /Interperiodica”

EFFECT OF THE SULFATE SULFUR CONTENT AND THE PREPARATION PROCEDURE
551
Table 1. Catalytic properties of zirconium oxide treated
a zirconium oxynitrate solution by precipitation with an
aqueous ammonia solution at 65–70°C and a constant
value of pH 9. The precipitate (zirconium hydroxide),
which was washed and dried at 110°C, was calcined at
500°C to obtain crystalline zirconium oxide.
with sulfuric acid
wiso(wdeh)*,
[SO3]0, Tcalcination
,
[SO3],
wt %
S, m2/g
mol (g Cat)–1 h–1
wt %
°C
The hydroxide was sulfated by the treatment with a
sulfuric acid solution on filter in accordance with a pub-
lished procedure [1] or by incipient wetness impregna-
tion using a solution containing a given amount of sul-
furic acid. Zirconium oxide was also sulfated by the lat-
ter method. In this case, the duration of the interaction
between the oxide and sulfuric acid was varied (2 and
30 h). In the subsequent text, the interaction for 30 h is
referred to as prolonged. After drying at 110°C, the
samples were calcined in a flow of dry air at different
temperatures (from 400 to 650°C).
Sulfated zirconium oxides were analyzed by the dis-
solution in hydrofluoric acid followed by the removal
of the acid and the analysis of the solution by induc-
tively coupled plasma atomic emission spectrometry
(ICP AES).
The X-ray diffraction analysis was performed on a
DRON instrument using monochromatic CuKα radia-
tion. Diffraction patterns were recorded on a recorder at
a goniometer rate of 1 deg/min. Phases were identified
using the ASTM file.
The specific surface areas of samples were deter-
mined by the low-temperature adsorption of nitrogen.
The catalytic activity of samples in the reactions
was studied by the flow-circulation method. The reac-
tion rate (wiso) under standard conditions at 220°C, 20%
conversion, and a butene concentration of 7 vol % in a
mixture with nitrogen was taken as a measure of activ-
ity in 1-butene isomerization. The activity in alcohol
dehydration was measured at 300°C and an initial alco-
hol concentration of >30 vol %. When the stationary
alcohol concentrations is higher than 30 vol %, the
reaction rate is independent of the concentration of the
component; that is, the reaction is of zero order. In this
case, the reaction rate constant is equal to the reaction
rate (wdeh). This value was taken to characterize activity
in the dehydration reaction. Before testing, the catalysts
were calcined in a flow of air at a given temperature and
then evacuated at 400°C and a pressure of ~10–2 torr.
Thereafter, the temperature was decreased to a reaction
temperature in a nitrogen flow.
1.1
400
500
650
400
500
650
250
400
500
650
400
500
650
400
500
650
250
400
500
650
400
600
400
600
0.5
0.5
–
100 0.2 (0.02)
102 0.25 (0.02)
83 0.1
3.9
5.2
1.8
–
83 1.9
101 0.6
1.1
–
100 0.9 (1.5)
79 1.6
3.3
2.8
2.4
7.7
4.9
1.5
12.2
9.8
2.0
–
93 3.9/2.9
97 2.5
94 1.4 (1.5)
39 2.0
10.4
15.3
20.1
74 2.2
100 1.7
50 2.7 (3.3)
52 2.7
100
24 1.7
–
20.3
19.0
2.7
5.7
3.6
9.0
5.7
27 2.2 (3.4)
33 1.6
110 1.8
63 1.3
6.0**
9.0**
68 0.9
55 0.85
49 0.84
* w is the reaction rate of isomerization at a 20% conversion of
iso
1-butene; w is the reaction rate of dehydration.
deh
** The catalysts were prepared by the prolonged interaction of zir-
conium oxide and sulfuric acid.
RESULTS AND DISCUSSION
of samples prepared based on crystalline zirconium
oxide. Sulfate was partially degraded in the course of
catalyst calcination; the higher the temperature of cal-
cination, the greater the loss of sulfate. After calcina-
tion at 650°C, the residual sulfate content was no higher
than 2.7 wt % SO3.
Samples containing from 1.1 to 20 wt % SO3 were
prepared based on zirconium oxide, and samples con-
taining up to 10 wt % SO3 were prepared based on zir-
conium hydroxide. Tables 1 and 2 summarize their
compositions and properties. The concentrations
[SO3]0 of sulfate sulfur added in the course of prepara-
The specific surface areas of samples dramatically
tion on a calcined zirconia basis are given in Tables 1 decreased as the SO3 content increased. The specific
and 2. As can be seen in Table 1, the temperature of cal- surface areas of samples calcined at different tempera-
cination significantly affected the sulfate sulfur content tures were almost equal at similar SO3 contents.
KINETICS AND CATALYSIS Vol. 43 No. 4 2002

552
SHMACHKOVA, KOTSARENKO
sulfate in a detectable amount was not observed. The
w
iso × 103, mol m–2 h–1
crystal structure of ZrO2 in catalysts depends on the
nature of the zirconium compound used for sulfation.
The monoclinic modification of ZrO2 was retained over
the tested range of the temperatures of calcination of
sulfated zirconium oxide prepared from the crystalline
oxide. The sulfated oxide prepared from zirconium
hydroxide up to calcination temperatures of 400°C was
X-ray amorphous. As was repeatedly mentioned in the
literature [4], thermal treatment at higher temperatures
resulted in the formation of highly dispersed tetragonal
ZrO2. This modification of ZrO2 was also retained at a
maximum of the tested calcination temperatures
(650°C).
I
50
1
1'
1''
2
3
3'
4
III
II
10
0
0.5
1.0
2.0
3.0
[SO3], mmol/g
Fig. 1. Effect of the concentration of sulfate sulfur in cata-
lysts prepared from (I) zirconium oxide or (II) zirconium
hydroxide and (III) by the prolonged interaction and cal-
cined at (1, 1', 1") 400, (2) 500, (3, 3') 600, or (4) 650°C on
the specific activity of the catalysts in the reaction of
1-butene isomerization.
Tetragonal and monoclinic zirconium oxides are
almost inactive in the reaction of 1-butene isomeriza-
tion. The samples exhibited a detectable activity after
supporting even a small amount of sulfate. The catalyst
activity depends on the SO3 content and the tempera-
ture of calcination. The total activity increased with the
SO3 content of the sample calcined at 400°C (Table 1);
it reached a weakly pronounced maximum at ~3.3 wt %
SO3 and then decreased. An increase in the calcination
temperature, which was accompanied by the decompo-
sition of a portion of sulfate, resulted in a decrease in
the activity of catalysts.
The calcination of catalysts based on zirconium
oxide at 650°C, which was accompanied by the degra-
dation of a portion of sulfate, resulted in the regenera-
tion of the initial oxide surface area. At an SO3 concen-
tration of 0.5–3 wt %, the surface area was 93–
110 m2/g; this value is consistent with the surface area
of the parent crystalline oxide.
The interaction of sulfuric acid with zirconium
hydroxide results in the formation of sulfate com-
pounds, which are more thermally stable and exhibit
more developed specific surface areas (Table 2).
Figure 1 demonstrates the specific activity as a func-
tion of the concentration of sulfate ions in the samples.
As can be seen, it is described by saturation curves for
all of the obtained series of samples. The maximum
activity depends on the preparation procedure. The sul-
fated oxides prepared by the short-time interaction of
the oxide with sulfuric acid exhibited the highest activ-
ity (curve I). The samples prepared by the interaction of
the oxide with sulfuric acid for a longer time were less
active (curve II). The samples prepared by the sulfation
of the hydroxide exhibited the lowest activity (curve III).
In a particular catalyst preparation procedure, the activ-
ity depended on only the concentration of sulfate sulfur.
The X-ray diffraction patterns of the resulting sam-
ples exhibited only reflections due to ZrO2 over the
entire ranges of SO3 concentrations and calcination
temperatures. The formation of crystallized zirconium
Table 2. Catalytic properties of zirconium hydroxide treat-
ed with sulfuric acid
Ssp,
wiso,
[SO3]0, Tcalcination
,
[SO3],
wt %
Figure 2 shows the activity per mole of sulfate sulfur
as a function of sulfate sulfur concentration in the test
samples. The activity of sulfated zirconium oxide
increased from 0 to 9 h–1 as the SO3 content was
increased up to ~0.15 mmol/g; it reached a maximum
value at an SO3 content of 1.4–2.5 wt % and then dra-
matically decreased (curve I). The extremal curve indi-
cates that not all of the sulfate groups are the sources of
protons. It is likely that, at low sulfate contents, the
oxide is sulfated in a surface layer with the polydentate
binding of sulfate anions. A decrease in the activity at
high sulfate contents can be explained by the assump-
tion that the surface composition remains unchanged,
whereas an excess of sulfate sulfur penetrates into near-
surface oxide layers and does not participate in the
reaction. In the region of the maximum activity, the sur-
face coverage with sulfate anions is ~75% on the
assumption that the anion occupies an area of 25 Å2 [14].
m2/g mol (g Cat)–1 h–1
wt %
°C
3.3
400
600
400
600
400
600
600
400
600
3.5
0.31
4.2
3.9
5.1
–
480
32
0.5
0.005
0.85
1.1
4.5
7.0
171
107
99
1.3
–
0.8
8.0
6.2
8.25
6.70
168
272
168
0.46
0.52
0.84
10.0
KINETICS AND CATALYSIS Vol. 43 No. 4 2002

EFFECT OF THE SULFATE SULFUR CONTENT AND THE PREPARATION PROCEDURE
553
wiso/[SO3] × 10–3, h–1
12
In the catalysts prepared by the sulfation of zirco-
nium hydroxide, the activity per mole of sulfate sulfur
plotted as a function of sulfate concentration has a sim-
ilar shape (Fig. 2, curve II); however, the activity
changes on a much smaller scale. Because the reactivity
of the hydroxide toward sulfuric acid is higher than that
of the oxide, it is believed that in this case bulk sulfates
are formed; however, at present, there is no evidence for
their occurrence. Actually, the highly dispersed oxide
in a low-temperature tetragonal modification is formed;
this is possible because of the stabilizing effect of sul-
fate anions. The samples resulting from the prolonged
interaction of zirconium oxide with sulfuric acid
approached the sulfated oxide based on the hydroxide
in their properties. The low activity of both catalysts
can indicate that the polydentate binding of sulfate ions
is predominant in these catalysts. This hypothesis is
supported by data on the strength and concentration of
proton sites in the catalysts of both sample series with
similar concentrations of sulfate sulfur [8]. At calcina-
tion temperatures equal to 400 and 600°C, the concen-
trations of proton sites in catalysts based on zirconium
oxide were higher by factors of 4 and 3.5, respectively.
However, the strengths of proton sites in these catalysts
were equal (PA = 1170 kJ/mol), and the reaction rates
of 1-butene isomerization referenced to the concentra-
tion of proton sites were similar. These rates were insig-
nificantly higher than those for amorphous and crystal-
line aluminosilicate catalysts [10].
1
1'
2
3
4
10
8
6
4
I
2
II
0
0.5
1.0
1.5
2.0
2.5
3.0
[SO3], mmol/g
Fig. 2 Effect of the concentration of sulfate sulfur in cata-
lysts prepared from (I) zirconium oxide and (II) zirconium
hydroxide and calcined at (1, 1') 400, (2) 500, (3) 600, or
(4) 650°C on the activity in the reaction of 1-butene isomer-
ization.
minosilicate catalysts by a factor of ~5 [10] and that
they are among the best catalysts for isobutanol dehy-
dration. However, their specific activity (0.015–
0.125 mol m–2 h–1) is close to that of sulfated aluminum
oxides (0.01–0.1 mol m–2 h–1) over the tested range of
SO3 concentrations [18]. The reaction products formed
on these catalysts exhibited the same composition,
which is different from the composition of products
formed on silicotungstic heteropoly acid (35–37%
n-butenes), which bears only proton sites. The above
data are indicative of the participation of the aprotic
sites of the sulfated oxide in alcohol dehydration; how-
ever, the contribution of these sites is the same as that
in sulfated aluminum oxides, which are not superacid
catalysts. For superstrong aprotic sites, a considerable
increase in the rate of dehydration and a decrease in the
concentration of n-butene in the reaction products
would be expected.
The results obtained in both of the reactions are
unambiguously indicative of the absence of both proton
and aprotic superstrong sites. This conclusion is consis-
tent with the results obtained by the IR spectroscopy of
adsorbed pyridine and CO [6, 8, 19].
We found that the catalytic properties of sulfated
zirconium oxide vary over a wide range depending on a
number of factors. The main factors affecting the prop-
erties of catalysts are the following: the composition of
catalysts, the nature of the parent oxide, the tempera-
ture of calcination, and the duration of the interaction of
a sulfuric acid solution and the oxide in the course of
sulfation.
Because aprotic sites occur on the surface of these
catalysts along with proton sites, it is of interest to fol-
low the behavior of the catalysts in a model reaction
that occurs at not only proton but also aprotic sites. The
dehydration of alcohols is a reaction of this kind. It
occurs either by a carbenium ion mechanism at proton
sites or with the participation of a pair of aprotic and
base sites. Information on reaction pathways can be
obtained from the products of isobutanol dehydration.
Previously, it was found using a wide range of catalysts
that a mixture of all butenes with an n-butene concen-
tration of 33–40% is formed at proton sites. Isobutylene
is formed at the paired sites [9, 15].
Zirconium oxide exhibits a low activity in isobu-
tanol dehydration; its specific activity is equal to 2 ×
10–5 mol m–2 h–1. The sulfated oxide exhibits a high
catalytic activity, which depends on the amount of sup-
ported sulfate and on the temperature of calcination
(see Table 1), as in the case of 1-butene isomerization.
This catalytic activity increases with the sulfate sulfur
concentration in the sample.
All isomeric butenes are the products of isobutanol
dehydration. The concentration of n-butenes is 26–28%
regardless of the composition of catalysts and the tem-
peratures of calcination. In this respect, the behavior of
sulfated zirconium oxide is analogous to that of all of
the bulk and supported sulfates tested [12, 17, 18].
The most probable sources of proton sites are SOH
groups. Because both of the reactions occur at proton
sites, the effects of the above factors are primarily due
The results suggest that the specific activity of sul- to changes in the concentration of SOH groups. Based
fated zirconium oxides is higher than the activity of alu- on this approach, processes that occur in the prepara-
KINETICS AND CATALYSIS Vol. 43 No. 4 2002

554
SHMACHKOVA, KOTSARENKO
3. Yamaguchi, T., Jin, T., Ishida, T., and Tanabe, K., Mater.
Chem. Phys., 1987, vol. 17, p. 3.
tion of sulfated zirconium oxide can be represented as
described below. On the treatment of zirconium oxide
with a sulfuric acid solution, the components react to
form various sulfates. Based on the catalytic behavior
and thermal stability, it is believed that at least normal
and acidic sulfates are formed. Because acid sulfates
are less thermally stable compounds, they undergo deg-
radation as the temperature of calcination is increased.
This is evident from a decrease in the concentration of
sulfate sulfur and in the activity with calcination tem-
perature (see Table 1). The behavior in the isomeriza-
tion reaction (Figs. 1, 2) suggests that the surface com-
position remained almost unchanged at the concentrations
of sulfate anions higher than 0.5–0.7 mmol SO3/(g Cat). A
further increase in the amount of supported sulfuric
acid results in either an increase in the thickness of a
surface sulfate layer or the formation of dispersed zir-
conium sulfates rather than changes the surface compo-
sition. The ratio between surface and bulk sulfates also
depends on the nature of the parent oxide. Because of a
higher reactivity of the hydroxide, the concentration of
bulk sulfates is higher in this series of catalysts, as fol-
lows from their lower specific activities.
The results demonstrated that the specific activity of
catalysts based on zirconium oxide in the model reac-
tions is higher than the activity of hydroxide-based cat-
alysts. Although the strengths of proton and aprotic
sites in the catalysts of both series are similar, their
behaviors in the reaction of n-butane skeletal isomer-
ization were different; catalysts based on zirconium
hydroxide are much more active [14, 20]. Thus, it fol-
lows that skeletal isomerization takes place at more
complex centers than single proton or paired (aprotic–
basic) sites at which the model reactions occur.
4. Song, X. and Sayari, A., Catal. Rev.–Sci. Eng., 1996,
vol. 38, p. 329.
5. Corma, A., Chem. Rev., 1995, vol. 95, no. 3, p. 597.
6. Morterra, C., Cerrato, G., Pinna, F., Signoretto, M., and
Strukul, G., J. Catal., 1994, vol. 149, p. 181.
7. Spielbauer, D., Mekher, G.A., Zaki, M.I., and Knoz-
inger, H., Catal. Lett., 1996, vol. 40, p. 71.
8. Shmachkova, V.P., Kotsarenko, N.S., and Paukshtis, E.A.,
React. Kinet. Catal. Lett., 2000, vol. 71, p. 385.
9. Paukshtis, E.A., Infrakrasnaya spektroskopiya v getero-
gennom kislotno-osnovnom katalize (Infrared Spectros-
copy in Heterogeneous Acid–Base Catalysis), Novosi-
birsk: Nauka, 1992.
10. Malysheva, L.V., Kotsarenko, N.S., and Paukshtis, E.A.,
Kinet. Katal., 1990, vol. 31, no. 1, p. 247.
11. Karakchiev, L.G., Kotsarenko, N.S., Dzis’ko, V.A., and
Gorshkova, T.P., Kinet. Katal., 1968, vol. 9, no. 2, p. 282.
12. Shmachkova, V.P., Kotsarenko, N.S., Malysheva, L.V.,
and Paukshtis, E.A., Kinet. Katal., 1990, vol. 31,
p. 1021.
13. Malysheva, L.V., Shmachkova, V.P., Kotsarenko, N.S.,
and Paukshtis, E.A., Kinet. Katal., 1991, vol. 32, no. 4,
p. 940.
14. Chen, F.R., Coudurier, G., Joly, J.-F., and Vedrine, J.C.,
J. Catal., 1993, vol. 143, p. 616.
15. Kotsarenko, N.S. and Malysheva, L.V., Kinet. Katal.,
1983, vol. 24, no. 4, p. 877.
16. Kotsarenko, N.S., Shmachkova, V.P., and Mastikhin, V.M.,
Kinet. Katal., 1998, vol. 39, no. 4, p. 575.
17. Shmachkova, V.P. and Kotsarenko, N.S., Kinet. Katal.
(in press).
18. Shmachkova, V.P., Cand. Sci. (Chem.) Dissertation,
Novosibirsk, 1994.
19. Olindo, R., Pinna, F., Strukul, G., Canton, P., Cerrato, G.,
Meligrana, G., and Morterra, C., Stud. Surf. Sci. Catal.,
2000, vol. 130, p. 2375.
REFERENCES
1. Arata, K., Adv. Catal., 1990, vol. 37, p. 165.
2. Jin, T., Yamaguchi, T., and Tanabe, K., J. Phys. Chem., 20. Shmachkova, V.P., Kotsarenko, N.S., and Paukshtis, E.A.,
1986, vol. 90, p. 4797.
Cat. Lett. (in press).
KINETICS AND CATALYSIS Vol. 43 No. 4 2002