On the superacidity of sulfated zirconia catalysts for low-temperature isomerization of butane

Full text of DOI:10.1021/ja9631460

12240
J. Am. Chem. Soc. 1996, 118, 12240-12241
On the Superacidity of Sulfated Zirconia Catalysts
for Low-Temperature Isomerization of Butane
Jose E. Tabora and Robert J. Davis*
Department of Chemical Engineering
UniVersity of Virginia
CharlottesVille, Virginia 22903
ReceiVed September 6, 1996
ReVised Manuscript ReceiVed October 21, 1996
Sulfated zirconia catalysts for low-temperature butane isomer-
ization reportedly possess acid sites with a Hammett acidity,
Ho, of less than -16. Since 100% sulfuric acid has an Ho
Figure 1. Effect of a butene adsorption bed on the n-butane
isomerization activity of sulfated zirconia: (a) (triangles) no adsorption
bed, (b) (open circles) adsorption bed of 0.5 g, and (c) (filled circles)
adsorption bed of 2.0 g. The inset shows run (c) on expanded axes.
Reaction conditions: 0.4 g of catalyst, T ) 373 K, YC4 ) 0.1, He flow
1
value of -12, sulfated zirconia is classified as a solid superacid
catalyst with acid strength 10 000 times greater than that of pure
sulfuric acid. However, determination of solid acid strength
by a color change of adsorbed indicators is fraught with
experimental difficulty. More recently, acidity measurements
for sulfated zirconia using spectroscopic evaluations of adsorbed
indicators are consistent with strengths not exceeding that of
3
-1
9
cm min STP.
sical carbonium ions which subsequently release dihydrogen.4,9
Carbenium ions are also formed by reactions of trace olefins
00% sulfuric acid.2 Thus, some researchers propose that
4,9
1
with acids of moderate strength. In this study, we attempted
sulfated zirconia is essentially mounted sulfuric acid instead of
a strong superacid.3 Catalytic test reactions are also used to
probe surface acidity. The high activity of sulfated zirconia
for butane isomerization has been used to establish its supera-
cidity since sulfuric acid does not catalyze isomerization of
to eliminate all of the trace butenes in the butane feed in order
to test the intrinsic activity of the acidic sites on sulfated zirconia
for butane isomerization.
One way to reduce the trace butene concentration in the
butane is to pass the feed stream over a platinum catalyst in
the presence of dihydrogen to hydrogenate the butenes to
butanes. Liu et al. attempted to lower the butene concentration
with this method and showed that the butane isomerization rate
4
n-alkanes. In this communication, we show that butane
isomerization on sulfated zirconia occurs exclusively through
the participation of trace butenes, present in the feed or generated
in situ, and not through alkane activation via superacid protolysis
reactions. Thus, previous rankings of acidity derived from the
activity level of sulfated zirconia catalysts are highly suspect.
The promotional effect of trace amounts of butenes on the
acid-catalyzed butane isomerization was first reported by Pines
and Wackher.5 Recent work in our lab and elsewhere has shown
that trace butenes play an important role in the isomerization
was lowered, but the catalyst was still active.8
Catalytic
hydrogenation does not completely eliminate the trace butenes,
but instead lowers their concentration to the equilibrium value.
In this work, an adsorption method was used to eliminate
butenes from the feed. An adsorption bed of calcined sulfated
zirconia, maintained at room temperature, was placed prior to
an active catalyst bed containing sulfated zirconia at 373 K.
The trace butenes adsorbed irreversibly on surface acid sites
since the adsorption bed temperature was too low for significant
reaction and desorption.
6-8
of n-butane to isobutane over sulfated zirconia catalysts. We
demonstrated that transition metal promoters like Fe, Mn, and
Pt supported on sulfated zirconia facilitate the low-temperature
isomerization reaction by increasing the surface concentration
The sulfated zirconia for both the adsorption bed and the
reactor was generated from calcination of sulfate-doped zirco-
nium hydroxide (Magnesium Electron Inc.) at 825-900 K in
flowing air, cooled to room temperature, and stored before use.
The calcined sulfated zirconia was reactivated in situ in the the
adsorption bed and the reactor by heating in flowing air to 773
6
of olefins. These active olefinic intermediates either react with
acid sites to form carbenium ions or alkylate surface carbenium
ions to form octyl (C8) intermediates. Butane isomerization
occurs by methyl shifts and â-scissions of C8 intermediates
followed by hydride transfers according to classical acid
catalysis.9 In all of these steps, superacidity of the catalyst is
not required.
2
K. These treatments yield sulfated zirconia with about 100 m
g-1 specific surface area.10 We also prepared a Pt-promoted
The bimolecular chain reaction for butane isomerization
discussed above occurs over transition metal-promoted and
unpromoted sulfated zirconia.6 The role of promoters is easily
rationalized in terms of olefin formation since transition metals
and metal oxides are recognized as dehydrogenation catalysts.
However, the mechanism of butene or carbenium ion formation
on unpromoted sulfated zirconia is less clear. Superacids form
classical carbenium ions by protonating alkanes to give nonclas-
sulfated zirconia catalyst by incipient wetness impregnation of
MEI sulfate-doped zirconium hydroxide with an aqueous
solution of hexachloroplatinic acid (Aldrich, ACS Reagent)
followed by calcination in air at 873 K. Elemental analysis by
Galbraith Laboratories (Knoxville, TN) showed S and Pt
loadings to be 2.0 and 1.8 wt %, respectively. The specific
surface area of the platinum-promoted sample was measured at
-8
98 m2 -1
g . In the absence of an adsorption bed, the butene
impurity level in the n-butane feed (Aldrich, 98%) was measured
by capillary column gas chromatography and found to be
∼0.4%, present mostly as trans-2-butene and isobutene. When
butene impurities were removed from the reactant feed, the trace
concentration fell below the detection limit estimated as
<0.001%.
Figure 1 shows the results from reactions carried out in the
absence of an adsorption bed (butenes present in the feed), and
with adsorption beds of two different sizes. The reaction
*
To whom correspondence should be addressed.
(
(
(
(
1) Hino, M.; Arata, K. J. Chem. Soc., Chem. Commun. 1980, 851.
2) Umansky, B.; Engelhardt, J.; Hall, W. K. J. Catal. 1991, 127, 128.
3) Babou, F.; Coudurier, G.; Vedrine, J. C. J. Catal. 1995, 152, 341.
4) Olah, G. A.; Molnar, A. Hydrocarbon Chemistry; John Wiley and
Sons: New York, 1995.
(
(
(
5) Pines, H.; Wackher. R. C. J. Am. Chem. Soc. 1946, 68, 599.
6) Tabora, J. E.; Davis, R. J. J. Catal. 1996, 162, 125.
7) Adeeva, V.; Lei, G. D.; Sachtler, W. M. H. Catal. Lett. 1995, 33,
1
1
1
35.
(
8) Liu, H.; Adeeva, V.; Lei, G. D.; Sachtler, W. M. H. J. Mol. Catal.
995, 100, 35.
(
9) Gates, B. C. Catalytic Chemistry; John Wiley and Sons: New York,
(10) Zirconium Compounds for Catalysis; MELCat. Doc. 1010; Mag-
nesium Elektron Inc.: Flemington, NJ.
992.
S0002-7863(96)03146-0 CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 48, 1996 12241
3
conditions were T ) 373 K, Ybutane ) 0.1, and He flow 9 cm
of rates. Apparently, the acid sites of the catalysts are not
significantly affected by trace amounts of H2S that may be
evolving from the adsorption bed.
-
1
min STP. For the run with no adsorption bed (Figure 1a),
the reaction proceeded through an induction period of increasing
activity followed by deactivation. The rate of reaction was in
the range of values reported in the literature for sulfated zirconia
catalysts. When an adsorption bed of 0.5 g is used (Figure 1b),
the level of butene impurities drops below the sensitivity of
our detector. Interestingly, the initial maximum in activity of
The selectivity to isobutane exceeded 92% in all of the runs
with sustained activity. The formation rates of disproportion-
ation products propane and pentanes (normal and iso) correlated
with the formation rate of isobutane, indicating that the
bimolecular pathway of isobutane formation described above
is relevant to all the cases. No disproportionation products were
detected during the run in Figure 1c due to the low conversion
level of butane.
-
8
-8
the catalyst at 373 K is reduced from 38 × 10 to 18 × 10
-1 -1
mol g s , presumably because of the lower butene concentra-
tion in the feed. However, at times later than 100 min, the rate
of n-butane isomerization starts to increase considerably. This
puzzling behavior was observed repeatedly when attempts were
made to eliminate the impurity level of butenes from the reactant
feed with moderately sized adsorption beds. Evidently, the
increase in rate after 100 min results from leakage or break-
through of butenes from the adsorption bed. When a relatively
large adsorption bed of 2 g was used (Figure 1c and inset),
sulfated zirconia exhibited an extremely low activity maximum
The small initial activity observed in Figure 1c could result
from trace amounts of olefins produced from oxidative dehy-
drogenation on sulfated zirconia. Farcasiu et al. suggest that
oxidation plays an important role in the initial catalysis of
1
2
hydrocarbons on sulfated zirconia.
The high sensitivity of sulfated zirconia to trace amounts of
olefins is probably responsible for the relatively poor agreement
in catalytic activity reported in the literature since the impurity
level of butenes is seldom, if ever, reported along with reactivity
data. More importantly, we find little evidence for superacidic
sites on sulfated zirconia. The lack of significant butane
isomerization activity at 373 K with a purified feed is consistent
with negligible carbenium ion formation via superacidic pro-
tolysis of butane. Even though the very small initial activity
of sulfated zirconia in Figure 1c may result from superacidic
sites, a more likely explanation is that redox sites initiate
carbenium ion formation. Regardless of the initiation mecha-
nism, the butane isomerization reaction is not sustained in the
absence of butenes.
-
8
-1 -1
of 5 × 10 mol g
s
for a short initial period before
becoming totally inactive. Two possibilities can account for
the results in Figure 1. Either sulfated zirconia does not catalyze
butane isomerization in the absence of butene, or the adsorption
bed alters the intrinsic activity of the catalyst.
A recent report by Ng and Horvat showed that small amounts
of dihydrogen sulfide are evolved from sulfated zirconia during
butane isomerization.11 Thus, small amounts of H2S may be
evolving from the large adsorption bed and poisoning the active
sites in the reaction zone. In order to disprove this hypothesis,
we tested a Pt-doped sulfated zirconia catalyst with a 2 g
adsorption bed. Since Pt enhances the butane isomerization
reaction by generating butene in situ,6 we expect that the Pt-
promoted catalyst will be active as long as the acid sites are
not poisoned. Indeed, the activity of our Pt-promoted catalyst
with purified butane was very high. The rate of butane
,8
Acknowledgment. This work was supported by a U.S. National
Science Foundation Young Investigator Award (Grant CTS-9257306)
and a DuPont Young Professor Award.
-
8
-1 -1
isomerization was 29 × 10 mol g
s
after 100 min on
stream. High conversions at earlier times prevented calculations
JA9631460
(11) Ng, F. T. T; Horvat, N. Appl. Catal., A 1995, 123, L197.
(12) Farcasiu, D.; Ghenciu, A.; Li, J. Q. J. Catal. 1996, 158, 116.