Hydrogenation of toluene over La0.8Sr0.2Fe1-x CoxO3-zz perovskite catalysts

Article abstract of DOI:10.1515/znb-1997-0807

The hydrogenation of toluene over perovskites of composition La_0.8Sr_0.2Fe_1-x Co_xCb_3-z (x = 0.0, 0.2, 0.5, 0.8 and 1.0) is reported. The catalyst material has been prepared by the coprecipitation metho

Full text of DOI:10.1515/znb-1997-0807

Hydrogenation of Toluene over Lao.sSro^Fej.xCoxO^ Perovskite Catalysts
Carsten Schinzer3, Riyadh A. Saymeh*b, Hassan M. Asfour0
a Institut für anorganische Chemie, Auf der Morgenstelle 18, D-72076 Tübingen, Germany
b Department of Chemistry, Yarmouk University, Irbid, Jordan
c Department of Applied Chemistry, College of Science, Applied Science University,
Amman, Jordan
Z. Naturforsch. 52 b, 927-934 (1997); received April 11, 1997
Perovskites, Mixed Metal Oxides, Stability, Catalytic Hydrogenation, Catalytic Reduction
The hydrogenation of toluene over perovskites of composition Lao.gSro.iFei-x Cox03_z
(x = 0.0, 0.2, 0.5, 0.8 and 1.0) is reported. The catalyst material has been prepared by the
coprecipitation method using maleic acid as the complexing agent. The activity for the gas
phase reaction of toluene and hydrogen at temperatures between 50 °C to 300 °C is reported
for as-prepared and activated materials.
The conversion has been found to be nearly temperature-independent over the as-prepared per-
ovskites with a maximum of 40% conversion found for a composition Lao.8Sro.2Feo.5Coo.5O2.86-
Activation by an oxygen treatment at 500 °C was successful to enhance the activity of the Co-rich
perovskites. All of these catalysts show a complete conversion of toluene at 100 °C, but activity
decreases by time due to a reduction of the catalyst material. Again, Lao.8Sro.2Feo.5Coo.5O2.86
exhibits the slowest loss of activity. A model for its special properties is mainly attributed to
the p-type conduction behaviour.
The conversion behaviour of the perovskite type catalysts investigated here is compared to
the results obtained with noble-metal supported catalysts and it can be concluded, that the
perovskites exhibit higher conversion rates, maximum temperatures and higher cracking onset
temperatures than these.
Introduction
in the case of exhaust aftertreatment [5] and hy-
drogenation of olefins [8 , 9]. The highest activity
was found for compositions Lan sAn oMnO^ [51 and
Lao.8Sro.2Co03- z [8],
Mixed metal oxides of the perovskite type have
been the object of extensive studies in the recent
years [1 ] because they show outstanding electronic
and magnetic properties. Among these, the catalytic
properties have become an important field of inves-
tigation [2, 3]. Perovskites and related oxides have
been found to be active in both, oxidation and re-
duction processes, such as the reduction of oxygen
in solid oxide fuel cells [3], the aftertreatment of
exhaust gases [4, 5] and the hydrogenation of hy-
drocarbons [6 , 7].
Gunasekran etal. investigated perovskites of gen-
eral formula Lao.8Sro.2M0 3 _z [7, 10] and found Co
to be the most active metal among M = Cr, Mn, Fe,
Co and Y while Fe exhibits the lowest maximum
temperature in the hydrogenation of propylene. In
comparison to Pt / AI2O3 which is a good cata-
lyst at relatively low temperature the perovskites
show good conversion at high temperatures (100-
300 °C). The study of hydrogenation of aromatic
compounds is of interest since it could be used for
decreasing the aromatic content in diesel oil which
makes pollution problems and lowers the quality
of the fuel. Perovskites might be suitable for this
purpose as refinery usually works at relatively high
temperatures.
A wide variety of compositions including solid
solutions such as A i_xAxB0 3 _z, A B i_xB'x0 3 _z, or
Ai_yAyBi_xB^0 3 _z can be used. The catalytic ac-
tivity of these perovskites is mainly controlled by
two factors: (1) the composition on the A-site and
(2) the electronic properties of the B-site cations.
Several investigators have studied the influence of
the A-site composition on the catalytic activity
We report here the catalytic properties of per-
ovskites Lao.8Sro.2 Fei_xCox0 3 _z for the gas phase
hydrogenation of toluene. Our interest is to observe
the stabiliy of the catalysts under reaction conditions
* Reprint requests to Dr. R. A. Saymeh.
0939-5075/97/0800-0927 $ 06.00 (c) 1997 Verlag der Zeitschrift für Naturforschung. All rights reserved.
K
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

928
C. Schinzer et al. • Toluene Hydrogenation over Perovskites
Table I. Experimental conditions.
Reactor
as a function of reaction temperature as well as the
influence of the B-site composition on the maxi-
mum of conversion.
Pyrex tube reactor with a silica frit
Frit diameter = 10.0 mm
Sample weight = 0.1 g
Experimental
The catalysts were prepared by the coprecipitation
method as follows. Stoichiometric amounts of LaCh •
7H20 , Sr(N03)2,Fe(N03)v 9H20 and C 0C 2O 3-2^0(311
reagent grade) were dissolved in 40 ml of a 1:1 mixture
of nitric acid and deionized water. Maleic acid was then
added to the solution in a threefold excess compared to
the amount of metal ions in the solution. After evaporat-
ing the solution to dryness, the remaining precipitate was
heated to about 150°C to start the decomposition of the
complexing agent. A final calcination for 2 h at 500 °C
completed the preparation.
Catalyst bed length = 3.2 - 3.8 mm
Temperature = 50 - 300 °C
Feed gas
Toluene concentration = 4.2 % (vol)
Balance = hydrogen
Flow rate = 30 ccm / min
Space velocity = 1500 - 1800 / h
GLC
Column = 3% OV-17, chrom W HP 80/100,
2m x 1/8" stainless steel
Column temperature = 100 °C
TCD-Filament = 250 °C
TCD-Block = 200 °C
Sample loop size = 100 ji\
The mean oxidation state of the cations (Fe and Co)
was determined for the as-prepared catalysts as well as
for the used material. The iodometric back titration was
applied for this purpose.
O
(i) X-ray diffraction
The diffraction patterns were recorded using a Philips
powder diffractometer connected to a PW 1710 controller.
The radiation used was Co KQ with a wavelength of
1.7890 A. The 2(9-values were corrected by means of
an alumina reference sample.
(ii) BET surface area
The surface area of the perovskites was determined by
means of the BET method using a Micromeritics Gemini
analyzer. An amount of ca. 50 mg of sample was weighed
into an analysing tube and heated for 30 min at 150°C
under nitrogen. After a subsequent weighing to determine
the weight loss the tube was fixed on the sample holder.
The standard tube of the analyser has been filled with
approximately the equal volume of glass pearls for the
measurement.
(Hi) Catalytic activity
ii
j<
The experimental conditions for the activity test are
summarized in Table I. The hydrogenation reaction was
done in a microreactor with a fritted disc under flow con-
ditions. The temperature of the reactor was varied by
means of an oven.
Hydrogen flow was fixed to 30 ml / min by using a
Matheson Flow Controller (Model 8274) passing through
a saturator filled with Toluene (Koch-Light Ltd., England:
purified by destination).
X
%rfuA-ir
610----------
1
---------- 1-----------1_______1______
1
_______1______
50
40
30
2 0 [°]
Fig. 1. X-ray diffraction profiles of the as-prepared sam-
ples in the series Lao.8Sro.2Fei_x Cox03_z. ‘x’ marks
La203-impurity peaks.
Analyses of the reactant and the reaction products were
done using a GLC connected to the flow system through
a 6-port-valve with a sample loop of 100 //l.
Activity was investigated for as-prepared and for acti-
vated samples in a way that starting from 100 °C a heating
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

C. Schinzer et aL ■Toluene Hydrogenation over Perovskites
929
Table II. Chemical and physical properties of the per-
ovskite catalysts.
X
BET surface area Lattice constant
z
0
OX
0
0
. 0
. 2
, 0
2
(
1
)
+3,16(2) 21,6 (5) m2/g
3,926 (5) A
3,917(5) A
3,882 (5) A
3,846 (5) A
3,822 (5) A
lw
V
16,3 (3) m2/g
14,9 (2) m2/g
13,2 (3) m2/g
0,05 (1) +3,11 (2)
0.5 0,14(1) +2,92 (2)
0
. 8
+2,89 (2)
+2 , 8 6 (2 )
0,15 (1)
0,17(1)
6
, 6 (2 ) m2/g
1 .0
Numbers in brackets represent the standard deviation of the last
digit.
!
Table III. Comparison of catalytic activity and properties
after the activity test for the as-prepared catalysts.
I I
Maximum conversion Onset of Oxidation Lattice
X
cracking state
constant
1°C]
[%1
270
240
275
a
+3,05 (2) 3,928(5) A
+2,99 (2)
+2,69 (2) 3,883(5)A
+2,53 (2) 3,839(5) A
0 .0
8
290 °C
270 °C
300 °C
250 °C
240 °C
0 .2 12
3,922(5) A
0.5 39
0.8
5
9
a
+2,27 (2)
3,830(5) A
1.0
a No maximum observed. Numbers in brackets represent the
standard deviation of the last digit.
60
40
30
50
20 [°
Fig. 2. X-ray diffraction profiles of the samples after ac-
tivity test in the series Lao.8Sro.aFei-x Cox03_z. ‘x’ marks
La^O.i-impurity peaks.
Table IV. Properties of the activated samples.
x
Light off Onset of Oxidation states
temp. cracking before15 after
Lattice const. (A)
before11 after
Results
0.5 180 °C
0.8 160 °C 315 °C +2,96 (2)+2,63 (2) 3,839(5) 3,84(1)
1.0 110°C 260 °C +2,81 (2) +2,53 (2) 3,830(5) 3,833(7)
a
+3,09 (2) +2,97 (2) 3,883(5) 3,868(3)
(i) X -ray diffraction
The XRD profiles of the series are presented in
Fig. 1. They clearly indicate the development of a
perovskite phase for all samples. As an impurity, a
small amount of La2 0 3 is detected in all composi-
tions. The perovskite phases are of low crystallinity
as deduced from the broad peaks with halfwidths
of ca. 0.5° 2 0 which corresponds to a particle
size around 25 nm according to the Scherrer equa-
tion. The determined lattice constants are shown in
Table II.
a No cracking observed up to 300 °C; b from similarly treated
samples, see text. Numbers in brackets represent the standard
deviation of the last digit.
rate of 1 K/min was applied. GLC runs were taken every
5-1 0 K. The activity measurement was stopped at the
time when the cracking temperature was reached.
As-prepared samples were heated to 300 °C in the
reaction atmosphere to get rid of possibly adsorbed water.
The sample was then cooled to 100 °C. During the cooling
some preliminray GLC runs were taken. The activated
samples were treated in an oxygen atmosphere for 1 h
at 500 °C prior to the activity test. After cooling in an
inert atmosphere (He) to 100 °C, activity was tested as
described before. The catalytic activity is presented by
percentage conversion which is calculated as follows:
The XRD profiles of the as-prepared catalysts
after the activity test are shown in Fig. 2. It can
be seen that no further impurities arose during the
catalytic hydrogenation reaction. The same results
are found for the activated samples.
The lattice constants of the used catalysts are
shown in the Tables referring to the catalytic prop-
erties of the perovskites (/. e. Tables III and IV).
While the lattice constants do change slightly, there
= A m / (A t + A m),
%conv
where At and Am represent the areas of the GLC peaks
of toluene and methylcylcohexane, respectively.
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

930
C. Schinzer et al. • Toluene Hydrogenation over Perovskites
50%
40%
30%
20%
10 %
0%
50 %
40%
30%
20 %
10%
0%
.
..................
1.
100
.
L I . . . . .
.
.
............... ...
1. . . . . . . . . . 1 .........................
50
150
200
250
300
50
100
150
200
300
250
(a)
(b)
50%
40 %
30%
20%
10%
0%
50 %
40%
30 %
20 %
10%
0%
/
r
\ J
\ f
---------
.........
,
.......... .......... ,..........
,
50
100
150
200
250
300
50
100
150
200
250
300
(C)
(d)
(e)
Fig. 3. Conversion to methylcyclohexane as a function of temperature over Lao.8Sro.2Fe02,98 (x = 0.0, curve a),
Lao.8Sro.2Feo.8Coo.2O2,95 (x = 0.2, curve b), Lao.8Sro.2Feo.5Coo.5O2.86 (x = 0.5, curve c), Lao.8Sro.2Feo.2Coo.sO2,85
(x = 0.8, curve d) and Lao.8Sro.2Co02,83 (x = l .0, curve e). Arrows indicate onset of cracking.
heavily affect the shape and the surface area of the
catalysts.
is no difference in the half widths of the reflections
before and after the screening.
(Hi) Oxidation state
(ii) BET surface area
The oxidation states of the samples were deter-
mined from iodometric titration and are also listed
in Table II. The mean oxidation state of Fe and Co
decreases with increasing x, i. e. the oxygen content
of the compounds, z, also decreases.
The BET surface areas were determined for the
as-prepared perovskites and are listed in Table II.
The weight loss during heating at 150 °C is around
4 % for all samples. A change in particle size cannot
be detected from the XRD reflections. Therefore, we
conclude that the hydrogenation reaction does not
After the hydrogenation reaction, all samples
show a decrease of the oxidation number, i. e.
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

C. Schinzer et al. • Toluene Hydrogenation over Perovskites
931
oxygen is lost during the reaction due to water for-
mation.
Lao.8Sr0.2 Fe0.8Co0.2O2.9 5 . Since we could not iden-
tify the products the results for these catalysts are
not presented. Methylcyclohexane is the only prod-
uct of the hydrogenation, when the reaction is per-
formed over the activated Co-rich samples (x = 0.5,
0.8 and 1.0). Also, cracking and water formation
occurred over these catalysts.
(iv) Catalytic activity of the as-prepared samples
The GLC runs generally showed three peaks for
H2 , C7 H8 and C7 H 14 at all temperatures. A large
increase of the area of the H? peak and some new
peaks indicated the cracking of toluene into lighter
hydrocarbons. For the iron rich compounds (x = 0.0
and 0 .2 ) an increase of the methylcyclohexane peak
was observed, too.
Water formation occurs in all cases due to the
release of oxygen from the perovskite surface. It
gave a broad peak in the chromatogram which was
impossible to quantify. The formation of water is
higher for the iron rich samples than for the cobalt
rich.
The conversion rate for Lao.8Sro.2 Feo.5 Coo.5 0 3 _z
is shown in Fig. 4 a. Total conversion is observed
from 100 to 180 °C. The conversion decreases
with further increase of temperature but rises
to about 90% on cooling. The cracking onset
temperature was not reached during the reaction
cycle. Lao.8 Sro.2Feo.2 Coo.8O2,85 shows a different
behaviour as can be seen from Fig. 4 b. Total con-
version is also seen but holds only up to 160 °C.
A sharp decrease of activity follows, and cracking
occurs starting from 315 °C. In the cooling regime
the activity does not rise up to former values but
decreases constantly with decreasing temperature.
The same holds for Lao.8Sro.2Co02,83 the activity
of which is shown in Fig. 4 c. Total conversion is
observed up to 110 °C and cracking occurs at tem-
peratures higher than 260 °C.
Lao.8 Sro.2 Fe0 2 ,9 8 does not exhibit a vigorous
conversion to methylcyclohexane production by in-
creasing the temperature as it is shown in Fig. 3 a.
A broad maximum could be detected at a tem-
perature of ca. 270 °C, although the conversion
of toluene does not coincide with this finding.
The cracking onset temperature is 290 °C. Fig. 3 b
shows the behaviour of Lao.sSro.2 Feo.8Coo.2O2,95
(x = 0.2) which exhibits a lower activity than
Lao.8Sro.2Fe02,98- The adsorption of toluene occurs
already below 110 °C. Cracking is observed at tem-
peratures higher than 270 °C.
The properties of the Co-rich activated catalysts
are summarized in Table IV. The oxidation states
of the perovskites after activation were determined
from samples that were treated in the same way
as the in situ activation but without performing the
hydrogenation reaction.
The conversion behaviour of
Lao.8 Sro.2 Feo.5 Coo.5 0 2 ,8 6 (x = 0.5) is presented in
Fig. 3 c. Again, a high amount of toluene is adsorbed
at temperatures below 80 °C and released within the
first 50 °C of heating. A slight maximum in methyl-
cyclohexane formation occurs at 260 °C while
cracking starts at 275 °C. Lao.8Sro.2Feo.2Coo.8O2,85
(x = 0.8; Fig. 3 d) shows a temperature independent
formation of small amounts of methylcyclohexane.
Also, an adsorption effect is not detected for temper-
atures as low as 90 °C. Cracking of toluene occurs
at temperatures higher than 250 °C. The conversion
behaviour of Lao.sSro.2Co0 2 ,8 3 (x = 1.0) is shown
in Fig. 3 e. Neither a maximum of conversion nor
an adsorption of toluene down to 80 °C is observed.
Cracking occurs at temperatures higher than 240 °C.
Discussion
Concerning the crystalline properties of the as-
prepared catalysts, the diffractograms can all be in-
dexed whithin a pseudocubic primitive cell. Indica-
tion of reflections in Figs 1 and 2 refer to this cubic
cell. Although Lao.sSro^FeOi and Lao.sSro.2Co0 3
obey orthorhombic and rhombohedral crystal struc-
tures when prepared by the solid state method [11,
12], these superstructures are not found here due to
the low crystallinity of the samples. In fact we ap-
ply the coprecipitation method to obtain particularly
low crystallinity and high surface areas.
The higher impurity concentrations in the case
of Lao.8 Sro.2Co0 3 _z can be attributed to the for-
mation of CoO. This compound has also been
reported to be an impurity when applying the
(v) Catalytic activity of the activated samples
The GLC runs showed five peaks in high-temperature solid-state reaction to obtain per-
ovskites Lao.8 Sro.2Coi_y Fey 0 3 [11].
the case of activated Lao.sSro.2Fe0 2 ,9 8 and
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

932
C. Schinzer et al. • Toluene Hydrogenation over Perovskites
100%
80 %
60%
40%
20 %
0%
100 %
\
\
80 %
60%
40%
20%
0%
:
VV
:
\
r
. . .
.
.............................1 .................
.
.
. . ± .
.
.
.
.
.
.
.
.
.
.
1 . . .
.
. .
.
1—1—1 . . . . . . . 1.
.
. . . . . . .
.
1.
.
.
. . . X-. W J_! . 11 111.1.1 11XXX.
.
. . . . -l-Ll 1
100
150
200
250
100
150
200
250
300
350
300
(b)
(a)
(c)
Fig. 4. Temperature dependence of conversion to methylcyclohexane (arrows indicate heating and cooling regime) over
the activated samples LaasSro^Feo.sCoo^Os-z (curve a), Lao.8Sro.2Feo.2Coo.8O3._z (curve b) and Lao.sSnuCoO.^-z
(curve c).
The BET surface areas found for the as-prepared
From the above-mentioned we conclude that the
catalysts show that the coprecipitation method for perovskite structure is stable under the reaction con-
the pure iron compound as well as for the cobalt-
containing compounds with x = 0.2, x = 0.5 and
x = 0.8 is successful. The pure cobalt compound
has a significantly lower surface area than the other
perovskites which may be linked to the relatively
high impurity concentration.
As it is seen from Table II, the value of the lat-
tice constant decreases with increasing x in the se-
ries. In the same manner the oxygen content (3-z)
and therefore the mean oxidation number of the B-
cations (OX) decreases (z is given in Table II, so
OX will decrease by increasing z). Since Co3+ is
smaller than Fe3+ (0.545 A the further, 0.655 A the
latter referring to the values given by Shannon [13])
the decrease of the lattice constant can easily be
explained in terms of the decreasing size of the B
cation.
ditions used in this study.
From a first general comparison it is obvious that
the catalytic activity of the investigated perovskites
is increased profoundly by the in-situ pretreatment
in oxygen atmosphere. After this pretreatment, the
mean oxidation number of the B-ions is found to be
higher than before. The oxidation state of the B site
cations correlates with the Lewis acidity [14], Chou
and Vannice [15] concluded that benzene readily
adsorbs on Lewis acidic sites. The same holds for
toluene.
Alcock et al. state that the conduction type is
heavily affecting the activity in hydrogenation, i.e.
the activity is generally low in n-type and high
in p-type conductors [16, 17], The type of con-
duction (p- or n-type) is connected to the oxida-
tion state of Co since Fe as a high spin cation is
expected to exhibit hopping conduction, whereas
Co2+ in the low spin configuration (d7, t2g6 eg')
or Co4+ in the low spin configuration (d5, t2g 5
e„°) will exhibit n-type or p-type conduction,
respectively. The activity of samples with an oxida-
The decreasing oxidation number does not affect
the lattice constants severely as shown by compar-
ing the values given in Tables II and III. The lattice
constants remain nearly unchanged within their es-
timated errors.
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

C. Schinzer et al. •Toluene Hydrogenation over Perovskites
933
tion state of Fe/Co less than +3.0 should therefore
be low.
The untreated Fe rich catalysts Lao.8Sro.2FeO2.98
(x = 0.0) and Lao.8Sro.2 Feo.8Coo.2O2,95 (x = 0.2)
do not show reasonable activity for the hydrogena-
tion compared to the other perovskites. This agrees
with what was found for Lao.sSro^FeO^ by Gu-
nasekran et al. [7] and could be attributed to the
smaller change in the oxidation state. Pretreatment
led to an enhancement of activity to give multiple
reaction products.
temperature [° C]
As for the Co rich catalysts
Fig. 5. Comparison of the catalytic activity of activated
Lao.sSro.2Feo.5Coo.fi0 3 -z (—), 1% Pt / AI2O3 (— ) and
1%Rh / AI2Ö3 (•••) (data from this work and from [18]).
Lao.8Sro.2 Feo.5 COo.5 O2,86 (x = 0.5),
Lao.8Sro.2 Feo.2Coo.8O2,85 (X = 0.8) and
Lao.sSro.2 Co0 2 ,8 3 (x = 1.0) we found that among the
untreated samples only the x = 0.5 catalyst shows
a maximum in conversion (Fig. 3). Also the on-
set of cracking shifts from 300 °C for the x = 0.5
compound (not indicated in Fig. 3 c) to lower tem-
peratures with increasing x. The pretreated samples
show a much higher activity compared to the un-
treated samples (cf. curves in Figs 3 and 4). Espe-
cially in the low temperature range (100-150 °C)
they show total conversion. This could be attributed
to the presence of higher oxidation numbers for the
activated samples (cf. Tables II and IV).
Co is partially oxidized by the pretreatment to a
high oxidation state (+3 or +4), and during re-
action it converts to the lower state of +2. The
loss is fastest for Lao.8Sro.2 Co0 3 _z and slowest for
Lao.8Sro.2 Feo.5 Coo.5 O3 ._z. This can also be deduced
from the behaviour in the cooling regime in which
the curve of Lao.8Sro.2 Feo.5 Coo.5 0 3 _z is identical
to the heating regime, while for the other catalysts,
conversion is much lower. This possibly explains
why the “light-off temperature” decreases with in-
creasing x for the pretreated samples.
Among the pretreated catalysts the highest ac-
tivity is found for the catalyst of stoichiom-
etry Lao.8Sro.2Feo.5 Coo.5 0 3 _z. This catalyst ex-
tends total conversion from 100 to 190 °C. This
could be due to the presence and stability of
The relatively fast loss of charge for x = 0.8 and
x = 1.0 is due to the high reduction potential of
the Co3+,4+ / Co2+ equilibrium. In the presence of
iron and since the reduction potential of the Fe3+,4+
/ Fe2+ equilibrium is less, it is quite reasonable that
the iron will stabilize the high oxidation state of Co.
This might also explain the low activity of the Fe
rich catalysts for hydrogenation.
oxidation states higher than +3.
A sample of
Lao.8Sro.2Feo.5 Coo.5 O2,86 treated similar to the in-
situ pretreatment without performing the reaction
possessed a mean oxidation number of +3.09.
In comparison to the other samples the decrease
of oxidation number is smaller, i. e. the decrease
for Lao.8 Sro.2Feo.5 Coo.5 0 3 _z is ca. 0.1 whereas
it is ca. 0.3 for Lao.8Sro.2Feo.2Coo.8O3._2 and
Lao.8Sro.2Co03_z.
For the Co rich catalysts we observe only methyl-
cyclohexane as a reaction product. From this re-
sult we can deduce that hydrogenation follows a
single step mechanism over these catalysts. A fur-
ther study of the mechanism of hydrogenation over
Lao.8Sro.2 Feo.5 Coo.5 0 3 _z is planned for the future.
The high activity of the pretreated
Lao.8Sro.2 Feo.5 Coo.5 0 3 _z perovskite can be ratio-
In comparing the catalytic activity of the per-
nalized as follows. Co centers on the surface are ovskites to that of noble metal supported catalysts
more active for the hydrogenation reaction than Fe
centers [7]. On the other hand, the Co-rich per-
such as 1%Pt / AI2O3 or 1%Rh / ALO3 [18], we find
that the perovskite La0.8Sr0.2 Fe0.5 Co0.5 O3_z works
ovskites are more likely to be reduced under the better at temperatures higher than 150 °C (Fig. 5
reaction conditions, as deduced from the oxida- and Table V). This behaviour makes the perovskites
suitable as hydrogenation catalysts at high temper-
ature. In practical use, however, the fact must be re-
spected, that minor contents of the feed may affect
tion numbers after the activity test (cf Table IV).
Therefore, an irreversible storage effect is consid-
ered for the pretreated Co-rich perovskites, i. e.
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM

934
C. Schinzer et al. • Toluene Hydrogenation over Perovskites
Table V. Comparison of the perovskites with other hydro-
genation catalysts.
• The perovskite structure is stable under reduc-
ing atmosphere at high temperatures and therefore
perovskites are acceptable as catalysts for the hy-
drogenation of olefinic and aromatic compounds.
• The most active composition is
Catalyst type
M axim um conv. Onset o f Ref.
rate light o ff cracking
<i03_ za 100% 180°C
this work
Lao.8Sro.2Feo.5Coo.
1% Rh / A l2Ch
1% Pt / AI2O 3
Lao.8Sro.2 Feo.5 Coo.5 0 3 _z after pretreatment with
oxygen.
65% 80°C
90% 140°C
2 2 0
|18]
118]
290
• The activity for hydrogenation is most probably
due to the mode of electronic conduction, which is
probably of the p-type in these perovskites.
a Activated sample.
the catalyst (e.g. the poisoning and decomposition
of perovskites by SO? [3]).
As known from previous work (e. g. [ 15] and liter-
ature cited therein) the hydrogenation demands not
only an active metal center for the reaction but also a
Lewis acidic support such as A L O 3 or SiOo-ALO^.
The Lewis acidity of the perovskites is controlled
by the charge carriers (/. e. holes or electrons) [7],
the type of which depends on the transition metal
ion on the B-site. The active center is obviously lo-
cated on the B-sites of the perovskites whereas the
role of the A-sites is restricted to a charge deposit
to provoke higher oxidation states of the transition
metal ions.
Acknowledgements
The authors wish to thank the European Community
and the Hashemite Kingdom of Jordan for support under
project no. SEM 03 / 628 / 033. Also, we express our grat-
itude to the Deanship of Scientific Research and Graduate
Studies at Yarmouk University for support under project
no. 15 / 96. We are indebted to Dr. K. A. Murad and
F. El-Ababneh of the Physics Department, Yarmouk Uni-
versity, Irbid for performing the XRD measurements. The
help of Prof. Dr. K. G. Nickel, C. Berthold and B. Geißler
of the Institute of Mineralogy and Petrology, University
of Tübingen, with the BET surface area measurements is
much appreciated. C. S. expresses his gratitude to Prof.
Dr. S. Kemmler-Sack, Institute of Inorganic Chemistry,
University of Tübingen.
Conclusions
The following conclusions are drawn from the
work:
[10] N. Gunasekran, N. Bakshi, C. B. Alcock, J. J. Car-
berry, Solid State Ionics 83, 145 (1996).
[11] L. -W. Tai, M. M. Nasrallah, H. U. Anderson, D.
M. Sparlin, S. R. Sehlin, Solid State Ionics 76, 259
(1995).
[12] A. N. Petrov, O. F. Konochuk, A. V. Andreev, V. A.
Cherepanov, P. Kofstad, Solid State Ionics 80, 189
(1995).
[1] F. S. Galasso, Structure, Properties and Preparation
of Perovskite-Type Compounds. Pergamon Press,
Oxford (1969).
[2] R. J. H. Voorhoeve, in J. J. Burton and R. L. Garten
(eds): Advanced Materials in Catalysis, p. 129, Aca-
demic Press, New York (1977).
[3] L. G. Tejuca, J. L. G. Fierro, J. M. D. Tascön, Adv.
Catal. 36, 237 (1989).
[13] R. D. Shannon, N. Prewitt, Acta Crystallogr. A 32,
751 (1976).
[4] B. T. Matthias, J. P. Remeika, R. J. H. Voorhoeve, U.
K. Patent, 1,434,609, June 1973.
[14] J. E. Germain, Catalytic Conversion of Hydrocar-
bons, Academic Press, London, New York (1969).
[15] P. Chou, M. A. Vannice, J. Catal. 107, 129 (1987).
[16] C. B. Alcock, J. J. Carberry, R. Doshi, N. Gu-
nasekran, J. Catal. 143, 533 (1993).
[17] R. Doshi, C. B. Alcock, N. Gunasekran, J. J. Car-
berry, J. Catal. 140, 557 ( 1993).
[18] R. A. Saymeh. H. M. Asfour, Mu’tah J. Res. Stud.
Ser. B. 10, 135 (1995).
[5] Y. Teraoka, H. Nii, S. Kagawa, K. Jansson, M. Ny-
gren. J. Mater. Chem. 36, 97 (1996).
[6] J. O. Petunchi, M. A. Ulla, J. A. Marcos, E. A. Lom-
bardo, J. Catal. 70, 356 (1981).
[7] N. Gunasekran, J. J. Carberry. R. Doshi, C. B. Al-
cock, J. Catal. 146, 583 (1994).
[8] S. Rajadurai, J. J. Carberry, B. Li, C. B. Alcock,
Catal. Lett. 4, 43 (1990).
[9] J. J. Carberry, C. B. Alcock, U. S. Patent, 5,028.404,
July 1991.
Brought to you by | University of Sussex Library
Authenticated
Download Date | 3/2/19 10:10 AM