THE POISONING EFFECT OF Co3O4 ON γ-Al2O3 AND MgO SUPPORTS DURING THE DECOMPOSITION OF ISOPROPYL ALCOHOL

Article abstract of DOI:10.1135/cccc19952057

Catalysts containihg 20 wt.percent Co3O4 supported on γ-Al2O3 and MgO were prepared.The catalyst precursors were analyzed by TG, DTG and DTA, and the calcined catalysts (400-700 deg C) were characterized by XRD.The amount of excess surface oxygen for the catalysts was estimated iodometrically.The vapour-phase decomposition of isopropylalcohol over these catalysts was studied.Co3O4/γ-Al2O3 catalysts are selective towards propene formation, whereas Co3O4/MgO catalysts are selective towrds acetone formation.The poisoning effect of Co3O4 on both supports and the activity of the catalysts is discussed.

Full text of DOI:10.1135/cccc19952057

The Poisoning Effect of Supported Co3O4
2057
THE POISONING EFFECT OF Co
3
O
4
ON γ-Al
2
3
O AND MgO SUPPORTS
DURING THE DECOMPOSITION OF ISOPROPYL ALCOHOL
Samih A. HALAWY, Mohamed A. MOHAMED and Suzan F. ABD EL-HAFEZ
Department of Chemistry,
Faculty of Science, Qena 83511, Egypt
Received May 25, 1994
Accepted May 22, 1995
Catalysts containing 20 wt.% Co O supported on γ-Al O and MgO were prepared. The catalyst
3
4
2
3
precursors were analyzed by TG, DTG and DTA, and the calcined catalysts (400–700 °C) were
characterized by XRD. The amount of excess surface oxygen for the catalysts was estimated
iodometrically. The vapour-phase decomposition of isopropyl alcohol over these catalysts was
studied. Co O /γ-Al O catalysts are selective towards propene formation, whereas Co O /MgO
3
4
2
3
3
4
catalysts are selective towards acetone formation. The poisoning effect of Co O on both supports
3
4
and the activity of the catalysts is discussed.
Cobalt oxide catalysts are currently used in petrochemical hydrodesulfurization pro-
1
2
cesses and in the partial oxidation of alkenes . Supported or mixed Co O /Al O cata-
3
4
2
3
3
lysts have also been used in the decomposition of H O (ref. ). A few papers have been
devoted to the characterization of cobalt oxide/supported catalysts . Co O /MgO cata-
lysts have been used in various chemical reactions such as the oxidation of propene .
Adsorption of O , NO and NO on CoO–MgO catalysts has been reported. Many
characterization studies have been made on magnesia supports , MgO–CoO solid solu-
tion and metal ions/MgO catalysts .
2
2
4
,5
3
4
6
7
2
2
8
9
10
The present work deals with the poisoning effect of Co O on two supports, γ-Al O
3
3
4
2
and MgO, during the decomposition of isopropyl alcohol.
EXPERIMENTAL
Materials and Techniques
1
1
The γ-Al O support was prepared following Murgulescu et al. from aluminium hydroxide Al(OH)3
2
3
precipitated by the reaction of AlCl (BDH, U.K.) with NH OH. The substance was then calcined at
3
4
7
50 °C in air for 10 h. The MgO support was prepared by direct calcination of MgCO (Prolabo,
3
France) in air at 900 °C for 4 h. The catalyst samples containing 20 wt.% Co O supported on γ-Al O
3
3
4
2
and MgO, respectively, were prepared as follows: 18.13 g of Co(NO ) . 6 H O (BDH, U.K.) was
3
2
2
dissolved in 50 ml of deionized water and 20 g of the support (γ-Al O or MgO) was suspended in the
2
3
solution. The mixtures were evaporated to dryness over a water bath with continuous stirring. The
Collect. Czech. Chem. Commun. (Vol. 60) (1995)

2
058
Halawy, Mohamed, Abd El-Hafez:
resulting materials were dried overnight, in an oven at 120 °C and finally calcined in air for 5 h at
4
00–700 °C.
Thermal analyses (i.e. thermogravimetry (TGA), derivative thermogravimetry (DTG) and differential
1
2
thermal analysis (DTA)) were carried out by the conventional methods described elsewhere . X-Ray
powder diffractograms were obtained on a JSX-60 X-ray diffractometer (JEOL, Japan) equipped with
a Zr-filtered MoKα radiation source (λ = 0.7093 Å). Excess surface oxygen for the catalysts under
1
3
investigation was determined iodometrically in acidic medium . Measurements of the catalytic activity
and selectivity of the catalysts towards the decomposition of isopropyl alcohol (IP) were conducted
in the continuous flow mode at atmospheric pressure. A mixture of 1.4% IP in N , obtained by pas-
2
sing the carrier gas through liquid IP held at 3 °C, was used. The total flow rate was kept constant
at 215 ml/min. Catalyst sample (0.5 g) in a glass reactor 16.5 cm long, 1 cm i.d., was stabilized for
2
h at 190 °C before measurement. The feed exit gases were analyzed by gas chromatography (Shi-
madzu GC-14A, equipped with a Shimadzu Chromatopac C-R4AD data processor) using a stainless
steel column (PEG 20M 20% on Chromosorb W, 60/80 mesh, 3 m × 3 mm) at 110 °C. Catalytic
activity was expressed as the degree of conversion of supplied IP to all products, and selectivity as
the mole ratio of a particular product in all decomposition products, respectively. The reaction rates
1
4
and reaction order were calculated as described elsewhere .
Surface area measurements of the samples were carried out by the BET method with N as the
1
5
2
1
6
adsorbate, by using a conventional volumetric gas adsorption apparatus . The basicity of the
Co O /MgO catalysts was determined by investigating the adsorption of acetic acid molecules by the
3
4
1
7
pulse method . An amount of 0.5 g of each catalyst was accommodated in the reactor at the reaction
temperature. Pulses of a heptane solution of 0.1 M acetic acid (10 µl) were injected into the sample
followed, after one minute, by passing the flow of IP. The concentration of unreacted IP was calcu-
lated after each injection and the pulses were stopped when the adsorbate (acetic acid) peak appeared
in the chromatogram. The number of pulses characterized the basicity of each catalyst. Similarly, the
acidity of the Co O /γ-Al O catalysts was determined by examining the adsorption of pyridine.
3
4
2
3
RESULTS AND DISCUSSION
Thermal Analyses
TGA, DTG and DTA curves of the 20 wt.% Co O /γ-Al O precursor are shown in Fig. 1a.
3
4
2
3
The TGA curve exhibits two overlapping weight loss steps between 75–179 °C, fol-
lowed by a third step between 179–231 °C, with a total weight loss of 30.9% of the
initial sample weight. The DTG curve of this sample displays three peaks with Tmax at
9
2, 135 and 198 °C corresponding to the three steps in the TGA curve. In the DTA of
the Co O /γ-Al O precursor, there are three endothermic peaks located at 90, 133 and
3
4
2
3
1
96 °C due to the dehydration and decomposition of cobalt nitrate supported on γ-Al O
2
3
1
8,19
(
refs
). The TGA curve of the 20 wt.% Co O /MgO precursor (Fig. 1b) shows one
3
4
step within the temperature range of 320–390 °C, associated with a weight loss of
7.5% of the initial sample. The DTA curve of this sample exhibits a peak at 379 °C
3
with a shoulder at 349 °C. The DTA curve (Fig. 1b) shows a large split endotherm with
Tmax at 366 and 378 °C, corresponding to the dehydration and decomposition processes
of cobalt nitrate and formation of Co O on MgO. The activation energies (∆E) for the
3
4
thermal decomposition of the two precursors (nitrates) were calculated from the TGA
Collect. Czech. Chem. Commun. (Vol. 60) (1995)

The Poisoning Effect of Supported Co3O4
2059
data using the Coats–Redfern equation20 as explained elsewhere . The values obtained
21
are ∆E = 63.1 ± 0.6 and 86.2 ± 6.1 kJ/mol for the two decomposition steps of γ-Al O
2
3
supported Co(II) nitrate, and ∆E = 218.3 ± 8.7 kJ/mol for MgO supported Co(II) nitrate
2
2
(
in a good agreement with the values found for pure cobalt nitrate ).
X-Ray Diffraction Analysis
XRD diffractograms of Co O /γ-Al O , calcined at 400–700 °C for 5 h in air (Fig. 2a)
3
4
2
3
agree well with published values for both γ-Al O (ASTM 10-425) and Co O (ASTM
2
3
3
4
9
-418). They consist of lines related to the γ-Al O support (d = 2.406, 1.985 and 1.413 Å;
2 3
2
3
24
ref. ) and lines belonging to Co O (d = 2.804 and 1.539 Å; ref. ). The XRD patterns
3
4
of the Co O /MgO catalyst calcined in air for 5 h are shown in Fig. 2b. The diffraction
3
4
lines are consistent with published data for MgO (ASTM 4-0829), Co O (ASTM 9-418)
3
4
and the (Co,Mg)O system (ASTM 2-1201). The lines at d = 1.216, 1.487 and 2.105 Å
2
5
are due to the MgO support , whereas those at d = 0.942, 1.055 and 2.431 Å are
2
4
attributed to Co O (ref. ). The line at d = 1.267 Å (with an intensity of 6%) is ascribed
3
4
a
DTA
0
loss, wt.%
1
0
1
96
1
33
2
0
1
98
9
0
135
TGA
DTG
3
0
92
b
DTA
0
loss, wt.%
1
0
2
0
3
66
378
79
3
4
0
0
3
49
TGA
DTG
3
30
130
230
330
430
530
T, °C
FIG. 1
TGA, DTG and DTA curves of the thermal decomposition of a 20 wt.% Co O /γ-Al O , b 20 wt.%
3
4
2
3
Co O /MgO precursors, carried out at 10 °C/min in a dynamic atmosphere of dry N (40 ml/min)
3
4
2
Collect. Czech. Chem. Commun. (Vol. 60) (1995)

2
060
Halawy, Mohamed, Abd El-Hafez:
to the formation of a new (Co,Mg)O phase (ASTM 2-1201). From the above results of
XRD analysis it can be concluded that Co O does not react with γ-Al O within the
3
4
2
3
temperature range of 400–700 °C, whereas it partially reacts with the support forming
(
Co,Mg)O.
Catalytic Activity Measurements
It should be noted that the γ-Al O and MgO supports used in the present study were
2
3
calcined at temperatures higher than the calcination temperatures of the supported
Co O oxide catalysts (e.g. γ-Al O was originally calcined at 750 °C while the MgO
3
4
2
3
support was calcined at 900 °C). Hence, it is reasonable to assume that neither of the
supports in the Co O /γ-Al O and Co O /MgO systems suffered any change during the
3
4
2
3
3
4
calcination of the systems from 400–700 °C. Consequently, all changes in the catalytic
activity of the supported catalysts can be attributed to Co O rather than to γ-Al O or
3
4
2
3
MgO.
a) Over Co O /γ-Al O catalysts: γ-Al O oxide is known to possess Lewis type
3
4
2
3
2
3
2
6
acidity which arises during its preparation from Al(OH) (ref. ). The activated alumina
can be rehydrated by water formed during the reaction . Therefore, the pure γ-Al O
3
2
7
2
3
support exhibits a very high catalytic activity and selectivity towards the dehydration
reaction of IP (to produce propene), see Table I.
The addition of Co O (which is known to possess both acidic and basic properties)
3
4
to γ-Al O generally resulted in a reduction of the activity of γ-Al O (Table I). The
2
3
2
3
catalytic activity of the four Co O /γ-Al O catalysts increased from 400 °C to reach a
3
4
2
3
700 °C
8
0
0
700 °C
80
40
0
0
I/I
I/I
4
6
00 °C
00 °C
00 °C
6
00 °C
8
0
0
80
40
4
5
5
00 °C
8
0
0
80
40
4
4
8
0
0
400 °C
80
40
4
0
0
1
.4
1.8
2.2
2.6
3.0
0.6
1.2
1.8
2.4
d, Å
3.0
d, Å
FIG. 2
XRD patterns of a 20 wt.% Co O /γ-Al O , b 20 wt.% Co O /MgO, calcined at different tempera-
3
4
2
3
3
4
tures for 5 h in air
Collect. Czech. Chem. Commun. (Vol. 60) (1995)

The Poisoning Effect of Supported Co3O4
2061
maximum at 600 °C. Subsequently, the catalytic activity decreased at 700 °C. This
trend could be correlated with surface excess charge data and acidity measurements
(
Table I).
The catalytic activity measurements give evidence that cobalt oxide blocks the active
centres of γ-Al O , thus reducing its activity while having no effect on the support
2
3
selectivity. Hence, the catalyst centres presumably form on alumina domains but not on
cobalt oxide domains.
b) Over Co O /MgO catalysts: The pure basic MgO oxide support is observed to
3
4
possess a very high catalytic activity and selectivity towards the dehydrogenation reac-
tion of IP (to produce acetone), see Table I.
The addition of Co O to MgO resulted in a general decrease in the catalytic activity
3
4
of the pure MgO support. The catalytic activity was found to be inversely proportional
to the calcination temperature. This behaviour could be correlated with the values of
their excess surface charge and basicity measurements (Table I). Similarly, the change
in the catalytic activity could be attributed to Co O rather than to MgO support be-
3
4
TABLE I
Results of decomposition of isopropyl alcohol over Co O /γ-Al O and Co O /MgO catalysts at dif-
3
4
2
3
3
4
ferent calcination temperatures (in °C). All catalysts were calcined in air for 5h
Catalyst
Parameter
2
0 wt.% Co O /γ-Al O
20 wt.% Co O /MgO
3
4
2
3
3
4
γ-Al O
MgO
2
3
support
support
4
00
500
600
700
400
500
600
700
Conversion, %
99.3
27.0
56.4
61.9
57.5
99.9
12.0
73.3
–
4.9
30.0
–
2.7
0.9
5.6
–
RTP
611.6 169.7 347.3 381.2 354.1 614.5
16.9
–
Rp
610.9
0.7
89.5 345.2 380.8 353.5
1.2
0.7 613.2
99.8 0.2
Ra
78.7
2.1
99.4
0.6
3.1
61.8
2.9
–
0.3
99.9
0.1
9.8
61.1
3.4
–
73.3
–
30.0
–
16.9
–
5.6
–
Sp
99.88 53.6
0.11 46.4
Sa
0.2
1.8
60.5
1.9
–
99.8 100
100
100
100
–
3
Oex, atom% . 10
–
2.6
60.4
1.7
–
–
15
–
46.3
34.3
32.3
–
12.2
23.8
–
2.6
16.2
–
2
SBET, m /g
42
38.1
–
a
Acidity
4.8
a
Basicity
–
69.3
24
22
18
3
RIP the total rate of removal of IP (mmol/h g cat.), R the rate of formation of propene (mmol/h g
p
cat), R the rate of formation of acetone (mmol/h g cat.), S the selectivity towards the formation of
a
p
propene, S the selectivity towards the formation of acetone, O the excess surface oxygen.
a
ex
a
2
–3
Acidity and basicity are expressed as mmol/m
. 10 .
cat
Collect. Czech. Chem. Commun. (Vol. 60) (1995)

2
062
Halawy, Mohamed, Abd El-Hafez:
cause MgO was initially calcined at 900 °C, i.e. much higher than the calcination tem-
perature of all the four catalysts.
The catalytic activity measurements (Table I) demonstrate that the poisoning effect
of Co O on the MgO support is twice as large as on γ-Al O . For example,
3
4
2
3
Co O /MgO calcined at 400 °C reduced the catalytic activity to about 1/8 with respect
3
4
to that of the pure MgO; at the same calcination temperature, Co O /γ-Al O reduced
3
4
2
3
the catalytic activity of pure γ-Al O to about 1/4 of its initial value.
2
3
The surface excess charges seem to be more effective in case of the Co O /MgO
3
4
catalysts as compared to the Co O /γ-Al O system.
3
4
2
3
The activation energies for the dehydration of IP over Co O /γ-Al O and over
3
4
2
3
Co O /MgO, both at 500 °C, were calculated to be 81.3 ± 3.4 and 77.5 ± 9.1 kJ/mol,
3
4
respectively, with the reaction order n = 0.24 and 0.26, respectively.
Finally, the catalytic activity of pure Co O is known to be influenced by the calci-
3
4
2
8
nation temperature, e.g. in the decomposition of H O (ref. ). Therefore, the effect of
2
2
calcination temperature of pure Co O on its catalytic activity during the decomposition
3
4
of IP will be dealt with in a separate study.
REFERENCES
1. Gates B. C., Katzer J. R., Schuit G. C. A.: Chemistry of Catalytic Processes, Chap. 5.
McGraw–Hill, New York 1979.
2. Linde V. R., Margolis L. Ya., Lubentsov B. Z., Echmaeva T. A., Shashkin D. P.: Oxid. Commun.
5
, 49 (1984).
3
4
5
6
7
. Goldstein J. R., Tseung A. C. C.: J. Catal. 32, 452 (1974).
. Castner D. G., Watson P. R., Chan I. Y.: J. Phys. Chem. 93, 3188 (1989).
. Castner D. G., Watson P. R., Chan I. Y.: J. Phys. Chem. 94, 819 (1990).
. Ilieva L., Matyshak V., Kotsev N., Kadushin A., Shopov D.: Appl. Catal. 50, 37 (1989).
. Indovina V., Cordischi D., Febbraro S., Occhiuzzi M.: J. Chem. Soc., Faraday Trans. 1 81, 37
(
1985).
8
9
. Sidjabat O., Trimm D. L., Wainwright M. S.: J. Chem. Technol. Biotechnol. 56, 241 (1993).
. Moretti G., Cimino A., Minelli G., De Angelis B. A.: J. Electron Spectrosc. Relat. Phenom. 40,
8
5 (1986).
1
1
0. Shi C., Xu M., Rosynek M. P., Lunsford J. H.: J. Phys. Chem. 97, 216 (1993).
1. Murgulescu I. G., Cismaru D., Costeanu E., Miu P.: Proc. 4th Int. Congr. Catal., Moscow 1968,
Vol. I, p. 137. Akad. Kiado, Budapest 1971.
1
1
1
2. Mohamed M. A., Halawy S. A., Ebrahim M. M.: J. Anal. Appl. Pyrolysis 27, 109 (1993).
3. Girgis B. S., Mourad W. E., Ghazi T. M.: Surf. Technol. 7, 367 (1978).
4. Bond G. C., Halawy S. A., Abd El-Salaam K. M., Hassan E. A., Ragih H. M.: J. Chem. Technol.
Biotechnol. 59, 181 (1994).
1
1
1
1
1
2
5. Brunauer S., Emmett P. H., Teller E.: J. Am. Chem. Soc. 60, 309 (1938).
6. Joy A. S.: Vacuum 3, 254 (1953).
7. Ai M.: J. Catal. 52, 16 (1978).
8. Simmons E. L., Wendlandt W. W.: Thermochim. Acta 3, 25 (1971).
9. Duval C., Wadier C.: Anal. Chim. Acta 23, 257 (1960).
0. Coats J. W., Redfern J. P.: Nature 201, 68, 101 (1964).
Collect. Czech. Chem. Commun. (Vol. 60) (1995)

The Poisoning Effect of Supported Co3O4
2063
2
2
2
2
2
2
1. Halawy S. A., Mohamed M. A.: Collect. Czech. Chem. Commun. 59, 2253 (1994).
2. Mu J., Perlmutter D. D.: Thermochim. Acta 56, 253 (1982).
3. Fouad N. E., Knozinger H., Zaki M. I., Mansour S. A. A.: Z. Phys. Chem. 171, 75 (1991).
4. Halawy S. A., Mohamed M. A., Bond G. C.: J. Chem. Technol. Biotechnol. 58, 237 (1993).
5. Hanuza J., Jezowska-Trzebiatowska B., Oganowski W.: J. Mol. Catal. 29, 109 (1985).
6. Peri J. B.: Actes Congr. Int. Catalyse, 2e, Paris 1960, Vol. I, p. 1333. Edition Technip, Paris
1
961.
2
2
7. Parera J. M., Figoli N. S.: J. Catal. 14, 303 (1969).
8. Mohamed M. A., Halawy S. A.: Bull. Fac. Sci., Assiut Univ. 19, 23 (1990); Chem. Abstr. 115,
1
73416 (1991).
Collect. Czech. Chem. Commun. (Vol. 60) (1995)