Kinetic characterization of the reduction of silica supported cobalt catalysts

Article abstract of DOI:10.1007/s10973-006-7901-y

The reduction process of silica supported cobalt catalyst was studied by thermal analysis technique. The reduction of the catalyst proceeds in two steps: Co₃O₄ + H₂ to 3CoO + H₂O, 3CoO + 3H₂ to 3Co + 3H₂O which was validated by the TPR and in-situ XRD experiments. The kinetic parameters of the reduction process were obtained with a comparative method. For the first step, the activation energy, E _a, and the pre-exponential factor, A, were found to be 104.35 kJ mol^-1 and 1.18?10⁶～2.45?10⁹ s^-1 respectively. The kinetic model was random nucleation and growth and the most probable mechanism function was found to be f(α)=3/2(1- α)[-ln(1-α)]^1/3 or in the integral form: g(α)=[-ln(1-α)]^2/3. For the second step, the activation energy, E _a, and the pre-exponential factor, A, were found to be 118.20 kJ mol^-1 and 1.75?10⁷～2.45 ? 10 ⁹s^-1 respectively. The kinetic model was a second order reaction and the probable mechanism function was f(α)=(1-α) ² or in the integral form: g(α)=[1-α]^-1-1.

Full text of DOI:10.1007/s10973-006-7901-y

Journal of Thermal Analysis and Calorimetry, Vol. 90 (2007) 2, 415–419
KINETIC CHARACTERIZATION OF THE REDUCTION OF SILICA
SUPPORTED COBALT CATALYSTS
Y. J. Wan, J. L. Li* and D. H. Chen**
Key Laboratory of Catalysis and Materials Science of Hubei Province, College of Chemistry and Materials Science,
South-Central University for Nationalities, Wuhan 430074, China
The reduction process of silica supported cobalt catalyst was studied by thermal analysis technique. The reduction of the catalyst
proceeds in two steps:
Co3O4 +H2®3CoO+H2O, 3CoO+3H2®3Co+3H2O
which was validated by the TPR and in-situ XRD experiments. The kinetic parameters of the reduction process were obtained with a
comparative method. For the first step, the activation energy, Ea, and the pre-exponential factor, A, were found to be 104.35 kJ mol–1
and 1.18·106~2.45·109 s–1 respectively. The kinetic model was random nucleation and growth and the most probable mechanism
function was found to be f(a)=3/2(1–a)[–ln(1–a)]1/3 or in the integral form: g(a)=[–ln(1–a)]2/3. For the second step, the activation
energy, Ea, and the pre-exponential factor, A, were found to be 118.20 kJ mol–1 and 1.75·107~2.45·109 s–1 respectively. The kinetic
model was a second order reaction and the probable mechanism function was f(a)=(1–a)2 or in the integral form: g(a)=[1–a]–1–1.
Keywords: cobalt catalyst, Fischer-Tropsch synthesis, kinetic parameters, reduction, thermal analysis
Introduction
mium catalysts using TPR data. Thermal analysis
technique was limited to the surface characterization
of solid state. However, it offers quantitative kinetic
information, which is a distinct advantage; it is thus
an important tool for the study of heterogeneous sys-
tem [12].
Fischer-Tropsch synthesis (FTS) is a process for hy-
drocarbon production from synthesis gas. The syn-
thetic fuel products via Fischer-Tropsch synthesis
have some advantages like high cetane numbers, low
sulfur and low aromatics contents. The fuel is suit-
able for diesel engines and environmental friendly
[1, 2]. Cobalt catalyst is normally used for Fischer-
Tropsch synthesis of long-chain paraffin because of
its high activity, low water-gas shift activity and
comparatively low cost [3]. Generally, the reducibil-
ity [4], dispersion [5] and interaction between cobalt
and the support were studied with TPR, XRD, XPS
and other characterization techniques [6, 7]. The
thermal analysis technique was seldom used to study
the reduction of cobalt catalyst for FTS, and the
analysis of the kinetic model and the kinetic parame-
ters for the catalyst reduction are rarely reported.
Fouad [8] reported the study of the reduction of
WO3 ®WO2 ®W 0 and Moura et al. [9] reported the
kinetic characterization of the reduction of copper
and nickel catalysts using thermal analysis tech-
nique. Kanerro [10] reported the kinetic parameters
of alumina-supported vanadium oxide and
Sarbak [11] reported thermal characterization of alu-
minum supported chromium and platinum-chro-
Thermal analysis technique has been widely used
in the study of kinetics of solid state reaction [13] for
the determination of the activation energy, Ea, the
mechanism function, g(a), and the pre-exponential
factor, A [14]. Many kinetic methods including the in-
tegral method of Coats-Redfern [15], differential
method of Achar [16], variable iso-conversional
method [17], accommodated model of Gao [18] and
the comparative method of Chen et al. [19] have been
used. In general, the first step is to obtain the reliable
activation energy by an Ozawa-iterative procedure or
the KAS method [20–26]. The second step is to use the
comparative method of Chen [19, 21] to determine the
most probable mechanism function of the reaction and
to calculate the pre-exponential factor A. In this paper,
we report the thermal behavior of the reduction of the
silica-supported
cobalt
catalysts
used
for
Fischer-Tropsch synthesis and the determination of the
kinetic parameters with the comparative method.
*
**
Author for correspondence: lij@scuec.edu.cn
Author for correspondence: chendh46@hotmail.com
1388–6150/$20.00
Akadémiai Kiadó, Budapest, Hungary
Springer, Dordrecht, The Netherlands
© 2007 Akadémiai Kiadó, Budapest

WAN et al.
Experimental
Results and discussion
Catalyst preparation
The analysis of reduction of Co3O4 in the
Co3O4/SiO2 catalyst
The Co(7.6%)/SiO2 catalyst prepared by incipient wet-
ness impregnation was used for the study. Before the
impregnation, the SiO2 support (Qingdao Meigao
Chemical Factory, China), with average pore size of
6.8 nm; pore volume of 0.78 cm3 g–1 and surface area of
457.3 m2 g–1, was calcined at 773 K for 6 h. Then the re-
quired amount of an aqueous solution of Co(NO3)2·
6H2O (AR grade, Guoyao Chemical Co., Ltd.,China)
was dissolved in an appropriate volume of deionized
water and added to the calcined silica support. Subse-
quently, the catalyst was dried in air at 393 K for 12 h,
and then calcined in air at 623 K for 5 h.
Four peaks at the temperature ranges of 300–391,
463–543, 543–704 and 704–840 K are observed in
the DTG curve (Fig. 1) demonstrating that the reac-
tion preceded in four steps. From the TPR curve
(Fig. 2), three hydrogen consumption peaks marked
peaks I, II and III are observed. This indicates that hy-
drogen consumption consisted of three steps.
Two reference experiments, one for TG of the
catalyst under nitrogen atmosphere, the other for TG
of the support under hydrogen atmosphere, were
done. The results are shown in Figs 3 and 4. No phase
change and reduction of the catalyst under nitrogen
and of the silica support under hydrogen are observed
except the dehydration process occurring at the tem-
perature range of 303–391 K. From the DTG and TPR
curves in Figs 1 and 2, it is observed that peaks II, III
and IV in the DTG curve corresponds to peaks I, II
and III of the TPR curve, indicating that the oxygen
mass loss process in the thermal gravimetric experi-
Thermogravimetry (TG)
Thermal gravimetric analysis (TG) was performed on a
Perkin-Elmer TGS-2 thermobalance with various heat-
ing rates. The catalyst (6.00~8.00 mg) was first heated
under a nitrogen flow at a rate of 60 mL min–1 from 303
to 423 K and held at 423 K for 30 min, cooled down to
room temperature, and heated again from 303 to 1073 K
under a hydrogen flow at a rate of 60 mL min–1.
The temperature was raised linearly with heating rates
of 5, 10, 15, 20, 25 K min–1. The mass loss was automat-
ically recorded using a computer.
Temperature programmed reduction (TPR)
The H2-TPR measurements were performed with a
Zeton Altamira AMI-200 catalyst characterization in-
strument. The sample (0.15 g) was placed in a quartz
tubular reactor. Prior to the TPR measurement, the
catalyst was flushed with high purity argon at 423 K
for 1 h, then cooled down to 323 K. High purity hy-
drogen was switched on and the flow rate through the
reactor was controlled at 60 mL min–1. The tempera-
ture was raised at a rate of 10 K min–1 from 323 to
1073 K. The H2 consumption (thermal conductivity
detector, TCD, signals) was recorded automatically.
Fig. 1 TG-DTG curves of Co3O4/SiO2 catalyst under hydro-
gen; heating rate 10 K min–1
80
60
I
II
X-ray powder diffraction (XRD)
40
20
TPR
III
The phase change of the catalyst with reduction temper-
ature was detected in situ by a Bruker-D8 X-ray
diffractometer with Cu-Ka radiation under hydrogen at-
mosphere. The scan range was 20–80° with 0.015°
steps. The catalyst was first heated under N2 atmosphere
from 303 to 423 K and held at 423 K for 30 min and
then cooled down to room temperature, then heated
again from 303 to 1073 K at a heating rate of 10 K min–1
under H2 atmosphere. The sample was scanned at the
temperatures of 383, 533, 603, 703, 823 and 873 K.
0
–20
300
400
500
600
700
800
900 1000
Temperature/K
Fig. 2 the TPR curve of Co3O4/SiO2 catalyst under hydrogen;
heating rate 10 K min–1
416
J. Therm. Anal. Cal., 90, 2007

KINETIC CHARACTERIZATION OF THE REDUCTION OF SILICA SUPPORTED COBALT CATALYSTS
Co3O4
CoO
Co
873 K
823 K
703 K
603 K
533 K
383 K
20
30
40
50
60
70
80
2 °
Fig. 3 TG-DTG curves of sample under nitrogen
Fig. 5 XRD patterns of the catalyst
experiments. We did not found the interaction phase
by XRD, probably due to the small crystal size.
Calculation of activation energy (Ea)
The TG-DTG curves at different heating rates show
that the reduction temperature increased with increas-
ing heating rate (Figs 6 and 7). The smaller the heat-
ing rates, and the lower the reduction temperature, it’s
clear that the lower heating rate was useful for the oc-
currence of reduction processes. According to the ba-
sic thermal theory for the KAS method and Ozawa
method, the parameters a, T, da/dT were used to cal-
culate the activation energy.
Fig. 4 TG-DTG curves of support under hydrogen
The average Ea values calculated using the itera-
tive procedure for the first and the second reduction
steps are 104.35 and 118.20 kJ mol–1 respectively.
Variations of activation energy, Ea, with reduction
reaction fraction, a, by the KAS method for the two re-
duction steps are shown in Figs 8 and 9. It can be seen
that the activation energy of both the reduction pro-
cesses changed little with a. The reduction proceed un-
der rather moderate conditions.
ment corresponds to the hydrogen consumption pro-
cess in the TPR experiments.
Thus, steps II, III and IV are due to the reduction of
the Co3O4/SiO2 catalyst. Quantitatively, the mass losses
at steps II and III are 0.887 and 2.093% respectively,
while the mass loss at step III is approximately three
times that at steps II. From the TPR profiles, the area of
peak II is also about three times the area of peak I. Thus,
the reduction of Co3O4/SiO2 proceeds as follow:
1st step:
nd step:
Co3O4 +H2 ®3CoO+H2O
3CoO+3H2 ®3Co+3H2O
2
This is consistent with previous studies
[4, 6, 29]. Steps IV at 703–840 K could be the reduc-
tion of the interaction between cobalt and silica sup-
port [7, 29].
From the XRD patterns (Fig. 5), under hydrogen
atmosphere only Co3O4 phase can be found after heat-
ing at 383 K. Increasing the temperature from 383 to
533 K, Co3O4 crystallites disappeared and CoO crys-
tals were formed. Increasing the temperature from
533 to 703 K the CoO is no longer detectable, being
replaced by Co° crystallites indicating that the reduc-
tion of Co3O4 has been completed at temperature be-
low 703 K, it is consistent with the TG and the TPR
Fig. 6 TG curve of catalyst under hydrogen at different heat-
ing rates
J. Therm. Anal. Cal., 90, 2007
417

WAN et al.
200
160
120
80
40
0
0.0
0.2
0.4
0.6
0.8
1.0
Fig. 9 The change in activation energy, Ea, with a for the sec-
ond reduction step
Fig. 7 The DTG curve of catalyst under hydrogen at different
heating rates
Estimation of the mechanism of the reduction reaction
200
160
120
80
The values of a, T and da/dT obtained from the TG
curves, were inserted into the 31 types of the mecha-
nism functions [21], the Ea, A and the linear correla-
tion coefficients (R) were calculated by the linear
least squares method. Furthermore, when the ‘real’
kinetic model is obtained, the activation energies ob-
tained from KAS equation or iterative procedure will
conform to Achar equation [27] and Coats-Redfern
equation [28] with better correlation coefficients. The
results from the different heating rates display the
same trend (Table 1).
40
0
0.0
0.2
0.4
0.6
0.8
1.0
Comparing the results of the 31 types of mecha-
nism function, the function No. 8, f(a)=
3/2(1–a)[–ln(1–a)]1/3, is the most probable mecha-
nism function of the first reduction process of
Fig. 8 The change in activation energy, Ea, with a for the first
reduction step
Table 1 Ea, A and R obtained by Achar method and Coats-Redfern method from different heating rates
Coats-Redfern method
Achar method
Reduction
steps
b/
Function
No.
K min–1
Ea/kJ mol–1
lnA/s–1
R
Ea/kJ mol–1
lnA/s–1
R
II
5
10
15
20
25
5
8
8
107.94
85.55
20.08
14.69
14.67
13.98
16.83
19.63
17.99
17.08
17.99
17.52
0.9989
0.9949
0.9957
0.9934
0.9986
0.9988
0.9986
0.9982
0.9978
0.9978
114.13
104.39
103.68
104.62
103.81
127.00
120.06
108.83
116.37
114.32
21.62
19.23
19.06
19.28
19.03
20.45
18.73
16.68
18.30
17.76
0.9978
0.9926
0.9913
0.9819
0.9758
0.9878
0.9886
0.9899
0.9838
0.9821
8
85.18
8
82.09
8
94.48
III
22
22
22
22
22
123.62
116.93
111.57
115.52
113.75
10
15
20
25
Table 2 The most probable mechanism function of different reduction process
Reduction steps
Function No.
Sign
Mechanism
f(a)
g(a)
II
8
A1.5
F2
Avrami-Erofee
Second order
3/2(1–a)[–ln(1–a)]1/3
(1–a)2
[–ln(1–a)]2/3
(1–a)–1–1
III
22
418
J. Therm. Anal. Cal., 90, 2007

KINETIC CHARACTERIZATION OF THE REDUCTION OF SILICA SUPPORTED COBALT CATALYSTS
authors also thank Prof. Kongyong Liew for helpful discus-
sions in the preparation of this manuscript.
Co3O4/SiO2 catalysts, its kinetic model was random
nucleation and growth model. The function No. 22,
f(a)=(1–a)2, is the most probable mechanism func-
tion for the second reduction process, its kinetic
References
model was a second order chemical reaction.
The most probable mechanism function of re-
duction process of Co3O4 in Co3O4/SiO2 are shown in
Table 2.
1 C. Knottenbelt, Catal. Today, 71 (2002) 437.
2 H. Schulz, Appl. Catal. A: Gen., 186 (1999) 3.
3 M. E. Dry, Catal. Today, 71 (2002) 227.
4 R. C. Reuel and C. H. Bartholomew, J. Catal., 85 (1984) 63.
5 E. Igesia, Appl. Catal. A: Gen., 161 (1997) 59.
Conclusions
6 I. Puskas, T. H. Fleisch, J. B. Hall, B. B. L. Meyers and
R. T. Roginski, J. Catal., 134 (1992) 615.
7 G. J. Haddad and J. G. Goodwin Jr., J. Catal., 157 (1995) 25.
8 N. E. Fouad, J. Anal. Appl. Pyrol., 44 (1997) 13.
9 A. Moura, A. S. Araujo, A. C. S. L. S. Coutinho,
J. M. F. B. Aquino, A. O. S. Silva and M. J. B. Souza,
J. Therm. Anal. Cal., 79 (2005) 435.
The reduction behavior and the reduction kinetic pa-
rameters of the Co3O4 in Co3O4/SiO2 catalyst for
Fischer-Tropsch synthesis were determined using the
thermal analysis technique and the thermal analysis
kinetic method. The conclusions are summarized as
follows:
10 J. M. Kanervo, M. Elina Harlin, A. Outi, I. Krause and
M. A. BaÔares, Catal.Today, 78 (2003) 171.
11 Z. Sarbak and W. Jóüwiak, J. Therm. Anal. Cal.,
85 (2006) 335.
The reduction process proceeds in two steps:
Co3O4 +H2 ®3CoO+H2O
12 H. Tanaka, Thermochim. Acta, 267 (1995) 29.
13 B. Malecka and D. E. Rozdz-Ciesla, J. Therm. Anal. Cal.,
68 (2002) 819.
3CoO+3H2 ®3Co+3H2O
14 J. P. Elder, Thermochim. Acta, 318 (1998) 229.
15 A. W. Coats and J. P. Redfern, Nature, 201 (1964) 68.
16 B. N. Achar, G. W. Brindley and J. H. Sharp, Proc. Int.
Clay Conf., 1 (1966) 67.
The reduction process proposed is consistent
with previous study, demonstrating that the thermal
analysis technique could be reliably used to study the
reduction of heterogeneous system.
17 C. Popescu, Thermochim. Acta, 285 (1996) 309.
18 Z. Gao, M. Nakada and I. Amassaki, Thermochim. Acta,
369 (2001) 137.
• The kinetic parameters and the kinetic model of the
reduction process were obtained. The kinetic
model of the first reduction process is random nu-
cleation and growth. The activation energy, Ea, and
the pre-exponential factor, A, was determined to be
104.35 kJ mol–1 and 1.18·106~2.45·109 s–1 respec-
tively. The most probable mechanism function in
the differential form is f(a)=3/2(1–a)[–ln(1–a)]1/3
or in integral form, g(a)=[–ln(1–a)]2/3. The kinetic
model for the second reduction process is a second
order reaction. The activation energy is
118.20 kJ mol–1, and the pre-exponential factor is
1.75·107~2.45·109 s–1. The most probable mecha-
nism function in differential form is f(a)=(1–a)2 or
in integral form g(a)=[1–a]–1–1.
19 G. Chunxiu, S. Yufang and C. Donghua, J. Therm. Anal.
Cal., 76 (2004) 203.
20 J. H. Flynn and L. A. Wall, J. Polym. Sci. Part B: Polym.
Phys., 4 (1966) 323.
21 X. Gao and D. Do. Llimore, Thermochim. Acta,
215 (1993) 47.
22 H. E. Kissinger, Anal. Chem., 29 (1957) 1702.
23 J. H. Flynm. Thermochim. Acta, 300 (1997) 83.
24 C. D. Doyle, Nature, 207 (1965) 290.
25 T. Ozawa, Chem. Soc. Jpn., 38 (1956) 1881.
26 L. A. Ortega, Perez-Maqueda and J. M. Criado,
Thermochim. Acta, 282 (1996) 29.
27 A. W. Coats and J. P. Redfern, J. Polym. Sci. Part B:
Polym. Phys., 3 (1965) 917.
28 B. N. Achar, Proc. Int. Clay Conf., 1 (1969) 6.
29 B. Ernst, S. Libs, P. Chaumette and A. Kiennemann, Appl.
Catal. A: Gen., 186 (1999) 145.
Acknowledgements
Received: July 28, 2006
The authors gratefully acknowledge the supports from the Na-
tional
Accepted: October 10, 2006
OnlineFirst: February 26, 2007
Natural
Science
Foundation
of
China
(20473114,20590360), Talented Young Scientist Foundation
of Hubei (2003ABB013), Excellent Young Teachers Program
of the Ministry of Education of China, the State Ethnic Affairs
Commission, PR China and Returnee Startup Scientific Re-
search Foundation of the Ministry of Education of China. The
DOI: 10.1007/s10973-006-7901-y
J. Therm. Anal. Cal., 90, 2007
419