Thermal decomposition study of silica-supported nickel catalyst synthesized by the ammonia method

Article abstract of DOI:10.1016/j.molcata.2007.11.033

Silica-supported nickel catalyst synthesized by the ammonia method and their precursors treated and untreated with ammonia solution were studied by temperature-programmed reduction (TPR), scanning electron microscopy (SEM), and simultaneous thermogravimetry, differential scanning calorimetry and differential thermal analysis (TGA-DSC-DTA) carried out in nitrogen atmosphere (100 ml/min) at different heating rates (5, 10, 15 and 20 K/min). Temperature effect on the catalyst structural composition was studied by X-ray diffraction (XRD). The results indicated that synthesized catalyst shows mainly a higher temperature metal phase (973 K), constituted by a superposition of bi-dimensional sheets similar to nickel phyllosilicate, resulting from the strong metal-support interaction (SMSI) between nickel aqua ammine complex and silica surface. At high temperature, such a compound degrades to stable silica-supported nickel catalyst. Activation energies and kinetic parameters for each decomposition process were calculated by Kissinger's method. Enthalpy values were determined from differential scanning calorimetric profiles. The activity tests of such a catalyst to methane steam reforming indicated that deactivation process caused by carbon deposition was less pronounced than deactivation of a similar nickel content catalyst prepared by impregnation.

Full text of DOI:10.1016/j.molcata.2007.11.033

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
Thermal decomposition study of silica-supported nickel
catalyst synthesized by the ammonia method
∗
´
Alfonso Loaiza-Gil , Marlin Villarroel, Jose F. Balbuena,
´
´
Marıa A. Lacruz, Sergio Gonzalez-Cortes
Laboratorio de Cine´tica y Cata´lisis, Departamento de Qu´ımica, Facultad de Ciencias,
Universidad de Los Andes, Merida 5101, Venezuela
Available online 5 December 2007
Abstract
Silica-supported nickel catalyst synthesized by the ammonia method and their precursors treated and untreated with ammonia solution were
studied by temperature-programmed reduction (TPR), scanning electron microscopy (SEM), and simultaneous thermogravimetry, differential
scanning calorimetry and differential thermal analysis (TGA–DSC–DTA) carried out in nitrogen atmosphere (100 ml/min) at different heating
rates (5, 10, 15 and 20 K/min). Temperature effect on the catalyst structural composition was studied by X-ray diffraction (XRD). The results
indicated that synthesized catalyst shows mainly a higher temperature metal phase (973 K), constituted by a superposition of bi-dimensional
sheets similar to nickel phyllosilicate, resulting from the strong metal–support interaction (SMSI) between nickel aqua ammine complex and silica
surface. At high temperature, such a compound degrades to stable silica-supported nickel catalyst. Activation energies and kinetic parameters
for each decomposition process were calculated by Kissinger’s method. Enthalpy values were determined from differential scanning calorimetric
profiles. The activity tests of such a catalyst to methane steam reforming indicated that deactivation process caused by carbon deposition was less
pronounced than deactivation of a similar nickel content catalyst prepared by impregnation.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Temperature-programmed reduction; Thermal studies; Nickel; Silica; Methane steam reforming
1. Introduction
973 K. Previous studies performed on silica-supported iron and
cobalt catalysts, by Fourier transformed infrared spectroscopy
(FTIR), X-ray photoelectron spectroscopy (XPS), scanning
electron microscopy associated with energy dispersive X-ray
spectroscopy (SEM-EDS), X-ray diffraction (XRD) and ther-
mal analysis (TG–DSC) suggested that iron and cobalt nitrates,
at basic pH, give rise to metal aqua ammine complex ions
formation that react with silica surface to form cobalt and
iron phyllosilicates like Co3(OH)4Si2O5, Co3(OH)2(Si2O5)2,
Fe2(OH)4Si2O5 and Fe2(OH)2(Si2O5)2. Their reduction has led
to highly loaded Co/SiO2 and Fe/SiO2 [2,5,6]. Owing to the sim-
ilarity of Ni2+ and Co2+ ions which both give rise to the same
type of phyllosilicates [7] it was thought to prepare a silica-
supported nickel catalyst using the same parameters employed
for the preparation of silica-supported cobalt catalysts. As the
first step, was considered more interesting to have a better under-
standing of the chemistry involved in the catalyst preparation
which was likely similar to cobalt catalyst than focusing the
attention on the preparation parameters influence, because of
the large number of studies already done by Martin and co-
workers focused to tuning the preparation parameters [1]. This
The ammonia method has been used to synthesize silica-
supported nickel and cobalt catalyst [1] in an effort to improve
the metal dispersion and the attrition resistance and decreases
the metal particle size and porous of the active phase. The
method consists to contact the silica support with a metal nitrate
solution to which ammonia solution was added. It was pro-
posed [1,2], that ammonia attacks the silica support more or less
strongly as revealed by the substantial increase of the specific
surface area of that precursor [3,4]. Furthermore, metal aqua
ammine complex ions formation like [M(H2O)6−n(NH3)n]m+
,
(m = metal oxidation state, n ≤ 6, coordination number), could
be favored and its can reacts with the silica surface originat-
ing bi-dimensional metal-support compounds of ferromagnetic
properties at low temperature [1]. The silica-supported nickel
catalyst was obtained by reduction under hydrogen flow at
∗
Corresponding author. Tel.: +58 2742401371.
E-mail address: loaizag@ula.ve (A. Loaiza-Gil).
1381-1169/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2007.11.033

208
A. Loaiza-Gil et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
strategy requires the characterization of the precursor along the
entire preparation route. Temperature-programmed reduction
and scanning electron microscopy were employed to achieve
information on the morphology and the number of metallic
phases and thermogravimetry, differential scanning calorime-
try and differential thermal analysis (TGA–DSC–DTA) were
used to study the chemistry involved in the catalyst preparation.
The last techniques demonstrated to be powerful tools to char-
acterize the conformational changes induced by temperature in
the sample under study, since they allow calculating the amount
of decomposed sample, energy changes and the amount of heat
involved in the decomposition reactions, besides the thermal
properties of the sample.
Among the differential methods [8] for the study and deter-
mination of thermal decomposition kinetic parameters, the
Kissinger’s method has been the most used [9]. This method
allows calculating the activation energy from calorimetric pro-
files obtained at different heating rates. The Kissinger equation
can be expressed as
cipitate keeping the pH around 11–12. After 1 h stirring, 5 g
of silica (aerosil 200 m2/g specific surface area, Degussa) was
added to the solution. The mixture was dried in an oven at 353 K
for 48 h after 24 h stirring. The silica-supported nickel cata-
lyst was obtained by reduction “in situ” under hydrogen flow
(10 mL/min) at 1073 K. The reduction time lasted 24 h. In order
to avoid unnecessary metal particles growing, the heating rate
was set at 1 K/min.
2.2. Experimental methods
Thermogravimetry, differential scanning calorimetry and dif-
ferential thermal analysis were carried out simultaneously on a
Thermal Analyzer TA Instruments, model SDT Q600 (1773 K)
under nitrogen flow (100 mL/min) using an average weight of
7.3 mg of sample and 5, 10, 15 and 20 K/min heating rate.
XRD measurements were performed in a powder diffractome-
˚
ter Phillips PW1050/25 equipped with a Cu K␣ (λ = 1.5406 A)
radiation source. The XRD data were taking at 2θ angular range,
step size of 0.002◦ and 10 s count time.
ꢀ
ꢁ
β
Tm2
Ea
Temperature-programmed reduction profiles (TPR) were
performed on a Micromeritics TPR/TPD 2900 from room tem-
perature to 1073 K at 10 K/min heating rate under 50 mL of 10%
hydrogen in argon. Samples of 150 mg were pretreated in air
before the TPR measurements. Micrographs were performed on
a scanning electron microscope Hitachi, S-2500.
ln
= −
+ C
(1)
RTm
where R is the universal constant of gases (J mol−1 K−1), β is
the heating rate (K s−1), Ea is the activation energy (J mol−1),
Tm is the peak temperature (K) and C is a constant. The activa-
tion energy is obtained from the slope (−Ea/R), of the straight
line obtained by plotting ln(β/Tm2 ) against 1/Tm. On the other
hand, the peak shape of the resulting thermal analysis can be
symmetrical or not. The peak symmetry is directly related with
the reaction order of the chemical process. It was found [9]
that when the reaction order n = 1, the peak has a symmetrical
shape with similar areas on the right and on the left of the max-
imum inflection point (Tm). When the reaction order falls, the
peak asymmetry increases. The left area of Tm is bigger than
that on the right. Therefore, it is possible to use the peak shape
as a variable to determine the reaction order from a differen-
tial thermal analysis (DTA). The shape index, S, is defined as
the ratio between the slopes of the peak tangents at the inflec-
tion point. Furthermore, it was found that plotting S versus n2,
both parameters S and n are related according to the following
equation:
Catalytic performance of silica-supported nickel catalyst for
methane steam reforming was tested in a continuous, fixed-bed
tubular reactor described previously [13]. A controlled heater
ensuredauniformbedtemperaturemonitoredbyathermocouple
placed in the reactor center. Carbon monoxide, carbon dioxide
and light hydrocarbons were analyzed by online gas chromatog-
raphy (Hewlett Packard 6890 Plus) equipped with a thermal
conductivity detector and a Porapak Q column (10 ft. × 1/8 in.,
80/100 mesh). 0.4 g of the catalyst was reduced “in situ” under
10 mL/min of hydrogen flow from room temperature to 1073 K,
at 1 K/min heating rate for 24 h. The reaction was performed at
973 K, atmospheric pressure and 500 h−1 gas hour space veloc-
ity (G.S.H.V.) using wet methane (CH4 + H2O). After run the
gas feed was closed and the reaction temperature was sharply
decreased to room temperature.
S = 0.63n2.
(2)
3. Results and discussion
2. Materials and methods
2.1. Catalyst preparation
Simultaneous thermal analyses TGA–DSC and DTA on hex-
ahydrate nickel nitrate carried out between room temperature
and 670 K and 15 K/min heating rate, are shown in Figs. 1 and 2
respectively. Fresh nickel salt precursor show five endothermic
peaks and 73.2% total mass lost due to water dehydration and
thermaldecompositionofnitratesgivingrisenickeloxideforma-
tion (II). The residual weight of 26.8% suggests that molecular
weight of nickel salt precursor was quite similar to that reported.
The molecular weight of thermal decomposed sample was cal-
culated by the following equation:
The silica-supported nickel catalyst was prepared according
to a method similar to that described by Martin and co-workers
[1]. The preparation of silica-supported 10% nickel catalysts
was as follows: 2.81 g of Ni(NO3)2·6H2O (98% purity, IQE)
was added to 56 mL of distilled water at room temperature and
vigorously stirred up to nickel salt dissolution. Nickel hydroxide
was precipitated by addition of 2–3 drops of ammonia solu-
tion (28% NH3, Fischer Scientific Company). A large excess
of ammonia solution (56 mL) was added to dissolve the pre-
M.W.NiO
residue(%) =
× 100
(3)
M.W.initial compound

A. Loaiza-Gil et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
209
The small weight difference (0.4%) found on commercial nickel
nitrate hexahydrate could be due to hydration water lost by salt
manipulation [10,11].
The first two decomposition steps occur along 331–364 K and
were assigned to hydration water lost. The dehydration temper-
ature difference was found to be dependent on the existence on
nitrates groups which several water bond types [10–14]. The
following two decomposition steps occur around 463–512 K
and are harder to explain. It was proposed [13,14] that these
two steps correspond to water and NOx losses and to the for-
mation of intermediate compounds type nickel hydroxynitrates
Ni2(NO3)2(OH)2 and nickel oxynitrates NiO.NiONO3. The last
step could be assigned to NOx and O2 lost from the intermediate
compounds originating stable nickel oxide. The enthalpy val-
ues of 290.1, 226.4 and 278.1 kJ/mol obtained for the first three
decomposition steps (Table 1) suggests that water molecules
with different bonding type were liberated. These results agreed
very well with the decomposition pattern of moisturized transi-
tion metal nitrates reported by Malecki et al. [13] and Zivkovic
et al. [14].
Fig. 1. Simultaneous (a) TGA and (b) DSC profiles (nitrogen flow, 100 mL/min)
of commercial Ni(NO3)2·6H2O at 15 K/min, heating rate.
Two roles were proposed [1] for the ammonia solution: the
first one was the functionalization of the silica support according
to the following reaction:
+
Si–OH + NH3 → SiO−NH4
(4)
The second one was the formation of nickel aqua ammin+e ions,
[Ni(H2O)6−n(NH3)n]2+ whichexchangewithsurfaceNH4 ions.
Taking account of two OH− groups for the electric charge bal-
ance of nickel aqua ammine ion, in solid state, the molecular
weightofnickelsalttreatedwithammoniasolutionwasexpected
to be between 200 g/mol (n = 0) and 194 g/mol (n = 6).
The TGA–DSC and DTA experiments on nickel salt treated
with ammonia solution (dried by solvent evaporation) can light-
ing the second approach and are shown in Figs. 3 and 4
respectively. The thermogravimetric profile show in this case
a decomposition pattern on three steps and 65.5% total mass
lost (35.5% residue). The calculated molecular weight (Eq.
(3)) of 217 g/mol resulted to be superior to the expected value.
The difference between the expected and calculated molecular
weight could be caused by an incomplete sample dried. The DSC
profile of this sample shows two endothermic peaks with mini-
Fig. 2. DTA profiles of commercial Ni(NO3)2·6H2O (nitrogen flow,
100 mL/min) at (a) 5 K/min, (b) 10 K/min, (c) 15 K/min and (d) 20 K/min heating
rate.
Table 1
Calculated kinetic decomposition parameters of synthesized silica-supported nickel catalyst and their precursor treated and untreated with ammonia solution
Sample
Weight losses
ꢀH (KJ/mol)
S
n
Ea (kJ/mol)
ln A
κ (s−1
)
Ni phyllosilicate
1
2
292.1
243.3
1.1
0.4
1.3
0.8
48.9 0.6
91 3.0
21.0
24.4
15.6
12.8
Ni(NO3)2·6H2O (ammonia treated)
Untreated Ni(NO3)2·6H2O
1
2
3
116.1
308.8
−505.1a, 254.6
290.1
226.4
278.1
80.81
691.8
1.3
2.6
2.9
1.4
2.0
2.1
57 1.0
151 1.8
220 3.0
24.4
48.2
49.1
18.5
30.2
21.9
1
2
3
4
5
2.3
0.9
0.8
0.4
0.6
1.9
1.2
1.1
0.7
1.0
125 1.2
127 1.2
236 8.0
57 0.8
49.2
45.7
64.9
15.6
25.9
37.9
33.0
37.7
8.0
115 3.0
11.4
a
Exothermic process.

210
A. Loaiza-Gil et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
Fig. 3. Simultaneous (a) TGA and (b) DSC profiles (nitrogen flow, 100 mL/min)
of nickel nitrate hexahydrate treated with ammonia solution at 15 K/min heating
rate.
Fig. 5. Simultaneous (a) TGA and (b) DSC (nitrogen flow, 100 mL/min) of
silica-supported nickel catalyst carried out at 15 K/min, heating rate.
first step and four in the second without any further structural
rearrangement observed. However, the calculated activation
energies (Table 1) of 48.92 and 90.92 KJ/mol for the first and
second steps in that order suggest that in addition of water
and/or ammonia lost, structural rearrangement, bonds broken
and/or formation occurs. On the other hand, comparing the cal-
culated activation energies values of synthesized catalyst and
treated nickel salt, it was found that the activation energy on
the second decomposition step of synthesized catalyst does not
correspond with any decomposition step of treated nickel salt
as confirmed by the different values of the calculated reac-
tion order n (Table 1), suggesting that silica and nickel aqua
ammine complex ions reacts between them originating a new
compound that in our case could be corresponding to nickel
phyllosilicate hydrate. The weight lost observed for the synthe-
sized catalyst (Fig. 5) suggested that the decomposition process
mum at 325 K (ꢀH = 116.1 kJ/mol) and 399 K (ꢀH = 308.8 J/g)
suggesting a loss of physisorbed and structural water respec-
tively in that order. The third DSC peak was exothermic
suggesting a great structural rearrangement before the occur-
rence of the last endothermic decomposition that give rise
to nickel oxide formation. Such a result indicated that the
reaction between ammonia and nickel nitrate hexahydrate to
form nickel aqua ammine complex ions has a good chance to
occur.
The next step in this work was to check out the chemical
composition of the synthesized silica-supported nickel catalyst
before the reduction process. The thermal analysis TGA–DSC
and DTA of this catalyst is shown in Figs. 5 and 6 respectively.
The TG profile show that 25.4% total mass lost occurs in two
steps giving rise NiO/SiO2 catalyst (10%Ni). The result sug-
gests that three water or ammonia molecules were lost in the
Fig. 6. DTA profiles (nitrogen flow, 100 mL/min) of silica-supported nickel
catalyst at (a) 5 K/min, (b) 10 K/min, (c) 15 K/min and (d) 20 K/min heating
rate.
Fig. 4. DTA profiles (nitrogen flow, 100 mL/min) of nickel nitrate hexahydrate
treated with ammonia solution at (a) 5 K/min, (b) 10 K/min, (c) 15 K/min and
(d) 20 K/min heating rate.

A. Loaiza-Gil et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
211
Fig. 7. Temperature-programmed reduction profile of silica-supported nickel
catalyst synthesized by the ammonia method. On the left corner (up):
temperature-programmed reduction profile of silica-supported nickel catalyst
synthesized by impregnation.
Fig. 8. SEM micrograph of silica-supported nickel catalyst synthesized by the
ammonia method. The sample was dried at 423 K.
support nickel catalyst diffraction patterns, fresh and after being
calcinated at 373, 473 and 773 K. It was observed basically the
same diffraction pattern as the temperature increases suggest-
ing that catalyst chemical structure was thermally stable up to
773 K. This last result agreed very well with the TPR results
that indicated that 973 K temperature or more was necessary to
collapse the catalyst structure.
Preliminary catalytic test of methane steam reforming per-
formed on the silica-supported nickel catalyst synthesized
by the ammonia method and by impregnation are shown in
Figs. 11 and 12. Typical deactivation curves, probably caused
by inactive carbon deposition were observed for both catalysts.
of the synthesized catalyst occur according to the following
reactions:
Ni3(OH)2(Si2O5)2·6H2O → Ni3(OH)2(Si2O5)2·3H2O
+ 3H2O
(5)
(6)
Ni3(OH)2(Si2O5)2·3H2O → 3NiO + 4SiO2 + 4H2O
Such a result agreed quite well with that obtained by Mar-
tin and co-workers [1] and suggests that the use of ammonia
method favor strong metal–support interaction (SMSI) forma-
tion. In order to verify the chemical and structural composition
of the catalyst, further experiments of temperature-programmed
reduction and scanning electron microscopy were performed.
Temperature-programmed reduction profiles of silica-supported
nickel catalyst synthesized by the ammonia method and by
impregnation are shown in Fig. 7. It was observed that the cat-
alyst synthesized by the ammonia method show mainly a high
temperature metal phase with maximum around 973 K corre-
sponding to SMSI and a small contribution of NiO at 473 K.
Instead, silica-supported nickel catalyst synthesized by impreg-
nationshowtworeductionpeaksat588and753 Kcorresponding
to NiO free and weakly associated to the support [15,16]. A
micrograph of silica-supported nickel catalyst, dried at 423 K
was shown in Fig. 8. It was observed a structure constituted by
superimposed sheets similar to nickel phyllosilicate in agree-
ment to TGA–DSC and TPR results.
Fig. 9showstheX-raydiffractionpatternsofsilicatreatedand
untreated with ammonia solution and silica support nickel cata-
lyst. It was observed for all of three patterns a shoulder around
2θ = 23◦ characteristic of amorphous substances. Besides, the
diffraction pattern of silica support nickel catalyst shows two
small shoulders around 37◦ and 63◦ probably corresponding to
NiO species from the lower temperature metal phase (473 K)
determined by TPR. A similar diffraction pattern was found for
cobalt phyllosilicate catalyst [17]. Fig. 10 compares the silica
Fig. 9. XRD of (a) commercial silica (200 m2/g), (b) commercial silica
(200 m2/g)treatedwithammoniasolutionand(c)silica-supportednickelcatalyst
dried at 353 K.

212
A. Loaiza-Gil et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
The conversion of methane versus time on the catalysts syn-
thesized by the ammonia method is displayed in Fig. 11(a). It
was observed a dramatically decreases of the conversion from
100 to 62% on the first 50 min followed by a further conver-
sion decrease after a period of 600 min of slowly conversion
decreases. After that the methane conversion was observed to
be constant around to 50% for the rest of the reaction time
tested. The selectivity to hydrogen was observed (Fig. 11(b)) to
decrease from 100 to 92% on the reaction time. As the same way
the selectivity to carbon monoxide was observed to increases up
to 8%. No carbon dioxide was detected. The performance of
this catalyst to the steam reforming was compared with a sim-
ilar catalyst synthesized by impregnation. In Fig. 12(a), it was
observed that, conversion decreases on the first 500 min from
100 to 56%. After that the conversion decreases with a differ-
ent deactivation rate arriving to 35% at the end of the test. The
selectivity to hydrogen decreased from 100 to 85% on the entire
reaction time while the selectivity to carbon monoxide was pro-
gressively increasing. Also in this case, no carbon dioxide was
detected.
The performance of the catalyst synthesized by the ammonia
method was better than that synthesized by impregnation. Such
different behavior could be caused by differences in the metal
particle sizes because of the reduction temperature of 1023 K
Fig. 10. XRD of silica-supported nickel catalysts dried at several temperatures.
Fig. 11. Methane steam reforming on silica-supported nickel catalyst synthe-
sized by the ammonia method performed at 973 K, atmospheric pressure and
500 h−1: (a) methane conversion (%) vs. time and (b) selectivity (%) to hydrogen
and carbon monoxide vs. time.
Fig. 12. Methane steam reforming on silica-supported nickel catalyst synthe-
sized by impregnation performed at 973 K, atmospheric pressure and 500 h−1
:
(a) methane conversion (%) vs. time and (b) selectivity (%) to hydrogen and
carbon monoxide vs. time.

A. Loaiza-Gil et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 207–213
213
employed in both catalysts. The reduction temperature was just
enough to collapse the high temperature metal phase observed
on the silica-supported nickel catalyst. In the case of the catalyst
synthesized by impregnation, such temperature was the longer
superior to that required for the catalyst reduction. Therefore,
the catalyst synthesized by impregnation was exposed to a high
temperature for a time longer than the catalyst synthesized by
the ammonia method favoring the growing of metal particle size.
by carbon deposition was less pronounced than deactivation of
a similar nickel content catalyst prepared by impregnation.
References
[1] A. Barbier, A. Hanif, J. Dalmon, G. Martin, Appl. Catal. A: Gen. 168 (1998)
333–343.
[2] A. Loaiza-Gil, J. Arenas, M. Villarroel, F. Imbert, H. del Castillo, B. Fontal,
J. Mol. Catal. A: Chem. 228 (2005) 339–344.
[3] G.A. Martin, Thesis, University Claude Bernard, Lyon (1966).
[4] G.A. Martin, B. Imelik, M. Prettre, J. Chim. Phys. 66 (1969) 1682.
[5] J.A. Dalmon, G.A. Martin, C.R. Acad. Sci. Paris 267C (1968) 1.
[6] J.A. Dalmon, G.A. Martin, B. Imelik, J. Chim. Phys. 2 (1973) 214.
[7] J.A. Dalmon, G.A. Martin, C.R. Acad. Sci. Paris 267C (1968) 610.
[8] S. Glastone, Text of Chemical Physics, second edition, Van Nostrand, New
York, 1946.
[9] H.E. Kissinger, Anal. Chem. 29 (1957) 1702.
[10] E. Mikuli, A. Migdal-Mikuli, R. Chyzy, B. Grand, R. Dziembaj, Ther-
mochim. Rec. 370 (2001) 65–71.
[11] M.A.A. Elmasry, A. Gaber, E.M.H. Khater, J. Therm. Anal. 52 (1998)
489–495.
[12] F. Bigoli, A. Braibanti, A. Tiripicchio, M. TiripicchioCamellini, Rec. Cryst.
B27 (1971) 1427.
[13] A. Malecki, R. Gajerski, S. Labus, B. Prochowska-Klisch, K.T. Woj-
ciechowski, J. Therm. Anal. 60 (2000) 20.
[14] Z.D. Zivkovic, D.T. Zivkovic, D.B. Grujicic, J. Therm. Anal. 53 (1998)
617.
4. Conclusions
Silica-supported nickel catalyst synthesized by the ammonia
method and their precursors treated and not with ammonia solu-
tion were characterized by TGA, DSC, DTA thermal analysis,
TPR, SEM micrographs and X-ray diffraction. Thermal decom-
position kinetic parameters were calculated by the Kissinger’s
method.
The results indicated that the synthesized catalyst shows
mainly a higher temperature metal phase (973 K), constituted by
a superposition of bi-dimensional sheets similar to nickel phyl-
losilicate, probablyresultingfromtheSMSIbetweennickelaqua
ammine complex and silica surface. Such a compound degrades
to stable silica-supported nickel catalyst at high temperature.
Activation energies and kinetic parameters for each decompo-
sition process were calculated by Kissinger’s method. Enthalpy
values were determined from differential scanning calorimetric
profiles. The activity tests of such catalyst to methane steam
reforming indicated that deactivation process probably caused
[15] R. Hyung-Seog, D. Wen-Sheng, J. Ki-Won, L. Zhomg-Wen, P. Sang-Eon,
O. Young-Sam, Bull. Kor. Chem. Soc. 23 (5) (2002) 669–673.
[16] J. Li, G. Lu, Appl. Catal. A: Gen. 273 (2004) 163–170.
[17] R. Trujillano, J. Grimoult, C. Louis, J.F. Lambert, Stud. Surf. Sci. Catal.
130 B (2000).