Thermal decomposition of nickel acetate tetrahydrate: An integrated study by TGA, QMS and XPS techniques

Article abstract of DOI:10.1016/j.molcata.2004.09.065

The thermal decomposition of nickel carboxylates is a feasible route to synthesize metal nanoparticles suitable for catalytic purposes. The aim of this work is the characterization of the thermal decomposition course of nickel acetate tetrahydrate, Ni(CH₃COO)₂·4H₂O. A thermogravimetric (TGA) decomposition study carried out in three different atmospheres (air, helium and hydrogen) showed that the dehydration of the parent salt occurs between 118 and 137°C. However, irrespective of the chosen atmosphere, the sample weight loss registered in this stage remains invariable, suggesting the formation of a an intermediate basic acetate with the formula 0.86Ni(CH₃COO)₂·0.14Ni(OH)₂. The dehydration step was followed at ca. 350°C by the subsequent one-step major decomposition of the acetate group, producing NiO and Ni, in treatment atmospheres of air and hydrogen, respectively, but there was some indication of an additional step when the thermolysis was conducted in helium. The conclusions possible from thermal analysis were confirmed by monitoring evolved gases employing quadrupole mass spectrometry (QMS), and a set of reactions linked to the decomposition of the acetate group has been proposed to account for most of the gas products detected. X-ray photoelectron spectroscopy (XPS) was used to investigate the solid phases obtained during the thermal decomposition of the salt in He atmosphere.

Full text of DOI:10.1016/j.molcata.2004.09.065

Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
Thermal decomposition of nickel acetate tetrahydrate:
an integrated study by TGA, QMS and XPS techniques
a,∗
b
b
b
Juan C. De Jesus , Ismael Gonz a´ lez , Angel Quevedo , Tito Puerta
a
PDVSA-INTEVEP, Apdo. 76343, Caracas 1070A, Venezuela
b
Escuela de Qu ´ı mica, Facultad de Ciencias, Universidad Central de Venezuela, Apdo. 47102, Caracas 1040A, Venezuela
Available online 24 November 2004
Abstract
The thermal decomposition of nickel carboxylates is a feasible route to synthesize metal nanoparticles suitable for catalytic purposes.
The aim of this work is the characterization of the thermal decomposition course of nickel acetate tetrahydrate, Ni(CH COO) O.
3
2
·4H
2
A thermogravimetric (TGA) decomposition study carried out in three different atmospheres (air, helium and hydrogen) showed that the
◦
dehydration of the parent salt occurs between 118 and 137 C. However, irrespective of the chosen atmosphere, the sample weight loss registered
in this stage remains invariable, suggesting the formation of a an intermediate basic acetate with the formula 0.86Ni(CH
3
COO)
2
·0.14Ni(OH)
2
.
◦
The dehydration step was followed at ca. 350 C by the subsequent one-step major decomposition of the acetate group, producing NiO and
Ni, in treatment atmospheres of air and hydrogen, respectively, but there was some indication of an additional step when the thermolysis was
conducted in helium. The conclusions possible from thermal analysis were confirmed by monitoring evolved gases employing quadrupole
mass spectrometry (QMS), and a set of reactions linked to the decomposition of the acetate group has been proposed to account for most
of the gas products detected. X-ray photoelectron spectroscopy (XPS) was used to investigate the solid phases obtained during the thermal
decomposition of the salt in He atmosphere.
©
2004 Elsevier B.V. All rights reserved.
Keywords: Nickel acetate tetrahydrate; Decomposition; X-ray photoelectron spectroscopy; Thermogravimetric analysis; Quadrupole mass spectrometry
1
. Introduction
studying the thermal decomposition of transition metal car-
boxylates, as potential sources of high surface area solids for
catalysis, which demands particles in the range 1–3 nm for
maximum efficiency and whose properties are strongly size
dependent in this range. As a result, decomposition of pure
nickel formate has been claimed recently to yield in a single
step finely divided nickel metal with mean particle diameter
of about 1.2 nm [8], and studies about the preparation and
characterization of nickel-supported alumina employing
aqueous solutions of nickel formate are now available
[9]. Therefore, nickel carboxylates present an interesting
alternative to conventional nitrate precursor, enabling
the preparation of nickel-supported catalysts essentially
through a one-step impregnation and decomposition mild
route. In this sense, contributions to the knowledge of the
surface structure of nickel nanoparticles obtained by thermal
decomposition of carboxylate precursors are necessary to
achieve the desired behavior on supported systems.
The favoured method for preparation of supported cata-
lysts is often referred as impregnation and drying, in which
porous supports are impregnated with a solution of the active
metal, followed by evaporation of the solvent and calcination
in air to generate the oxidic phases that are precursors of the
active (usually reduced) centers. Supported nickel catalysts
comprise one of the most important class of heterogeneous
catalysts, due to the widespread applications of these systems
in a variety of applications, like methanation [1-3], partial
oxidation [4,5], and steam reforming [6,7]. Nickel nitrate,
which is widely used for the preparation of supported nickel
catalysts, decomposes quantitatively to nickel oxide during
the calcination step. There has been significant interest in
∗
Corresponding author.
E-mail address: dejesusjc@pdvsa.com (J.C. De Jesus).
1
381-1169/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2004.09.065

2
84
J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
Thermal decomposition of nickel acetate has been the sub-
accessible with the quadrupole mass spectrometer employed.
However, a careful analysis of published cracking patterns
[18]justifythefollowingsuppositions, whichtoagreatextent
assist the identification of products and quantitative analysis
of the desorption spectra [19]:
ject matter of several previous studies [10–17]. Atmospheres
employed so far include vacuum [10], flowing N2 [11–17],
dry oxygen [11], air [10,14], water [15], and self-generated
atmospheres [16]. Solid residues covering intermediates and
final products were characterized mainly by combining XRD
and/or IR [10–15], and exceptionally employing NMR/SEM
16]. On the other hand, gases evolved concurrently during
the thermal decomposition of nickel acetate have received
limited consideration, they have been studied mainly by GC
1
. The signals at m/e = 2, 18, 25, 31, 37, 44, 56, 58, and 60
come up exclusively from the desorption of hydrogen, wa-
ter, acetylene, ethanol, C3H4, CO2, isobutylene, acetone,
and acetic acid, respectively.
[
2
. The signal at m/e = 43 receives contributions from both
acetone and acetic acid, and is also characteristic of acetic
anhydride, but after subtracting expected contributions it
was determined that acetic anhydride production during
thermal decomposition of NiAc in He is virtually zero.
. The m/e = 16 peak is directly related to CH4, after sub-
tracting contributions from both water and CO2.
. After subtracting contributions from CH4 and acetic acid,
the m/e = 14 peak can be associated directly to ketene.
. Formic acid can be related to m/e = 46, after subtracting
ethanol contribution.
[
11], IR [14], and MS [16,17]. In spite of the available infor-
mation, there are still significant qualitative and quantitative
differences reported for the mechanisms, solid intermediates
and gases evolved during the thermal decomposition of nickel
acetate [14].
In this paper we demonstrate the applicability of X-
ray photoelectron spectroscopy (XPS), combined with
studies of thermogravimetry (TGA) and temperature pro-
grammed decomposition studies with a quadrupole mass
spectrometer (TPD-QMS), to the further characterization of
the pyrolitic decomposition of nickel acetate tetrahydrate,
Ni(CH3COO)2·4H2O.
3
4
5
6
. The m/e = 28, formally associated with CO, needs to be
corrected by contributions arising from CO2, formic acid
and isobutylene.
7
. The peak at m/e = 30 arises from both formaldehyde and
ethane.
2
. Experimental
2
.1. Materials
The application of accepted relative ionization gauge sen-
sitivity factors, allowed the adjustment of intensity signals
(Amps) acquired during the TPD runs, to obtain mol frac-
tions as presented in Table 2.
Commercially available nickel acetate tetrahydrate
[
Ni(CH3COO)2·4H2O] was used directly as the initial pre-
cursor for TGA, TPD and XPS studies.
2.4. XPS analysis
2
.2. Thermal analysis
XPS spectra were recorded using a Leybold–Heraeus
TG measurements were carried out using a Cahn 2000
commercial surface analysis apparatus (LHS 11), equipped
with a single channel detector, and employing AlK␣ radiation
(1486.6 eV)at360 W(powersettings:12 kVand30 mA). The
100 mm radius hemispherical analyzer was set in the constant
pass energy mode (pass energy = 200 eV). For XPS measure-
ments, approximately 1 g of parent nickel acetate is placed in
the quartz reactor, and heated linearly to the final temperature
microbalance. Between 10 and 15 mg of sample were spread
thinly over the quartz sample pan to minimize bed effects dur-
ing decomposition of the nickel acetate samples. The heating
◦
rate was 30 C/min, and the purge gas employed in all cases
was He, at a flow of 45 ml/min. The treatment gas (He, Air
and H2) was admitted at 40 ml/min.
◦
at 30 C/min, in a dynamic atmosphere of He (40 ml/min).
2
.3. TPD analysis
The sample is maintained over 60 min in the He atmosphere
at the desired temperature, and cooled down to room temper-
ature under flowing gas. The powdered samples are quickly
removed, pressed, and mounted employing double sided tape.
The normal operating pressure inside the turbo-pumped anal-
ysis chamber was kept below 5 × 10 Torr during data col-
lection. Each spectral region was signal-averaged for a given
number of scans to obtain good signal-to-noise ratios. Sur-
face charging was observed on most samples, and accurate
bindingenergies(BE)weredeterminedbychargereferencing
by means of either adventitious carbon at 284.6 eV or with
the reported value for metallic nickel component of 852.3 eV
[20]; the BE were reproducible within ± 0.2 eV. The photo-
electron spectra were collected and stored in an interfaced
TPD measurements were made in a purpose built quartz
reactor of approximately 400 ml, connected through a leak
valve and a heated capillary to an independently pumped In-
ficon Transpector quadrupole mass spectrometer operating
under computer control. A nickel acetate tetrahydrate sam-
ple (10 mg), charged in the quartz reactor, was heated up to
−
8
◦ ◦
1
000 C at 30 C/min in He at a flow rate of 40 ml/min. The
experiment was repeated several times to assure reproducibil-
ity of TPD data.
Preliminary analysis of the desorption process showed
there are a significant amount of reaction products that con-
tribute to a wide range of signals in the 0–100 AMU range,

J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
285
personal computer, using the SPECTRA 6.0 software from
Specs GmbH.
Table 1
Weight losses measured during TGA of NiAc (wt.%)
◦
Temperature ( C)
H2
Air
He
2
.5. Additional characterization
250
66.5
24.0
24.0
66.1
30.7
30.2
66.3
28.0
25.0
3
5
80
00
The total carbon content of the residue obtained after com-
plete decomposition of 1 g of nickel acetate in the quartz
reactor in a flow of He of 40 ml/min, heated from room tem-
perature to 500 C (30 C/min) and maintained 15 min at the
final temperature was measured by a C-400 (LECO Corp.)
elemental analyser, having a carbon sensitivity of 0.01%.
No matter of the treatment atmosphere employed during
◦
◦
the thermal decomposition of NiAc, a broad peak with a
◦
Tmax = 118–137 C is registered in all the DGA curves; it
can be assigned with a high degree of certainty to the dehy-
dration of crystallized water in the initial precursor. After the
◦
dehydration is completed, the average for the WR at 250 C
3
. Results
recorded under the three different atmospheres (66.3%) is
clearly below the value expected for the release of four wa-
ter molecules (71.1%), indicating an evolution of additional
amounts of material to the gas phase. Earlier reports suggest
that hydrolysis of surface acetate groups is feasible during
the dehydration of NiAc, resulting in simultaneous evolution
of acetic acid to the gas phase, and generating basic nickel
acetate in the solid phase, (1 − x)Ni(CH3COO)2·xNi(OH)2)
3
.1. Thermogravimetric analysis
The results from the thermogravimetric measurements in
flowing atmospheres of H2, He and air are presented in the
derivative form (DGA) in Fig. 1.The heating program in all
◦
◦
cases was 25–700 C at a rate of 30 C/min. The values in
percentage of weight retained (WR) at selected temperatures
during the thermogravimetric analysis of the decomposition
in a flow of He of nickel acetate tetrahydrate, are presented in
Table 1. We can see that, under our experimental conditions,
NiAc decomposition undergoes a maximum of three thermal
processes, named as I–III in the weight derivative curves de-
picted in Fig. 1. Steps I and II are common to all treatments,
but step III is only observable during inspection of the DGA
curve obtained under He atmosphere (Fig. 1b).
[
11,13]. As will be shown below, our QMS results confirm
the production of acetic acid during the dehydration step, as
well as small amounts of additional by-products.
◦
The step (II) in Fig. 1, with Tmax = 345–350 C, corre-
sponds to the major decomposition of the dehydrated inter-
◦
mediate. Under H2 atmosphere, the WR at 380 C of 24.0%
is consistent with the theoretical value for the formation of
Ni (23.6%), while under air, the WR of 30.7% is compatible
with the value expected for the formation of NiO (30.0%).
◦
At temperatures up to 500 C, no additional changes in the
weight of the residue are recorded in treatments of either H2
or air atmospheres.
◦
Interestingly, a third feature at Tmax = 395 C (step III,
Fig. 1b) is clearly observed when the decomposition of NiAc
is carried out in flowing He at atmospheric pressure. TGA
experiments recorded at slightly higher pressures (data not
shown) indicate that this process is pressure sensitive. The
◦
intermediate WR value of 28.0% measured at about 380 C,
suggests that a mixture of NiO and Ni is most likely ob-
◦
tained under an inert atmosphere. Further heating to 500 C
in He leads to a WR value of 25.0, slightly higher than the
value calculated for the formation of pure Ni (23.6%). As will
be discussed below, we believe this is due to carbonaceous
species present on the surface of the metallic residue.
3.2. Gas evolved analysis
Information regarding the decomposition pattern of
species generated during the pyrolisis of nickel acetate
was obtained from temperature programmed decomposition
(TPD) studies with a mass quadrupole (QMS). Mass spectra
collected on-line during the decomposition of NiAc in He
are shown in Figs. 2–5. Fig. 2 exhibits the curve obtained
after adding the individual contributions for all the chemi-
Fig. 1. Derivative thermogravimetric curves for the decomposition pf nickel
acetate tetrahydrate, collected at 30 C/min in He at 40 ml/min. Gas atmo-
spheres employed are shown over each trace.
◦

2
86
J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
Fig. 4. Temperature-programmed decomposition spectra collected during
linear heating of nickel acetate tetrahydrate, at a ramp of 30 C/min in a flow
of He at 40 ml/min (H2 and hydrocarbons, see text).
Fig. 2. Synthetic temperature-programmed decomposition trace, obtained
after adding contributions for all gases identified by QMS during the thermal
decomposition of nickel acetate tetrahydrate at 30 C/min, in flowing He at
◦
◦
4
0 ml/min.
evident in Fig. 2. However, the step III, clearly noticed pre-
viously during the TG analysis of the NiAc decomposition
process in He, is not distinguished in the envelope of Fig. 2.
All the gases identified by QMS during the thermal de-
composition of NiAc were conveniently detached in three
main groups, namely oxygenates (Fig. 3), hydrogen + hy-
drocarbons (Fig. 4), and carbon oxides (Fig. 5). The water
desorption trace is not reported because the corresponding
signal at m/e = 18 is very broad, due to adsorption and possi-
ble some retention on the steel capillary. The relative product
yields, based on the calibration of the TPD peak areas after
correcting for coincident m/e values and conversion to partial
pressures, are summarized in Table 2 and depicted in Fig. 6.
Ketene (CH2CO), ethanol (CH3CH2OH), formic
acid (HCOOH), acetic acid (CH3COOH), and acetone
cal species identified by means of QMS. As a visual aid to
help in the comparison with TGA data depicted in Fig. 1b,
dotted lines are included in Fig. 2 at the same temperatures
previously assigned for I–III decomposition steps. The curve
resembles reasonably well the DGA displayed in Fig. 1b,
corroborating most of the trends discussed previously for the
thermal decomposition of NiAc under He atmosphere. The
◦
broad peak observed between ca. 50 and 240 C in Fig. 2
can be related to the dehydration of NiAc, as explained be-
fore (step I, Fig. 1). As will be discussed below, it comprises
different products besides water and acetic acid expected to
form due to the surface hydrolysis of the acetate groups.
The major decomposition of NiAc, as indicated by a sec-
◦
ond broad peak extending from ca. 270 to 500 C, is also
(CH3COCH3), were unambiguously identified during the
thermal decomposition of NiAc under He (see Fig. 3). As
was settled before, hydrolisis of surface acetate groups most
likely account for the observed production of acetic acid at
Fig. 3. Temperature-programmed decomposition spectra collected during
linear heating of nickel acetate tetrahydrate, at a ramp of 30 C/min in a flow
Fig. 5. Temperature-programmed decomposition spectra collected during
linear heating of nickel acetate tetrahydrate, at a ramp of 30 C/min in a flow
◦
◦
of He at 40 ml/min (oxygenates, see text).
of He at 40 ml/min (carbon oxides, see text).

J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
287
◦
◦
Table 2
ca. 325 C. At about 350 C, an equivalent amount of acetic
acid is desorbed, and there is also noticeable the releasing of
minute amounts of ketene, formic acid, and ethanol.
Product yields for nickel acetate decomposition in He
◦
Desorption product
Hydrogen
m/e
2
Yield fraction
0.113
Temperature ( C)
394
Mass spectra of hydrogen and hydrocarbons evolved dur-
ing the decomposition of NiAc in He (Fig. 4), disclose that
trace amounts of methane (CH4) and ethylene (C2H4) are
produced concurrently with oxygenated hydrocarbons at ca.
60 C. During the major decomposition of the acetate group,
a small amount of H2 is also evolved to the gas phase at about
90 C. Minute amounts of isobutilene were also detected
during both the dehydration and the decomposition stages.
Carbon oxides are also produced during the thermal de-
composition of nickel acetate tetrahydrated (see Fig. 5).
While minute amounts of COx species are evolved during
the dehydration step at ca. 150 C, relatively high amounts of
carbon oxides are produced between 280 and 500 C, during
the major decomposition step of the acetate group. Particu-
larly, it is of interest to note that the the CO2 trace in the left
panelofFig. 5presentsadistinguishingdoublestructure, with
Ketene
14
0.025
162
344
0
.195
0.119
.195
7.440
.044
0.019
.170
1.000
.164
0.013
.044
0.006
.031
Methane
16
18
25
28
30
31
182
359
◦
1
0
Water
162
373
◦
3
2
C2H2
162
339
0
Carbon monoxide
HCHO + C2H6
Ethanol
350
939
0
◦
165
350
◦
0
162
350
0
◦
C3H4
37
44
0.277
0.189
327
peak maximum located at ca. 330 and 365 C. Data showed
in Table 2 and depicted in Fig. 6, clearly indicated that CO2 is
quantitatively the main product detected during the thermal
decomposition of nickel acetate tetrahydrated. On the other
hand, CO production is observed mainly at 350 C, and inter-
estingly is the only product detected as a trace at ca. 940 C,
Carbon dioxide
168
339/365
6
.289
0.006
.075
0.003
.011
Formic acid
Isobutilene
46
56
162
365
◦
0
◦
197
356
as shown in the right panel of Fig. 5.
0
Acetone
58
60
1.541
0.226
327
3.3. XPS analysis
Acetic acid
162
353
1.390
The powders that result from the thermal decomposition
of parent nickel acetate under He atmosphere were charac-
terized employing X-ray photoelectron spectroscopy (XPS).
Fig. 7 shows the relevant XPS spectra for pure nickel acetate
◦
ca. 150 C. At this point, it is worth to mention that acetone
production is not detected during the dehydration step.
Additionally, from the analysis of the spectra presented in
Fig. 3, it is clear that during the major decomposition of the
basic nickel acetate, acetone production is the first process at
Fig. 7. X-ray photoelectron spectra collected during the thermal decom-
position of nickel acetate tetrahydrate. The samples are heated in a quartz
◦
reactor in flowing He (40 ml/min) at a linear ramp of 30 C/min, and main-
tained 60 min at the temperatures displayed over each trace, cooled at room
temperature in the carrier gas, and transferred immediately to the instrument.
Parent nickel acetate without any treatment is displayed as trace at 25 C.
Fig. 6. Relative product yields for the thermal decomposition of nickel ac-
etate tetrahydrate at 30 C/min in a flow of He at 40 ml/min (see Table 2).
◦
◦

2
88
J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
◦
◦
at 25 C and residues collected at 300, 350 and 450 C. The
first observation to be addressed here is in fact that the C 1s
XPS spectra displayed in the right panel of Fig. 7 display
only two distinct peaks at binding energies around 288.2 and
more challenging, because an additional step suggests the
involvement of a NiO/Ni mixture as a likely intermediate.
Therefore, we focus our attention to the events associated
with the thermal decomposition process of the nickel acetate
precursor under an inert He atmosphere. Table 1 reveals that
the average mass value retained after step I under different
atmospheres (66.3%) is well below that expected for stoichio-
metric loss of four water molecules (71.1%). Our values are
lower than the ones reported elsewhere in a static atmosphere
of air (69.3%) [14], and under dynamic atmospheres of N2,
(68%) in [11] and (72%) in [12]. On the other hand, Galwey
et al. [16] measured a mass loss almost identical to the one
reported in our work, after heating in vacuum a commercial
sample of nickel acetate for 2 h at 470 K; they interpreted the
result as if the precursor salt were pentahydrated instead of
tetrahydrated.
Our understandingof the thermogravimetric resultsallows
a supplementary reading of the extra weight loss obtained
during the dehydration step. Dor e´ mieux [17] first suggested
the possible formation of a basic acetate as intermediate com-
pound [(1 − x)NiAc2·xNi(OH)2] in his pioneering work, and
in considering the value of WR obtained systematically in our
work for the residue of step (I), we can assume the stoichiom-
etry of the hydrolisis reaction to allow roughly an estimation
of the x value for the formation of basic nickel acetate, per-
mitting the following chemical reaction to be proposed:
2
84.6 eV. We can easily associate these two features with
the carboxylate group and adventitious carbon, respectively.
The carboxylate component diminishes strongly after heat-
ing to 300–350 C, and virtually disappears on final heating
to 450 C. Notice, however, that the carbide component ex-
pected at around 283.3 eV [21] is not apparent upon exami-
nation of the C 1s XPS spectra displayed on Fig. 7, ruling out
the formation of surface nickel carbides, as reported previ-
ously by others accordingly to XRD analysis of the residues
◦
◦
[11].
The changes perceivable on the corresponding Ni 2p spec-
tra are, however, much more subtle. The overall shape of the
Ni 2p XPS line proved to be virtually unaffected upon heating
◦
of the nickel acetate to temperatures up to 350 C, as showed
in the left panel of Fig. 7. However, TGA and TPD data sug-
gest that very important chemical changes must be accounted
for in this temperature range. On the other hand, intense color
changes are observable upon treatment, from bright green for
the parent nickel acetate tetrahydrated, to green-yellowish to
◦
the basic nickel acetate (300 C) and black for both residues
collected at 350 and 450 C. In these series of experiments,
◦
the Ni 2p spectra are characterized mostly by a main peak
at around 855.5 eV, as well as very intense satellites at the
higher binding energy side of the main peaks, both character-
Ni(CH3COO)2 · 4H2O=0.86Ni(CH3COO)2 · 0.14Ni(OH2)
istic of Ni2 ion. A large amount of Ni species are reported
around the obtained value of 855.5 eV, including nickel ac-
etate, NiCO3, Ni(OH)2, and NiO [22]. Therefore, assuming
+
2+
+
0.28CH3COOH + 3.72H2O
(1)
◦
that the overall XPS signal of nickel in the 25–350 C tem-
perature range may originate from multiple Ni2+ species, the
identification of the nickel states is not as straightforward as
in the case of the C 1s XPS.
Therefore, the production of 0.28 mol of acetic acid would
explain the difference in weight loss observed experimentally
here, and may also account very well for the evolution of
an additional water molecule, as interpreted alternatively by
Galwey et al. [16].
The best approach is just to focus on the changes observ-
able in the Ni 2p region for the sample treated at 450 C
◦
(Fig. 7, top). There, a new component at around 852.5 eV
clearly develops, in agreement with the formation of metallic
nickel as proposed before from the analysis of TGA and TPD
data.
4
.2. Gas evolved analysis
One of the key observations reported in this paper is the
fact that our TPD-QMS studies corroborate on-line the si-
multaneous production of both water and acetic acid during
the dehydration step of nickel acetate tetrahydrated. In addi-
tion, other oxygenated hydrocarbons, as ketene, formic acid,
4
. Discussion
4
.1. Thermogravimetry
◦
and ethanol, were also detected at ca. 120 C, as shown in
Fig. 3. The similar temperature and peak shape for all these
four products suggest that they were derived from the same
surface species through equivalent reaction channels, accord-
ingly with the following trend (see Table 2 and Fig. 6):
Our quantitative TGA results confirm previous findings
indicating that the decomposition of nickel acetate tetrahy-
drated is critically controlled by the reaction atmosphere [14].
As shown in Table 1, by comparing the results obtained for
the thermal decomposition of the parent salt in H2 with those
obtained in air, it is possible to propose the formation of ei-
ther metallic nickel or NiO, respectively. Nonetheless, from a
mechanistic point of view, the characterization of the decom-
position process under an inert atmosphere of He is much
CH3COOH > CH2CO > CH3CH2OH ꢀ HCOOH
Alternatively, the oxygenates produced during the dehy-
dration step may be due to the following set of reactions,
presumably involving gas phase homogeneous transforma-

J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
289
tions of the acetic acid:
for the CO2 trace depicted in the left panel of Fig. 5, can be
conveniently explained with the following set of reactions:
CH3COOH = CH2CO + H2O
CH3COOH + H2 = CH3CH2OH
(2)
(3)
(4)
◦
Ni(CH3COO2 = NiCO3 + CH3COCH3 (330 C)
(7)
(8)
(9)
◦
NiCO3 = NiO + CO2 (330 C)
2
CH3COOH = 2HCOOH + C2H4
◦
NiO + CO = Ni + CO2 (373 C)
Whichever mechanism is responsible for the production of
The peak temperatures reported here for the production
of CO2 are in good agreement with previous results obtained
during the thermal decomposition of nickel acetate in air [14].
Acetonegeneration, ontheotherhand, supportstheformation
in the solid phase of NiCO3, as suggested in a previous work
ketene, ethanol and formic acid during the dehydration step,
it is clear that only a fraction of all the acetic acid formed by
hydrolysis is detected. The water to acetic acid ratio measured
during the dehydration step (see Table 2) is well above the one
expected through the fulfillment of reaction (2) (33 versus
[14]. We believe NiCO3 is a short-lived intermediate that
1
3). This figures imply that approximately 60% of all the
decarboxylates immediately to give NiO, as proposed in 8.
On the other hand, it is difficult to understand by what
mechanism the acetate group decomposes to yield gaseous
acetic acid as a product [12]. However, it has been pointed
out that the major gaseous product of the decomposition of
acetates of reducible metal ions (e.g. nickel acetate) is acetic
acid [24].
acetic acid generated initially by surface acetate hydrolysis is
transformed to by-products through decomposition reactions
like (2)–(4).
It should nevertheless be pointed out here that it is well
knownthatundesirablesidereactionsbetweengasesmayalso
take place at some extent at the ionizer of mass quadrupoles,
andquantificationwiththisinstrumentsisnotalwaysstraight-
forward; for instance, the purpose of the gas analysis pre-
sented in this study is to speculate on possible mechanisms
and achieve a reasonable comparison with experimental data
available in the literature.
Onemoreimportantobservation, evidentfromtheanalysis
of the acetone trace depicted in Fig. 3, needs to be highlighted
here. The fact that acetone production during the dehydration
step is virtually zero, rules out the occurrence at low temper-
atures of the following ketonysation reaction, reported else-
where during the acid catalyzed decomposition of acetic acid,
and feasible at temperatures close to the one reported here for
the dehydration step [23]:
Autocatalytic NiO reduction to metallic nickel is the key
observation reported in the literature in the case of nickel
carboxylate decomposition [16]. In the present work, CO re-
duction of NiO is proposed through reaction (9), but, the for-
◦
mation of H2, observed in Fig. 4 at temperatures above 300 C
◦
with a peak maximum around 390 C, may also contribute to
the quantitative reduction of intermediate NiO, through the
following reaction:
NiO + H2 = Ni + H2O
(10)
We also believe that this reaction is responsible for the step
III (see Fig. 1b), noticed during themogravimetric analysis
of the decomposition of nickel acetate tetrahydrated in He,
because a broad water signal is produced at high temperatures
2
CH3COOH = CH3COH3 + CO2 + H2O
(5)
◦
with a peak temperature of around 373 C (see Table 2).
In addition to oxygenates, minute amounts of methane and
On the other hand, the water detected at higher tempera-
tures could be also in part the result of the dehydration of the
basic portion of the proposed intermediate:
◦
(11)
CO2 are also observed concurrently during the dehydration
step, although in much lower concentrations, as shown in
Fig. 4 and left panel of Fig. 5, respectively. In fact, methane
◦
and CO2 production at ca. 160–190 C can be explained
Ni(OH) = NiO + H O (370–390 C)
2
2
through the following reaction:
The TPD results summarized in Table 2 and depicted in
Fig. 6, suggest that carbon oxides and oxygenates are the
main products of the thermal decomposition of nickel acetate
tetrahydrated, with the following trend:
CH3COOH = CH4 + CO2
(6)
Following the same approach as before, next we will ad-
dress the issue of the major decomposition process, observed
◦
CO2 > CH3COCH3 > CH3COOH > CO > CH2CO
in this work around 340 C by TGA (Fig. 1) and corrobo-
rated by means of TPD-QMS studies (Fig. 2), in very good
agreement with data reported elsewhere [11,14].
This is in contrast with the results of Mohamed [11] who
reported a very different trend, carrying out the decomposi-
tion in a similar manner under an inert N2 atmosphere, but
employing gas chromatography for the characterization of
the gases evolved:
As seen in Fig. 3, acetone is the first oxygenated produced
◦
at ca. 325 C during the major decomposition of the acetate
◦
group. Then, at about 350 C, acetic acid is detected, as well
as minute amounts of ketene, formic acid and ethanol, virtu-
ally the same products observed previously during the dehy-
dration step. CO2 is clearly the main product observed during
thethermaldecompositionofnickelacetate(see Table2). The
acetone production, as well as the double structure observed
CH3COOH > CO2 ꢀ CH3COCH3
We believe that this divergence is more likely related to
differences in the technique employed for gas-evolved anal-
ysis (i.e. gas chromatography versus mass quarupole).

2
90
J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
4
.3. X-ray photoelectron spectroscopy
The discussion provided so far suggests some possible in-
5. Conclusions
On linear heating, nickel acetate tetrahydrate first releases
water at ca. 120 C, but acetic acid is produced concurrently
◦
termediates and products for the thermal decomposition of
nickel acetate, based solely on the interpretation of weight
losses and the complementary analysis of evolved gases by
means of TPD-QMS. Further insight into the chemical com-
position of the solids generated at different stages can be ob-
tained taking advantage of the power of XPS as a surface sen-
sitive technique, as compared to XRD, habitually employed
for solid characterization.
Besides the parent nickel acetate, our TGA and TPD-QMS
findings suggest the involvement of at least Ni(OH)2, NiCO3,
NiO, and Ni, during the course of the thermal decomposition
of nickel acetate. Furthermore, there is some uncertainty as
to whether nickel carbide is as well a reaction intermediate
and/or retained in the final decomposition product during the
pyrolisis of NiAc [11,16].
due to the surface hydrolysis of the constituent anion. During
the dehydration step, gas phase reactions of the acetic acid
produce mainly ketene, and minute amounts of ethanol and
formic acid. Basic nickel acetate, with an estimated formula
0.86Ni(CH COO) ·0.14Ni(OH) , is proposed as an interme-
3
2
2
diate following the dehydration step, as suggested by inter-
pretation of thermogravimetric data. On continued heating,
the subsequent decomposition of the basic nickel acetate at
◦
ca. 340 C lead directly to the formation of either NiO or
Ni, under treatment atmospheres of air and H , respectively.
In an inert atmosphere, however, an additional step suggests
that NiO is intermediate to the final production of metallic
Ni with an estimated 4–5% of carbonaceous residues. While
the evolution of acetone during the major decomposition step
2
The XPS data reported in the results section reveal that
surface carbides are not detected as steady species along
the thermal decomposition treatment, in contrast with pre-
vious XRD results. It is well known that the surface sensi-
tivity of XPS allows the detection of chemical states other-
wise imperceptible by bulk analysis techniques as XRD, and
this may account for the differences observed in the present
work. We nevertheless favor the results suggested here, be-
cause it has been pointed out that nickel carbide is metastable
suggests the involvement of NiCO as a short-lived interme-
diate, acetic acid production may be an indication of the final
reduction to metallic nickel. In contrast to previous reports,
carbides were not detected in the surface as intermediates
during the thermal decomposition of nickel acetate by XPS
analysis.
3
Acknowledgements
and quickly decomposed on heating close or above
◦
Finantial support for this research was provided by
PDVSA-Intevep.
4
00 C [25].
TGA and TPD-QMS analysis indicated an almost com-
plete reduction of nickel acetate to metallic nickel, when the
◦
thermolysis was carried out at 450 C in He atmosphere. On
References
the other hand, XPS complementary analysis, as shown in
the left side panel of Fig. 7 (top), still indicate important
contributions of Ni2+ species on the surface of the metallic
residue.
Our samples were not specially preserved, and in this con-
dition natural passivation of elemental metals is expected
even after brief exposures to the atmosphere [26]. On the
other hand, metallic nickel nanoparticles are expected to be
very reactive, and the build-up of a relatively thick nickel ox-
ide/hydroxyde film is feasible immediately upon air exposure
[1] M. Agnelli, H.M. Swaan, C. Marquez-Alvarez, G.A. Martin, C.
Mirodatos, J. Catal. 175 (1998) 117.
[2] A.E. Aksoylu, Z. Onsan, Appl. Catal. A Gen. 164 (1997) 1.
[3] I. Alstrup, J. Catal. 151 (1995) 216.
[4] R. Jin, Y. Chen, W. Li , W. Cui, Y. Ji, C. Yu, Y. Jiang, Appl. Catal.
A Gen. 201 (2000) 71.
[5] Z. Liu, K. Jun, H. Roh, S. Baek, S. Park, J. Mol. Catal. A: Chem.
189 (2002) 283.
[6] H.S. Bengaard, J.K. Norskov, J. Sehested, B.S. Clauser, L.P. Nielsen,
A.M. Molenbroek, J.R. Rostrup-Nielsen, J. Catal. 209 (2002) 365.
[
7] T. Borowiecki, A. Golebiowski, B. Stasinska, Appl. Catal. A. Gen.
53 (1997) 141.
8] A.B. Edwards, C.D. Garner, K.J. Roberts, J. Phys. Chem. B 101
1997) 20.
[27].
1
XPS analysis confirm that highly reactive metallic nickel
is obtained directly as the final product of the thermal de-
composition of nickel acetate, under an inert He atmosphere
[
(
[9] A.N. Kharat, P. Pendleton, A. Badalyan, M. Abedini, M.M. Amini,
J. Catal. 205 (2002) 7.
◦
and at temperatures higher than 400 C. Independent carbon
[
[
10] P. Baraldi, Spectrosc. Acta A 38 (1982) 51.
11] M.A. Mohamed, S.A. Halawy, M.M. Ebrahim, J. Anal. Appl. Pyrol-
ysis 27 (1993) 109.
analysis showed that the product contained 4–5% of carbon,
while TPD data depicted in Fig. 5 (right side panel) suggests
that the carbonaceous species present on the surface of the
[12] M. Afzal, P.K. Butt, H. Ahmad, J. Therm. Anal. 37 (1991) 1015.
◦
nickel particles react at 940 C, evolving minute amounts of
[13] A.M. Gadalla, H.F. Yu, Thermochim. Acta 164 (1990) 21.
[
[
[
14] G.A.M. Hussein, A.K.H. Nohman, K.M.A. Attyia, J. Therm. Anal.
2 (1994) 1155.
15] N. Nishizawa, T. Kishikawa, H. Minami, J. Sol. State Chem. 146
1999) 39.
16] A.K. Galwey, S.G. McKee, T.R.B. Mitchell, React. Sol. 6 (1988)
173.
CO to the gas phase.
4
We are at present investigating the catalytic activity of
the nickel particles obtained after the thermal decomposition
of nickel acetate tetrahydrate, and also exploring alternative
ways of characterizing their physicochemical properties.
(

J.C. De Jesus et al. / Journal of Molecular Catalysis A: Chemical 228 (2005) 283–291
291
[
17] J.L. Doremieux, Bull. Soc. Chim. Fr. 5 (1969) 1508.
[22] L.J. Matienzo, L.I. Yin, S.O. Grim, W.E. Swartz, Inorg. Chem. 12
(12) (1973).
[23] O. Kresnawahjuesa, R.J. Gorte, D. White, J. Mol. Catal. A 208
(2004) 175.
[24] N.E. Fouad, M.A. Mohamed, M.I. Zaki, H. Knozinger, J. Anal. Appl.
Pyrolysis 53 (2000) 185.
[25] J.P. Coad, J.C. Riviere, Surf. Sci. 3 (1961) 609.
[26] T.L. Barr, J. Phys. Chem. 16 (1982) 1801.
[27] X. Phung, J. Groza, E.A. Stach, L.N. Williams, S.B. Ritchey, Mater.
Eng. A 359 (2003) 261.
[
18] The Mass Spectrometry Data Centre (Eds.), Eight Peak Index of
Mass Spectra, The Royal Society of Chemistry, Nottingham, 1983.
19] M.L.M. Bonati, R.W. Joyner, M. Stockenhuber, Catal. Today 81
[
[
(2003) 653.
20] C.D. Wagner, W.M. Riggs, L.E. Davis, J.F. Moulder, G.E. Muilen-
berg (Eds.), Handbook of X-Ray Photoelectron Spectroscopy,
Perkin–Elmer, Eden Prairie, 1978.
[21] N. Laidani, L. Calliari, G. Speranza, V. Michelli, E. Galvanetto, Surf.
Coat. Tech. 100–101 (1998) 116.