Nanostructured cobalt oxides (Co3O4 and CoO) and metallic Co powders synthesized by the solution combustion method

Article abstract of DOI:10.1016/j.materresbull.2010.03.001

The combustion synthesis technique using glycine and urea as fuels and cobalt nitrate as an oxidizer is capable of producing well-crystallized Co₃O₄, CoO, as well as metallic Co powders. An interpretation based on the thermodynamic viewpoint and the measurement of the combustion temperatures during the reactions occurring for various fuel-to-oxidant ratios was proposed for a study of the nature of combustion and its correlation with the characteristics of as-synthesized powders. The largest measured specific surface area of the powders was 36 m²/g at a 0.14 glycine-to-nitrate ratio. The crystallites were nano-sized ranging from approximately 23 to 90 nm.

Full text of DOI:10.1016/j.materresbull.2010.03.001

Materials Research Bulletin 45 (2010) 672–676
Contents lists available at ScienceDirect
Materials Research Bulletin
journal homepage: www.elsevier.com/locate/matresbu
Nanostructured cobalt oxides (Co O and CoO) and metallic Co powders
3
4
synthesized by the solution combustion method
J.C. Toniolo *, A.S. Takimi, C.P. Bergmann
Department of Material Engineering, Federal University of Rio Grande do Sul, Av. Osvaldo Aranha 99, 705, Downtown, 90035190 Porto Alegre, RS, Brazil
A R T I C L E I N F O
A B S T R A C T
Article history:
The combustion synthesis technique using glycine and urea as fuels and cobalt nitrate as an oxidizer is
Received 17 February 2009
Received in revised form 9 February 2010
Accepted 3 March 2010
3 4
capable of producing well-crystallized Co O , CoO, as well as metallic Co powders. An interpretation
based on the thermodynamic viewpoint and the measurement of the combustion temperatures during
the reactions occurring for various fuel-to-oxidant ratios was proposed for a study of the nature of
combustion and its correlation with the characteristics of as-synthesized powders. The largest measured
Available online 9 March 2010
2
specific surface area of the powders was 36 m /g at a 0.14 glycine-to-nitrate ratio. The crystallites were
Keywords:
nano-sized ranging from approximately 23 to 90 nm.
A. Ceramics
ß 2010 Elsevier Ltd. All rights reserved.
B. Chemical synthesis
C. X-ray diffraction
D. Thermodynamic properties
1
. Introduction
process varies. Besides, the knowledge on the influence of these
variables on the SCS method of preparation of these powders is
In recent years, attention has increasingly been focused on the
development of nano-sized cobalt oxide powders (Co and CoO).
These powders have high potential for use as catalysts [1], gas
sensors [2], magnetic materials [3], rechargeable batteries [4],
pigments for ceramics [5], solar energy absorbers [6], and
electrochromic devices [7–9].
scarce.
For this reason, in the current work, we report the synthesis of
cobalt oxides (Co and CoO) and metallic Co powders, using
glycine and urea as fuels, and cobalt nitrate as oxidizer. Besides,
different fuel-to-oxidant molar ratios were employed. The effect of
the fuel in controlling particle size and microstructure of the
product for combustion with different fuel-to-oxidant ratios was
investigated here.
3 4
O
3 4
O
3 4
Cobalt (II, III) oxide (Co O ) and cobalt (II) oxide (CoO) particles
are largely prepared by a wide assortment of wet chemical
techniques like freeze-drying, spray-pyrolysis [10], co-precipita-
tion [11], solubility-controlled synthesis [12], sol–gel [13], and
combustion synthesis. In contrast, there are not many studies of
cobalt oxides particles obtained by the solution combustion
synthesis (SCS) method.
2. Experimental procedure
Cobalt nitrate Co(NO
Synth Qu ı´ mica, Brazil) of (vendor specification) 98, 98.5, and
3
)
3
ꢀ6H
2
O (Aldrich, USA), glycine, and urea
(
To the best of our knowledge, there are only a few reports on the
99.5% purities, respectively, were employed as starting materials.
3 4 2 3
formation of cobalt (II, III) oxide by the SCS method. Co O /g-Al O
The components were carefully dissolved in deionized water,
catalysts were prepared by Zavyalova et al. [14] to investigate their
catalytic performance during the total oxidation of methane. In the
work cited, a series of chemical precursors such as cobalt acetate,
cobalt acetate/cobalt nitrate mixtures, cobalt nitrate with citric
acid, glycine, and urea fuel mixtures was employed. Additionally,
and the solution was transferred to a stainless steel container,
which was uninterruptedly heated with a Bunsen-type burner
without any prior thermal dehydration. After some time, the fuel-
nitrate solution turned into a transparent viscous gel, which is
auto-ignited, giving voluminous foam as product of combustion.
The flowchart of this process is shown in Fig. 1.
Under continuous intense heating, the precursor mixture auto-
ignited roughly in the 130–180 8C range and resulted in
spontaneous ignition forming powder that contained crystalline
phases. In fact, all the fuel-to-oxidant ratios evaluated, upon
combustion, resulted in a grayish voluminous product that was
identified by X-ray diffraction (XRD) to be crystalline phases of
Co
3
O
4
incorporating Li ions was obtained [15].
At present, there are no studies on the fuel-to-oxidant ratio in
the formation of cobalt oxides, and information concerning the
*
Corresponding author. Tel.: +55 51 33083405; fax: +55 51 33083406.
E-mail addresses: juliano_toniolo@hotmail.com,
juliano.toniolo@marcopolo.com.br (J.C. Toniolo).
3 4
Co O , CoO or Co, which suggests that a complete combustion
0025-5408/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.materresbull.2010.03.001

J.C. Toniolo et al. / Materials Research Bulletin 45 (2010) 672–676
673
The stoichiometric composition of the glycine–nitrate redox
mixture demanded the presence of 1(ꢁ10) + n(+9) = 0, or
n = 1.11 mol in the reaction. Thus, the chemical precursors were
combined in the molar proportion of 1:1.11 (or glycine-to-nitrate
molar ratio of 0.56). For urea–nitrate it entailed 1(ꢁ10) + n(+6) = 0,
or n = 1.67 mol. Likewise, reactants were added in the molar
proportion of 1:1.67 (or urea-to-nitrate molar ratio of 0.84). SCS
reactions with glycine involving stoichiometric, fuel-lean, and
fuel-rich combinations can be represented correspondingly, as
follows:
1
:0CoðNO
3
Þ2 ꢀ 6H
2
O þ 1:11NH
2
CH
2
COOH ! 0:33Co
3
O
4
þ 1:55N
2
þ 2:22CO
2
2
þ 8:77H O
ðEq: stoichiometric; 0:56Þ
1
:0CoðNO
þ 0:25CoO þ 2:66N
þ 14:32H ðEq: fuel-rich; 1:67Þ
3
Þ2 ꢀ 6H
2
O þ 3:33NH
2
CH
2
COOH þ 5:07O
2
3 4
! 0:25Co O
2
þ 6:66CO
2
2
O
1
:0CoðNO
3
Þ2 ꢀ 6H
2
O þ 0:28NH
2
CH
2
COOH ! 0:33Co
3
O
4
þ 1:14N
2
þ 0:56CO
2
þ 6:70H
2
O þ 1:70O
2
ðEq: fuel-lean; 0:14Þ
The reactions with urea were calculated in the same manner as for
glycine. To understand the variations of the adiabatic flame
temperature with respect to the fuel-to-oxidant ratio, the
following equation was employed [18].
Fig. 1. Flowchart for the preparation of Co and cobalt oxide powders.
0
0
p
DHr ꢁ
D
H
T ¼ T
0
þ
probably occurred due to the characteristics of chemical reagents
such as the fuels consumed and the oxidant nature of nitrate.
The as-synthesized powders were characterized by XRD and
scanning electron microscopy (SEM). The combustion reaction was
investigated with thermocouple measurements. The temperatures
of combustion were measured using an adjustable type K
thermocouple.
C
p
0
0
where T
0
is 25 8C,
D
H , and DH are the enthalpies of formation of
r p
p
the reactants and products, respectively, and C is the heat capacity
of the products at constant pressure. Available thermodynamic
data in literature [19,20] for various reactants and products are
presented in Table 1. Substituting the thermodynamic data from
Table 1 in the equation, the enthalpy of combustion and the
theoretical adiabatic flame temperatures can be calculated.
Thus, all the equations can be simply expressed by some effects
shown in Table 2.
The thermodynamic calculations of the combustion reaction
show that as the fuel-to-oxidant ratio increases the amount of gas
produced, the enthalpy, and the adiabatic flame temperature also
increase.
X-ray diffraction was carried out on the combustion-synthe-
sized powders for phase characterization at a rate of 18/min and for
crystallite size determination at a rate of 0.48/min, using Cu K
a
radiation on a Philips X-ray diffractometer (model X’Pert MPD).
The single-line integral-breadth method of crystallite size
evaluation (designated as the single-line method [16]) was applied
with the Pearson VII profile, assuming characteristic peaks
intended for each studied phase, such as Co
2 0 0), and Co (1 1 1). Data were analyzed using the WinFit
software (version 1.2.1). SEM micrographs were recorded on a Jeol
model JSM-5800) instrument after coating the samples with gold.
3 4
O (2 2 0), CoO
1
(
3.2. Phase formation and morphology
(
In order to verify the effect of fuel type on combustion behavior,
different fuels were employed. Proportions with urea or glycine led
to Co formation in stoichiometric reactions. The glycine-rich
reactions corroborate the simultaneous formation of two oxide
3
. Results and discussion
3 4
O
3.1. Thermodynamic modeling
3 4
phases Co O and CoO that combine with Co in the 0.77 fuel-to-
Spontaneous redox reactions are generally exothermic in
nature and often lead to explosions when materials are not
carefully chosen. Both cobalt nitrate–glycine and cobalt nitrate–
urea mixtures underwent a self-propagating reaction showing that
they were capable of being heated without causing any explosive
exothermic reaction.
The initial composition of the solution containing cobalt nitrate
and glycine/or urea was derived from the total oxidizing and
reducing valences of the oxidizer and fuel using the concepts of
propellant chemistry [17]. Carbon, hydrogen, and cobalt were
considered as reducing species with corresponding valences of +4,
Table 1
Relevant thermodynamics data.
a
ꢁ1
ꢁ1 ꢁ1
(cal mol K )
Compound
D
H
f
(kcalmol
)
C
p
Co(NO
NH CH
CO(NH
Co (c)
3
)
2
ꢀ6H
COOH (c)
(c)
2
O (c)
ꢁ111.07
ꢁ94.03
ꢁ56.34
ꢁ0.232
ꢁ0.059
0.001
–
–
–
2
2
2
)
2
b
3
O
4
131.65 + 0.0660T
b
CoO (c)
Co (c)
45.26 + 0.0107T
b
4.469 + 0.0299T
CO
2
(g)
(g)
(g)
ꢁ94.051
0
10.34 + 0.00274T
6.50 + 0.0010T
5.92 + 0.00367T
7.20 + 0.0036T
–
N
O
2
+1, and +2, respectively. Oxygen was considered an oxidizing
2
0
element with valence of ꢁ2, and nitrogen was considered to be 0.
The total calculated valences of metal nitrates by arithmetic
summation of the oxidizing and reducing valences was ꢁ10. The
calculated valence of glycine was +9 and of urea +6.
2
H O (g)
ꢁ57.796
ꢁ33.2
NO
2
(g)
a
(
c) = crystalline, (g) = gas.
b
T = absolute temperature.

6
74
J.C. Toniolo et al. / Materials Research Bulletin 45 (2010) 672–676
Table 2
Effects of fuel type and composition based on thermodynamics.
Fuel-to-oxidant molar
ratio (% fuel)
Total
D
H8
Adiabatic
flame
ꢁ
1
number
of mol of
gases
(kcal mol
)
temperature
(8C)
GLY
U
GLY
U
GLY
U
GLY
U
1
1
0
0
0
0
0
.67 (+200) 2.51 (+200) 23.67 27.04 ꢁ1097 ꢁ1069 2910
2850
2762
2716
2618
2524
2314
1902
.11 (+100) 1.67 (+100) 18.10 20.36
ꢁ830
ꢁ696
ꢁ566
ꢁ499
ꢁ435
ꢁ327
ꢁ812 2816
ꢁ688 2744
ꢁ557 2653
ꢁ495 2544
ꢁ430 2335
ꢁ327 1902
.84 (+50)
.56
1.26 (+50)
0.84
15.30 17.04
12.55 13.68
12.03 12.46
11.06 11.44
.42 (ꢁ25)
.28 (ꢁ50)
.14 (ꢁ75)
0.63 (ꢁ25)
0.42 (ꢁ50)
0.21 (ꢁ75)
9.42
9.55
oxidant molar ratio, as seen in Fig. 2. On the other hand, for urea-
rich reactions the oxide phases were barely obtained (Fig. 3).
Inasmuch as the fuel-to-oxidant ratio increases, variation in the
relative concentrations of different chemical species of Co also
increases. This augment may be attributed mainly to the
differences in flame temperature and as well as to the nature of
reactions in the process of combustion.
Fig. 4. Variations of specific surface area of powders for different proportions of
glycine-nitrate and urea-nitrate. *The error estimate based on instrument accuracy
was 0.2%.
2
+
Fig. 4 shows the variation of the specific surface area of powders
in relation to the glycine-nitrate and urea-nitrate proportions.
Lean-fuel L(ꢁ75%) reactions in the 0.14 glycine- and 0.21 urea-to-
oxidant molar ratio yielded the highest specific surface area, while
the fuel-rich reactions yielded the lowest.
Besides, cobalt oxides are known to release oxygen upon
3 4
heating, converting Co O into CoO [21].
Fig. 5. Effects of crystallite size and phase formations of cobaltic oxide, cobaltous
oxide, and metallic cobalt for different proportions of glycine-nitrate and urea-
nitrate. *The error estimate based on instrument accuracy was 5.0%.
Fig. 2. Overlay of X-ray diffraction patterns of as-synthesized powders with various
glycine-to-oxidant molar ratios: (a) 0.14 fuel-lean (ꢁ75%); (b) 0.56 stoichiometric;
(
c) 0.70 fuel-rich (+25%); (d) 0.77 fuel-rich (+37.5%); and (e) 1.67 fuel-rich (+200%).
Fig. 3. Overlay of X-ray diffraction patterns of as-synthesized powders with various
urea-to-oxidant molar ratios: (a) 0.21 fuel-lean (ꢁ75%); (b) 0.84 stoichiometric; (c)
Fig. 6. Effects of fuel-to-oxidant ratios (Co
3 4
O phase) and maximum temperatures
1
.15 fuel-rich (+37.5%); (d) 1.26 fuel-rich (+50%); and (e) 2.51 fuel-rich (+200%).
of combustion for different proportions of glycine-nitrate and urea-nitrate.

J.C. Toniolo et al. / Materials Research Bulletin 45 (2010) 672–676
675
Fig. 7. SEM micrographs of cobalt oxide ceramic powders produced from glycine-lean reaction (0.56 ratio): (a) at low-magnification (5000ꢂ) and (b) at high-magnification
30,000ꢂ).
(
There is a probable correlation between the increase in
crystallite size and the reduction of specific surface area with
regard to the fuel content.
rendering it more reactive than the amino and carboxylic acid pairs
of glycine, in spite of the combustion reactions with urea yield
lower maximum temperatures.
In fuel-rich reactions R(200%) in the 1.67 glycine-to-oxidant
molar ratio, it was possible to observe the crystallite size
contributions of CoO and metallic Co formation by means of the
single-line technique (Fig. 5). Specifically, the metallic Co
formation was not originated with urea reactions.
The presence of metallic Co and CoO indicates that organic
compounds reduce partly or completely cobalt oxide, which is
formed by decomposition of cobalt nitrate. Moreover, some
studies developed by Luisetto et al. [22] revealed a correlation
of the reducibility behavior of Co with the particle size
The SEM morphology of the agglomerates of the cobalt oxide is
shown in Fig. 7. It exhibited foamy agglomerated particles with a
wide distribution (low-magnification) and presence of large voids
in its structure (high-magnification). The formation of these
features is attributed to the large volume of gas evolved during
combustion.
No significant differences came out of the fuel-to-oxidant ratios
from all morphologies examined with the SEM.
4. Conclusions
(
Co
3
O
4
! CoO ! Co).
Fig. 6 shows the formation of smaller crystallites for fuel-lean
Glycine-nitrate and urea-nitrate combustion syntheses present
reactions which followed an expected reducing trend as soon as
the temperature decreased for these proportions.
The reactions showed a similar linear behavior in proportion to
the fuel added; the temperature increases and there is a growth of
crystallite sizes in accordance with an increase in the fuel-to-
oxidant ratio.
The values obtained with thermocouple showed to be
significantly different from the calculated adiabatic flame tem-
peratures for maximum temperatures. These results are an
approach due to a probable presence of inaccuracy measurement.
Thermocouple has inertia and it is measured on the punctual form.
Nevertheless, the measured flame temperatures are typically
much lower than calculated values as a result of radioactive losses,
incomplete combustion, and heating of air.
a remarkable potential for producing Co and cobalt oxide powders.
The metallic Co powder was formed only in fuel-rich reactions
with glycine, while the cobaltous oxide phase was formed in fuel-
rich reactions of both fuels. This difference may be attributed to the
chemical nature of the fuels.
The smallest crystallite size obtained was 23 nm and the
2
highest specific surface area was 36 g/m under fuel-lean reactions
of glycine. Thermodynamic modeling of the combustion reaction
shows that when the fuel-to-oxidant ratio increases the amount of
gas produced and the temperature of the adiabatic flame also
increase. The same type of increment was obtained in the actual
flame temperature and crystallite size.
Acknowledgments
As can be seen, the estimated flame temperature was system-
atically higher for glycine as a fuel compared to the case when urea
was used, and this conclusion agrees well with the experiment. At
the same time, the crystallite size was systematically larger for
samples prepared from mixtures containing urea even though one
should expect better crystallinity for the powders prepared at
higher temperature.
Thanks are due to undergraduate student Frederico Wallauer
for combustion synthesis measurements and graduate student
Cibele M. Halmenschlager for providing support in scanning
electron microscopy.
This result may indicate a strong chelation of Co by glycine,
while in the case of urea one deals with the decomposition of plain
cobalt nitrate in the presence of fuel without a substantial
complexation.
References
[
[
[
1] J.H. Choy, et al. Chem. Mater. 14 (2002) 3823–3828.
2] M. Ando, T. Kobayashi, S. Iijima, M. Haruta, J. Mater. Chem. 7 (9) (1997) 1779.
3] M. Verelst, et al. Chem. Mater. 11 (1999) 2702–2708.
Although this transition metal is satisfactory and is mostly
effectively complexed by the amino groups and carboxylic acid of
the glycine, it is possible that a steric hindrance occurs caused by
the zwiterionic character (amino and carboxylic acid presence) of
the glycine molecule that could serve as a barrier to the growth of
the crystallites. Another factor that can explain the reason for the
lower formation of crystallites with glycine than with urea is the
vigorous activity generated by the two amino ions from urea,
[4] S.A. Needham, Development of Advanced Electrode Materials for Use in Recharge-
able Lithium Batteries, Doctoral Thesis, University of Wollongong, 2007.
[5] J.D. Albani, A.C. Antunes, S. Cava, A.S. Pianaro, S.R.M. Antunes, 3rd Brazil MRS
Meeting, 2004, H-O0222.
[
6] E. Barrera, I. Gonz a´ lez, T. Viveros, Sol. Energy Mater. 51 (1998) 69–82.
[7] S. Weichel, P.J. Moller, Surf. Sci. 399 (1998) 219.
[8] C.G. Granqvist, Handbook of Inorganic Electrochromic Materials, Elsevier,
Amsterdam, 1995.
[9] F. Svegl, B. Orel, M.G. Hutchins, K. Kalcher, J. Electrochem. Soc. 143 (5) (1996)
1532–1539.

6
76
J.C. Toniolo et al. / Materials Research Bulletin 45 (2010) 672–676
[
10] Z.W. Zhao, K. Konstantinov, L. Yuan, H.K. Liu, S.X. Dou, J. Nanosci. Nanotechnol. 7
4) (2004) 861–866.
[17] V.C. Sousa, A.M. Segad a˜ es, M.R. Morelli, R.H.G.A. Kiminami, Int. J. Inorg. Mater. 1
(1999) 235–241.
(
[
[
[
[
[
[
11] J.B. Wu, J.P. Tu, X.L. Wang, W.K. Zhang, Int. J. Hydro. Energy 32 (5) (2007) 606–610.
12] T. He, D. Chen, X. Jiao, Y. Wang, Y. Duan, Chem. Mater. 17 (2005) 4023–4030.
13] N.R.E. Radwan, M.S. El-Shall, H.M.A. Hassan, Appl. Catal. A 331 (2007) 8–18.
14] U. Zavyalova, P. Scholz, B. Ondruschka, Appl. Catal. A 323 (2007) 226–233.
15] F. Gu, C. Li, Y. Hu, L.J. Zhang, J. Cryst. Growth 304 (2007) 369–373.
[18] J.C. Toniolo, M.D. Lima, A.S. Takimi, C.P. Bergmann, Mater. Res. Bull. 3 40 (2005)
561–571.
[19] R.H. Perry, C.H. Chilton, Chemical Engineers Handbook, 5th ed., McGraw-Hill,
New York, 1973, p. 197.
[20] O.V. Dorofeeva, O.N. Ryzhova, J. Chem. Thermodyn. 41 (2009) 433–438.
[21] R. Flamand, High Temp.–High Pres. 3 (16) (1984) 323–328.
[22] I. Luisetto, F. Pepe, E. Bemporad, J. Nanopart. Res. 10 (2008) 59–67.
16] R. Delhez, T.H. Keijser, E.J. Mittemeijer, Fresen. Z. Anal. Chem. 312 (1982)
1
–16.