Sintering kinetics of submicron sized cobalt powder

Article abstract of DOI:10.1016/j.tca.2009.01.017

The paper details the results of sintering kinetics studies on submicron sized fine cobalt metal powders prepared through oxalate decomposition route using both conventional sintering method as well as stepwise isothermal dilatometry (SID) technique. Powd

Full text of DOI:10.1016/j.tca.2009.01.017

Thermochimica Acta 488 (2009) 54–59
Contents lists available at ScienceDirect
Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca
Sintering kinetics of submicron sized cobalt powder
Bhaskar Paula, Dheeraj Jainb, A.C. Bidayea, I.G. Sharmaa, C.G.S. Pillaib,∗
a Materials Processing Division, Bhabha Atomic Research Centre, Mumbai 400 085, India
b Chemistry Division, Bhabha Atomic Research Centre, Mumbai 400 085, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The paper details the results of sintering kinetics studies on submicron sized fine cobalt metal pow-
ders prepared through oxalate decomposition route using both conventional sintering method as well as
stepwise isothermal dilatometry (SID) technique. Powder preparation and processing parameters such
as decomposition temperature, compaction pressure and sintering atmosphere, etc. were optimized to
achieve highest sintered density with lowest activation energy through conventional sintering. Role of
submicron sized powder and surface contamination layers on initial shrinkage was explained towards
early shrinkage of cobalt compacts by dilatometric analysis. Measured step isothermal shrinkage data
were analyzed by Makipirtti–Meng method. The shrinkage data were found to fit well with the rate
equation proposed in this method and its validity was established for the metallic systems also. Kinetic
parameters were evaluated and sintering was found to occur through three major mechanisms operative
successively, which are grain boundary diffusion, lattice diffusion and plastic/viscous flow with the energy
of activation as 135, 234 and 373 kJ/mol, respectively. The results were well supported by microstructural
analysis of sintered specimens.
Received 19 August 2008
Received in revised form 1 January 2009
Accepted 7 January 2009
Available online 23 January 2009
Keywords:
Sintering
Kinetics
Cobalt
Dilatometry
SID
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
etc. Understanding of the sintering mechanism of these powders is
therefore very essential from both applied as well as research point
The major use of cobalt metal continues to be in super alloys
followed by its use in hard materials, pigments, magnets, batteries,
orthopedic alloys, catalysts and coloring agents, etc. Besides this,
extensive usage of its radioisotope 60Co, as a gamma radiation
source for a variety of industrial and medical applications, self
powered neutron detectors (SPNDs), flux mapping, fuel manage-
ment, etc. makes it a material of applied interest for the researchers
[1,2]. As the industrial applications of cobalt are increasing, its
reprocessing/recovery from scrape materials such as spent catalyst,
used cutting tools, etc. is gaining importance. An efficient flow
sheet for recovering high purity cobalt metal from spent ammonia
cracker catalyst has previously been established in authors’ lab
using hydrometallurgical route [3,4]. In this process, the spent
catalyst is converted to cobalt oxalate dihydrate Co(OOC)2·2H2O,
which then is thermally decomposed in reducing atmosphere to
produce submicron sized pure cobalt powder. During decompo-
sition of oxalate, six foreign atoms leave its lattice for each cobalt
atom and therefore produce highly porous metal powder with a
large specific surface area and fine crystallite size.
of view. Sintering kinetics of cobalt metal powders has been inves-
tigated earlier [5–7]. However, for submicron sized powders, which
are inherently superior due to their large surface area and higher
reactivity as compared to coarse ones, no such report is available.
To address this, sintering studies on submicron sized cobalt
powder were carried out using conventional sintering and stepwise
isothermal dialatometry (SID), a relatively new approach that has
proven its utility in analyzing the sintering mechanism of ceramic
powders [8–11]. Using this approach it is possible to evaluate both,
the sintering mechanism and activation energies using a single
shrinkage data obtained from dialatometry. Observed results from
SID were analyzed by a model given by Makipirtti–Meng [8–11].
It has been found that the model fits well to explain the sintering
kinetics of metallic powder systems also.
2. Experimental
2.1. Preparation and characterization of submicron sized cobalt
powder
These powders are then sintered into different shapes by powder
metallurgy and their sintering is mainly governed by powder prop-
erties such as particle size, shape/morphology and distribution,
Cobalt oxalate dihydrate Co(OOC)2·2H2O, produced by the
method reported earlier [3,4] was used for obtaining cobalt metal
powder. Dihydrate nature of the oxalate was confirmed by thermo
gravimetric analysis. Decomposition of the oxalate was carried
out in a specially designed vertical silica reactor with porous plate
∗
Corresponding author. Tel.: +91 22 25595355; fax: +91 22 25505151/25519613.
E-mail address: cgspil@barc.gov.in (C.G.S. Pillai).
0040-6031/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2009.01.017

B. Paul et al. / Thermochimica Acta 488 (2009) 54–59
55
for gas flow acting as a base for the charge material. The reactor
was placed in a resistive tubular furnace and decomposition was
performed in flowing Argon–hydrogen atmosphere (80:20 by
volume) with a flow rate of 20 ml/min at different temperatures
(723, 773, 823 and 873 K) with a soak period of 30 min in each
preparation. The contents were cooled to ambient temperature
under the same reducing atmosphere and passivated by flowing
CO2 gas (10 ml/min) over the powder before taking the samples
out. Elemental analyses of both oxalate and metal powders were
carried out on an atomic absorption spectrometer (GBC Spectra
Avanta, Model 908). The metal powders were phase characterized
by room temperature powder X-ray diffraction patterns recorded
on a powder diffractometer (Philips, Model PW 1820 controlled by
PW 1710 microprocessor). Silicon was used as an external standard.
Particle size of the as-prepared cobalt powder was measured by
a laser particle size analyzer (CILAS 1064) using distilled water as
dispersant. Powder morphology and particle size were examined
in a scanning electron microscope (JEOL, Model JSM-T330A).
2.2. Sintering studies on cobalt metal powder
For conventional sintering studies, the as-prepared metal pow-
ders were pressed into cylindrical pellets (6 mm diameter, 4 mm
height) at compaction pressures of 150, 200, 250 and 300 MPa and
sintered up to 1573 K in Ar–4% H2 atmosphere with a flow rate of
20 ml/min. Heating rate of 10 K/min and a soak period of 1 h at
the peak temperature was kept for all the samples. Densities of
sintered pellets measured geometrically, were examined as a func-
tion of compaction pressure and powder processing temperature.
The highest density was achieved for powder decomposed at 823 K
and compacted with a pressure of 250 MPa. These powder process-
ing parameters were adopted for all dilatometric experiments to
study sintering kinetics. Green pellets were also sintered in pure
argon atmosphere to see the effect of sintering atmosphere on the
sintered density.
All dilatometric measurements were carried out in a vertical-
type thermo mechanical analyzer (Setsys Evolution TMA 1600,
M/s SETARAM, France) with minimum load (1 g) conditions in
reducing atmosphere (Ar–4% H2, 20 ml/min). Shrinkage profiles
were recorded by heating the green compacts at 10 K/min. up to the
programmed temperature and soaked at that temperature for 1 h
followed by cooling to ambient. A sample already sintered at 1573 K
for 1 h was also measured separately up to the same temperature
with same heating cycle to evaluate the linear thermal expansion
behavior of cobalt. The same was subtracted from the as-measured
shrinkage data of green compacts to determine true shrinkage
behavior. A special sintering schedule was employed for recording
the sintering behavior through stepwise isothermal dilatometry
(SID). In this experiment, the shrinkage was continuously mea-
sured while heating the green pellet from 298 to 823 K at a heating
rate of 10 K/min. and then onwards up to 1223 K with the same
rate of heating, with an isothermal hold of 30 min at each 50 K
interval (step). Green densities (ꢀo) of the pressed samples were
measured geometrically where as that of the sintered samples
(ꢀs) were measured both dimensionally as well as by Archimedes
principle using water as immersion liquid. Microstructures of
samples sintered at different temperatures were studied using
scanning electron microscopy.
Fig. 1. SEM image of cobalt powder prepared from oxalate decomposition at 823 K.
Inset shows the higher magnification image of agglomerated particle to show sub-
micron nature of the powder.
0.005% Cu, as other metal impurities. XRD analysis of as-prepared
cobalt powder revealed a mixture of both HCP and FCC structure.
No detectable peaks for surface carbonates/oxides were observed
in the diffraction pattern, which must be due to amorphous nature
[6] of these surface monolayers, formed during passivation of
the metal powder. Fig. 1 shows the SEM image of cobalt powder
prepared at 823 K, which reveals a fibrous morphology. The micron
sized maze/sweet corn-like morphology of the powder is mainly
due to the spherolitic growth of small spherical individual crys-
tallites as is evident from the inset of Fig. 1. The observed fibrous
morphology is possibly due to the tendency of ferromagnetic
cobalt crystallites to align in a particular direction and form long
chain-like structures. Also, as the powder is prepared above 673 K,
adhesion between individual particles results into agglomeration
of the powder. The average agglomerate size measured from SEM
images is 23 ( 3) ␮m, close to that evaluated through laser particle
size analyzer, i.e. 12, 18 and 23 ␮m for powders prepared at 773,
823 and 873 K, respectively. It is to be noted that unavailability
of suitable dispersant for the particle size analysis resulted in
average agglomerate size determination and not the crystallite
size. However, the average crystallite size, clearly visible in the
inset of Fig. 1 is around 200 nm and reveals the submicron nature
of the metal powder.
3.2. Sintering by conventional method
Shrinkage analysis of a green compact during its initial stage of
sintering is given by the well-known Arrhenius function [7,12],
ꢀ
ꢁ
ꢀ
ꢁ
ꢀ
ꢁ
ꢁl
lo
−2Q
−Qa
T2/p
= K exp
= Kꢀ exp
(1)
pRT
RT
where
Bot
(2pDqR)2/p
3. Results and discussion
K =
(2)
(3)
3.1. Powder morphology
pQa
Q =
2
The analysis of as-prepared cobalt oxalate dihydrate
(Co(COO)2·2H2O) showed 0.052% Ni, 0.005% Al, 0.016% Fe and
and T2/p(ꢂl/l0) = shrinkage parameter

56
B. Paul et al. / Thermochimica Acta 488 (2009) 54–59
Fig. 2. Arrhenius plots for the shrinkage parameter of cobalt powder compacts
sintered up to 1523 K under different preparation temperatures and sintering atmo-
spheres. Inset shows the variation of sintered density as a function of powder
preparation temperatures and compaction pressures.
Fig. 3. Dilatometric shrinkage curves recorded on cobalt green compacts up to differ-
ent sintering temperatures in Ar–4% H2 atmosphere. Inset shows the early shrinkage
due to fine particle size and surface contamination reduction.
gen to react with surface layer of carbonates/oxides formed during
passivation and convert it to metal again. This in turn minimizes the
diffusion barrier present due to these surface layers. Also, hydrogen
molecules diffuse into the pores interstitially and therefore facili-
tates the densification process via reaction with surface layers [13].
The second factor is the physical capability of hydrogen to act as
a lattice defect catalyst [14] that facilitates the movement of point
defects and dislocations during sintering and therefore lowers the
activation energy.
Suitability of powders decomposed at 823 K for attaining lowest
activation energy of sintering can be explained as follows: regarding
the powder properties, two opposing factors are operative simul-
taneously (i) particle size and (ii) surface layer on these particles.
Smaller the particle size, faster is the diffusion resulting into low
activation energy. On the other hand, larger surface area available
with these smaller particles will lead to more surface layers (car-
bonates/oxides) acting a diffusion barrier and therefore slow down
sintering. Powder obtained at 723 K with a smaller average particle
size of 12 ␮m possesses higher surface contamination and therefore
yields lower sintered densities with higher activation energies. On
the other hand, powder decomposed at 873 K is coarse in nature
with a larger particle size and lesser surface contamination, again
leading to poor sintering kinetics. Therefore powder processed
at 823 K has an optimum compromise between particle size and
surface contamination and lead to faster sintering with lowest acti-
vation energy. The above reasoning is supported by results shown
in the inset of Fig. 2 where maximum sintered density is achieved
with powders processed at 823 K and compacted with a pressure
of 250 MPa.
Here, ꢂl is the change in sample length due to shrinkage, lo, the
initial length, Bo, a constant dependent upon material properties,
t, the sintering time, D, the particle diameter, R, the gas constant,
T, the absolute temperature and Kꢀ, a constant. Q is the activation
energy of mass diffusion and depends upon the sintering mech-
anism operative whereas Qa is apparent activation energy that is
determined experimentally. Parameters p and q depend upon the
sintering mechanism operative at a certain stage and take up differ-
ent values as (p, q) = (2, 1), (5, 3), (6, 4) and (7, 4) for viscous/plastic
flow, lattice/volume diffusion, grain boundary diffusion, and sur-
face diffusion, respectively.
Above methodology was adopted to analyze the conventional
sintering data in the present work and the value of (ꢂl/lo) was taken
as (ꢀs − ꢀo)/ꢀs where ꢀs and ꢀo represents the measured densi-
ties of sintered and green samples, respectively. All possible values
of p were tested and linear fit of the plot between ln{T2/p(ꢂl/lo)}
and 1/T was evaluated. The best fit resulted with p = 5, i.e. for lat-
tice/volume diffusion as the dominant process during initial stage
of sintering. This is further supported by SEM micrographs recorded
on samples sintered at 1023 K (Fig. 7) as discussed later. Fig. 2 shows
the Arrhenius plots drawn using Eq. (1) with p = 5 for metal pow-
ders prepared at different temperatures and sintered in different
atmospheres. The values of apparent activation energy ‘Qa’, eval-
uated from the slopes of the linear fits were used to calculate the
actual activation energy ‘Q’ using Eq. (3). The values are presented
in Table 1. It is evident that the lowest activation energy is required
for powder prepared at 823 K and sintered in Ar–4% H2 atmosphere
whereas sample sintered in pure argon atmosphere required max-
imum energy of activation.
The above results of conventional sintering studies clearly
establish the influence of powder processing conditions, sintering
atmosphere and compaction pressure over the sintering kinetics
and energy of activation. The results helped us in defining the
optimum powder parameters for further dilatometric studies to
investigate sintering mechanism.
The role of reducing atmosphere in lowering the activation
energy as compared to inert atmosphere (pure argon) can be
explained in a twofold manner. First is the chemical ability of hydro-
Table 1
Activation energy ‘Q’ of sintering for cobalt green compacts for powders prepared at
different temperatures and sintered in different atmospheres.
3.3. Sintering studies by dilatometry: evidence of early shrinkage
Decomposition
temperature (K)
Sintering atmosphere
Activation energy (kJ/mol)
Fig. 3 shows the shrinkage profiles of cobalt metal powder
compacts sintered at different temperatures in thermo mechani-
cal analyzer. It can be seen that maximum shrinkage is achieved for
sample heated up to 1573 K yielding nearly 98% theoretical density
823
873
823
Ar–4% hydrogen
Ar–4% hydrogen
Pure argon
223
245
365

B. Paul et al. / Thermochimica Acta 488 (2009) 54–59
57
cobalt pellets. However, an interesting observation, as shown in the
inset of Fig. 3 was noted in shrinkage profiles recorded in the earlier
stages of heating. Instead of initial expansion, normally expected
due to usual thermal expansion of any material before actual sin-
tering commences, small amount of shrinkage was observed when
sintered in reducing atmosphere (curve-a). On the other hand, sam-
ple sintered in pure argon atmosphere neither showed expansion
nor contraction up to 823 K (curve-b). To check for the initial shrink-
age observed below 823 K in the reducing atmosphere, a green
pellet preheated up to 823 K in Ar–4% H2 atmosphere for 1 h was
subjected to dilatometer. The preheated compact yielded a nor-
mal shrinkage curve with expansion up to 900 K followed by actual
sintering (curve-c). Also, it can be seen from the inset that actual
sintering starts at quite low temperature at 830 K as compared to
preheated compact that showed sintering only above 900 K. The
initial shrinkage observed at temperatures below 823 K before the
actual sintering can be explained as due to (i) reduction of surface
contamination (carbonates/oxides) layers to metal and thus reduc-
ing the volume [6] and (ii) very high surface energy of the submicron
sized powders that facilitate their aggregation at relatively lower
temperatures and hence exhibit shrinkage.
Fig. 4. SID shrinkage profile of cobalt green compact along with the shrinkage rate
as a function of time.
3.4. Sintering kinetics using stepwise isothermal dialatometry
(SID)
SID shrinkage data obtained from cobalt powder compacts have
been analyzed by the method proposed by Makipirtti and Meng
[8–11]. Before detailing the results, a brief description of this model
is presented.
(i) Makipirtti–Meng method: according to Makipirtti–Meng
equation for an isotropic sintering behavior, the fractional densi-
fication function, Y is expressed as
3
3
Vo − Vt
Vo − Vf
Lo − Lt
Y =
=
(4)
3
3
Lo − Lf
where Vo (Lo), Vt (Lt), and Vf (Lf) are the initial volume (length),
volume (length) at time t, and volume (length) of the fully-dense
specimen, respectively. A normalized rate equation as suggested by
Makipirtti–Meng method can be written as
Fig. 5. Plots of ln{(dY/dt)/[Y/(1 − Y)]}versus ln[(1 − Y)/Y] according to
Makipirtti–Meng equation. Variation of temperature is also shown on the
right hand scale.
ꢀ
ꢁ
1/n
dY
dt
1 − Y
= nk(T)Y(1 − Y)
(5)
(6)
Y
ꢀ
ꢁ
−Q
in Fig. 5 as ln(dY/dt)/[Y/(1 − Y)] versus ln[(1 − Y)/Y]. A near straight
line behavior for each isothermal hold zone except for 1050 K indi-
cates that the model agrees well with the present sintering data
of cobalt. A discontinuity in linearity for isothermal hold zone of
1050 K, indicated by an arrow further supports a probable change
in the dominant mechanism of sintering during this period as was
indicated by the shrinkage rate behavior in Fig. 4. The values of
slope ‘n’ and intercept ‘ln k(T)’, evaluated by least-squares linear
fitting for each straight line segment of the curve, are presented
in Table 2. For the isothermal hold zone at 1050 K however, the
slope calculated was from the linear fitting of the curve before the
start of the discontinuity. Slope could not be calculated after the
k(T) = ko exp
RT
Here k(T) is specific rate constant, ko, the frequency factor, Q, the
energy of activation, R, the gas constant and n, a parameter related
to the sintering mechanism.
Fig. 4 shows the shrinkage profile and shrinkage rate of the green
compact that was subjected to SID sintering schedule. It is clear
from the curve that maximum shrinkage during a single isothermal
soak period occurs at 1050 K with the maximum rate of shrink-
age. The nature of shrinkage rate curve also indicates a probable
change in the dominant mechanism of sintering during this isother-
mal period. The shrinkage data, after fitting into Eq. (5) are plotted
Table 2
Kinetic analysis of SID shrinkage data of cobalt green compacts according to Makipirtti–Meng equation.
Temperature (K)
1000/T (K−1
)
1/n
n
ln(n)
ln K(T)
Regression parameter
790
845
900
0.00126
0.00118
0.00111
0.00105
0.00100
0.00095
0.00091
0.00087
4.62
4.54
3.17
3.13
2.81
2.66
2.77
2.71
0.216
0.220
0.315
0.319
0.355
0.375
0.361
0.369
−1.5304
−1.5138
−1.1537
−1.1410
−1.0338
−0.9788
−1.0188
−0.9969
−18.47
−17.67
−15.06
−13.66
−11.44
−9.31
0.9921
0.9937
0.9876
0.9924
0.9906
0.9942
0.9925
0.9941
950
1000
1050
1100
1150
−7.12
−3.97

58
B. Paul et al. / Thermochimica Acta 488 (2009) 54–59
completion of discontinuity at the same temperature due to the
scattered nature of the data. Interestingly, parameter ‘n’, that relates
to the sintering mechanism takes nearly three different average val-
ues in the measured temperature range as n ∼ 0.218 (790–850 K),
0.317 (900–950 K) and 0.365 (1000–1150 K). These values suggest
that there are three different sintering mechanisms, operative as
dominant processes over the above temperature intervals.
Fig. 6 shows the Arrhenius plot between ln k(T) and 1/T for SID
shrinkage data. The curve can be fitted into three linear segments
in the temperature ranges 790–850, 850–950 and 950–1150 K. The
apparent activation energy ‘Q’ was calculated from the slopes of
each segment and compared with that of various sintering mech-
anisms reported in the literature [5,7]. The values are presented
in Table 3. It is evident that sintering of cobalt powder compacts
can be divided into three main stages. The first stage (790–850 K)
relates to sintering within the aggregates which are formed upon
coalescence of the individual crystallites and thereby leading to
neck formation through grain boundary diffusion. This is further
supported by the SEM image recorded on 850 K sintered sample
that shows initial stage of neck formation (Fig. 7a). The higher value
of activation energy (234 kJ/mol) for temperature range 850–950 K
is an indicative of lattice (volume) diffusion as the dominant pro-
Fig. 6. Arrhenius plot of ln k(T) versus 1000/T based on the data in Table 3.
Fig. 7. SEM micrographs of fractured surfaces of cobalt samples sintered up to different temperatures: (a) 850 K, (b) 950 K, (c) 1100 K, (d) 1273 K, (e) 1373 K, and (f) 1473 K.
Table 3
Values of activation energies along with the literature values and corresponding sintering mechanism as evaluated from the Arrhenius plot obtained from Makipirtti–Meng’s
method.
Temperature range (K)
Activation energy (kJ/mol)
Mechanism of sintering
Literature value (kJ/mol)
790–850
850–950
950–1150
135
234
373
Grain boundary
Lattice/volume diffusion
Viscous/plastic flow
163 [5]
289 [5]
261–304 [7]

B. Paul et al. / Thermochimica Acta 488 (2009) 54–59
59
cess for mass transport. The same is supported by SEM micrograph
recorded on a sample sintered up to 950 K as shown in Fig. 7b. The
relatively lower values of activation energies for these two stages
as compared to literature reported values is due to fine sized parti-
cles rendering larger interfaces, more defects, etc. [15]. In the third
stage (950–1150 K), sintering mainly occurs by viscous flow/plastic
flow along with other mechanisms also taking part and therefore
required maximum energy of activation. The viscous/plastic flow
of cobalt can be seen from the SEM images recorded on 1100 K
sintered sample (Fig. 7c). Actually, due to the low stacking fault
energy of cobalt, it has a tendency to show dynamic recrystalliza-
tion in the temperature range from 900–1200 K [16]. A large number
of dislocations that were generated during the cold compaction
of cobalt powder get annihilated in this stage of plastic flow and
result into recrystallized form of the parent structure. The micro-
graph recorded on sample sintered at 1273 K (Fig. 7d) clearly reveals
the same. Successive grain growth with increasing temperature can
be seen in Fig. 7e and f recorded on samples sintered at 1373 and
1473 K, respectively.
method to evaluate the sintering mechanism and activation ener-
gies. The method was found to fit well with the shrinkage data
of cobalt and hence validated its applicability for sintering stud-
ies of metallic powders also. Sintering was found to occur through
three dominant mechanisms with average activation energies of
135, 234 and 373 kJ/mol corresponding to grain boundary diffusion,
lattice diffusion and plastic/viscous flow, respectively. The results
were found to be consistent with the microstructural evaluation as
studied by SEM analysis.
Acknowledgements
The authors wish to acknowledge Dr. A.K. Suri, Director, Mate-
rials Group, BARC and Dr. D. Das, Head, Chemistry Division, BARC
for their constant encouragement and support during the course of
this work.
References
[1] I.G. Sharma, Studies on preparation and characterization of cobalt powder
and its fabrication into cobalt rods, Ph.D. Thesis, Mumbai University, India,
2006.
[2] J.A. Ylikerala, M. Gasik, Powder Report 59 (2004) 36.
[3] M.V. Rane, V.H. Bafna, R. Sadanandam, A.K. Sharma, K. Ramadevi, N.K. Menon,
M.F. Fonseca, S.K. Tangri, A.K. Suri, Hydrometallurgy 77 (2005) 247.
[4] R. Sadanandam, M.F. Fonseca, K. Srikant, A.K. Sharma, S.K. Tangri, A.K. Suri,
Hydrometallurgy 91 (2008) 28.
From the above discussion, it is therefore quite clear that the
SID approach can well be used to evaluate the sintering kinetics of
metallic systems and can give fairly reliable values of the activa-
tion energies involved with different mechanism of mass transport
using the dilatometry data of a single experiment. The results are
found to be consistent with the microstructural evaluation as stud-
ied by SEM analysis.
[5] J.P. Jernot, J.L. Chermat, A. Deschanvres, B.L. Mordike, Materials Research Bul-
letin 17 (1982) 815.
[6] Y. Sakka, Journal of Less Common Metals 168 (1991) 277.
[7] K. Saitou, Scripta Materiallia 54 (2006) 875.
4. Conclusion
[8] R. Yan, F. Chu, Q. Ma, X. Liu, G. Meng, Materials Letters 60 (2006) 3605.
[9] G.Y. Meng, O.T. Sorensen, in: T. Han (Ed.), Advanced Structural Materials, vol. 2,
Elsevier Science Publishers B.V., Amsterdam, Netherlands, 1991, p. 369.
[10] H.T. Wang, X.Q. Liu, F.L. Chen, G.Y. Meng, Journal of American Ceramic Society
81 (1998) 781.
[11] Y.F. Liu, X.Q. Liu, S.W. Tao, G.Y. Meng, O.T. Sorensenb, Ceramic International 28
(2002) 479.
[12] R.M. German, Powder Metallurgy Science, 2nd ed., MPIF, Princeton (NJ), 1994,
250.
[13] L. Liu, N.H. Loh, S.B. Tor, Y. Murakoshi, R. Maeda, Scripta Materialia 55 (2006)
1103.
[14] O. Dominguez, J. Bigot, Nanostructured Materials 6 (1995) 877.
[15] V.V. Dabhade, T.R. Rama Mohan, P. Ramkrishnan, Materials Science and Engi-
neering A 452 (2007) 386.
Sintering kinetics of submicron sized fine cobalt metal powder
prepared through oxalate decomposition route was studied by both
conventional sintering technique and stepwise isothermal dilatom-
etry (SID) technique. Experimental results demonstrate that factors
like powder preparation temperature, compaction pressure and
sintering atmosphere play an important role in attaining optimum
sintered density and fast sintering with lowest activation energy.
Small amount of shrinkage observed at low temperatures before
the start of actual sintering was due to fine sized nature of pow-
der and reduction of surface carbonate/oxide layers in reducing
atmosphere. SID shrinkage data were analyzed by Makipirtti–Meng
[16] B. Paul, R. Kapoor, J.K. Chakravartty, A.C. Bidaye, I.G. Sharma, A.K. Suri, Scripta
Materialia 60 (2009) 104.