Microstructure and thermal conductivity of surfactant-free NiO nanostructures

Article abstract of DOI:10.1016/j.jssc.2012.01.052

High purity, nanometer sized surfactant-free nickel oxide (NiO) particles were produced in gram scale using a solution combustion method and characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), gas pycnometry and gas adsorption analysis (BET). The average particle size of the as-synthesized NiO increases significantly with the preheating temperature of the furnace, while the specific surface area decreases. A BET specific surface area of ～100 m²/g was obtained for NiO nanoparticles with size as small as 3 nm synthesized at 300 °C. The thermal conductivity (κ) of pressed pellets of the synthesized NiO nanoparticles obtained using spark plasma sintering (SPS) and uniaxial hot pressing is drastically decreased (～60%) compared to that of NiO single crystal. This strong reduction in κ with particle size suggests the suitability of the synthesized surfactant-free NiO nanoparticles for use as nanoinclusions when designing high performance materials for waste heat recovery.

Full text of DOI:10.1016/j.jssc.2012.01.052

Journal of Solid State Chemistry 190 (2012) 29–35
Contents lists available at SciVerse ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Microstructure and thermal conductivity of surfactant-free
NiO nanostructures
a,b,c
c
d
a,c
c
Pranati Sahoo
, Dinesh K. Misra , Jim Salvador , Julien P.A. Makongo , Girija S. Chaubey ,
a,c
b,c
a,b,c,
n
Nathan J. Takas , John B. Wiley , Pierre F.P. Poudeu
a
Laboratory for Emerging Energy and Electronic Materials, Materials Science and Engineering Department, University of Michigan, Ann Arbor, MI 48109, USA
Department of Chemistry, University of New Orleans, New Orleans, LA 70148, USA
b
c
The Advanced Materials Research Institute, University of New Orleans, New Orleans, LA 70148, USA
d
Chemical Sciences and Materials Systems Laboratory, General Motors R&D Center, Warren, MI 48090, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
High purity, nanometer sized surfactant-free nickel oxide (NiO) particles were produced in gram scale
using a solution combustion method and characterized by X-ray diffraction (XRD), transmission
electron microscopy (TEM), gas pycnometry and gas adsorption analysis (BET). The average particle
size of the as-synthesized NiO increases significantly with the preheating temperature of the furnace,
Received 14 September 2011
Received in revised form
2
1 January 2012
Accepted 23 January 2012
Available online 30 January 2012
2
while the specific surface area decreases. A BET specific surface area of ꢀ100 m /g was obtained for NiO
nanoparticles with size as small as 3 nm synthesized at 300 1C. The thermal conductivity (k) of pressed
Keywords:
pellets of the synthesized NiO nanoparticles obtained using spark plasma sintering (SPS) and uniaxial
Nanostructures
hot pressing is drastically decreased (ꢀ60%) compared to that of NiO single crystal. This strong
Thermal conductivity
Combustion synthesis
Inorganic compounds
Electron microscopy
Surface properties
reduction in
k with particle size suggests the suitability of the synthesized surfactant-free NiO
nanoparticles for use as nanoinclusions when designing high performance materials for waste heat
recovery.
&
2012 Elsevier Inc. All rights reserved.
1
. Introduction
fractions of independently synthesized nanoparticles of the inclu-
sion phase [12,13]. However, for this approach to become a viable
strategy for nanocomposite thermoelectric materials, synthetic
techniques for production of surfactant-free nanoparticles of the
desired inclusion phase must be readily available. In this paper, we
discuss a simple method for the synthesis of surfactant-free NiO
nanoparticles for use as inclusion phases in high temperature
nanocomposite thermoelectric materials.
Nickel oxide, in bulk or nanoscale form, has been an attractive
materials for variety of applications ranging from solar cells [14],
electrochromic devices [15], gas sensing [16] to data storage [17].
In addition, nanostructured NiO, because of its large chemical and
thermal stability, also shows great promise for use in energy
conversion applications. For instance, p-type nanocrystalline NiO
can be used as an inclusion phase to promote additional phonon
scattering in high temperature p-type thermoelectric composite
materials such as half-Heusler phases (Zr0.5Hf0.5CoSb0.8Sn0.2 [18]
Since the concept of nanostructuring as a promising approach to
high performance thermoelectric materials was introduced two
decades ago [1], tremendous effort has been invested in the design
of new nanocomposite materials with high figures of merit. The
nanocomposites are generally fabricated by assembling fine nano-
particles of the desired thermoelectric material prepared via
various synthetic techniques such as hydrothermal methods [2],
wet chemical methods [3] or mechanical alloying [4,5]. Another
successful strategy to nanocomposite thermoelectric materials
consists of in situ growth of secondary inclusion phases within a
bulk thermoelectric material [6–11]. The motivation behind the
design and implementation of nanostructuring is to take advantage
of the reduction in lattice thermal conductivity arising from phonon
scattering at grain boundaries and interfaces between the inclusion
phase and the matrix. Reduction in lattice thermal conductivity can
also be achieved in nanocomposite materials fabricated by direct
combination of bulk thermoelectric materials with various volume
x
and Zr0.5Hf0.5Co1ꢁxRh Sb0.99Sn0.01 [9]) or silicon germanium
materials (Si80Ge20 and Si80Ge20 [19,20]). However, the perfor-
B
x
mance of devices and materials fabricated using nanocrystalline
NiO particles is strongly influenced by the microstructure (size
and morphology) and the accessibility of the surface of the
nanoparticles which, in turn, depend on synthetic strategies.
Several chemical routes for the synthesis of surfactant-free, highly
n
Corresponding author at: Laboratory for Emerging Energy and Electronic
Materials, Materials Science and Engineering Department, University of Michigan,
Ann Arbor, MI 48109, USA. Fax: þ1 734 763 4788.
E-mail address: ppoudeup@umich.edu (P.F.P. Poudeu).
0022-4596/$ - see front matter & 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2012.01.052

3
0
P. Sahoo et al. / Journal of Solid State Chemistry 190 (2012) 29–35
disperse particles of various metal oxides have been previously
developed and used for decades [21–25]. However, these methods
are severely limited when it comes to the preparation of particles
within the nanometer range. Not only do the synthesized parti-
cles showed poor crystallinity, but the synthetic protocols are
generally complex, difficult to reproduce, and challenging to scale
up to industrial quantities due to the high reactivity of the metal
precursors. The ability to produce nanometer scale surfactant-free
metal oxides with controlled particle shape and size is a corner-
stone for their use in nanoscience and nanotechnology.
Recently, the solution combustion method [26,27] and controlled
thermal decomposition of metal nitrates [28] were introduced as
optimized techniques to synthesize surfactant-free NiO nanoparticles.
The solution combustion technique allows homogeneous mixing in
aqueous solution, of reactants at the molecular level yielding precise
control over the stoichiometry of the final product. The high
temperatures associated with the synthetic process insures homo-
geneous products with excellent crystallinity, without calcinations
steps [29]. The formation of the crystalline NiO phase as well as the
size of the synthesized particles is controlled by the fuel to oxidizer
Therefore, one should therefore take the necessary precautions
to contain the gases and also to prevent any mess during the
boiling step. After the predetermined reaction time elapses,
the beaker was removed from the furnace (at the preheated
temperature) and allowed to cool to room temperature and the
NiO products were collected. Reaction parameters such as the
preheating temperatures and durations were varied to study their
effects on nanoparticle size, color and morphology.
Specimens for thermal conductivity measurements were fab-
ricated by spark plasma sintering or hot pressing fine powders of
the synthesized NiO nanoparticles under a dynamic vacuum of
ꢁ
4
ꢀ10 Torr. Pellets obtained from spark plasma sintering was
fabricated by cold pressing the NiO nano-powder under a uniaxial
pressure of 50 MPa in a graphite die, followed by heating at a rate
of 100 1C/min from 25 1C to 700 1C and at 50 1C/min from 700 1C
to 900 1C. The total heating time was 12 min including the
heating ramp. After consolidation was complete as indicated by
no net displacement of the pressing piston, the sample was
allowed to cool to room temperature with pressure released.
Pellets obtained by uniaxial hot press were fabricated by simul-
taneously applying a maximum pressure of 100 MPa and a
maximum temperature of 900 1C. The samples were placed into
a graphite die with an inner diameter of 10 mm and then sintered
(metal nitrate) ratio (c) [26,29]. In this paper, we explore the
synthesis of surfactant-free NiO nanoparticles with various sizes
using the combustion method. The synthesis parameters such as
the preheating temperature of the furnace and duration of the
reaction are systematically varied to gain control over the size range
of the synthesized nanoparticles. The structure and morphology of
the synthesized nickel oxide nanoparticles were characterized by
X-ray powder diffraction (XRPD) and electron microscopy (SEM,
TEM). Their specific surface area was measured using gas adsorption
analysis (BET method). NiO nanoparticles obtained from reactions
under various preheating temperatures and duration were compacted
using spark plasma sintering (SPS) or uniaxial hot press techniques
and the effects of particle size and pressing parameters on the
thermal conductivity of the resulting pellets were investigated in
the temperature range from 300 to 750 K.
ꢁ
5
under vacuum (10 Torr). The final working set point of pressure
(100 MPa) was first reached with a ramp of 2 MPa/min, followed
by a hold of 4 h while the temperature was still rising at a heating
o
rate of 5 C/min to 900 1C. The maximum temperature and
pressure were kept for 1 h to achieve higher strengthening and
good compaction of the pellet. The sample was then cooled to
room temperature at a cooling rate of 5 1C/min simultaneously
while the pressure released (2 MPa/min). The density of the
resulting pellet was determined by measuring the dimensions
and mass. NiO pellets were examined using transmission electron
microscopy (TEM) to probe grain growth during spark plasma
sintering or hot pressing.
2
. Experimental
2.2. Characterization
2.1. Synthesis
2.2.1. X-ray powder diffraction
The resulting NiO nanoparticles were characterized by X-ray
All the reagents used in the synthesis are commercially
powder diffraction for phase identification using CuK
a radiation
available. Nickel nitrate hexahydrate (Ni(NO
9%), was used as the oxidizer and nickel precursor and urea
(NH CO; EM Science) was used as fuel for the combustion.
3
)
2
ꢂ 6H
2
O; Aldrich,
(l
¼1.5406 A˚ ) on an X’Pert Pro, PANalytical system, equipped with
9
(
a curved graphite crystal monochromator and a scintillation
2
)
2
counter, operating in geometry with tube settings of 40 kV
y–y
Aqueous solutions of nickel nitrate and urea were thoroughly
mixed by stirring using a magnetic stirrer. The stoichiometric
composition of nickel (II) nitrate hexahydrate (oxidizer) and urea
and 40 mA. The average particle sizes of the synthesized NiO
nanoparticles were estimated from the XRD patterns using
Scherrer analysis (Table 1).
(fuel) was calculated using the total oxidizing and reducing
valencies of the components. The mixture is stoichiometric and
the heat of combustion is maximized when the oxidizer to fuel
2
.2.2. Electron microscopy
ratio (
fixed at the stoichiometric value (
00 1C were performed into a Pyrex beaker whereas a Quartz
c) equals 1 [30]. In this study, the oxidizer to fuel ratio was
The determination of particle size and morphology of the
c
¼1). Synthesis at 300, 400 and
as-synthesized NiO nanoparticles as well as the investigation of
the microstructure of spark plasma sintered or hot pressed pellets
of NiO nanoparticles were performed using a JEOL 2010 micro-
scope in both bright field image and selected area electron
diffraction (SAED) modes operated at 200 kV.
TEM specimens of hot pressed and spark plasma sintered
pellets were prepared by dicing 3 mm disks using an ultrasonic
disk cutter (model 170, Fischione). The disks were mechanically
polished using a dimple grinder with a load of 12 g (Gatan 1656),
6
beaker was used for reactions with furnace preheated at 800 1C.
Equal volume (100 ml) of 0.1 N nickel nitrate hexahydrate and
0
.166 N urea solutions were homogeneously mixed, transferred
into the appropriate beaker, and placed in a box furnace pre-
heated to the desired reaction temperature. The mixture under-
goes a flameless combustion during which the solution initially
boils and the mixture gradually dehydrates and finally decom-
poses to produce NiO nanoparticles with the evolution of a large
þ
followed by Ar -ion milling (FISCHIONE 1010). For the charac-
2 2 2
amount of gases such as N , CO and H O as follows:
terization of the as-synthesized NiO nanoparticles by TEM,
suspensions of the powders in alcohol were dropped onto a holey
carbon grid.
3
Ni(NO
3
)
2
ꢂ 6H
2
O
(aq)þ5NH
2
CONH2(aq)
(g)
-
3
NiO(s)þ8N2(g)þ5CO2(g)þ28H
2
O
(1)

P. Sahoo et al. / Journal of Solid State Chemistry 190 (2012) 29–35
31
Table 1
Particle size range, BET specific surface area and density of NiO nanoparticles synthesized at different temperature. *NiO-300-2 corresponds to NiO nanoparticles
synthesized at 300 1C in 2 h.
o
a
b
c
d
2
e
3
Sample code
T ( C)
t (h)
d (nm)
D (nm)
A (m /g)
Density (g/cm )
Color
*
NiO-300-2
300
400
600
600
800
800
2
Z3
Z7
24–34
24–50
ꢀ13
ꢀ15
99.70(3)
52.13(3)
8.59(2)
4.88(2)
2.73(2)
1.85(2)
6.177(8)
6.226(2)
6.729(2)
6.731(4)
6.845(4)
6.829(4)
Black
NiO-400-1
NiO-600-0.5
NiO-600-1
NiO-800-0.5
NiO-800-1
1
Black
0.5
1
Dark-green
Green
39
0.5
1
140–150
125–350
Green
300
Green
a
Furnace preheating temperature.
Reaction duration.
b
c
Particle size from TEM.
d
e
Average particle size from XRD (Scherrer analysis).
BET surface area.
2.2.3. Gas adsorption analysis (BET)
Specific surface area measurements were performed using the
2
gas adsorption BET method on a Micromeritics ASAP 2020. The
N
specific surface areas were calculated by comparing the adsorp-
tion capacity of the sample at room temperature and low
2
temperature (using liquid N ). The gas was dosed 11 times and
the average of the values obtained was reported with their
standard deviations.
2.2.4. Gas pycnometry
Powder density measurements were performed using helium
gas pycnometry on a Micromeritics Accupyc II 1340. Reported
data with their standard deviations are the average of 10
measurements. Synthetic parameters, specific surface area, parti-
cle size range, powder density and color of the synthesized NiO
nanoparticles are summarized in Table 1.
2.2.5. Thermal conductivity
Thermal diffusivity data were measured using a Netzsch LFA
4
57 laser flash system. A pyroceram reference material was
measured alongside each sample. Measurements were made
under flowing N gas (430 mL/min) from 300 K to 775 K at
increments of 25 K. C values for thermal conductivity calcula-
tions were extracted from the laser flash data and are comparable
to the C values of Bulk NiO reported earlier [31].
Fig. 1. X-ray diffraction pattern of NiO nanoparticles synthesized at: (a) 300 1C for
2 h; (b) 400 1C for 1 h; (c) 600 1C for 30 min; (d) 600 1C for 1 h; (e) 800 1C for
2
30 min and; (f) 800 1C for 1 h. The theoretical pattern of NiO is shown for
p
comparison.
p
3
to the bulk counterpart (6.8 g/cm ) [34]. The small particle size
and high surface area are favorable for the oxidation of the surface
3
. Results and discussion
of NiO (green) nanoparticles to Ni
2 3
O (black). The conversion rate
of NiO to Ni decreases with increasing average particle size
2 3
O
3
.1. Synthesis and X-ray diffraction
and decreasing specific surface area resulting in the observed
gradual change in the color of the synthesized NiO powders.
Fig. 1 shows a comparison of X-ray diffraction patterns of NiO
nanoparticles synthesized at various temperatures with the
theoretical pattern calculated from NiO structural data [35]. All
diffraction peaks can be indexed with the NiO fcc (face-centered
cubic) lattice. Except for the broad feature between 151 and 251
Surfactant-free NiO nanoparticles can be synthesized in fairly
large quantities via combustion of a mixture of nickel nitrate
oxidizer) and urea (fuel). However, the formation of the crystal-
(
line NiO phase strongly depends on the scale of the reaction
mixture as well as the temperature at which the furnace was
preheated. For instance, to produce NiO nanoparticles at 300 1C
with particle sizes below 7 nm, a stoichiometric mixture of nickel
nitrate and urea required a minimum of 2 h for completion. When
the furnace is preheated to 400 1C, NiO nanoparticles with a
similar size range can be obtained within 1 h. NiO nanoparticles
with a size range between 100 and 150 nm could be synthesized
within 30 min with the furnace preheated at 800 1C (Table 1). The
color of the synthesized NiO nanoparticles changes gradually
from black to dark green to green as the furnace preheating
temperature and reaction time increase. The observed change in
the color of the synthesized NiO nanoparticles can be explained
(2ꢁ
NiO-600-0.5, which is attributed to amorphous Ni
tional peaks of any other impurity phase could be detected on the
diffraction patterns. The presence of amorphous Ni in NiO-
300-2 and NiO-600-0.5 samples was also confirmed by X-Ray
photoelectron spectroscopy (XPS) (Fig. 2). The broad O 1s (Ni
y) on the diffraction patterns of NiO-300-2, NiO-400-1 and
2 3
O
, no addi-
2 3
O
2 3
O )
and Ni 2p3/2 (þ3) peaks in the XPS spectra of the NiO-300-2
sample (Fig. 2A and B), suggests the presence of a large fraction of
2 3 2 3
Ni O . However, the fraction of Ni O is sufficiently smaller in the
NiO-600-0.5 sample as can be appreciated by the large decrease
in the intensity of O 1s (Ni
spectra.
2
O
3
) and Ni 2p3/2 (þ3) peaks in the XPS
by the presence of various concentration of Ni
2
O
3
impurity phase
in the sample [32,33]. As can be seen from Table 1, NiO powders
with small average particle size showed lower density compared
The intensities of (111), (200) and (220) peaks on the XRD
patterns of the synthesized NiO nanoparticles are comparable

3
2
P. Sahoo et al. / Journal of Solid State Chemistry 190 (2012) 29–35
with those of corresponding peaks on the theoretical pattern
suggesting that well crystalline NiO nanoparticles could be
synthesized via the solution combustion method. The broadening
of diffraction peaks of the samples produced from lower reaction
temperatures indicates the formation of NiO nanoparticles with
smaller sizes.
3.2. Particle size and specific surface area
Selected transmission electron microscopy images of the
as-synthesized NiO nanoparticles are shown in Fig. 3. Aggregates
of spherical NiO nanoparticles with particle size as small as 3 nm
are obtained from combustion of nickel nitrate and urea at 300 1C
over 2 h (Fig. 3A). The specific surface area of the synthesized
2
material, determined using N gas adsorption analysis (BET) is
2
3
ꢀ
100 m /g. The measured density of 6.177(8) g/cm (Table 1)
corresponds to about 91% of the theoretical density of bulk NiO
6.8 g/cm ) [34]. The reduced density again suggests that the
3
(
3
synthesized NiO nanoparticles contained some Ni
impurity phase. The estimation of the fraction (x) of Ni
impurity using the equation d¼(1ꢁx) d(NiO)þxd(Ni ), where
d is the measured density of the sample, indicates that NiO
powders synthesized at 300 1C contain approximately 32% Ni
2 3
O (4.84 g/cm )
2 3
O
2 3
O
2 3
O
as an amorphous impurity phase. Increasing the furnace preheat-
ing temperature to 400 1C resulted in well crystalline NiO nano-
particles with particle sizes as small as 7 nm (Fig. 3B). The small
increase in particle size results in a decrease of the specific surface
2
area of the synthesized materials to 52 m /g. Aggregates of NiO
nanoparticles with increasingly larger particle sizes and broader
size range are obtained when the furnace is preheated to 600 and
800 1C. NiO nanoparticles synthesized at 600 1C for 1 h show
particle sizes as small as 24 nm (Fig. 3D) while particles synthe-
sized at 800 1C for 1 h are about 125 nm (or higher) in size
(Fig. 3E). This increase in the particle sizes of the synthesized
NiO materials observed from TEM studies is consistent with the
NiO average particles size estimated from the XRD patterns using
Scherrer analysis. As a result of the large increase in the particle
size of the synthesized materials, the corresponding specific
surface area is drastically reduced (Table 1).
3.3. TEM and thermal conductivity of NiO nanoparticle pellets
As stated in Section 1, independently synthesized NiO
nanoparticles can be used as additional phonon scatters at high
temperatures in p-type thermoelectric composite materials
such as the half-Heusler or silicon germanium phases. To probe
the suitability of the synthesized NiO nanoparticles for such
Fig. 2. XPS spectra of NiO nanoparticles synthesized at 300 1C for 2 h and 600 1C
for 30 min: (a) Ni 2p and (b) O 1s.
Fig. 3. Selected transmission electron microscopy images of nickel oxide nanoparticles synthesized at: (A) 300 1C for 2 h; (B) 400 1C for 1 h; (C) 600 1C for 30 min;
D) 600 1C for 1 h; (E) 800 1C for 30 min; (F) 800 1C for 1 h. Inset in (B) corresponds to high resolution TEM of a single NiO nanoparticle synthesized at 400 1C for 1 h.
(

P. Sahoo et al. / Journal of Solid State Chemistry 190 (2012) 29–35
33
applications, we have investigated the microstructures and the
thermal conductivity of the synthesized NiO nanoparticles after
compaction using spark plasma sintering or uniaxial hot pressing
techniques. The density and the particle size range of the result-
ing pellets are given in Table 2.
Pellets obtained from NiO nanoparticles synthesized at 400 1C
for 1 h also showed large grains after spark plasma sintering. The
grain size is between 535 and 800 nm which represents 70–80
times the size of the as-synthesized NiO nanoparticles. The grains
are also randomly oriented and densely packed. The pellet
obtained from NiO nanoparticles synthesized at 800 1C showed
the smallest grain growth after spark plasma sintering with the
particle size ranging from 565 to 820 nm. The grains are densely
packed with clean boundaries, consistent with the pellet 95% NiO
theoretical density. The pellet fabricated from NiO nanoparticles
synthesized at 400 1C for 1 h via uniaxial hot pressing showed the
largest grain growth (up to 130 times the size of initial nano-
particles) and a broader particle size range (from 413 nm to
1300 nm). This combination of small and large grains results in
more effective packing of the particles (Fig. 4D), which is
consistent with the pellet 98% NiO theoretical density (Table 2).
Despite the drastic grain growth of the synthesized NiO
nanoparticles after processing by spark plasma sintering and
uniaxial hot press, the grain size for the pellets fabricated using
NiO nanoparticles synthesized at 300 1C for 2 h are still in the size
range (115–500 nm) suitable for effective interface phonon scat-
tering when embedded in high temperature thermoelectric
materials such as the silicon germanium alloys. For instance, the
average phonon mean free path in heavily doped Si80Ge20 alloy
is in the range of 200–300 nm at room temperature [37]. This
suggests that NiO inclusions with size range up to 300 nm can
significantly reduce the phonon thermal conductivity of such
alloys. However, phonon scattering at grain boundaries is more
effective when the inclusion phases are considerably smaller than
the phonon mean free path in the host matrix. Therefore,
surfactant-free NiO nanoparticles with size ranges between 10
and 150 nm, which can be synthesized at up to 10 g/h via
combustion of nickel nitrate and urea at temperatures between
Fig. 4 shows selected bright field TEM images of NiO nano-
particles after spark plasma sintering (Fig. 4A–C) and hot pressing
(
Fig. 4D). The fcc (Fmꢁ3m) structure of NiO with lattice para-
˚
meter a¼8.35 A, was confirmed by a selected area electron
diffraction pattern obtained from one of the grains (see inset of
Fig. 4A). After spark plasma sintering, the grain size of NiO
nanoparticles synthesized at 300 1C for 2 h is between 115 and
5
00 nm (Table 2). This is 30–70 times the size of the initial
nanoparticles, indicating drastic grain growth during the spark
plasma sintering process. The grains are highly crystalline, ran-
domly oriented and densely packed with well defined grain
boundaries (Fig. 4A). The observed packing of grains is consistent
3
with the pellet (6.285 g/cm ) 92% NiO theoretical density
3
(
when compared to the 6.8 g/cm theoretical density of NiO).
The increase in the density of the sample after SPS, when
3
compared to the measured powder density of 6.177 g/cm for the
as-synthesized NiO nanoparticles, suggests a significant reduction
of the fraction of Ni
Ni is known to decompose to NiO with loss of O
00 1C [36]. This decomposition reaction presumably occurs
during the spark plasma sintering process at 900 1C resulting in
2
O
3
impurity phase in the resulting pellet.
2
O
3
2
around
6
2 3
NiO pellets containing a smaller fraction of the Ni O phase.
Table 2
Density and grains size (TEM) of NiO pellets fabricated by hot pressing and spark
plasma sintering. *NiO-300-2-SPS corresponds to pellet of NiO-300-2 sample
fabricated by SPS.
4
00 1C and 800 1C, are suitable for use as inclusions in the
Pressing technique
Sample code
Pellet density
Grain size from
TEM (nm)
3
fabrication of silicon germanium nanocomposites.
(
g/cm )
Although, spark plasma sintering and hot pressing of free-
standing NiO nanoparticles result in large grain growth, we
anticipate that by embedding the nanoparticles in high melting
point matrices such as silicon germanium alloys, the matrix–
nanoparticle interactions will significantly suppress the particle
growth during thermal processing. Therefore, Si80Ge20/NiO(nano)
composites with nanometer scale NiO inclusions can be obtained
using spark plasma sintering or hot pressing. Interestingly, the
average electron mean free path in heavily doped Si80Ge20 alloy is
about 5 nm [37], suggesting that by carefully tailoring the size
range of the NiO inclusions, electron flow throughout the Si80Ge20
matrix will not be significantly affected, thus resulting in nano-
composites with high figures of merit (ZT).
SPS
SPS
HP
*NiO-300-2-SPS
NiO-400-1-SPS
NiO-400-1-HP
NiO-800-0.5-SPS
6.285(3)
6.253(4)
6.665(3)
6.443(4)
115–500
535–800
413–1300
565–820
SPS
Fig. 5 shows the temperature dependence of the thermal
conductivity of the synthesized NiO nanoparticles after spark
plasma sintering or uniaxial hot pressing. Temperature depen-
dence of the thermal conductivity of NiO single crystal [38] is
shown for comparison. Regardless of the synthesis and pressing
conditions, the thermal conductivity of all samples decreases with
rising temperature. All thermal conductivity curves showed a
change in slope between 510 and 530 K which is the magnetic
transition (N e´ el temperature) region of bulk NiO. At lower
temperatures (300–500 K), the temperature dependence of the
thermal conductivity for all samples follows the power law
ꢁ
l
k
ꢀT suggesting that the thermal conductivity in this tempera-
ture range is dominated by acoustic phonon scattering. Similar
behavior of the thermal conductivity was observed for NiO single
crystal (Fig. 5). The power exponent (
average grain size within the specimen which in turn is a function
of the sample processing conditions. The observed values of
are 1.47 for NiO-300-2-SPS, 1.73 for NiO-400-1-SPS, and 1.93 for
l) increases with the
Fig. 4. Selected TEM images of pellets NiO nanoparticles fabricated by Spark
l
plasma sintering or uniaxial hot pressing: (A) NiO-300-2-SPS; (B) NiO-400-1-SPS;
(C) NiO-800-0.5-SPS; and (D) NiO-400-1-HP.

3
4
P. Sahoo et al. / Journal of Solid State Chemistry 190 (2012) 29–35
scattering mechanism resulting in even larger decreases in the
thermal conductivity of the NiO nanoparticle pellets compared to
that of the NiO single crystal. At 750 K, a thermal conductivity of
ꢀ
5 W/m K was measured for the NiO-300-2-SPS specimen. By
extrapolation the current thermal conductivity temperature
dependence, a value of ꢀ4 W/m K can be anticipated at 1000 K.
The observed dependence of the thermal conductivity of NiO on
the grain size could be beneficial in the design of bulk nanocom-
posite thermoelectric materials with low thermal conductivity
and high figures of merit. Assuming, small grain growth of the
synthesized NiO nanoparticles when embedded in a bulk matrix,
nanocomposites with significantly lower thermal conductivity
can be designed with carefully chosen matrices.
4. Conclusions
Surfactant-free NiO nanoparticles with various and controllable
size ranges can be produced in a relatively short time via the solution
combustion method using nickel nitrate as an oxidizer and urea as
the fuel. The particle size range and the crystallinity of the final
product can be controlled by adjusting the furnace preheating
temperature and the duration of the reaction. For example, a
minimum of 2 h is required to produce NiO nanoparticles with sizes
as small as 3 nm at a combustion temperature of 300 1C. However,
NiO nanoparticles with size range of 140–150 nm can be obtained
within 30 min when the furnace preheated at 800 1C. The thermal
conductivity of pressed pellets of the synthesized NiO nanoparticles
are reduced by more than 60% compared to that of NiO single crystal.
The synthesized NiO nanoparticles, because of their chemical stability
at high temperature and the strong dependence of the thermal
conductivity on the particle size, are suitable for use as inclusions
in the design of nanocomposite materials for high temperature
energy conversion.
Fig. 5. Temperature dependence of the thermal conductivity of NiO nanoparticles
pellets obtained by spark plasma sintering or uniaxial hot pressing. The thermal
conductivity of NiO single crystal adapted from Ref. [38] is plotted for comparison.
8
00-0.5-SPS samples. A rather steeper dependence (l¼2.11) was
observed for the NiO-400-1-HP specimen. Around the magnetic
transition temperature, all thermal conductivity curves show a
continuous minimum contrasting with the sharp change in slope
observed for the thermal conductivity of NiO single crystal
(
Fig. 5). This phenomenon was described as resulting from critical
scattering of phonons by the spin system at the N e´ el temperature,
in addition to the existing acoustic, boundary and impurity
scattering mechanisms [38,39]. As can be seen from the inset of
Fig. 5, the abruptness of the change in slope of the thermal
conductivity curves decreases and the temperature of minimum
thermal conductivity slightly increases when moving from
NiO-800-0.5-SPS to NiO-300-2-SPS. This suggests that grain sizes
within the pellets play a key role in how phonons are scattered by
the spin system at the N e´ el temperature. The broad range of NiO
grain sizes and the random orientation of the grains within the
pellets, as revealed by the TEM images (Fig. 4), lead to regions of
different N e´ el temperatures at which the critical phonon scatter-
ing by the spin system occurs. The density of such regions
increases with decreasing average grain size within the pellets
resulting in the weakening of the anomaly in the thermal
conductivity curves. At higher temperatures (i.e. above the
magnetic transition temperature), the thermal conductivities of
the synthesized materials continue to decrease at a slower slope,
again contrasting with the thermal conductivity observed in bulk
NiO, which increases with temperature above the N e´ el tempera-
ture (Fig. 5).
Acknowledgments
The authors gratefully acknowledge the financial support from
DARPA (contract # HR 0011-08-1-0084). This work made use of
the laser flash diffusivity apparatus (LFA457) purchased with funds
from the Louisiana Board of Regents (Grant # LEQSF(2008-09)–
ENH-TR-58).
References
[1] L.D. Hicks, M.S. Dresselhaus, Phys. Rev. B 47 (1993) 16631–16634.
[
[
2] W.Z. Wang, B. Poudel, D.Z. Wang, Z.F. Ren, Adv. Mater. 17 (2005) 2110.
3] L. Bertini, C. Stiewe, M. Toprak, S. Williams, D. Platzek, A. Mrotzek, Y. Zhang,
C. Gatti, E. Muller, M. Muhammed, M. Rowe, J. Appl. Phys. 93 (2003) 438–447.
[4] B. Poudel, Q. Hao, Y. Ma, Y.C. Lan, A. Minnich, B. Yu, X. Yan, D.Z. Wang,
A. Muto, D. Vashaee, X.Y. Chen, J.M. Liu, M.S. Dresselhaus, G. Chen, Z. Ren,
Science 320 (2008) 634–638.
[5] D.M. Rowe, V.S. Shukla, N. Savvides, Nature 290 (1981) 765–766.
[6] A. Gueguen, P.F.P. Poudeu, C.P. Li, S. Moses, C. Uher, J.Q. He, V. Dravid,
K.A. Paraskevopoulos, M.G. Kanatzidis, Chem. Mater. 21 (2009) 1683–1694.
[7] K.F. Hsu, S. Loo, F. Guo, W. Chen, J.S. Dyck, C. Uher, T. Hogan,
E.K. Polychroniadis, M.G. Kanatzidis, Science 303 (2004) 818–821.
The most interesting finding in the study is the strong
suppression of the thermal conductivity of NiO when grain size
is on the nanometer scale. The overall decrease in the thermal
conductivity of NiO nanoparticle pellets compared to that of the
NiO single crystal is higher than 60%. At room temperature, the
thermal conductivity dropped from ꢀ50 W/m K for bulk NiO
to ꢀ20 W/m K for the NiO-800-0.5-SPS specimen. This sharp
decrease in the thermal conductivity likely arises from the
increased phonon scattering at the multiple grain boundaries
within the pellets. At high temperatures, phonon scattering at
grain boundaries is added to the existing acoustic phonon
[8] P.F.P. Poudeu, J. D’Angelo, H.J. Kong, A. Downey, J.L. Short, R. Pcionek,
T.P. Hogan, C. Uher, M.G. Kanatzidis, J. Am. Chem. Soc. 128 (2006)
14347–14355.
[9] P. Maji, N.J. Takas, D.K. Misra, H. Gabrisch, K. Stokes, P.F.P. Poudeu, J. Solid
State Chem. 183 (2010) 1120–1126.
[10] P.F.P. Poudeu, J. D’Angelo, A.D. Downey, J.L. Short, T.P. Hogan, M.G. Kanatzidis,
Angew. Chem.-Int. Ed. 45 (2006) 3835–3839.
[
11] J.R. Sootsman, D.Y. Chung, M.G. Kanatzidis, Angew. Chem.-Int. Ed. 48 (2009)
616–8639.
[12] L.D. Chen, X.Y. Huang, M. Zhou, X. Shi, W.B. Zhang, J. Appl. Phys. 99 (2006)
64305.
8
0
[
[
13] X.Y. Huang, Z. Xu, L.D. Chen, Solid State Commun. 130 (2004) 181–185.
14] Y. Wang, Y.F. Zhang, H.R. Liu, S.J. Yu, Q.Z. Qin, Electrochim. Acta 48 (2003)
4253–4259.

P. Sahoo et al. / Journal of Solid State Chemistry 190 (2012) 29–35
35
[
[
15] C.R. Magana, D.R. Acosta, A.I. Martinez, J.M. Ortega, Sol. Energy 80 (2006)
61–169.
[26] K.V. Rao, C.S. Sunandana, J. Nanosci. Nanotechnol. 8 (2008) 4247–4253.
[27] K.G. Chandrappa, T.V. Venkatesha, K.O. Nayana, M.K. Punithkumar, Mater.
Corros. (2011). doi:10.1002/maco.201005966.
[28] J.R.A. Sietsma, J.D. Meeldijk, J.P. den Breejen, M. Versluijs-Helder, A.J. van Dillen,
P.E. de Jongh, K.P. de Jong, Angew. Chem.-Int. Ed. 46 (2007) 4547–4549.
[29] A.S. Mukasyan, P. Epstein, P. Dinka, Proc. Combustion Institute 31 (2007)
1789–1795.
1
16] I. Hotovy, J. Huran, L. Spiess, H. Romanus, S. Capone, V. Rehacek, A.M. Taurino,
D. Donoval, P. Siciliano, J. Phys.: Conf. Ser. 61 (2007) 435–439.
17] R.H. Kodama, A.E. Berkowitz, Phys. Rev. B 59 (1999) 6321–6336.
18] S.R. Culp, J.W. Simonson, S.J. Poon, V. Ponnambalam, J. Edwards, T.M. Tritt,
Appl. Phys. Lett. 93 (2008) 022105.
[
[
[
[
[
[
19] G. Joshi, H. Lee, Y.C. Lan, X.W. Wang, G.H. Zhu, D.Z. Wang, R.W. Gould, D.C. Cuff,
M.Y. Tang, M.S. Dresselhaus, G. Chen, Z.F. Ren, Nano Lett. 8 (2008) 4670.
20] C.B. Vining, W. Laskow, J.O. Hanson, R.R.V. der Beck, P.D. Gorsuch, J. Appl.
Phys. 69 (1991) 4333.
21] A. Dessombz, D. Chiche, P. Davidson, P. Panine, C. Chaneac, J.P. Jolivet, J. Am.
Chem. Soc. 129 (2007) 5904–5909.
22] J.P. Jolivet, C. Froidefond, A. Pottier, C. Chaneac, S. Cassaignon, E. Tronc,
P. Euzen, J. Mater. Chem. 14 (2004) 3281–3288.
23] E. Matijevic, Annu. Rev. Mater. Sci. 15 (1985) 483–516.
24] E. Matijevic, Chem. Mater. 5 (1993) 412–426.
[30] J.J. Kingsley, K.C. Patil, Mater. Lett. 6 (1988) 427–432.
[31] J.E. Keem, J.M. Honig, Center for Information and Numerical Data (CINDAS)
Report 52 (1978), Purdue University.
[32] M. Prasad, M.G. Tendulkar, J. Chem. Soc. 1 (1931) 1403.
[33] G.A. Niklasson, C.G. Granqvist, J. Mater. Chem. 17 (2007) 127–156.
[34] S. Sasaki, K. Fujino, Y. Takeuchi, Proc. Jpn Acad. 55 (1979) 43–48.
[35] R.W. Cairns, E. Ott, J. Am. Chem. Soc. 55 (1933) 527–533.
[36] N.N. Greenwood, A. Earnshaw, in Chemistry of the Elements, 2nd edition,
Butterworth, UK, 1997.
[
[
[
[37] D.M. Rowe, L.W. Fu, S.G.K. Williams, J. Appl. Phys. 73 (1993) 4683–4685.
[38] F.B. Lewis, N.H. Saunders, J. Phys. C: Solid State Phys. 6 (1973) 2525–2532.
[39] H. Stern, J. Phys. Chem. Solids 26 (1965) 153–161.
25] D. Portehault, S. Cassaignon, E. Baudrin, J.P. Jolivet, Chem. Mater. 19 (2007)
5
410–5417.