Materials Sciences and Applications, 2011, 2, 1416-1420
doi:10.4236/msa.2011.210191 Published Online October 2011 (http://www.SciRP.org/journal/msa)
Influence of Sintering Temperature on
Densification, Structure and Microstructure of Li
and Sb Co-Modified (K,Na)NbO3-Based Ceramics
Rashmi Rani*, Seema Sharma
Ferroelectrics Research Laboratory, Department of Physics, A N College, Patna, India.
Received July 21st, 2011; revised September 20th, 2011; accepted September 29th, 2011.
ABSTRACT
Polycrystalline samples of Lead free (K0.5Na0.5)1−x(Li)x(Sb)x(Nb)1−xO3 ceramics with nominal compositions (x = 0.040 to
0.060) have been prepared by high temperature solid state reaction technique. X-ray diffraction (XRD) pattern shows
that the crystal structure transforms from orthorhombic to tetragonal as Li and Sb content increases. Normal sintering
process yield compounds with density ~98.2% of the theoretical value. Densification of the Li and Sb co-doped KNN
ceramics might be explained by the liquid-phase sintering. This may be attributed to the low melting temperature of Li
compounds that appears to promote the formation of a liquid phase during sintering.
Keywords: KNN, Perovskite, XRD, SEM
1. Introduction
The speedy development of piezoelectric devices ur-
gently calls for environment-friendly materials substitut-
ing for the widely used lead zirconate titanate (Pb Zr,
Ti)O3, (PZT)). Among various lead-free piezoelectric
materials, ceramics based on potassium sodium niobate
(KNbO3-NaNbO3) are most promising [1-5]. Highly
dense (Na0.5K0.5) NbO3 (KNN) ceramics was fabricated
using hot pressing which possessed high piezoelectric
constants [2]. It was even reported that KNN based ce-
ramics were sensitive to moisture and only hot-pressed
samples could be sufficiently densified [1,2]. Recently
developed spark plasma sintering is another sintering
method to produce dense bulk samples of KNN ceramics
with a relative density higher than 99% [6,7]. Although
pressure-assisted sintering processes are effective in
consolidating KNN ceramics by lowering the sintering
temperature and hence suppressing the volatilization of
alkali components, it is obvious that normal or pressure-
less sintering of these materials is more suitable for mass
production. Most recent studies have concentrated on the
development of KNN-based lead-free ceramics with en-
hanced piezoelectric properties through doping or texture
control [8-10].
On the basis of the reported work and the analogy of
PZT-based ceramics, it is noted that the key approach for
improving the piezoelectric properties of KNN-based
ceramics is to lower the ferroelectric tetragonal-ferro-
electric orthorhombic phase transition (TO-F), forming
coexistence of the tetragonal and orthorhombic phases at
room temperature. This could be achieved by partial sub-
stitutions of A-site ions (Na0.5K0.5)+ and B-site ion (Nb)5+
by other analogous ions in the ABO3-type (KNN) pe-
rovskite structure. In the present work, KNN ceramics
partially substituted with Li+ and Sb+ were prepared by
ordinary solid-state sintering, and their structure and mi-
crostructure properties were studied.
2. Experimental Section
2.1. Experimental Procedure
(K0.5Na0.5)1−x(Li)x(Sb)x(Nb)1−xO3 ceramics with nominal
compositions (x = 0.040, 0.045, 0.050, 0.055 and 0.060)
were prepared by conventional ceramic technique using
analytical grades metal oxides or carbonate powders:
K2CO3 (99%), Na2CO 3 (99.8%), Li2CO3 (99.9%), Sb2O3
(99.9%), and Nb2O5 (99.5%).These powders were used
as starting raw materials. For each composition, the
starting materials were weighed according to the
stoichiometric formula and ball-milled in acetone using
zirconia balls for 24 h. After drying, the calcination was

Influence of Sintering Temperature on Densification, Structure and Microstructure of Li and Sb 1417
Co-Modified (K,Na)NbO3-Based Ceramics
carried out at 900˚C, 890˚C and 870˚C for 4 h. The cal-
cined powders were then pressed into 10mm diameter
disks at 300 MPa. The disks samples were finally sintered
between 1060˚C - 1120˚C for 4 h - 2 h in air. Silver elec-
trodes were applied on the top and bottom surfaces of the
samples for the measurements.
2.2. Sample Characterization
The formation and quality of compounds were verified
with x-ray diffraction (XRD) technique. The XRD pat-
terns of the compounds were recorded at room tempera-
ture using x-ray powder diffractometer (Rigaku Minifiex,
Japan) with CuKα radiation (λ = 1.5405 Å) in a wide
range of Bragg angles 2θ (20˚ ≤ 2θ ≤ 60˚) at a scanning
rate of 2˚ min−1. The microstructures were observed us-
ing a scanning electron microscope (SEM). These mi-
crographs were obtained with a JEM-2000FX (JEOL
Ltd.) scanning electron microscope operated at 20 KeV.
Density of samples was determined using the Archime-
des’ method. The differential thermal analysis (DSC-TG),
Setaram Labsys, Setaram Instrumentation, Caluire, Fran-
ce) of ceramics powders was carried out in air from room
temperature to 800˚C with different heating rates of 2˚
min−1.
3. Results and Discussion
3.1. XRD Analysis
Figure 1 shows the room temperature XRD patterns of
the (1 − x)K0.5Na0.5NbO3–xLiSbO3 ceramics with 0.04 < x
< 0.06. It can be seen that a perovskite structure with a
small amount of secondary phase K3Li2Nb5O15 with the
Figure 1. Room temperature XRD pattern of (1 − x)K0.5
Na0.5NbO3–xLiSbO3 ceramics.
tetragonal tungsten bronze structure is formed. The crys-
tal structure changes from orthorhombic to tetragonal
with increase in x. As the crystal structure of KNN
(perovskite structure) is very different from LiSbO3 (il-
menite structure), our results suggest that Li+ and Sb5+
have diffused into the KNN lattices, with Li+ entering the
(Na0.5K0.5)+ sites and Sb5+ occupying the Nb5+ sites, to
form a homogeneous solid solution. Figure 2 shows the
enlarged XRD patterns of the ceramics in the ranges of
2θ from 44˚ to 48˚, respectively. It can be seen that at x <
0.05, the ceramic has an orthorhombic perovskite struc-
ture (the corresponding XRD patterns can be indexed by
Ref [11]).
As x (i.e., the concentration of LiSbO3) increases, a
tetragonal phase appears and increases continuously. At x
> 0.05, the ceramic begin to possess a tetragonal phase
(the XRD patterns can be indexed by Ref. [12]).
These suggest that the (Perovskite) orthorhombic and
tetragonal phases coexist in the (1 − x)K0.5Na0.5NbO3–x
LiSbO3 ceramics with 0.045 < x < 0.055. Similar to Ba-
TiO3, the orthorhombic phase of the ceramics demon-
strates a nonprimitive cell, while the tetragonal phase is
primitive. It is also noted that the diffraction peaks shift
slightly towards high diffraction angles as x increases.
This may be attributed to the smaller ionic radii of Li+
and Sb5+ than those of (Na0.5K0.5)+ and Nb5+.
In the orthorhombic region, the lattice constants c and
a have very close values, which can explain why there
were only two peak splitting of (2 0 0) reflections at a 2θ
of 45.5˚ for an orthorhombic structure, rather than three
peak splitting. In the tetragonal phase, the c/a ratio for
the ceramics with x = 0.055 and 0.06 were 1.0070 and
1.0074, respectively, which indicated that the tetragonal-
Figure 2. XRD patterns of the ceramics in the ranges of 2θ
from 44˚ to 48˚ of (1 − x)K0.5Na0.5NbO3–xLiSbO3 ceramics.

Influence of Sintering Temperature on Densification, Structure and Microstructure of Li and Sb
Co-Modified (K,Na)NbO3-Based Ceramics
1418
ity of the ceramics increased with further increasing x. In
addition, it was seen that the cell volumes of the ceramics
decreased gradually with the increasing x, which can
explain why the diffraction peaks shifted slightly towards
high diffraction angles as x increased.
3.2. SEM Analysis
The microstructure of the KNN ceramics containing x
mol% of Li and Sb (0.40 < x < 0.060) sintered at 1060˚C
for 2 h was investigated using SEM. Note that instead of
the fractured cross sectional SEM images, the surface
morphology images have been used to elucidate the mi-
crostructure features of KNN ceramics, simply because
they were confirmed to be very similar but more distinct
than the fractured cross-sectional SEM images.
The KNN ceramics sintered at 1060˚C had a porous
microstructure with small grains, as shown in Figure 3(a)
for x = 0.050 composition. However, when the sintering
temperature increased to 1080˚C a dense, uniform micro-
structure with enlarged grains was developed (Figure
3(b)). These enlarged grains were angular with a flat
interface and the specimen showed abnormally grown
grains, as indicated by the arrow in Figure 3(b). An an-
gular grain with a flat surface is typically observed in the
abnormal grain growth in the presence of the liquid phase.
Therefore, it was considered that densification of the Li
and Sb co-doped KNN ceramics might be explained by
the liquid-phase sintering. This may be attributed to the
low melting temperature of Li compounds that appears to
promote the formation of a liquid phase during sintering.
However, it was very difficult to find liquid phase, im-
plying that the liquid phase formed during the sintering
could be a transient liquid phase, with a high solubility in
the KNN ceramics that led to its eventual disappearance
with sintering time. Figures 3(c) and 3(d) shows the
KNN ceramics sintered at 1100˚C and 1120˚C; enlarged
grains can be observed showing an abnormal grain
growth (AGG). Their relative density was found to be
much less as compared to the ceramics sintered at
1080˚C. It is generally agreed that AGG is caused by the
existence of a liquid phase. Chun et al. [13] found that
extensive AGG occurred on the surface of a sintered Ba-
TiO3 specimen and concluded that the BaO evaporation
from the surface and consequent formation of eutectic
(a) (b)
(c) (d)
Figure 3. SEM micrographs of fractured surface of (1 − x)K0.5Na0.5NbO3–xLiSbO3 pellets (x = 0.045) sintered at (a) 1060˚C; (b)
080˚C, (c) 1100˚C and (d) 1120˚C. 1

Influence of Sintering Temperature on Densification, Structure and Microstructure of Li and Sb 1419
Co-Modified (K,Na)NbO3-Based Ceramics
liquid is the cause of AGG. Thus the ceramics can be
well sintered at ~1080˚C in order to obtain a dense mi-
crostructure, which is about 200˚C lower than the sinter-
ing temperature for PZT ceramics.
Although all alkali components may be volatilized
during the sintering, their volatilization rates must be
different at the same temperature. As a result, the final
actual compositions may become different from the
starting one. By referring to the report [14] for KNN ce-
ramics with different K/Na ratios, and comparing the
standard XRD patterns of KNbO3 (#32-0822) and
NaNbO3 (#33-1270), it is likely that the Na content de-
creased with increasing temperature. Considering that the
mass transportation for the volatilization from the interior
of the sample may occur through the grain surface or
grain boundary, it is reasonable to think that the compo-
sitions at some grain boundaries change toward a high
K/Na ratio. According to the phase diagram of the
KNbO3-NaNbO3 system, the solidus line temperature
decreases with increasing K/Na ratio, so that the liquid
phase is easy to be formed when volatilization occurs.
3.3. Density Analysis
Figure 4 shows the density of the ceramics as a function
of x. After the addition of LiSbO3, the density increases
and has a high value of 4.399 g/cm3 which is about
98.2% of the theoretical value for x = 0.045 at a sintering
temperature of 1080˚C.
Then, the bulk density gradually decreased with in-
creasing x. The decrease in density at higher x may be
attributed to the formation of the secondary phase
K3Li2Nb5O15 of which the theoretical density is low. The
densities of other ceramics were in the range of 4.023 -
4.38 g/cm3. These results indicated that the optimum Li
and Sb addition can promote sintering and thus improve
the density of ceramics. However, excess Li addition can
also result in the decrease of the bulk density.
3.4. Thermal Analysis
The solid state synthesis of the KNN samples (x = 0.045)
from alkaline carbonates and niobium oxide was fol-
lowed by thermal analysis (Figure 5).
The sample loses 12 wt% upon heating to 700˚C. A
weight loss of about 1% upon heating up to 200˚C ac-
companied by an endothermic peak is a result of H2O
removal. The presence of water content in the carbon-
ate-oxide powder mixture is due to the hygroscopic na-
ture of both carbonates, particularly K2CO3, which ab-
sorbs easily a few wt% of water in normal atmosphere
where the manipulation of the precursor powder takes
place. Between 380˚C and 700˚C there is a weight loss of
11% due to decomposition of carbonates. It occurs in two
0.040 0.045 0.050 0.055 0.060
4.00
4.05
4.10
4.15
4.20
4.25
4.30
4.35
4.40
4.45
4.50
x
den[g/cm3]
1060 C
1080 C
1100 C
1120 C
Figure 4. Density of (1 − x)K0.5Na0.5NbO3–xLiSbO3 ceramics
as a function of x.
0200 400 600 8001000
86
88
90
92
94
96
98
100
0200 400 600 8001000
-0.25
-0.20
-0.15
-0.10
-0.05
0.00
DSC (mW/mg)
Figure 5. DSC and TGA curves of (1 − x)K0.5Na0.5NbO3–x
LiSbO3 with x = 0.045.
TG (wt%)
tem p C
steps, 5% between 380˚C to 540˚C and 5% between
380˚C to 700˚C. As both alkaline carbonates are stable at
700˚C [15,16], the lowering of the decomposition tem-
perature of the carbonates is associated with the synthesis
of the alkaline niobate.
4. Conclusions
Li and Sb modified K0.5Na0.5NbO3 ceramics with single
perovskite phase have been synthesized by normal sin-
tering at 1060˚C - 1120˚C. The effects of the sintering
temperature on structure and microstructure of the ce-
ramic compounds are investigated. Using conventional
solid-state processing techniques perovskite phase was
achieved for all the compositions. The crystal structure
changes from orthorhombic to tetragonal with increase in
Li and Sb content. At a sintering temperature of 1080˚C,

Influence of Sintering Temperature on Densification, Structure and Microstructure of Li and Sb 1420
Co-Modified (K,Na)NbO3-Based Ceramics
a dense, uniform microstructure with enlarged grains was
developed for all compositions of KNN compounds. Af-
ter the addition of LiSbO3, the density increases and was
found to be maximum (98.2% of the theoretical value)
for x = 0.045. Densification of the Li and Sb co-doped
KNN ceramics might be attributed to the liquid-phase
sintering present during the sintering process.
5. Acknowledgements
Authors wish to acknowledge Department of Science and
Technology, Govt of India for the financial support under
a research project scheme.
REFERENCES
[1] G. H. Haerting, “Properties of Hot-Pressed Ferroelectric
Alkali Niobate Ceramics,” Journal of the American Ce-
ramic Society, Vol. 50, No. 6, 1967, pp. 229-230.
[2] R. E. Jaeger and L. Egerton, “Hot Pressing of Potassium-
Sodium Niobates,” Journal of the American Ceramic So-
ciety, Vol. 45, No. 5, 1962, pp. 209-213.
doi:10.1111/j.1151-2916.1962.tb11127.x
[3] L. Egerton and D. M. Dillon, “Piezoelectric and Dielec-
tric Properties of Ceramics in the System Potassium-So-
dium Niobate,” Journal of the American Ceramic Society,
Vol. 42, No. 9, 1959, pp. 438-442.
doi:10.1111/j.1151-2916.1959.tb12971.x
[4] M. Kosec and D. Kolar, “On Activated Sintering and
Electrical Properties of NaKNbO3,” Materials Research
Bulletin, Vol. 10, No. 5, 1975, pp. 335-340.
doi:10.1016/0025-5408(75)90002-1
[5] S. Narayana Murty, K. V. Ramana Murty, K. Umakan-
tham and A. Bhanumathi, “Modified (NaK)NbO3 Ceram-
ics for Transducer Applications,” Ferroelectrics, Vol. 102,
No. 1, 1990, pp. 243-247.
doi:10.1080/00150199008221484
[6] J.-F. Li, K. Wang, B. P. Zhang and L. M. Zhang, “Ferro-
electric and Piezoelectric Properties of Fine-Grained
Na0.5K0.5NbO3 Lead-Free Piezoelectric Ceramics Prepar-
ed by Spark Plasma Sintering,” Journal of the American
Ceramic Society, Vol. 89, No. 2, 2006, pp. 706-709.
doi:10.1111/j.1551-2916.2005.00743.x
[7] R. P. Wang, R. J. Xie, T. Sekiya and Y. Shimojo, “Fabri-
cation and Characterization of Potassium-Sodium Niobate
Piezoelectric Ceramics by Spark-Plasma-Sintering Me-
thod,” Materials Research Bulletin, Vol. 39, No. 11, 2004,
pp 1709-1715. doi:10.1016/j.materresbull.2004.05.007
[8] S. Tashiro, H. Nagamatsu and K. Nagata, “Sinterability
and Piezoelectric Properties of KNbO3 Ceramics after
Substituting Pb and Na for K,” Japanese Journal of Ap-
plied Physics, Vol. 4, No. 11B, 2002, pp. 7113-7118.
doi:10.1143/JJAP.41.7113
[9] E. Hollenstein, M. Davis, D. Damjanovic and N. Setter,
“Piezoelectric Properties of Li- and Ta-Modified (Na0.5
K0.5) NbO3 Ceramics,” Applied Physics Letters, Vol. 87,
No. 18, 2005, pp. 182905-182907.
doi:10.1063/1.2123387
[10] M. Matsubara, T. Yamaguchi, K. Kikuta and S. Hirano,
“Effect of Li Substitution on the Piezoelectric Properties
of Potassium Sodium Niobate Ceramics,” Japanese Jour-
nal of Applied Physics, Vol. 44, No. 8, 2005, pp. 6136-
6142. doi:10.1143/JJAP.44.6136
[11] International Centre for Diffraction Data, JCPDS-ICDD
2001, File No. 71-2171.
[12] International Centre for Diffraction Data, JCPDS-ICDD
2001, File No. 71-0945.
[13] J. S. Chun, N. M. Hwang, D. Y. Kim and J. K. Park,
“Abnormal Grain Growth Occurring at the Surface of a
Sintered BaTiO3 Specimen,” Journal of the American
Ceramic Society, Vol. 87, No. 9, 2004, pp. 1779-1781.
doi:10.1111/j.1551-2916.2004.01779.x
[14] B.-P. Zhang, J. F. Li and K. Wang, “Compositional De-
pendence of Piezoelectric Properties in NaxK1–xNbO3
Lead-Free ceramics Prepared by Spark Plasma Sintering,”
Journal of the American Ceramic Society, Vol. 89, No. 5,
2006, pp. 1605-1609.
doi:10.1111/j.1551-2916.2006.00960.x
[15] Gmelins Handbuch der Anorganischen Chemie, Verlag
Chemie, Weinheim, 1969, Ch. 22, p. 834.
[16] Gmelins Handbuch der Anorganischen Chemie, Verlag
Chemie, Weinheim, 1969, Ch. 21, p. 1317.