Microscopy
Microanalysis
Microsc. Microanal. 11, 572–580, 2005
DOI: 10.1017/S1431927605050683
AND
© MICROSCOPY SOCIETY OF AMERICA 2005
Electron Microscopy Studies of Potassium Sodium
Niobate Ceramics
Darja Jenko,* Andreja Benčan, Barbara Malič, Janez Holc, and Marija Kosec
“Jožef Stefan” Institute, Electronic Ceramics Department, Jamova 39, SI-1000 Ljubljana, Slovenia
Abstract: Using electron microscopy, K0.5Na0.5NbO3 ~KNN! ceramics sintered at 10308C for 8 h and 11008C for
2 and 24 h was studied. The scanning electron microscopy and X-ray spectrometry revealed that the materials
consisted of a matrix phase in which the ~Na⫹K!/Nb ratio corresponded closely to the nominal composition
and a small amount of Nb-rich secondary phase. A bimodal microstructure of cube-shaped grains was revealed
in the fracture and thermally-etched surfaces of the KNN. In the ceramics sintered at 11008C, the larger grains
~up to 30 mm across!, contained angular trapped pores. The transmission electron microscopy analysis revealed
that the crystal planes of the grains bordering the intragranular pore faces were of the $100% family with respect
to the simple perovskite cell. Ferroelectric domains were observed in the grains of this material.
Key words: potassium sodium niobate, synthesis, sintering, scanning electron microscopy, transmission electron microscopy, cube-shaped grains, angular pores, ferroelectric domains
I NTR ODUCTION
Piezoelectric materials based on the Pb~Zr,Ti!O3 solid solution have been widely used because of their piezoelectric,
pyroelectric, and ferroelectric properties. One of the major
drawbacks of these materials, however, is their high lead
content: the ceramics contain ;60 wt% lead and therefore represent a possible ecological hazard. As a result,
most of the current research is oriented toward more
environmentally-friendly, lead-free materials. A group of
lead-free ferroelectric materials is one based on potassium
sodium niobate @~K,Na!NbO3 #. The solid solution with the
composition K/Na 50/50 close to the morphotropic phase
boundary exhibits a moderate dielectric constant e ⫽ 300–
800 and an optimum piezoelectric response d33 ⫽ 80–100
~Jaeger & Egerton, 1962; Kosec & Kolar, 1975; Jenko et al.,
2003; Malic et al., 2003!. The K0.5Na0.5NbO3 ~KNN! ceramics are suitable for applications in ultrasonic diagnostics
because of their low density of ;4.51 g/cm 3 ~Jaeger &
Egerton, 1962!, high velocity of sound at ;6000 m/s, and
high electromechanical coupling coefficient k T of ;0.4
~Tran-Huu-Hue et al., 2003!.
One of the major problems of KNN is sintering. According to early reports, the stoichiometric material is extremely
difficult to consolidate ~Jaffe et al., 1971; Kosec & Kolar,
1975!. The fabrication of these materials should be carried
out carefully, taking into account the hygroscopic nature of
Received June 15, 2004; accepted June 27, 2005.
*Corresponding author. E-mail: [email protected]
alkaline carbonates, especially that of potassium carbonate,
and the volatility of the alkaline species during thermal
treatment/annealing ~Jaffe et al., 1971; Kosec & Kolar, 1975;
Flückiger et al., 1977!.
In a previous study, we observed that the KNN ceramics densify in a narrow temperature interval: intensive shrinkage is a few 1008C below the melting point at 11408C ~Kosec
& Kolar, 1975; Malic et al., 2003!. The aim of this work was
to analyze, using scanning electron microscopy ~SEM!, the
microstructure and specifically the stoichiometry of KNN
ceramics sintered at 10308C and 11008C, and to study the
microstructural details of KNN ceramics using transmission electron microscopy ~TEM!. As far as we know, there
are no data regarding other studies of KNN available in
open literature.
M ATERIALS
AND
M ETHODS
The solid solution with the nominal composition KNN was
prepared by solid-state synthesis from alkaline carbonates
and niobium oxide. The synthesis details are described
elsewhere ~Jenko et al., 2003!. The particle size distribution
of the KNN powder was determined by laser granulometry
using an Alcatel Cilas 850.
The density was calculated from the mass and the
dimensions of the ceramic pellets. The phase composition
was determined by X-ray powder diffraction ~XRD! with a
Philips PW 1710 using CuKa radiation ~2u: 20–708, step:
0.028!, and Si as the internal standard. The cell parameters
EM Studies of KNN Ceramics
were refined by the least-squares method. Ceramic samples
were characterized by SEM in secondary-electron and backscattered electron image ~BEI! modes with a JEOL JSM5800. These were equipped with an Oxford-Link ISIS 300
energy-dispersive X-ray spectrometer and TEM with a JEOL
JEM-2010F.
The fracture surfaces of the samples were sputtered with
gold and analyzed in the SEM. Samples for microanalysis
were polished with 3-mm and 0.25-mm diamond paste using
standard metallographic preparation techniques, and some
of them were thermally etched and coated with carbon to
ensure electrical conductivity. Standardless quantitative X-ray
spectrometry ~EDS! analysis was performed using the SEMQuant program within the Oxford-Link ISIS 300 system,
with the virtual standard package ~VSP! data library and the
ZAF ~Z—atomic number, A—X-ray absorption, F—X-ray
fluourescence! matrix-correction method. The samples were
analyzed using an acceleration voltage of 20 kV, a spectrum
acquisition time of 100 s, a 358 take-off angle, and a 08 tilt of
the specimen. The Na-Ka, K-Ka, and Nb-La spectral lines
were used for the analysis. All of the composition calculations were determined in weight percent by the software and
converted to atomic percent. Single crystals of KNbO 3 and
NaNbO3 , which are also end-members of the solid solution
~K,Na!NbO 3 , were used as reference materials for the
sintered-ceramic samples of KNN to improve the accuracy
of the quantitative EDS analysis, especially when analyzing
sodium ~Samardžija et al., 2004!. Glasses containing Na
and/or K are often used as standards, but they can be unstable under the electron beam. Natural minerals with Na and/or
K ~albite, orthoclase! are available, but they are not reliable
enough because their compositions can vary from one specimen to another, and within a single specimen. For this reason, and because of their perovskite structure and similar
composition, single crystals of KNbO 3 and NaNbO3 appear
to be the best choice as reference materials for a quantitative
EDS analysis of potassium sodium niobates.
The quantitative characterization of the thermallyetched surfaces was carried out using the computerized
image-analysis UTHSCSA ImageTool program. Using this
program, the average grain size was determined by measuring the surface of each grain and transforming its irregularlyshaped area into a circle of equivalent diameter.
Samples for TEM analysis were prepared by mechanical
thinning, dimpling, and ion milling using 3.8-keV argon
ions from both sides and a milling angle of 108. The total
time of ion milling was 4 h.
R ESULTS
The particle-size distribution of the KNN powder is shown
in Figure 1. It has a median value of 0.65 mm, with the
largest particles up to 10 mm. Powder compacts were uniaxially pressed with 100 MPa and subsequently sintered at
10308C for 8 h or 11008C for 2 h and 24 h in air.
573
Figure 1. The particle size distribution of the KNN powder. Median particle size was 0.65 mm, diameter for 10% was 0.33 mm and
for 90%, 3.26 mm.
Table 1. The Density of KNN Ceramics Sintered at
1030 and 11008C
T sinter ~8C!/t ~h!
1030/8
1100/2
1100/24
r ~gcm⫺3 !
TD* ~%!
3.92
4.28
4.29
86.9
94.9
95.1
*Theoretical density of KNN is 4.51 gcm⫺3 ~Jaeger & Egerton,
1962!.
Densities of the KNN ceramics sintered at 10308C and
11008C were 86.9% and about 95% of theoretical value,
respectively, as shown in Table 1 ~Jenko et al., 2003!. Fracture surfaces of these samples are shown in Figure 2. The
sample sintered at 10308C consisted of fine, submicron
grains and a population of large grains ;5 mm across with
a distinct cubic morphology. After sintering for 2 h at
11008C, we observed an increase in size for both populations of grains. After 24 h at 11008C, the largest grains were
nearly 30 mm in size. For these grains, predominantly
transgranular fracture was observed, whereas for the fine
grains, intergranular fracture was predominant. On the
fracture surfaces of the large grains, we observed angular
trapped pores with sizes of up to 1 mm ~inset of Fig. 2!.
Densification and intensive grain growth was observed as
the temperature was increased from 10308C to 11008C.
The SEM micrographs of the polished surfaces of KNN
ceramics sintered at 10308C and 11008C are shown in
Figure 3. We encountered a problem of pull-outs during
grinding and polishing that we ascribed to the cubic morphology of the grains and to a rather large-grain size in the
case of the samples sintered at 11008C. In all samples, we
observed light gray inclusions of a secondary phase marked
574
Darja Jenko et al.
Figure 2. The SEM micrographs of the fracture surfaces of KNN ceramics sintered at 10308C and 11008C.
with an arrow on the micrographs in Figure 3. The amount
and the size of these inclusions increased with increasing
sintering temperature and time. The matrix and secondaryphase inclusions in the samples sintered at 10308C and
11008C were further analyzed by EDS.
To improve the accuracy of the quantitative EDS analysis of the ceramics, single crystals of KNbO 3 and NaNbO3
were used as reference materials. The EDS analysis results
on both single crystals in atomic percent were determined
at different randomly-selected locations on the samples and
were averaged with standard deviation included ~Table 2!.
The atomic percent of Na, K, and Nb were normalized
assuming the ABO3 stoichiometry in all cases. The measured value for sodium was 30% lower than the nominal
value. The values for potassium and niobium did not deviate much from the nominal composition.
The EDS analysis on both single crystals helped us to
calculate and determine the composition of the KNN sintered ceramics. Some of the locations where the EDS data
for the matrix ~嘸! and the secondary phase ~⫻! were taken
are shown in Figure 3. The results of the analyses are
collected in Table 3. The atomic percent of Na, K, and Nb
was normalized assuming the ABO3 stoichiometry in all
cases. The atomic percent of Na in the matrix was slightly
lower than the nominal value at 9.8 6 0.5 and 9.8 6 0.1 for
samples sintered at 10308C for 8 h and 11008C for 2 h,
respectively. After sintering at 11008C for 24 h, the Na
content was slightly higher, that is, 10.4 6 0.3 at.%. The
amount of Na in the secondary phase in the sample sintered
at 11008C for 24 h was 3.2 6 0.1 at.%, the amount of K was
10.7 at.%, while the amount of Nb was 23.0 at.%.
The XRD patterns of the sintered samples revealed only
the KNN perovskite phase ~Fig. 4!. This phase was indexed
based on a simple orthorhombic perovskite cell as proposed
by Stannek ~1970!. However, for the composition KNN, a
JCPDS-ICDD powder diffraction card did not exist. The
Table 2. Elemental Composition of the KNbO3 and NaNbO3
Single Crystals Determined by SEM-EDS Analysis*
Atomic percent
Nominal composition
NaNbO3
KNbO3
Na or K
Nb
20.0
15.5 6 0.1
~0.7%!
21.8 6 0.1
~0.5%!
20.0
21.9 6 0.1
~0.5%!
21.6 6 0.1
~0.5%!
Na/Nb or
K/Nb
1.00
0.71
1.01
*The atomic percent of Na, K, and Nb were normalized assuming the
ABO3 stoichiometry in all cases. The results are the average of different
randomly-selected locations on the sample with standard deviation included. The relative standard deviation is given in parentheses.
material was, according to the phase diagram ~Jaffe et al.,
1971!, orthorhombic and isostructural with orthorhombic
KNbO3 ~Katz & Megaw, 1967; JCPDS-ICDD 71-2171! with
a unit cell derived from the simple perovskite cell by rotating two axes by 458. The K0.65Na0.35NbO3 ~Ahtee & Hewat,
1978; JCPDS-ICDD 77-0038! was monoclinic, based on a
doubled, simple perovskite cell. The cell parameters of the
sintered KNN ceramics based on a simple orthorhombic
perovskite cell are listed in Table 4.
The thermally-etched surfaces of the KNN sintered at
11008C for 2 and 24 h revealed a broad distribution of
angular grains with trapped pores ~Fig. 5!. The average
grain size in the sample sintered for 2 h was 1.71 6 1.78 mm,
and for 24 h, 2.61 6 2.85 mm ~Fig. 6!. We were unable to
distinguish the two separate phases on the etched surfaces.
Microstructural details of the KNN ceramics sintered at
11008C for 2 h were further analyzed by TEM. Figure 7
EM Studies of KNN Ceramics
575
Figure 3. The SEM-BEI micrographs of the polished surfaces of KNN ceramics sintered at 10308C and 11008C. The
arrows denote the secondary phase inclusions. 嘸 and ⫻ denote some of the locations of the EDS analyses.
Table 3. Elemental Composition of the KNN Sintered at 10308C for 8 h and 11008C for 2 h and 24 h Determined by
SEM-EDS Analysis*
Atomic percent
Nominal composition
10308C/8 h
Matrix
11008C/2 h
Matrix
11008C/24 h
Matrix
Secondary phase
Na
K
Nb
Na/K
~Na⫹K!/Nb
10.0
10.0
20.0
1.00
1.00
9.8 6 0.5 ~5.1%!
11.1 6 0.1 ~0.9%!
20.9 6 0.1 ~0.5%!
0.88 6 0.05
1.00 6 0.03
9.8 6 0.1 ~1.0%!
10.8 6 0.2 ~1.9%!
21.0 6 0.1 ~0.5%!
0.91 6 0.02
0.98 6 0.01
10.4 6 0.3 ~2.9%!
3.2 6 0.1 ~3.1%!
11.0 6 0.1 ~0.9%!
10.7 6 0.1 ~0.9%!
20.7 6 0.1 ~0.5%!
23.0 6 0.1 ~0.4%!
0.95 6 0.03
0.30 6 0.01
1.03 6 0.01
0.60 6 0.01
*The atomic percent of Na, K, and Nb were normalized assuming the ABO 3 stoichiometry in all cases. The results are the average of at
least three analyses of different randomly-selected locations on the sample with standard deviation included. The relative standard
deviation is given in parentheses.
shows a TEM micrograph of a grain, marked with an arrow,
and a trapped pore. The pore has a cubic morphology. We
arbitrarily chose this grain to be in the zone @001# and the
crystal planes of the grain to which the pore is attached are
in the $100% family with regard to the simple perovskite cell.
Planar defects were found in the sample sintered at
11008C for 2 h. With selected-area electron diffraction and
using bright-field dark-field experiments ~two-beam case!,
it was determined that these defects were the d-type ~Edington, 1974!. Figure 8 shows TEM micrographs of d bound-
576
Darja Jenko et al.
Figure 4. The XRD spectra of the KNN sintered
at 10308C for 8 h and 11008C for 2 h and 24 h.
The hkl indices based on a simple orthorhombic
perovskite cell were proposed by Stannek ~1970!.
The peak denoted with “Al” is because of the
sample holder.
Table 4. The Cell Parameters of Sintered KNN Ceramics Based
on a Simple Orthorhombic Perovskite Cell
a ~nm!
b ~nm!
c ~nm!
V ~nm 3 !
1030/8
1100/2
0.4015~7!
0.4040~7!
0.3951~2!
0.3952~4!
0.4005~2!
0.4006~4!
0.0635~2!
0.0640~2!
1100/24
0.4044~6!
0.3949~4!
0.4004~4!
0.0639~2!
T sinter ~8C!/t ~h!
aries in the KNN ceramics. When tilted, d boundaries
showed asymmetric fringes in the bright field ~Fig. 8a! and
symmetric fringes in the dark field ~Fig. 8b!.
D ISCUSSION
Single crystals of NaNbO 3 and KNbO3 were used as reference materials for the quantitative SEM-EDS analysis of
KNN ceramics. The EDS analysis on NaNbO3 single crystal
revealed a 30% lower value for sodium for two reasons
~Table 2!: ~1! the high absorption of the relatively “soft”
Na-Ka X-ray line ~E ⫽ 1.04 keV! and insufficient absorption correction within the SEMQuant-ZAF program and
~2! use of albite mineral as a standard for VSP reference ~see
Materials and Methods!. As a result, the measured value for
sodium was lower than the nominal value. It could also be
possible that the lower value for sodium is caused by its
mobility under the electron beam as shown in glasses
composed mostly of SiO2 , Na2O, CaO, MgO, and Al2O3
~Jbara et al., 1995!. However, Samardžija et al. ~2004!
showed that the lower value was not the consequence of Na
volatility and/or mobility, whereas a NaNbO 3 single crystal
was stable under the electron beam. The values for potassium for KNbO3 single crystal and niobium for both single
crystals did not deviate much from the nominal composition ~Table 2!.
We also analyzed the KNN sintered ceramics. The atomic
percent of Na in the matrix phase was 9.8 6 0.5 for sample
sintered at 10308C for 8 h and slightly lower than the
nominal value ~Table 3!. We observed a slight increase of
the Na content with temperature and time above the level of
uncertainty that we connected with enhanced homogenization of Na within the matrix phase. The values in all three
samples did not vary much and the difference could be
related to the systematic error of the EDS analysis. The
values for K and Nb in all cases were approximately 11 and
21 at.%, respectively. This was even more pronounced in the
samples, sintered at 11008C that we connected with a more
homogeneous distribution of K and Nb within the matrix
grains. The Na/K ratio in the matrix was always slightly
lower than the nominal ratio of 1 ~between 0.88 and 0.95!,
and it increased with temperature and time. Analogously,
this trend was reflected in the ~Na⫹K!/Nb ratio.
The secondary-phase inclusions were not analyzed in
the samples sintered at 10308C for 8 h and 11008C for 2 h
because the areas of the secondary phase were too small and
would not provide an exact analysis. For that reason we
analyzed the inclusions only in the sample sintered at 11008C
for 24 h. The amount of Na in the inclusions, that is, 3.2
at.%, was much lower than in the matrix phase. On the
other hand, the amount of K was only slightly lower than in
the matrix phase at 10.7 at.%, whereas the amount of Nb
was higher at 23.0 at.%. The Na/K and the ~Na⫹K!/Nb
ratios corresponded to 0.30 and 0.60, respectively.
In addition to the perovskite KNbO3 , there are a number of other potassium niobate phases that are unstable, and
EM Studies of KNN Ceramics
577
Figure 5. The SEM-BEI micrographs of the thermally-etched surfaces of the KNN ceramics sintered at 11008C for 2 h
and 24 h.
Figure 6. The grain size distribution of KNN ceramics sintered at 11008C for 2 h and 24 h. Quantitative characterization of the microstructures was carried out on more than 1000 grains. N denotes number of grains, D size of grains.
Inset of both graphs shows a range of 6–30 mm ~logarithmic scale!.
they exhibit significant volatility of K2O upon annealing
~Jaffe et al., 1971; Flückiger et al., 1977; Kodaira et al., 1982!.
A similar event probably occurs in NaNbO3 , but there are
no data about this. There are also no literature data about
secondary phases in KNN, but the research on alkaline-earthdoped KNN shows secondary phase or phases that could be
ascribed to various alkaline or alkaline-earth alkaline polynio-
bates ~Powel, 1971; Ahn & Schulze, 1987; Malic et al., 2005a,
2005b!.
From the XRD data we calculated the cell parameters of
the sintered KNN ceramics based on a simple orthorhombic perovskite cell ~Table 4!. With increasing sintering temperature and time, we observed a small increase in one of
the cell parameters, whereas the other two remained con-
578
Darja Jenko et al.
Figure 7. The TEM micrograph along the @001# zone axis of a
KNN grain with a cube-shaped pore after sintering at 11008C for
2 h. The grain was indexed with a simple perovskite unit cell.
stant ~within the uncertainty range!. This was reflected in
the increase of the perovskite cell volume. We attributed this
increase to the observed increase of the sodium content
within the KNN matrix phase with annealing temperature
and time as determined by the EDS ~Table 3!.
In the XRD spectra of our samples ~Fig. 4!, we did not
detect any secondary phase, either because of the small
quantity or because this secondary phase may have been
isostructural with the matrix phase, but with a different
composition.
The etched microstructure of KNN sintered at 11008C
for 2 h revealed the presence of some large grains of ;20
mm across within otherwise fine-grained matrix, that is, the
average grain size was 1.71 mm. After sintering for 24 h, the
average grain size was 2.61 mm with the largest grains
reaching 30 mm ~Figs. 5 and 6!. These results are in agreement with the observations of the fracture surfaces ~Fig. 2!.
The observed bimodal microstructures of KNN ceramics
could have been a consequence of the broad particle size
distribution of the starting KNN powder ~Fig. 1!. Although
the powder was fine with the median size of 0.65 mm, there
was a fraction of powder particles with sizes up to 10 mm;
these could have represented seeds for exaggerated grain
growth. However, other mechanisms for exaggerated grain
growth were also possible.
We observed trapped angular pores within the population of large grains. The pore size increased with sintering
temperature and time and reached almost 1 mm across in
the sample sintered for 24 h at 11008C ~Figs. 2 and 5!.
Coalescence of the intragranular angular pores as a result of
rapid grain growth was also observed during the sintering
of NaCl ~Sata, 1994!.
In a TEM micrograph of a grain, the cubic morphology
of the intragranular pore was clearly shown ~Fig. 7!. The
crystal planes of the grain to which the pore was attached
are of the $100% family with regard to the simple perovskite
cell. Intragranular angular pores in NaCl were also observed
by Sata ~1994!. The pores were oriented parallel to the grain
faces as in our sample. In NaCl the $100% surfaces had the
Figure 8. Experimental ~a! bright- and ~b! dark-field TEM images of ferroelectric domains of KNN sintered at 11008C
for 2 h. Notation: B: bright fringe, D: dark fringe.
EM Studies of KNN Ceramics
lowest activation energies for vaporization ~Sata, 1992!. In
studies related to thin films and based on lead-based perovskites, it has been observed that the $100% faces have
the lowest surface energies ~Okuwada et al., 1989, 1991;
Tani et al., 1993!. This is valid also for SrTiO 3 ~Sano et al.,
2003!. We found no data on the surface energy of KNN or
related compounds; however, we assume that in this alkalinebased perovskite the $100% surfaces with regard to the simple perovskite cell had a lower surface energy than the other
crystal faces.
S UMMARY
The microstructures of KNN ceramics sintered at 10308C
for 8 h and 11008C for 2 h and 24 h were studied by
electron microscopy. The SEM-EDS revealed that the materials consisted of a matrix phase in which the ~Na⫹K!/Nb
ratio corresponded closely to the nominal composition of 1.
In the sample sintered at 11008C for 24 h, a small number of
Nb-rich secondary phase regions were found with a Na/K
ratio of 0.30 and a ~Na⫹K!/Nb ratio of 0.60. The fracture
and thermally-etched surfaces revealed the cubic morphology of the grains and two grain-size populations: fine,
micron-sized grains and large grains of a few tens of 10 mm.
In the latter, trapped cube-shaped pores were observed. In
KNN sintered at 11008C for 2 h, ferroelectric domains were
found.
A CKNOWLEDGMENTS
The Ministry of Higher Education, Science and Technology
of the Republic of Slovenia ~project P2-0105! and the European Commission ~LEAF project, G5RD-T-2001-00431! are
gratefully acknowledged for financial support. We wish to
thank Prof. P. Guenther ~ETH, Zuerich, Switzerland! and
Dr. J. Dec ~University of Silesia, Inst. Phys., Katowice, Poland! for providing the KNbO 3 and NaNbO3 single crystals.
University of Texas Health Science Center at San Antonio,
Texas developed the computerized image analysis free UTHSCSA ImageTool program ~available from the Internet from
http://ddsdx.uthscsa.edu/dig/itdesc.html!. Dr. Goran Dražić
is gratefully acknowledged for fruitful discussions concerning analytical electron microscopy.
R EFER ENCES
Ahn, Z.S. & Schulze, W.A. ~1987!. Conventionally sintered
~Na0.5 ,K0.5 !NbO3 with barium additions. J Am Ceram Soc 70
~1!, C18–C21.
579
Ahtee, M. & Hewat, A.W. ~1978!. Structural phase transitions in
sodium-potassium niobate solid solutions by neutron powder
diffraction. Acta Crystallogr A 34, 309–317.
Edington, J.W. ~1974!. 3: Interpretation of transmission electron
micrographs. In Monographs in Practical Electron Microscopy in
Material Science, p. 45. London, Basingstoke, UK: The Macmillan Press Ltd.
Flückiger, U., Arend, H. & Oswald, H.R. ~1977!. Synthesis of
KNbO3 powder. Am Ceram Soc Bull 56 ~6!, 575–577.
Jaeger, R.E. & Egerton, L. ~1962!. Hot-pressing of potassiumsodium niobates. J Am Ceram Soc 45 ~5!, 209–213.
Jaffe, B., Cook, W.R., Jr. & Jaffe, H. ~1971!. Perovskite niobates
and tantalates and other ferroelectric and antiferoelectric perovskites. In Piezoelectric Ceramics, Roberts, J.P. & Popper, P. ~Eds.!,
pp. 185–212. London, New York: Academic Press.
Jbara, O., Cazaux, J. & Trebbia, P. ~1995!. Sodium diffusion in
glasses during electron irradiation. J Appl Phys 78 ~2!, 868–875.
JCPDS-ICDD 71-2171, PCPDFWIN Version 2.02, 1999.
JCPDS-ICDD 77-0038, PCPDFWIN Version 2.02, 1999.
Jenko, D., Malič, B., Bernard, J., Cilenšek, J. & Kosec, M.
~2003!. Synthesis and sintering of KNN 50/50 ceramics. Mater
Technol 37, 49–52.
Katz, L. & Megaw, H.D. ~1967!. The structure of potassium
niobate at room temperature: The solution of a pseudosymmetric structure by Fourier methods. Acta Crystallogr 22, 639–684.
Kodaira, K., Shioya, J.S., Shimada, S. & Matsushita, T. ~1982!.
Sintering and dielectric properties of KNbO 3 . J Mat Sci Lett 1,
227–278.
Kosec, M. & Kolar, D. ~1975!. On activated sintering and electrical properties of NaKNbO 3 . Mat Res Bull 10, 335–340.
Malic, B., Jenko, D., Bernard, J., Cilensek, J. & Kosec, M. ~2003!.
Synthesis and sintering of ~K,Na!NbO3 based ceramics. In Solid
State Chemistry of Inorganic Materials IV, Alario-Franco, M.A.,
Greenblatt, M., Rohrer, G. & Whittingham, M.S. ~Eds.!, vol. 755,
pp. 83–88. Warrendale, PA: Materials Research Society.
Malic, B., Bernard, J., Holc, J. & Kosec, M. ~2005a!. Strontium
doped K0.5Na0.5NbO 3 based piezoceramics. Ferroelectrics 314,
149–156.
Malic, B., Bernard, J., Holc, J., Jenko, D. & Kosec, M. ~2005b!.
Alkaline-earth doping in ~K,Na!NbO3 based piezoceramics. J
Europ Ceram Soc 25, 2707–2711.
Okuwada, K., Imai, M. & Kakuno, K. ~1989!. Preparation of
Pb~Mg 1/3Nb2/3 !O3 thin film by sol-gel method. Jpn J Appl Phys
28 ~7!, L1271–L1273.
Okuwada, K., Nakamura, S., Imai, M. & Kakuno, K. ~1991!.
High-resolution electron microscopy on epitaxial Pb~Mg 1/3
Nb2/3 !O3 film prepared by sol-gel method. Jpn J Appl Phys 30
~6A!, L1052–L1055.
Powel, B.R. ~1971!. Processing of sodium-potassium niobate ceramics, M.Sc. Thesis. Berkeley: University of California.
Samardžija, Z., Bernik, S., Marinenko, R.B., Malič, B. & Čeh,
M. ~2004!. An EPMA study of KNbO 3 and NaNbO3 single
crystals—Potential reference materials for quantitative microanalysis. Mikrochim Acta 145 ~1–4!, 203–208.
Sano, T., Saylor, D.M. & Rohrer, G.S. ~2003!. Surface energy
anisotropy of SrTiO 3 at 14008C in air. J Am Ceram Soc 86 ~11!,
1933–1939.
Sata, T. ~1992!. Vaporization rates from sintered bodies and single
crystals of NaCl in flowing air. J Mat Sci 27 ~11!, 2946–2951.
Sata, T. ~1994!. Expansion during sintering of NaCl powders.
Ceram Int 20, 39–47.
580
Darja Jenko et al.
Stannek, W. ~1970!. Characterization of sintering phenomena of
Na0.5K0.5NbO3 , M.Sc. Thesis. Berkeley: University of California.
Tani, T., Xu, Z. & Payne, D.A. ~1993!. Preferred orientations
for sol-gel derived PLZT thin layers. In Mat. Res. Soc. Symp.
Proc.: Ferroelectrics thin films III, Myers, E.R., Tuttle, B., Desu,
S.B. & Larsen, P.K. ~Eds.!, vol. 310, pp. 269–274. Pittsburgh:
Materials Research Society.
Tran-Huu-Hue, L.P., Feuillard, G., Loyau, V., Ringgaard, E.
& Lethiecq, M. ~2003!. Nonlinear behaviour of piezoelectric materials with graded electromechanical properties.
In World Congress on Ultrasonics 2003 Proceedings, Paris, September 7–9, 2003, pp. 543–546.