Chapter 1 Chapter 1 Chapter 1 Chapter 1 Introduction Introduction Introduction Introduction
Chapter 1Chapter 1Chapter 1Chapter 1
IntroductionIntroductionIntroductionIntroduction
Chapter 1Chapter 1Chapter 1Chapter 1
1
CHAPTER 1
INTRODUCTION
1.1. Dielectric Materials
A material is something with properties that give it the potential for a particular
application, either structural, as with a building material; functional, as with materials
used to make devices (electronic, optical, or magnetic); or biological, with biomedical
applications. Every class of material has distinctive properties that reflect the
differences in the nature of bonding and hence conductivity. Based on bonding and
conductivity, materials can be classified as metals (conductors), semiconductors and
insulators (dielectrics). The term “dielectric” was coined by William Whewell (from
“dia-electric”) in response to a request from Michael Faraday.
Dielectric materials include covalent and ionic bonded materials, such as polymers,
glasses and ceramics. Due to strong bonds between valence electrons and atoms,
dielectrics are good electric insulators. Hence dielectrics and insulators can be defined
as materials with high electrical resistivity. When a dielectric is placed in an electric
field, electric charges do not flow through the material, as in a conductor, but only
slightly shift from their average equilibrium positions causing dielectric polarization.
Thus a dielectric is an electrical insulator that may be polarized by an applied electric
field. A good dielectric is, of course, necessarily a good insulator, but the converse is by
no means true.1
Although the term “insulator” refers to a low degree of conduction, the term
“dielectric” is typically used to describe materials with a high polarizability. The latter
is expressed by a number called the dielectric constant (ε). The dielectric constant (ε)
of a material indicates the degree to which the medium can resist the flow of electric
charge and is defined as ε = D/E where D is the electric displacement and E is the
electric field strength. Dielectric constant (ε) is an essential piece of information for
designing Capacitors, Radio Frequency(RF) Lines, Optical fibres (e.g. waveguide to
form filters), Printed Circuit Boards(PCBs), and Microwave Communication Systems.
Chapter 1Chapter 1Chapter 1Chapter 1
2
1.2. Microwaves
Microwaves are electromagnetic radiations. The electromagnetic spectrum ranges
from 1 MHz to 106 GHz and can be broken down into ranges as shown in figure 1.1
The radio frequency is in the range of 300 KHz to 300 MHz. Radio waves are used to
carry information in amplitude modulation (AM) and frequency modulation (FM)
broadcast radio, short wave radio, and very high frequency (VHF) television channels.
These signals can be carried using traditional transistors, tubes and circuits.
Microwaves lie in the region from 300 MHz to 300 GHz. The microwave region can
be broken down further into the ultra high frequency (UHF) region from 300 MHz to
3 GHz, the super high frequency (SHF) region from 3 GHz to 30 GHz, and the
extremely high frequency (EHF) region from 30 GHz to 300 GHz.
Fig. 1.1: Electromagnetic region in that microwaves lie in the region of 10 -1 to 10 -3 m.
MSR CMR ATCT GPS ST MH MR SCD PR SCU MS
0.3GHz 1GHZ 2 3 4 6 8 10GHz 20 30 80 100GHz 200 300
UHF SHF EHF
FREQUENCY
MSR= Military Search Radar CMR = Cellular Mobile Radio ATCT= Air Traffic Control Transponder
GPS = Global Positioning System ST= Space Telemetry MH= Microwave Heating
MR= Microwave Relay SCD= Satellite Communication Downlink
SCU= Satellite Communication Uplink PR = Police Radar MS = Missile Seeker
Fig. 1.2. The microwave spectrum and some of the applications utilizing microwaves..2
Chapter 1Chapter 1Chapter 1Chapter 1
3
Microwave energy has emerged as the most versatile form of energy applicable in
numerous diverse fields. Since, its first use in radar in World War II, it is now being
applied in communication, chemistry, rubber vulcanization, drying, food processing,
medical treatment and diagnosis and variety of materials processing fields. The most
recent development of microwave research is the use of separation of E and H fields
at 2.45 GHz in materials processing that has produced dramatic enhancements in
reaction kinetics and phase transformations.3 Recently, wireless communication
systems using microwaves to carry information have become increasingly popular.
Microwaves are able to carry more information than radio waves because of their
higher frequency. The higher frequency gives a wider bandwidth capacity, enabling it
to carry more information. Modern communication systems which include cellular
phones and satellite communications have moved to the microwave (MW) frequency
region, where advanced dielectric ceramics are frequently used in resonators and
filters. New communication systems have been investigated extensively for the
practical uses, such as wideband code division multiple access (W-CDMA),
multimedia mobile access communication (MMAC), intelligent transport system
(ITS), home radio frequency (RF), and high-speed wireless i.e. Local Area Network
(LAN). Because of current increase in demand for these systems, investigation of
microwave dielectrics used in RF devices has become more important.4 The demand
for microwave dielectric resonators increase rapidly due to the development of
microwave technique in recent years.5
1.3. Dielectric Resonator
A dielectric resonator is an electronic component that exhibits resonance for a narrow
range of frequencies in the microwave band. The electronic component (ceramic
element) acts as a resonator due to multiple total internal reflections at the high
dielectric constant material and air boundary. In 1939, Richtmyer6 showed that
unmetallized dielectric objects could function similarly to metallic cavity resonators;
he called them dielectric resonators. The dielectric resonator generally consists of a
‘Puck’ of ceramic which is able to resonate at the frequency of the carrier signal to
allow that signal to be efficiently separated from other signals in the microwave band;
Chapter 1Chapter 1Chapter 1Chapter 1
4
this frequency is called the resonant frequency (fo). The resonant frequency depends
on the dielectric material and the size of the resonator.
Ceramic pucks suitable for single mode resonator applications are cylindrical and
based on the original concepts of Richtmyer. The puck is designed to sustain a
standing wave within its body of a specific resonant frequency and may therefore act
as either a filter or a transmitting resonator.6 Ceramic resonators of differing
geometries are shown in Fig. 1.3, in which single mode pucks are indicated.
Multimode resonators are also manufactured, often with unusual geometries to induce
several resonant MW modes within their body at any one time. These materials find
extensive applications after the growth of the mobile phone market in the 1990s.
Mobile communications have been extended rapidly all over the world. Mobile phone
networks allow communication from cell to cell via antennas located on masts and
associated base stations. Within a cellular network, the average base station coverage
is a diameter of 35 and 18 km at 900 and 1800 MHz, respectively. Each base station
houses microwave (MW) resonators that are used to carry signals of a specific
frequency and remove (filter) spurious signals and sidebands, that interfere with the
quality of the transmitted/received frequency band.7
Fig. 1.3: Various geometries of ceramic puck used as single and multimode resonators and filters.7
In service, the ceramic conventionally rests within a silver-coated square cavity, as
illustrated in Fig. 1.4. Typically, many hundreds of these cavities will reside in the
base stations of a cellular network.7
Chapter 1Chapter 1Chapter 1Chapter 1
5
Fig. 1.4: Silver-coated cavity with pucks used in resonator technology.7
Fig. 1.5 (below) shows the most commonly utilized three dominant modes8 for dielectric
resonators. The TEM mode dielectric resonator is characterized by a guided mode field
distribution of a TEM mode with standing wave of a quarter wavelengths. This mode
dielectric resonator causes significant size-reduction of the component. The TM010 mode
dielectric resonator is characterized by a TM mode field distribution. This mode resonator
has the middle levels of unloaded Q and size reduction effect between the TE01δ and TEM
mode resonators. The TE01δ mode dielectric resonator is characterized by a dominant TE
mode field distribution, the field of which leaks in the direction of wave propagation. A
high unloaded quality factor can be achieved using this mode.
Fig. 1.5. Three dominant modes for dielectric resonators.8
Chapter 1Chapter 1Chapter 1Chapter 1
6
The quality of the resonator depends mainly upon the dielectric property of the
material.9 There are three main, very important properties of a dielectric resonator
material: relatively high dielectric constant, low dielectric loss, and temperature
stability. Thus for a ceramic to be usable as a dielectric resonator/filter, the three key
parameters need to be optimized10 as such:
i. Relative permittivity (εr): 20 < εr < 50
ii. Temperature coefficient of resonant frequency: τf ≈ + 0 ppm/0C and
iii. Low dielectric loss i.e. high quality factor which means Q > 30,000 at 1 GHz
Temperature-stable, medium-permittivity dielectric ceramics have been used as
resonators in filters for microwave (MW) communications for several decades and
perovskites have been extensively investigated11for use as MW dielectric materials
because of the outstanding and tuneable dielectric properties present in a number of
compositions. Complex perovskite ceramics with a general formula as A(B’1/3B”2/3)O3
have been reported to be excellent for MW dielectric resonators with sufficient high
dielectric constant, very low dielectric loss, and small temperature coefficient of resonant
frequency.12, 13, 14
1.4. Microwave Dielectric Properties
The various microwave dielectric properties are described as below.
1.4.1. Dielectric Constant or Relative Permittivity (εr)
Basically, dielectric constant is used to express polarizability of materials and it is
independent of frequency.8 In the modern usage dielectric constant refers to relative
permittivity (εr) and it may be either the static or the frequency-dependent relative
permittivity depending on context.
The relative static permittivity, εr, can be measured for static electric fields as follows:
first the capacitance of a test capacitor, C0, is measured with vacuum between its
plates. Then, using the same capacitor and distance between its plates the capacitance
Cx with a dielectric between the plates is measured. The relative dielectric constant
can be then calculated as
εr = Cx / C0 … (1)
Chapter 1Chapter 1Chapter 1Chapter 1
7
Thus permittivity is ratio of the amount of stored electrical energy in capacitor at an
applied voltage relative to that in vacuum.
This quantity becomes frequency dependent for time-variant electromagnetic fields
and in general it is regarded as relative permittivity. Thus the relative permittivity is
related to the resonant frequency, f0, by the following equations;
… (2)
where c is the speed of light in a vacuum and λd is the wavelength of the standing
wave along the diameter (D) of a resonator. Consequently, if the permittivity is
increased, the size of the resonator may be decreased while still maintaining a specific
resonant frequency, i.e. larger permittivities enable miniaturization.7 The resonant
frequency can be changed by manipulation of the volume of the resonating body and
the dielectric constant. Frequency selection can be achieved by altering the sample
volume and dielectric constant. Higher dielectric constant materials are preferred due
to the ability to make smaller volume components, which is an important
consideration when expensive materials are (e.g. Ta2O5) are required. However, the
relative permittivity (εr) is expected to be small for higher frequency millimetre wave
region, because of reducing the delay time of electronic signal transmission and
improvement of accuracy for production.
Dielectric materials are purposely doped with impurities so as to control the precise
value of εr within the cross-section. This controls the modes of transmission. It is
technically the relative permittivity that matters, as they are not operated in the
electrostatic limit.
1.4.2. Quality Factor (Q)
Quality factor (Q) is a measure of the selectivity of a resonator to a given frequency
and hence can be considered as the definition of how good our component is. Two
definitions are employed, depending on the frequency range of the measurement. At
sub-microwave frequencies the quality factor is a measure of the efficiency or power
loss. At microwave frequencies a peak occurs in the transmitted signal amplitude at
Chapter 1Chapter 1Chapter 1Chapter 1
8
the resonant frequency and has a finite width in a microwave dielectric system. A
high Q means a narrow peak, giving high selectivity to a given frequency. Higher
selectivity enables an increased density of channels in a given frequency band. Peak
resonant frequency (fo) divided by the peak width (∆f) is equal to Q or in sub-
microwave frequency measurements is equal to 1/tan δ.
… (3)
Evidently, the quality factor (Q) is approximately the inverse of the loss tangent
(tanδ). Thus, high Q implies low loss and vice versa. Higher Q values reduce the risk
of cross-talk within a given frequency range.
Q decreases with increasing frequency; Fig. 1.6 shows some of the results. It should be
stressed that ceramics of Ba(Mg1/3Ta2/3)O3 (BMT) and Ba(Zn1/3Ta2/3)O3 (BZT) still
keep a high value of Q more than 6000 even at 20 GHz.
Fig. 1.6: Q vs. frequency 9
The theoretical relationship between the two is such that Q x f0 should be constant for
any given material, and, often, Q x f0 values are quoted when comparing ceramics.10
Chapter 1Chapter 1Chapter 1Chapter 1
9
The radio frequency devices require narrow frequency selectivity and low power
consumption which are determined by the Q-factor of the materials. It is, therefore,
significant to develop and study dielectrics with the high Q-factor in the microwave
regime.4 The Q value of ceramics is structure-sensitive and strongly depending upon
the sintering condition.9 The quality factor (Q) is affected by intrinsic factors such as
crystal structure and by extrinsic factors such as grain growth, impurity, and so on.
Higher Q values can be obtained in materials by ordering through prolonged sintering
at high temperature.13 The various factors responsible for high Q values, for e.g., in
Ba(Zn1/3Ta2/3)O3 are:
i. Inducing 1:2 cation
ii. Volatilization of ZnO at extended high temperature
The possible mechanisms for ZnO loss as proposed by Desu et al.: 14
Ba3(ZnTa2)O9 Ba3Ta2O8 + ZnO ↑ ... (i)
Ba3(ZnTa2)O9 (1-x)Ba3(ZnTa2)O9 + xBa3Ta2O8 + xZnO ↑ ... (ii)
Ba3(ZnTa2)O9 (1-x)Ba3(Zn1-xBaxTa2)O9 + 4/3xZnO ↑ + x/3Ta2O5 ... (iii)
iii. Stabilization of the domain boundary and lowering of the free energy of the
antiphase boundary15
iv. Increase in the grain size and relative density16
Degradation of Q arises due to the creation of oxygen vacancies (VO) and possibly the
onset of electronic conduction. The formation of vacancies is a plausible explanation
from the point of view of phonon–photon interactions and it is intuitively easy to imagine
a distribution of VO increasing the anharmonicity of vibrations and dampening of phonon
modes, both of which are classic explanations for extrinsic dielectric loss. A direct link
between electronic conductivity and less Q is clear from a mechanistic perspective.7
1:2 Ordering in Perovskites
The perovskite related oxides are disordered (cubic) at low temperatures but gets
ordered (-B’-B”-B”-B’-) to a hexagonal structure at high temperatures.17,18 Ordering
behaviour in A(B’1/3B”2/3)O3-type complex perovskite ceramics has been extensively
Chapter 1Chapter 1Chapter 1Chapter 1
10
studied and it has been reported that the microwave dielectric properties of the
ceramics are strongly dependent on the cationic ordering in the perovskite structure.19
The ability of A(B’ B”)O3 perovskites to order depends mainly on the differences in
size and valency of the two B-site ions: large differences in these favour order.20 For
A(B’1/3B”2/3)O3 in general, when charge differences ≥4 the cations order but when the
charges differ by less than 4 disordered structures are found unless there is a large
difference in radii.21
A study of Ba(B’0.5Nb0.5)O3-type compounds demonstrated that size and charge
difference of the B position ions had an important, effect on their ordering.22, 23 In an
investigation of A(B’0.33Ta0.67)O3-type compounds, where A is a barium or strontium
ion and B’ is a smaller divalent ion, the results not only were substantiated but it also
was found that the degree of long-range ordering in the B position ions decreased as
the difference in the size of these ions became smaller.24
There are two kinds of ordering of the B-site cations.25 One is a complete ordering
between B’II
and B”v as can be found in Ba(Zn1/3Ta2/3)O3 which is often called 1:2
ordering and the space group is P-3m1 (D33d). The other which is often called 1:1
ordering, is ordering between B’II
and B”v
as can be found26 in Pb(Mg1/3Nb2/3)O3
and27 Pb(In1/2Nb1/2)O3; the space group is Fm3m (O5
h). The A(B’1/3B”2/3)O3 -type
complex perovskite may also have a disordered structure. In the case of complete
disordering 25 between B’
II and B”
v, the space group is Pm3m (O
1h).
B-site cation ordering of A(B’1/3B”2/3)O3 systems has been investigated for its effect
on the microwave Q-factor (Table1.1). In general, it is found that higher ordering
correlates with a higher Q-factor. 4 Measurements of microwave dielectric properties
showed that permittivity (εr) and temperature coefficient of resonant frequency(τf)
decreased with ordering, and quality factor (Q) increased with ordering.28 However,
Seung-Hyun Ra and Pradeep P. Phule,29 reported that the B-site cation ordering is not
a primary factor that influences the observed microwave loss in BMT ceramics.
Chapter 1Chapter 1Chapter 1Chapter 1
11
Table 1.1: Structure and Microwave dielectric properties of Ba(B’1/3B”2/3)O3 ceramics.4
Materials εεεεr Qf (GHz) τf (ppm/oC) Structure
Ba(Zn1/3Ta2/3)O3 30 168000 0 ORD
Ba(Mg1/3Ta2/3)O3 25 360000 4 ORD
Ba(Zn1/3Nb2/3)O3 41 86900 31 DIS/ORD
Ba(Mg1/3Nb2/3)O3 32 55400 33 ORD
Ba(Co1/3Nb2/3)O3 31 60000 -6 DIS/ORD
Ba(Co1/3Ta2/3)O3 25 46200 -16 ORD
Ba(Mn1/3Nb2/3)O3 39 900 27 DIS
ORD= Ordered, DIS= Disordered
Most of the studies on order–disorder transitions in perovskite structure have been
focussed on the transitions due to compositional variations. But, of the late, in some
A(B’1/3B”2/3) O3-type ceramics it has been observed that their ordering structure may
be changed locally by sintering temperatures.19
The ordering in the A(B’1/3B”2/3)O3 –type complex perovskite compounds can be
investigated from superlattice reflections either by an X-ray diffraction (XRD)
technique30 or by an electron diffraction technique. The electron diffraction technique
provides us evidence for local ordering which is more microscopic than XRD.31 The
other feasible tools25for the probing of the ordering in A(B’1/3B”2/3)O3 –type complex
perovskite compounds might be those based on vibrational spectroscopy, which is
highly sensitive to the short-range ordering. The 1:2 ordering might also be detected
by using Raman spectroscopy.
1.4.3. Dielectric Loss (or Dissipation Factor)
The dielectrics are associated with loss.32 It is the ratio of the energy dissipated to the
energy stored in the material. Dissipated energy is typically turned into heat from
conduction of electrons flowing through the material or through the anharmonicity of
the lattice vibrations.33 It is also known as the loss tangent or dielectric loss. A lower
dissipation factor is optimal in order to retain maximum signal and to prevent
excessive heat generation in devices.
Chapter 1Chapter 1Chapter 1Chapter 1
12
Mathematically, it can be expressed as the ratio of the imaginary portion of
permittivity (ε′′) and the real portion of permittivity (ε′).
tan δ = ε″/ε′ … (4)
The dielectric loss tangent increases proportionately to frequency8 in the frequency
range from 109 to 1011 Hz. Since more customers can be accommodated at a given
frequency range at low loss, the dielectric loss (tanδ or Q-1 [Refer Eqn (3)] needs to be
minimized to retain maximal frequency resolution.34
It is observed that the loss increases with increasing permittivity.35 Since the Q-factor
generally varies inversely with frequency (f) in the microwave region, the product Q x f
rather than Q alone is used to evaluate dielectric loss.
For a filter, the effect of losses (1/Qd) in the dielectric material used for the resonator,
radiative losses (1/Qr), conductive losses in the cavity walls (1/Qc), losses occurring in
the support used to mount the dielectric resonator (1/Qsup), and losses encountered in
coupling and tuning (1/Qk,t) must be considered. [The subscript to Q refers to the
source, e.g., d in Qd refers to the dielectric.] These Q factors are related by the
following equation (5):36, 37
… (5)
As the dielectric characteristics at microwave frequencies mainly depend upon an
ionic polarization brought about by lattice vibration, information concerning lattice
vibration is indispensable for understanding dielectric loss in a crystal.38 The
mesoscopic scale of disorder of the charge distributions plays a major role in
dielectric loss properties of complex perovskite oxides at microwave frequencies.39
Loss has also been attributed to defects and grain boundaries in polycrystalline
materials.9 Loss increases with porosity (P), which is defined as P = 1 – D where D is
the fraction of the material’s theoretical density. Losses and dielectric dispersion are
divided into two categories, intrinsic and extrinsic. Intrinsic losses are related to the
crystal structure (e.g. atomic masses, atomic charges and bond strengths) and the
interaction of the phonon with the AC electric field. Extrinsic losses are caused by
Chapter 1Chapter 1Chapter 1Chapter 1
13
imperfections in the crystal including impurities, grain boundaries, porosity, oxygen
vacancies, microstructural defects and random crystallite orientation.40
Dielectric loss in sintered pellets is dominated by extrinsic factors and thus is highly
dependent on sintering conditions. Hence the extrinsic contribution can be minimized
by using proper processing conditions. Ordering of the octahedral cations in
perovskites leads to weak coupling and a low energy leakage between the modes,
which can result in lowered intrinsic losses and a large increase in Q.33
The studies on dielectric properties of the materials are sometimes controversial41 which
is primarily due to the interference of extrinsic dielectric loss on the measured dielectric
characteristics. The most promising approach to understand the intrinsic microwave
dielectric properties of the materials is to study their higher frequency response,
including the submillimeter and infrared regime, since the intrinsic losses are
overwhelmingly stronger than the extrinsic ones in the far-infrared regime. 42-45
1.4.4. Temperature Coefficient (τf)
Temperature coefficient of the resonant frequency (τf), the permittivity (τε), or the
capacitance (τc) can be used to indicate the thermal stability of dielectric properties of
a material. The temperature coefficient of resonant frequency (τf ) determines how
well a resonator will function when there are fluctuations in temperature. Thus τf is a
measure of the ‘‘drift’’ with respect to the temperature of the resonant frequency. It is
self-evident that a material with a significantly non-zero τf is useless in an MW
circuit. Thus the temperature coefficient of resonant frequency (τf) is expected to be
near zero ppm/oC for receiving the RF signals in all places in the world.
The temperature coefficient of the resonant frequency (τf) is defined as
τf = - (1/2τε + αL ) … (6)
where τε is the temperature coefficient of permittivity and αL is the linear expansion
coefficient. It is related10 to the temperature coefficient of capacitance (τc) by
τf = - ½(τc + αL ) … (7)
Chapter 1Chapter 1Chapter 1Chapter 1
14
The signs of τf and τε are opposite and the τε controls the τf neglecting the small change
in the linear thermal expansion coefficient. Microwave circuits normally have some
low characteristic τf, so resonator components ideally counteract the inherent drift
with a small τf.33 Obtaining a material with a zero temperature coefficient is one of the
most difficult parts in the development of dielectric ceramic materials. Barium- and
strontium-based complex perovskites generally show positive and negative τf,
respectively, at room temperature.46 Hence Ba and Sr are employed to obtain τf values
zero or near to zero.
From an engineering perspective, the premise used to obtain a temperature-stable
material appears quite simple; a solid solution or composite is created of two materials,
each having opposite signs of τf. Although this approach adequately explains tuning τf
from the perspective of a composite, it does not describe the complexities associated
with solid solutions.7 Within perovskite solid solutions, the true tuning mechanism(s)
is/are a combination of the reduction in the polarizability per unit volume (decrease in
permittivity)47 and inducing a phase transition above room temperature usually
associated with rotations of the O-octahedra.46,48
1.5. Perovskite Structure
A perovskite is any material with the same type of crystal structure as calcium titanium
oxide (CaTiO3) – known as perovskite structure.32 Perovskites take their name from this
compound which was first discovered in the Ural Mountains of Russia by Gustav Rose
in 1839 and named after Russian mineralogist, Lev Aleksevich Von Perovski (1792-
1856). The first synthetic perovskites were produced by Goldschmidt (1926) of the
University of Oslo led to the use of the term perovskite as a description of a class of
compounds sharing the same general stoichiometry and connectivity found in CaTiO3.
Thus the perovskite structure is a term which is used to describe an arrangement of
anions and cations that is isomorphous with CaTiO3 but where the symmetry of each
phase can be quite different.48
It is a category of mixed metal oxide crystal which constitutes a large family of
crystalline ceramics. The ideal perovskite has a primitive cubic structure with the
formula ABO3 where ‘A’ is a (Lanthanide and, or alkaline earth metal) cation of larger
Chapter 1Chapter 1Chapter 1Chapter 1
15
size than B (a transition metal cation). It is considered a Face Centred Cubic (FCC) -
derivative structure in which the larger A cation and oxygen together form an FCC lattice
while the smaller B cation occupies the octahedral interstitial sites in the FCC array. There
is only the oxygen being B cation’s nearest neighbour. The structure is a network of corner-
linked oxygen octahedra, with the smaller cation filling the octahedral holes and the large
cation filling the dodecahedral holes.49
The symmetry of the ideal perovskite structure is cubic in space group Pm3m. The unit
cell of perovskite cubic structure is shown below in Figure 1.7.
Fig. 1.7(a): The cubic ABO3 Perovskite structure.50
Fig. 1.7(b): The cubic ABO3 perovskite 3D-structure. Grey spheres represent the A-site cations, blue spheres the B-site cations, and smaller red spheres are the anions.51
Chapter 1Chapter 1Chapter 1Chapter 1
16
Fig. 1.7(c): Perovskite structure52
Fig. 1.8: 1:2 ordering along the [111] in A(B’1/3B”2/3)O3.
In Figure 1.7(b) and 1.7(c), we can see that the coordination number of A is 12, while
the coordination number of B is 6. In most cases, the above figure is somewhat
idealized. In fact, any structure consisting of the corner-linked oxygen octahedra with a
small cation filling the octahedral hole and a large cation (if present) filling the
dodecahedral hole is usually regarded as a perovskite, even if the oxygen octahedra are
slightly distorted.53 Also, it is unnecessary that the anion is oxygen. For example,
fluoride, chloride, carbide, nitride, hydride and sulphide perovskites are also classified
Chapter 1Chapter 1Chapter 1Chapter 1
17
as the perovskite structures. As a result, we can say that perovskite structure has a wide
range of substitution of cations A and B, as well as the anions, but the substitution must
maintain charge balance and keep sizes within the range for particular coordination
number. Because the variation of ionic size and small displacements of atoms in
perovskite structure lead to the distortion of the structure and the reduction of symmetry
which have profound effects upon the electrical and magnetic properties of the crystals
eventually playing important role in dielectric ceramic.
The perovskite structure accommodates most of the metallic ions in the periodic table and a
significant number of different anions. Perovskites have been reported with all naturally
occurring cations in the periodic table except boron and beryllium. Phosphorus and the
noble gases are also not observed in the perovskite structure. The majority of the perovskite
compounds are oxides or fluorides, but the perovskite structure is also known for the
heavier halides54, 55, sulphides56, hydrides57, cyanides58, 59, oxyfluorides60 and oxynitrides.61
There are varieties of ordered structures known in perovskite oxides, among them the
double-perovskite (A2BB’O6 or AA’BB’O6) and the triple-perovskite (A3BB’2O9)
structures are the most popular. Many of these ordered perovskites contain Ti, Nb, or
Ta as the B-cation, and are interesting due to their optimum dielectric characteristics for
applications in the electronic industry.62 A wide range of multiple ion substitutions
(Fig.1.9) compositions and properties studied by R. Roy.63 provides plausible clues to a
further dielectric materials research.
Fig. 1.9: Chart illustrating types of multiple ion substitution in the perovskite lattice.63
Chapter 1Chapter 1Chapter 1Chapter 1
18
1.5.1. Distortions in Cubic Symmetry of Perovskites
The parent structure type is referred to as the ‘aristotype’, and distorted perovskites are
designated ‘hettotypes’. Three different types of distortions identified include:
distortions of BO6 octahedral units, B-cation displacements within the octahedra, and
octahedral tilting distortions.64 Distortions of BO6 octahedral units often result from
electronic factors. In octahedral coordination, the 3d transition metal cations Mn3+ and
Cu2+ (high-spin) have electron configurations (t2g)3(eg)
1 and (t2g)6(eg)
3, respectively.
Electronically, this is an unfavourable situation and the Jahn-Teller (J-T) theorem states
the occurrence of a distortion in the geometry. J-T distortions typically occur without a
large deviation of the O-B-O bond angles from ideal values of 90° and 180°.
Interestingly the mineral perovskite, CaTiO3, does not adopt the aristotype cubic
structure. The symmetry of CaTiO3 is lowered from cubic (Pm3m, Z = 1) to
orthorhombic (Pnma, Z = 4) by a cooperative tilting of the titanium centred
octahedral.65 This distortion is driven by the mismatch between the size of the cubo-
octahedral cavity in the corner-sharing octahedral network and the undersized ionic
radius of the Ca2+ ion. The octahedral tilting distortion lowers the coordination number
of Ca2+ from 12 to 8 in order to reduce the tension in the remaining Ca-O bonds66 and
increase the lattice energy. However, there is very little perturbation of the local
octahedral coordination of the Ti4+ ion.
Octahedral tilting is the most common type of distortion the perovskite structure type
undergoes. Displacements of octahedral cations are often observed due to the
combination of structural influences and electronic factors.
In the perovskite structure (ABO3) the A-site cation is located in the cavity formed by
the corner-sharing network of [BO6] octahedra. The fit of the A-site cation is described
by the tolerance factor (Ref: Eqn. 8). If the tolerance factor is greater than unity, often
no octahedral tilting is observed, whereas compositions with a tolerance factor less than
unity typically undergo octahedral tilting distortions.67 Oversized A-cations stretch the
octahedral bonds, resulting in an increase in octahedral volume and reduction of B-O
bonding. Distortion from ideal cubic symmetry of the aristotype cubic perovskite
(Pm3m) occurs by a practically rigid tilting of the octahedral units while maintaining
the corner-sharing connectivity. Distortions in the octahedral angles are minor (typically
Chapter 1Chapter 1Chapter 1Chapter 1
19
less than 4° for O-B-O). Octahedral tilting allows greater flexibility in the coordination
of the A-site cation while maintaining a regular coordination environment for the
octahedral cation. Octahedral tilting reduces the symmetry of the A-site cation
coordination environment and leads to a significant change in the A-O bond lengths. B-
site cation ordering causes cell distortion and more stable phase of ordered structure can
enhance cell distortion.4
1.5.2. Tolerance Factor (t)
The tolerance factor or Goldschmidt tolerance factor (Goldschmidt, 1926) is a measure
of the fit of the A-site cation to the cubic corner-sharing octahedral network. Thus it is a
parameter to predict the structural distortion in perovskites. In a cubic perovskite twice
the B-O bond length is the cell edge, and the twice the A-O bond length is equal to the
face diagonal. Thus the tolerance factor is expressed as,
… (8)
where RA, RB, and RO are the radius of the A, B, and O ions respectively7, 64
The tolerance factor value is an approximate guide to the structural stability of the
perovskite phase. In general, when the value of t is close to 1, the perovskite phase will
be formed. If t is very far from 1, then the perovskite phase will not form.48 It has been
found that21 anion-deficient phases have low values of tolerance factor, 0.85< t < 0.99.
An upper limit of the tolerance factor is approximately 1.04, whereas a tolerance factor
less than 0.87 is near the range where the ilmenite (FeTiO3) structure type becomes
more stable compared to the perovskite structure type.7
1.6. Synthesis of Perovskites
Rational synthesis of materials requires knowledge of crystal chemistry besides
thermodynamics, phase equilibria and reaction kinetics. A variety of inorganic solids
have been prepared in the past several years by the traditional ceramic method. A wide
range of conditions, often bordering on the extreme, such as high temperatures and
pressures, very low xygen fugacities and rapid quenching have all been employed in the
Chapter 1Chapter 1Chapter 1Chapter 1
20
matrial synthesis. The present day trend is to avoid brute-force methods in order to get
a better control of the structure, stoichiometry and phase purity.
Several methods are known in the literature for the synthesis of perovskites: Solid State
Reaction (mixed oxide) method, Sol-gel method, Molten Salt Synthesis (MSS) or Flux
method, Microwave Processing, Citrate gel method, Ceramic injection moulding
(CIM) method, Spray Pyrolysis Technique, Homogeneous Precipitation Process, Co-
precipitation process, Inverse micro-emulsion process, combustion method,
intercalation, ion-exchange, electrochemical method, and Hydrothermal Synthesis.
Among various methods listed above, the solid state reaction method and molten salt
synthesis method have been used extensively in our studies.
1.6.1. Solid State Reaction (Mixed Oxide) Method
It is one of the simplest and most widely used methods to synthesize perovskites. The
different steps involved in this method are shown in the flow chart (Fig. 1.10).
Fig. 1.10: Flow Chart of the Solid State Reaction Method
Chapter 1Chapter 1Chapter 1Chapter 1
21
The most common method of preparing metal oxides and other solid materials is by the
solid state method or ceramic method, which involves grinding powders of oxides,
carbonates, oxalates or other compounds containing the relevant metals manually by
mortar and pestle or mechanically by a grinder. The grinding decreases the particle size
and increases the surface area of the powder, thereby increasing the intimacy between
the reactants, which will facilitate the reaction by diffusion. These milled powders are
then calcined at desired temperatures to bring about effective decomposition
(heterogeneous reaction). The entire reaction has to occur in the solid state, initially by
a phase boundary reaction at the points of contact between the components and later by
the diffusion of the constituents through the product phase. With the progress of the
reaction, diffusion paths become increasingly longer and the reaction rates slower. The
product interface between the reacting particles acts as a barrier. The reaction can be
speeded upto some extent by intermittent grinding between heating cycles.
1.6.2. Molten Salt Synthesis (Flux Method)
Molten salt synthesis (MSS) has been reported to be one of the simplest techniques to
prepare pure, stoichiometric ceramic powders of multicomponent oxides since the
diffusivities of the components are much higher by comparison to the solid state
reaction.68 The following are the basic requirement for selecting the salt in the molten
salt synthesis:
i. The melting point of any given salt must be appropriately low compared to the
formation temperature of the main phase.
ii. The solubility of a salt must be sufficient to be eliminated by a simple washing
step.
iii. There must be no undesirable reactions between salt and constituent oxides.
The advantages of this method are significant reduction in the powder formation
temperature and time. Furthermore, control of powder morphology was found to be
much easier using this technique.
Chapter 1Chapter 1Chapter 1Chapter 1
22
Factors influencing the final particle
Characteristics at each step
Initial particle size, shape, geometry and morphology
Purity and identity of precursors.
Chemical nature and quantity of salts
Purity of salts
Molar ratio between salt and precursors
Chemical nature and Synthesis temperature/ Reaction time
Heating and cooling rates
Remaining impurities from salt and
precursor components
Presence of agglomerates
Precursor Materials
Synthesis of desired products at melting temperature of salts
Washing of salts with aliquots of deionized water
Drying
Mixing of precursor molecules with salt(s) and/or surfactant
Fig. 1.11: Flow chart illustrating factors influencing the MSS of transition-metal oxide Materials.69
In the synthesis of all oxide materials two steps, calcinations and sintering are referred
to frequently.
Calcination
It is an endothermic decomposition reaction in which an oxy-salt, such as a carbonate or
hydroxide, decomposes, leaving an oxide as a solid product and liberting a gas. The
calcined oxide is finely divided. It is easy to obtain a particle size much finer than 1 µm
and the specific surface are in the range of 100 m2/g. The surface area and the size are
easily modified. Low calcinations temperatures produce very fine, high surface area
powders. Heating at temperatures well above the decomposition temperature reduces the
surface area (and increases the particle size). Calcination is often the final step in the
production of high- purity ceramic powders, for example a homogeneous mixture of
several elements can be made by co-precipitation of solutes using hydroxides, carbonates,
or oxalates, which are later, converted to oxide compounds by calcinations.
Sintering
The pellets are sintered at higher temperatures to achieve a high density. This process
involves the heating of the pellets in a furnace below its melting point until its particles
adhere to each other. During sintering there occurs shrinkage of the calcined powder. It
Chapter 1Chapter 1Chapter 1Chapter 1
23
consists of solid particles bonding or neck formation, followed by continuous closing of
pores from a large open porosity to essentially pore-free bodies. Thus it converts the
green microstructure to the microstructure of the dense ceramic component. In this
process the important parameters such as sintering temperature, atmosphere, heating
and cooling rates, impurity concentrations, particle size etc., have to be exactly
controlled to obtain reproducible results. After sintering, the ceramic material is built up
of closely packed grains with random crystallographic orientations. While the sintering
process has an influence on the microstructure of the resulting material, the cooling rate
has a strong influence on the electrical properties of the ceramic material.1
It is the last of ceramic processing step where the ceramist has an influence on micro-
structural development. This influence is limited, however, as the worst in-homogeneities
that pre-exist in the compact are usually exaggerated during sintering; for example, flaws
will persist or even grow, while large particles may induce abnormal grain growth.71
1.7. Characterization Techniques for Perovskites
The Powder X-ray diffraction can be used to characterize the phases formed in the
polycrystalline powder, Newtron diffraction and Transmission Electron Microscope can
also be used to study the structural information.
1 0 2 0 3 0 4 0 5 0 6 0
2 θθθθ
Fig. 1.12: Powdered x-ray pattern showing disordered cubic perovskite A(B’1/3B”2/3)O3.
Chapter 1Chapter 1Chapter 1Chapter 1
24
1 0 2 0 3 0 4 0 5 0 6 0
2 θθθθ
Fig. 1.13: The arrows show super structure reflections corresponding to 1:2 ordering in A(B’1/3B”2/3)O3.
1.8. Materials for Dielectric Resonator Applications
Until the late 1960’s the applications of dielectric resonator were limited due to a lack
of suitable materials. Rutile phase (TiO2) was often used but its large resonant
frequency fluctuations with temperature made it impractical for many applications.
Single crystal alumina (sapphire) was also used but found very expensive. By the
1970’s, research and development for temperature stable dielectric resonator materials
had begun worldwide.32
Today a wide variety of dielectric ceramics is available for communications applications.
Among them the complex perovskites with general formula A(B1/3B’2/3)O3, where
A=Ba2+, B=Mg2+, Zn2+, or Ni2+, B’ =Ta5+ or Nb5+ show interesting and commercially
important properties at microwave frequencies. These ceramics typically have high
relative permittivity (or dielectric constant), low dielectric loss (also described in terms of
high dielectric Q value), and small temperature coefficient of resonance frequency
(τf).These materials are commonly used as resonators in microwave devices and systems,
particularly for communications applications. A high dielectric Q-factor combined with
small or preferably zero τf is critical for such applications.
Chapter 1Chapter 1Chapter 1Chapter 1
25
Dielectric materials, particularly those used in the microwave frequency applications
are generally grouped into three categories72
1. Containing compounds with highQ, Qxf(GHz)=190,000–350,000 and εr= 25-30
e.g. Ba(M1/3Ta2/3)O3 [where M=Mg, Zn]
2. Containing compounds with intermediate properties Qxf(GHz)=7400–50,000 and
εr ≈ 40, e.g. Ba2Ti9O20 and (Sn, Zr) TiO4.
3. Containing compounds with Q ≤ 7000 and high εr (80-100), e.g. BaO-A2O3-TiO2
[where A=La, Nd etc.]
The lattice constants, theoretical density and sintering temperature of some ceramics are
listed in Table 1.2.
Table 1.2: Lattice constants, theoretical density (ρ) and sintering temperature of perovskite-type ceramics.9
Compounds Structure Lattice constant (Å)
ρ
(g/cm3) Sintering
temperature (oC)
a c
Ba(Mg1/3Nb2/3)O3 (BMN) Hexagonal 5.776 7.089 6.211 1550
Ba(Mg1/3Ta2/3)O3 (BMT) Hexagonal 5.774 7.095 7.637 1550 - 1600
Ba(Zn1/3Nb2/3)O3 (BZN) Cubic 4.093 6.515 1500
Ba(Zn1/3Ta2/3)O3 (BZT) Hexagonal 5.787 7.087 7.944 1550
Ba(Mn1/3Nb2/3)O3 (BMnN) Pseudo-cubic 4.113 6.337 1550
Ba(Mn1/3Ta2/3)O3 (BMnT) Hexagonal 5.814 7.156 7.709 1600
Sr(Mg1/3Nb2/3)O3 (SMN) Hexagonal 5.638 6.920 5.378 1500
Sr(Zn1/3Nb2/3)O3 (SZN) Hexagonal 5.658 6.929 5.698 1500
Rapid progress in electromagnetic simulation methods has enabled a significant
improvement in the performance of state-of- the-art microwave devices. A growing
number of these devices require intricately shaped microwave materials. This has
placed increasing demand on the ceramic research community to develop a low-cost
process that can manufacture high performance microwave materials in complex-
shapes. The use of dielectric ceramics with low loss, high dielectric constants, and near-
Chapter 1Chapter 1Chapter 1Chapter 1
26
zero temperature coefficient of resonant frequency is also essential in achieving the
required microwave performance for the devices of interest.73 Although there has been
considerable amount of work in the area of dielectric oxides in the past 50 years,
dielectric materials continue to remain a significant area of scientific research due to
their wide range of applications such as resonators, filters and tuners. The perovskites
found to exhibit the highest Q values are based on the formula Ba(M2+1/3Ta5+
2/3)O3,
where M =Zn, Mg. For the Zn system, εr ≈ 30, and for the Mg system, εr ≈ 25. These
‘Tantalum’ based ceramic systems are: Ba(Zn2+1/3Ta5+
2/3)O3 (BZT) and
Ba(Mg2+1/3Ta5+
2/3)O3 (BMT).
1.8.1. Tantalum (Ta)-based Systems
Ba(Zn2+1/3Ta5+
2/3)O3 (BZT) and Ba(Mg2+1/3Ta5+
2/3)O3 (BMT) are the most common
‘Tantalum’ based ceramic materials with excellent microwave characteristics currently
used for dielectric resonator applications. The BZT has a high dielectric constant (εr
≈30), low temperature coefficient of resonant frequency (τf ≈ 4 ppm/C), and high
quality factor (Q x f ≈80 000–150 000 GHz).13 BMT has permittivity εr ≈ 24, high
quality factor (Q x f) ≈ 250 000 GHz and the low temperature coefficient of the
resonant frequency (τf) ≈ 0 ppm/oC.7 Single phase BMT has, of course, the highest Q
value among microwave materials studied to date,73,74 however, it is difficult to sinter.
Based on the Ginstling–Brounshtein model, the activation energy for the formation
process of BMT was estimated to be 257kJ/mol. Raising sintering temperatures resulted
in an increase in the ordering degree and bulk density of BMT.75 But a decrease in the
1:2 cation ordering and increase of dielectric loss in BMT occurred at sintering
temperatures above 1590 °C.76 Reducing the barium content in BMT substantially
resulted in improved densification and enhanced ordering structure. On the other hand,
an excess barium content in specimens hindered the progress of sintering, and also
induced the disordering structure.75
Both the BZT and BMT belong to a class of compounds often referred to as 1/3:2/3 i.e.
1:2 complex perovskites.77, 78 Since there are two different B-site cations, there are two
different structures that can exist: the disordered (space group Pm3m) and the 1:2
ordered states (i.e. for e.g. –Mg–Ta–Ta–Mg–Ta–Ta–) (space group P¯3m1).79
Chapter 1Chapter 1Chapter 1Chapter 1
27
However, the crystal structures reported in the literature indicate significant differences
in the reported O fractional coordinates and cation coordination environments of BZT
and BMT.11 The B-site bond lengths (Å) in BZT21: [Zn-O 1.977(4) x 6, Ta-O 2.023(7) x
3, and 2.141 (10) x 3] whereas that in BMT80: [Mg-O 2.07(2) x 6, Ta-O 2.04(2) x 3 and
2.033(6) x 3.
1.8.1.1. Barium Zinc Tantalate Ba3ZnTa2O9 or Ba(Zn1/3Ta2/3)O3 (BZT)
Neutron diffraction patterns (Fig. 1.14) and the data show the local B cation
environment of hexagonal phase of BZT in detail.
Fig. 1.14: Neutron diffraction profile21 for Ba3Ta2ZnO9.
Table 1.3: Interatomic distances and angles in Ba3Ta2ZnO9 from the neutron diffraction data.21
Ba(1)-O(1) 2.830(8)Å O(1)-O(1) 2.890(1)Å
Ba(1)-O(2) 2.895(12) O(1)-O(2) 2.976(4)
Ba(1)-O(2) 2.906(12) O(2)-O(2) 2.851(8)
Ba(2)-O(1) 2.890(1) O(2)-O(2) 2.741(8)
Ba(2)-O(2) 2.956(4) Ta-O(1) 2.023(7)
Zn-O(2) 1.977(4) Ta-O(2) 2.141(10)
O(2)-Zn-O(2) 92.2(5)o O(1)-Ta-O(1) 91.2(4)o
O(2)-Zn-O(2) 87.8(5) O(1)-Ta-O(2) 91.1(6)
O(2)-Ta-O(2) 83.4(8)
Chapter 1Chapter 1Chapter 1Chapter 1
28
Fig. 1.15: The B-cation environment in Ba3Ta2ZnO9.21
The long-range cation ordering observed in Ba3Ta2ZnO9 appears to originate because
of the difference in the ionic size of the two cations; Ta5+ =0.64, Zn2+ = 0.74Å
(Shannon & Prewitt, 1969) 21 Zn and Ta compete (Table 1.3) for the same corner-
sharing O atom; the Zn-O distance = 1.98Å (Zn2+ + O2- = 2.09Å) and the O-Ta
distance 2.14Å (Ta5+ + O2- =1.99Å). For this, a conceivable explanation is that the
high electron affinity of Ta5+ (and considerable covalence in the Ta-O bond) modifies
the O ligand field so that some 3d electron donation becomes possible from Zn to O
(in effect into a bond arising from Ta-O overlap). Combined with back donation from
O to the s-p orbitals of Zn this could give rise to an abnormally high Zn-O covalence
and a correspondingly short distance. The effect of delocalizing the d electrons of Zn
would be to transfer electron density to the d-orbital anti-bonding regions of Ta thus
weakening the bonding and increasing the Ta-O distance observed. The reduced
contribution to the lattice energy from an overall increase in the Ta-O distance is
partially compensated in a familiar way by an off-centre Ta displacement which leads
to three short and three long bonds with O.
A cubic perovskite structure belongs to the space group O1h of the cubic system, which
has no Raman-active mode and 3 infrared-active modes.38 The Ba(Zn1/3Ta2/3)O3 crystal
with a hexagonal superstructure, however, has 9 Raman-active modes and 16 infrared-
active modes. Fig. 1.16 and 1.17 show the Raman and infrared spectra of
Ba(Zn1/3Ta2/3)O3 ceramics.
Chapter 1Chapter 1Chapter 1Chapter 1
29
Fig. 1.16: Raman spectra38 of ceramic Ba(Zn1/3Ta2/3)O3.
Fig. 1.17: Measured and calculated far-infrared reflectivity38 of ceramic Ba(Zn1/3Ta2/3)O3.
Chapter 1Chapter 1Chapter 1Chapter 1
30
The dielectric constant and its temperature coefficient depend mainly on the
composition of the material. But loss quality varies from sample to sample of the same
composition, indicating that the loss quality is sensitive to slight differences in
crystallographic structure and microstructure of the material.13
Since the first report of the outstanding dielectric performance of complex perovskite BZT
many investigations have been conducted on the effect of different processing schemes and
chemical additives on the crystal chemistry, microstructure, and properties of BZT based
ceramics. Both ordered-type with a trigonal space group of P¯3m1 and disordered-type
with a cubic phase space group of Pm3m are formed depending on the synthesis conditions
of Ba(Zn1/3Ta2/3)O3.13 The B-site cations in Ba(Zn1/3Ta2/3)O3 are stoichiometrically ordered
in a hexagonal cell, with one Zn2+ layer and two Ta5+ layers repeat along the <1 1 1>
direction of the parent perovskite cubic cell. This ordering is a matter of considerable
interest because ordering is believed to play an important role in the quality factor (Q) of
BZT.13 The factors influencing Q values of perovskites Ba(M2+ 1/3Ta5+2/3)O3 (M = Mg, Zn)
have been considered to be long-range ordering (LRO) of cations, zinc oxide
volatilization/evaporation, point defects and stabilization of micro-domain boundaries.14,38
Different studies using Ba(Mg1/3Ta2/3)O3 (BMT) with minimized effect of volatilization
of ZnO have shown that the changes in Q originate from cation- ordering13, and the
cation-ordering enhances Q 81 While the other reports describe that the improvements
in Q are not related to cation-ordering, but to volatilization14 of ZnO, stabilization of the
domain boundary and lowering of the free energy of the antiphase boundary. The
deviations from stoichiometry induced by the volatilization of ZnO have been reported
to promote the growth of the ordered domains82 increase the extent of cation ordering83
and the c/a lattice distortion14 and improve the dielectric loss properties.13
Among all these different possibilities, the crystal structural ordering together with the
ceramic microstructure have been found to influence the Q factor of BZT strongly.84 It
has been found that Q improvement corresponds with increased Zn and Ta ordered
structures in the ceramics.13 Usually a significant ceramic microstructure difference
between the ordered and the disordered perovskite has been found.84 Dense ceramics
(density ≈7.76 g/cm3) has been obtained in the ordered perovskite with the relatively
larger grain (1.0–1.5µm). Whereas the disordered perovskite has the lower density
Chapter 1Chapter 1Chapter 1Chapter 1
31
(density ≈5.0 g/cm3) with the smaller grain size (0.4–0.6 µm). Thus these two factors
strongly relate each other in the BZT system. Therefore, their influence on the Q factors
can’t be discussed separately.
E. Koga et al., 85 studied large amount of Q factor variation within dense, highly ordered
region of the BZT system by means of crystal structure analysis, micro-structural
analysis and electrical measurements using samples around stoichiometric BZT. The
influences of composition deviation from the stoichiometric BZT on the structural order
and the Q factor were studied and the findings listed as below:
1. The crystal structure phases strongly depend on the slight composition deviation
from the stoichiometric BZT.
2. Single phase ordered perovskite is obtained only in the vicinity of the
stoichiometric BZT. [In this region, an improvement in Q factor (Q×f = 133,000
GHz was found in extended sintering up to 400 h at 1400C.]
3. In the other regions, the single phase of the disordered perovskite or the ordered
perovskite with the secondary phase is formed. [In these regions the Q factor was
found to be low.]
4. Presence of small amount of secondary phase (e.g., less than 1%) affects Q factor
to decrease. This suggests that the structural order and the presence of the
secondary phase play an important role on Q factor in the BZT system. Therefore,
suppressions of small amount of secondary phase (e.g., less than 1%) and increase
of ordering ratio by strict composition control can provide further Q factor
improvements in the BZT system.85
Min-Han Kim et al.,86 however, investigated that the increase in the Q value was not
related to the relative density or the 1:2 ordering, in contrast, the improvement in the Q
value occurred in specimens in which grain growth occurred.
M. Thirumal and P. K. Davies87 have reported the synthesis of hexagonal perovskite
Ba8ZnTa6O24 in single-phase by solid-state technique. High-density ceramics of
Ba8ZnTa6O24 was prepared at temperatures considerably lower (1400oC) than those
used to sinter pure Ba(Zn1/3Ta2/3)O3, and that exhibited very good microwave dielectric
properties with εr =30.5, Q x f = 62300 at 8.9 GHz; however, additives would be needed
to lower the temperature coefficient τf = +36 ppm/oC.87
Chapter 1Chapter 1Chapter 1Chapter 1
32
Use of Dopants
The use of additives is seen to improve the microwave dielectric properties of ceramics;
however, their effect on the BZT system is sometimes tricky. For an example, addition
of ZrO2 [i.e. Ba(Zn1/3Ta2/3)O3 + xZrO2 with 0.0 ≤ x ≤ 4.0 mol%] was investigated16 on
crystallographic order, microstructure, and microwave dielectric properties of BZT
ceramics. A small amount of ZrO2 disturbed the 1:2 cation ordering. The average grain
size of the BZT significantly increased with the addition of ZrO2. The relative density
increased with the addition of a small of ZrO2, but it decreased when the ZrO2 content
was increased. Variation of the dielectric constant with ZrO2 addition ranged between
27 and 30, and the temperature coefficient of resonant frequency increased abruptly as
the ZrO2 amount exceeded 2.0 mol%. The Q value of the BZT significantly improved
with the addition of ZrO2, which could be explained by the increased relative density
and grain size. The maximum Q x f value achieved in this investigation was ≈164000
GHz for BZT with 2.0 mol% ZrO2 sintered at 1550oC for 10 hours.16
Nomura and co-workers successfully9 prepared dense ceramics comprised BMT, BZT
with a small amount of Mn addition. The Mn doping not only favourably affected
sinterbility in lowering the required sintering temperature but also advantageously
heightened the unloaded Q value of the resulting sinter in an SHF band. They also
applied this doping method to Ba2Ti9O20 and Zr0.8Sn0.2TiO4 ceramics.9 The values of ε,
Q and τf in 10 GHz band are listed in Table 1.4.
Table 1.4: Values of ε, Q and τf of doped perovskite-type ceramics.9
Compound εεεε Q τf
(ppm/oC)
Remarks
BMN 32 5600 33 Mn 2 mol % , 9.9 GHz
BMT 25 16800 4.4 Mn 1 mol % , 10.5 GHz
BZN 41 9150 31 Anneal in N2, 9.5 GHz
BZT 30 14500 0.6 Mn 1 mol % , 11.4 GHz
BMnN 39 100 27 9.3 GHz
BMnT 22 5100 34 Anneal in N2, 11.4 GHz
SMN 33 2300 -14 Mn 2 mol % , 10.3 GHz
SZN 40 4000 -39 9.2 GHz
Chapter 1Chapter 1Chapter 1Chapter 1
33
It is interesting to note that the Ba compound has a positive τf, while the Sr has a
negative one, which corresponds to a negative and a positive temperature coefficient of
the dielectric constant, respectively. The most important result is that the Q values of
BMT and BZT ceramics with 1 mol % Mn exceed 104 in 10 GHz band. Generally, the ε
was not so sensitive to the Mn doping, while the Q showed a strong dependence on it. A
typical example is shown in Fig. 1.18.
Fig. 1.18: Q vs. Mn Concentration.9
The addition of small amount of BaZrO315or ZrO2
16 has been found to decrease the
ordering but increase the Q. Similarly, Ni-doped Ba(Zn1/3Ta2/3)O3, is commonly used
for many microwave materials as it exhibits excellent high frequency properties with a
high dielectric constant (εr ~ 30), a low loss tangent (<2×10−5 at 2 GHz) and has a near-
zero temperature coefficient of resonant frequency (τf).88 Zr is often alloyed with this
compound since it has been found to improve the manufacturability of the material by
allowing shorter annealing times to achieve low loss.89 The doping by BaWO4 has been
reported90to increase the sinterability of BZT ceramics. The highest Qxf values (15,000-
200,000 GHz) were obtained at 0.5-1.5 mol% BaWO4 in samples sintered at 1570-
1580oC for 3h in air. BZT doped with 5 mol% SrGa1/2Ta1/2O3 (SGT)91 has a temperature
Chapter 1Chapter 1Chapter 1Chapter 1
34
coefficient of the resonant frequency (τf) tunable through zero, a dielectric constant of
29 and a unloaded quality factor (Q) of 50–80,000 at a resonant frequency (fo) of ≈2
GHz(Q x fo=100–150,000 GHz).
Nanosized phase pure powder of Ba(Zn1/3Ta2/3)O3 (BZT) were obtained by sol–gel
method and subsequent pyrolysis of the dried gel.92 The powders were pyrolyzed at
different elevated temperatures and the particle sizes were found to increase with
temperature. To improve the sintered density, several sintering aids/dopants, in the form
of oxides of metallic elements were added by 1 mol%, to BZT nanopowders and the
powders were subsequently sintered. The effect of different dopants on the microwave
dielectric properties of BZT can be better explained from the following experimentally
observed plots.
Fig. 1.19: Variation of Q x f and τf with ionic radius of the dopants.92
Dopants like Ti, Zr and Sb, showed a reduction of Q x f whereas the dopants like Cr,
Ga, Mn, In, Ce and Bi showed an improvement in Q x f. All dopants were found to
reduce temperature coefficient of resonant frequency (τf) as can be seen in Fig.19.
When the average ionic radius of the dopant was about 0.615Å, τf was found to be zero.
Most of the dopants were found to promote grain growth of sintered microstructure.92
Temperature has been found one of the decisive factors to increase the degree of order
whereby improving/affecting Q. Therefore, knowing the maximum temperature at
which BZT orders prior to disordering is of technological importance. I. M. Reaney et
al., 93 studied and performed a series of quenching experiments to ascertain the
Chapter 1Chapter 1Chapter 1Chapter 1
35
approximate temperature of the order–disorder phase transition of BZT. The XRD (Fig.
1.20) revealed in their experiments a sudden decrease (maximum to zero) in the degree
of order in the BZT samples between 1600 and 1625°C. Hence it was assumed that a
reversible order–disorder phase transition in BZT had occurred between 1600 and
1625°C.
Fig. 1.20: XRD patterns from BZT (a) sitered at 1475oC and annealed & quenched from (b)1575oC, (c)1600oC and (d) 1625oC.93, 94
In another example of solid-state reaction, the BZT ceramics being sintered at 1400-
1600oC for 4 hours had low loss in microwave domain.95The increase of the sintering
temperature was found to lead to a normal granular growth, with a polyhedral grains
size development.96 For sintering temperatures higher than 1500°C, the XRD (Fig. 1.21)
patterns revealed the BZT multiphase compositions with the presence of trigonal
supercell peaks and low Zn content secondary phases. The last two disappeared at
temperatures higher than 1600°C. 96
The formation of secondary phases can introduce point defects into the matrix of
sintered body. The probable defects are vacancies at the A and/or the B sites and
interstitials in the perovskite structure.74
Chapter 1Chapter 1Chapter 1Chapter 1
36
Fig. 1.21: XRD patterns of Ba(Zn1/3Ta2/3)O3 system vs. sintering temperature.96
The analysis of the order–disorder transition of the BZT material showed a long-range
order with a 2:1 ratio of Ta and Zn cations on the octahedral positions95 of the
perovskite structure with the increase of the sintering temperature. There is a strong
correlation between the cation ordering, domain growth, zinc loss and sintering
parameters. The investigations on dielectric properties revealed 96 that the increase of
the sintering temperature had decreased the dielectric loss especially when an annealing
of 10 hours at 1400°C was performed. The quality factor Q was observed to depend
strongly on the BZT crystalline structure, i.e. on the unit cell distortion and cationic
order. Dielectric loss decreased with the increase of ordering degree in the structure and
Chapter 1Chapter 1Chapter 1Chapter 1
37
with the disappearance of secondary phases. Lowest loss was obtained for a Zn and Ta
completely ordered BZT ceramic with a strongly distorted unit cell. Porosity had small
effect on dielectric loss of BZT material.
As a matter of fact the solid state reaction of BaCO3, ZnO, and Ta2O5 is not the most
appropriate method, because a high sintering temperature is required to achieve high-
density Ba(Zn1/3Ta2/3)O3 (BZT) materials (>1500C),which is too high for industrial
applications.13,15 Many efforts have been made to lower the sintering temperature of
BZT ceramics by the addition of sintering agent 97 Wen-Cheng Tzou et al. used BZT
powder, calcined at 1200C for 3 h, as the precursor and 0.1 mol BaTi4O9 (BT4) to
improve the sintering characteristics of BZT ceramics.
Fig. 1.22: (a) The bulk densities and the dielectric constants of BT4-BZT ceramics, as a function of sintering temperature; (b) The quality factor(values) and temperature coefficients of resonant frequency of BT4-BZT ceramics, as a function of sintering temperature.
The quality values (Q × f) of BT4-BZT ceramics were investigated as a function of
sintering temperature [results shown in Fig. 1.22(b)]. As the sintering temperatures were
increased from 1240 to 1320C, the Q × f values also increased critically. The Q × f
values of BT4-BZT ceramics reached a saturation value of 68500 at 1320C sintered
ceramics, then the Q × f values of BT4-BZT ceramics slightly increased with the further
increase of sintering temperature to 1360C. The τf values of BT4- BZT ceramics are also
shown in Fig. 1.22(b). As the sintering temperatures were changed from 1240 to1320C,
the τf values of BT4-BZT ceramics linearly changed from −1.5 ppm/C to 4.1 ppm/C.
(a) (b)
Chapter 1Chapter 1Chapter 1Chapter 1
38
Thus the addition of 0.1 mol BaTi4O9 into the Ba(Zn1/3Ta2/3)O3 composition would
improve the sinterability of Ba(Zn1/3Ta2/3)O3 ceramics. The needed sintering temperature
of BT4-BZT ceramics was about 1320C which is much lower than that of
Ba(Zn1/3Ta2/3)O3 ceramics. The BT4-BZT ceramic sintered at 1320C has the microwave
dielectric property of εr = 31.5, Q × f= 68500, and τf = 4.1 ppm/C.98
A technological process has been investigated97 with respect to milling conditions,
calcinations parameters and also lithium salt (Li2CO3, LiF, BaLiF3 and LiNO3)
additions so as to lower the sintering temperature of BZT. First, milling conditions
investigation permits to optimise powder reactivity authorising a sintering temperature
reduction of nearly 100C. Secondly, the lowering of the ceramics sintering temperature
has been explored using the lithium salt additions. The addition of BaLiF3 permitted to
reach a sintering temperature of around 1200C. Permittivities of 29 for the LiF added
sample and 38 for the BaLiF3 added sample were measured with (τε(LiF) =137 ppmC−1
and τε(BaLiF3) =156 ppmC−1, respectively). Thus except for the temperature coefficient
which was increased, the addition of lithium salt did not modify or alter the dielectric
properties (tanδ and ε).97 A drastic decrease of BZT sintering temperature had been
reported99 by lithium salts and glass phase addition. A. Chaouchi et al. studied BZT
perovskite with the objective to decrease their sintering temperature close to 900oC. The
different types of sintering aids (glass phase combined or not with lithium salts) were
employed in order to decrease the BZT sintering temperature below the silver melting
point (961oC). The glass phases based on the SiO2 – ZnO – B2O3 (ZSB) system and LiF
lithium salt were used as the sintering agents which showed a good efficiency on the
sintering temperature of BZT material. The combined addition of 5% wt. of ZSB and
1% wt. of LiF to BZT showed a significant lowering of sintering temperature. The
materials achieved good dielectric properties: εr = 32, τε = -10 ppm/oC, low losses factor
(Tan δ < 10-3), high resistivity (1013Ω.cm). This is partially explained by the fact that
glass added samples had higher density.99
Despite of various efforts, the detailed mechanism of high Q factor of BZT system has
not been fully understood, namely, quite large amount of Q factor variation within
dense, highly ordered single phase region of the BZT system. If we could understand
Chapter 1Chapter 1Chapter 1Chapter 1
39
the origin of this large amount of the Q factor variation, then we would suggest methods
for further Q factor improvements of the BZT system. Depending on the conditions of
synthesis, both ordered and disordered-type were formed in case of Ba(Zn1/3Ta2/3)O3
(BZT) and Ba(Mg1/3Ta2/3)O3 (BMT).13 Thus the analysis of the exact relationships
between the composition and the dielectric properties is necessary.
There have been many attempts to explain the material’s excellent microwave dielectric
properties and further to improve its properties. A clear relationship between the
composition and properties will be helpful in investigation of new ceramics with less
costly raw materials. Although tantalate high Q perovskites such as BZT and BMT
have been commercialized, in the last few years, economic factors associated with the
high cost of Ta2O5 have increased the focus on their niobate, a low cost counterpart.
The low cost materials with high performance are in urgent need for development.
1.8.1.2. Barium magnesium tantalate Ba3MgTa2O9 or Ba(Mg1/3Ta2/3)O3 (BMT)
Nomura and co-workers12 reported Ba(Mg1/3Ta2/3)O3 in 1982. Since then there have been
several reports of BMT by solid sate method e.g. H. Matsumoto et al., 100, S.Nomura, 9 M.
Sugiyama et al., 101; M.-H. Liang, et al, 102; wet chemical methods e.g. O. Renoult, et al.;
103, S. Katayama et al., 104 sol-gel powders for microwave dielectric resonators, Ceramics
today- tomorrow ceramics105 P.Vincezini.(Ed.,) Elsevier Publishers B.V. (1991-1997).
There are several reports on the microwave dielectric properties of BMT with ε 24-25,
Q x f upto 430000GHz and τf close to zero ppm/0C.
1.8.2. Niobium (Nb)-based Systems: BZN, BMN, SZN
In A(B’1/3B”2/3)O3, the B-site substitution of the identical ionic radius106 (0.78 Å) Nb5+
for Ta5+ results in crystallization of isostructural 1:2 ordered perovskites; however, the εr
is higher, Q is lower, and τf is more positive. The origins of the dielectric property
differences between the isostructural compounds are not well understood.11
Entanglement of synthetic variables (e.g., chemical composition, reagent purity,
annealing and sintering temperatures and times, initial particle size, processing
conditions, and partial O2 pressure) and experimental observables (e.g., crystal structure,
Chapter 1Chapter 1Chapter 1Chapter 1
40
density, cation order inside of a domain, ordered domain size, domain boundaries,
defects, and impurity phases) that influence the dielectric properties (i.e., εr , τf, and Q)
complicate formation of structure-property relationships.
Nb-based perovskites are looked upon as candidates with comparable dielectric
properties to substitute the expensive Ta-based perovskites Ba(Zn1/3Ta2/3)O3 (BZT) and
Ba(Mg1/3Ta2/3)O3 (BMT). The Nb-based ceramics, e.g. Ba(Zn1/3Nb2/3 )O3 (BZN), and
Sr(Zn1/3Nb2/3)O3 (SZN) are cheaper than Ta-based complex perovskite ceramics and
have simple perovskite cubic structures (pm 3– m) with lattice constant a = 4.094 Å at
room temperature.107 The next important Nb-based ceramic is Ba(Mg1/3Nb2/3)O3
(BMN) which has interesting dielectric properties (εr = 32, τf = 33 ppm/C, and Q x f =
56 THz at 10 GHz) but because of its high sintering temperature (Ts) >1500C it is less
used than BZN, which exhibits a lower Ts (1350C) and similar dielectric properties
(εr = 41, τf = 30 ppm/C, Q x f = 54 THz at 10 GHz).108 However, the resonance
frequency temperature coefficient (τf) of the BZN ceramic is relatively high (30
ppm/C), and this limits BZN ceramics also to be used in microwave applications.109
The analysis of the permittivity (εr) and intrinsic loss110 of 1:2-ordered BZN and BMN
shows that the difference in permittivity between BZN and BMN is caused by the
significant influence of the dipole interaction on the nucleus oscillation in BZN.
Furthermore, such influence is attributed to the covalent interaction among Zn, O, and
Nb which is correlated with the Zn 3d orbital. The investigation of intrinsic loss shows
that the measured loss is close to the intrinsic loss in both BZN and BMN, and that the
difference in the measured loss is mainly caused by the difference in the intrinsic loss.
Since the intrinsic loss increases rapidly with the permittivity, it seems that the
difference in permittivity between BZN and BMN is the origin of the difference in the
measured loss.110
Research works are in progress to obtain the near-zero temperature coefficient of resonant
frequency (τf) and also improve other dielectric properties of BZN. A good combination
of microwave dielectric properties: εr=36, Q x f =16,170 GHz, and τf = −12 ppm/°C has
been reported111 in the composition (Ca1−xBax)(Zn1 / 3Nb2 / 3)O3 when x =0.1.
Chapter 1Chapter 1Chapter 1Chapter 1
41
Fig. 1.23: τf and εr values variation as a function of x for (Ca1−xBax)(Zn1 / 3Nb2 / 3)O3 ceramics.111
The τf value was found to increase sharply corresponding to the increased εr at x= 0.1,
where the primitive cubic structure of Ba(Zn1/3Nb2/3)O3 distorted to hexagonal structure.
For x=0.3–0.9, the τf value decreased when the Ca-rich secondary phase appeared. The
secondary phase with orthorhombic structure as Ca (Zn1/3Nb2/3)O3 (τf = −35 ppm/°C)
had a negative effect on the τf value. The rattling of Ca2+ and the cell volume variation
led to the anomalous variation of εr and τf in the present system for the Ba-rich
compositions where the decreased Q x f value was obtained.111
Various methods have been applied to improve the properties of BZN ceramics112 such as
doping with different elements like Ti 113or making composite of BZN ceramics with other
Ba(B’1/3B”2/3)O3 microwave dielectric ceramics (B’ = Mg, Zn, Ni; B” =Ta, Nb) like
Ba(Zn1/3Nb2/3)O3 – Ba(Ni1/3Nb2/3)O3.114 0.35[Ba(Ni1/3Nb2/3)O3] – 0.65[Ba(Zn1/3Nb2/3)O3]
composite microwave dielectric material which was sintered at 1450oC for 4 hours and
annealed for at 1300oC for 72h exhibited good dielectric properties: τf = +0.6ppmoC-1, εr =
35 and quality factor (Q) in excess of 25,000GHz.114 Various amounts of different dopants
such as oxides of monovalent, divalent, trivalent, tetravalent, pentavalent and hexavalent
elements were employed in the synthesis of BZN. Effect of these dopants on microwave
dielectric properties of BZN was investigated.115 Some of the dopants were found to
increase quality factor Q×f and slightly alter the temperature coefficient of resonant
Chapter 1Chapter 1Chapter 1Chapter 1
42
frequency (τf). The quality factor was found to depend on the dopant ionic radii and its
concentration. The quality factor was observed to increase when the ionic radius of the
dopant was close to the ionic radii of the B site ions Zn and Nb.115 In a synthesis of Bi, Ce
and In doped Ba(Zn1/3Nb2/3)O3(BZN), by conventional mixed oxide technique, the effect
of each dopant on the sintering temperature and dielectric properties was investigated.116
‘In’ doping between 0.2 and 4.0 mol% increased the density of BZN at 1300oC, but Ce
doping caused a decrease in density at 1250oC. Levels of Bi2O3 up to 1.0 mol% had
negative effect on densification, while high level doping significantly improved the
densification of the specimens. It was found that In, Ce and Bi doping resulted in single
phase formation at all concentrations, except 0.5 mol% Bi. In and Ce doping increased
the dielectric constant from 41 to around 66 at 1MHz, however, Bi doping decreased
the dielectric constant to about 37 at 0.2 mol%, and then higher doping116 led to
increase to about 63.
Calcinations conditions also play an important role in the formation of an ordered phase
and on the quality factor (Q) of BZN. Table 1.5 summarizes the microwave dielectric
properties of BZN ceramics under various calcination and sintering conditions. The
relative permittivity varies little around 35–41, but the Qxf values were significantly
affected by the calcinations and sintering conditions.
Table 1.5: Microwave Dielectric Properties of BZN Ceramics under different Calcination & Sintering Conditions 117
Calcination condition
Sintering condition
Relative density
(%)
Second phase εr Q x f (GHz)
1000oC/6h
1250oC/4 h 97.2 None 39.4 5749 1300oC/4 h 97.6 None 40.9 7094 1350oC/4 h 97.8 None 40.2 8915 1350oC/48 h 96.2 Ba5Nb4O15 39.2 11560 1350oC/4 h† 97.5 None 40.1 3720
1250oC/12h 1350oC/4 h 85.5 Ba5Nb4O15 35.5 9027 1450oC/4 h 96.5 Ba5Nb4O15 39.9 66278 1550oC/4 h 93.8 Ba5Nb4O15 38.2 41639
1000oC/6h and
1250oC/12h
1300oC/4 h 96.7 None 40.6 47101
1350oC/4 h 96.4 None 39.4 112286
Double calcinations
1400oC/4 h 95.4 None 41.1 69675 1350oC/48 h 95.0 Ba5Nb4O15 39.7 236 753
1350oC/48 h‡ 93.8 Ba5Nb4O15 37.7 67 714
†Annealing at 1250oC/12 h after sintering of 1350oC/4 h. ‡Without muffling powder during sintering. BZN, barium zinc niobate (Ba(Zn1/3Nb2/3)O3).
Chapter 1Chapter 1Chapter 1Chapter 1
43
For BZN specimens single calcined at 1000oC/6 h and sintered for 4 h, there was no
second phase formation, and all had a high relative density (>97%) but relatively low
Q x f values. The Q x f value did increase at higher sintering temperature (1250o–
1350oC) but not by much. Prolonged soaking time (1350oC/ 48 h) also increased the Q
x f value slightly. For sintered BZN specimens after 1250oC/12 h of single calcination,
the relative density decreased due to second phase formation, but the Q x f value
increased markedly. For the specimens double calcined at 1000oC/6 h and 1250oC/12 h,
the Q x f values became much higher than those with single calcination. The highest Q x
f of 236,753 GHz was observed for the specimen sintered at 1350oC/48 h. Raising the
sintering temperature (1300o–1350oC) or prolonging the soaking time could increase the
Q x f value, but not when sintered at 1400oC or without powder muffling during
sintering to prevent ZnO loss.
Fig. 1.24: Dark-field transmission electron microscopy (TEM) images of the barium zinc niobate (BZN) specimen after double calcinations and sintering at (a)1350oC/4h and (b)1350oC/48h.117
Figure 1.24 shows a dark-field TEM micrograph for the sintered specimen of double-
calcined BZN powder (at 1000oC/6 h and 1250oC/12 h). The dark regions did not
correspond to the disordered regions but rather to the ordered domains whose super
lattice reflections were not enclosed by the objective aperture of TEM. Figure 1.24(a)
shows that ordered domains, typically 50–100 nm in size, were present in the specimen
of a shorter soaking time (4 h). The ordered domains grew by an order of magnitude
Chapter 1Chapter 1Chapter 1Chapter 1
44
when the soaking time was increased from 4 to 48 h at 1350oC, as shown in Fig. 1.24(b)
of lower magnification. Apparently, the synthesis condition had a pronounced effect on
the growth of ordered domain in the sintered BZN specimen. Thus by examining the
effects of calcination condition on the ordered structure and the quality factor for BZN
microwave dielectric ceramics, the above study verified that the calcination condition
had played an important role in the formation of an ordered phase and the growth of an
ordered domain. The quality factor was found closely related to the size of ordered
domain, the degree of 1:2 ordering, and relative density. A BZN specimen sintered at
1350oC/48 h from the double-calcined powder possessed an εr of 39.7, a τf of
15 ppm/oC, and a Q x f value of 236753 GHz117
As a part of investigation of BZN-based ceramics, cooling rate after sintering has also
been found a factor to affect dielectric properties significantly. R. Freer et al., examined
the effect of the cooling rate (2°C–240°C/h) after sintering on the microwave dielectric
properties of the ceramics 0.35Ba(Ni1/3Nb2/3)O3–0.65Ba(Zn1/3Nb2/3)O3 being prepared
by mixed oxide route. The perovskite ceramics Ba(Zn1/3Nb2/3)O3 (BZN) and
Ba(Ni1/3Nb2/3)O3 (BNN) exhibited similar relative permittivities (≈40 and 29,
respectively), but τf values of opposite polarity. When BZN and BNN were combined
in 0.65:0.35 ratio (i.e., 0.35Ba(Ni1/3Nb2/3)O3 0.65Ba(Zn1/3Nb2/3)O3, a temperature-stable
product was achieved, with τf ≈ 0 ppm/ °C, although the dielectric losses varied
significantly with processing conditions.114
The variation in the microwave dielectric properties of 0.35BNN-0.65BZN ceramics
was examined as a function of cooling rate after sintering. While the extrinsic factors
(porosity, Inhomogeneity) were found to dominate the dielectric properties in the low-
frequency regime, they had very limited effect in the terahertz frequency regime.
A slower cooling rate resulted in enhanced Q x f values for the materials, but had
comparatively little effect on the relative permittivity. This correlates very well with the
increase in B-site occupancy, while the cell volume changed little as the cooling rate is
decreased. These phenomena are consistent with the facts that, as the sample cools
more slowly after sintering, the Raman shifts are changed insignificantly.118
Chapter 1Chapter 1Chapter 1Chapter 1
45
In an another example, solid solutions of the (x)Ba(Zn1/3Nb2/3)O3 – (1-
x)Ba(Co1/3Nb2/3)O3 (BZCN) system were produced by the mixed oxide route.119
Compositions near to 0.6[Ba(Zn1/3Nb2/3)O3]– 0.4[Ba(Co1/3Nb2/3)O3] exhibited zero
temperature dependence of resonant frequency (τf ≈ 0).Sintering temperatures of 1300–
1500oC with cooling rates of 3–180oC/h were used. Sintered densities of ≤95% were
achieved. A second phase deficient in Co and Zn was detected. Slow cooling after
sintering, down to 900oC, induced B-site ordering which increased the dielectric Q
values. The material would offer a new design option for filter engineers in the
development of 3G and other similar telecommunication systems. BZCN ceramics
would offer a high Q solution for base station filtering at a greatly reduced cost
compared to BZT ceramics.119
Small deviations in the stoichiometry of Ba(Zn1/3Nb2/3)O3 (BZN) and other related
microwave perovskites have a profound effect on their ordering, sintering, and
dielectric properties. The cubic perovskite structure is well known to accommodate
large degrees of non-stoichiometry and vacancy formation on the A-site and anion
positions. While the B sites readily accept multiple cations and the degree of vacancy
formation on this position is limited to concentrations less than 2%. (The exact relations
between the B-site ions and the ceramics dielectric properties are still unclear.120 The
bulk crystal chemistry of BZN has been extensively investigated, and it undergoes a
transition from a 1:2 ordered to disordered B-site arrangement at 1375oC.25,76,121-123
Much higher Q factor has been observed and reported in the disordered phase sintered
below 1350C than the ordered-phase sintered above1350C in case of Ba(Zn1/3Nb2/3)O3
(BZN), unlike BZT.124 The influence of the crystal structural ordering, microstructure,
secondary phase and lattice defect has been investigated on the high Q factor of the
BZN, using series of the samples with their compositions near the stoichiometric
BZN.125 The observations can be listed as:
1. The structural order and the crystal phases depended on the slight composition
deviation from the stoichiometric BZN and on the heat treatment.
2. For the stoichiometric BZN, the order-disorder phase transition occurred between
1300oC and 1400C (as evident from the plots in the Fig. 1.25).
Chapter 1Chapter 1Chapter 1Chapter 1
46
3. Regardless the crystal structural ordering the stoichiometric BZN with high
density and large grain size exhibited the significant high Q factors (Q x f) ≈ 9-10
× 104 GHz). This result suggests that the pore and grain boundary around the
crystal grain should affect the variation of the Q factor than the crystal-structural
ordering.
4. In the other regions, by the slight composition deviation from stoichiometry, the
ordered BZN with the secondary phase or the non-stoichiometric disordered BZN
with low density were formed.
Fig. 1.25: Q x f, density and average grain size dependence of the stoichiometric BZN as functions of various heat treatments. Ord:Ordered perovskite(trigonal), Dis:Disordered perovskite (Cubic). 125
Chapter 1Chapter 1Chapter 1Chapter 1
47
Since the Q factors were found to be low, the presence of the secondary phases, the
lattice defect and low density were considered to be causes of the decrease of the Q
factors in this system. From these results for BZN system, the highly dense ceramics
with large grain size in the microstructure and the suppressions of the secondary phase
and of the lattice defect by strict composition control would provide the further Q factor
improvements in BZN system.125 Thus the influence of the crystal-structural ordering
and the microstructure on the high Q factor seems far from fully understanding.
The dielectric loss properties of perovskite microwave materials are notoriously
sensitive to a number of different processing and structural variables. The losses can
vary from sample to sample even when they have the same nominal composition due to
small differences in the intrinsic crystal structure, microstructure, density, and impurity
concentration. In an investigation126 2 wt. % of ZnO–SiO2–B2O3 glass phase and
1 wt.% of LiF-added BZN sample sintered at 900oC exhibited a relative density higher
than 95% and attractive dielectric properties: a dielectric constant (εr) of 39, low
dielectrics losses (tan(δ) < 10-3) and a temperature coefficient of permittivity (τε) of 45
ppm/oC. The 2 wt.% ZnO–SiO2–B2O3 glass phase and 1 wt.% of B2O3-added BZN
sintered at 930oC showed also attractive dielectric properties (εr=38, tan(δ) < 10-3) and
it was more interesting in terms of temperature coefficient of the permittivity (τε = -5
ppm/oC).126 The lowest microwave losses have been reported in alkaline-earth based
niobate and tantalate members of the’’1:2’’ A2+ (B1/32+B2/3
5+)O3 family of mixed-metal
perovskites (A2+ = Ca, Sr, Ba; B2+ = Mg, Ca, Sr, Mn, Fe, Co, Ni, Cu, Zn; and B5+ = Nb,
Ta).89 The ionic radii of Sr2+ is 0.116 nm and that of Ba2+ is 0.136 nm127, the difference
between them are less than 15%, moreover these two ceramics have opposite signal of
the temperature coefficient of the capacitance (τc); negative and positive respectively.
Therefore, a temperature stable solid solution can be fabricated by the substitution of Sr2+
to Ba2+. A study 107 of the microwave dielectric properties of (Ba1-xSrx)(Zn1/3Nb2/3)O3
(BSZN) solid solutions revealed that the samples with (Ba0.35Sr0.65)(Zn1/3Nb2/3)O3 and
(Ba0.30Sr0.70)(Zn1/3Nb2/3)O3 had near zero τc values (-6.9 x 10-6 and -2.1 x 10-6ppm/oC
respectively), suitable dielectric constants (εr ≈ 42.4 and 39.4 respectively) and
relatively high values of Q (3000 and 3365, respectively at about 3.85 GHz), indicating
the suitability of these ceramics for the applications as microwave components.
Chapter 1Chapter 1Chapter 1Chapter 1
48
Another family of perovskites128 with the general stoichiometry (A2+2/3La1/3)
(B1+1/3B
5+2/3)O3
at 1 MHz (Sr2/3La1/3) (Li1/3Nb2/3)O3 showed εr =30, τc=146 ppm/°C;
(Sr2/3La1/3) (Li1/3Ta2/3)O3 εr =25 and τc=156 ppm/°C. Whereas at microwave
frequencies (Sr2/3La1/3) (Li1/3Nb2/3)O3 showed εr =29, and τf = -75.5 ppm/°C and quality
factor (Q x f ) = 6300 at 8.9 GHz; (Sr2/3La1/3) (Li1/3Ta2/3)O3 showed εr =25, and τf = -
25.1 ppm/°C and Q x f = 25 200 at 10.2 GHz. Measurements on (Ca2/3La1/3)
(Li1/3Nb2/3)O3 at microwave frequencies yielded Q x f =26 500 at 8.7 GHz, εr =30 and τf
= -25.7 ppm/°C. While the quality factors of these dielectrics are not as high as those of
the known 1:2 ordered high Q perovskite microwave dielectrics, they are quite
respectable and their negative values of τf coupled with their lower sintering
temperatures may make them attractive candidates for tuning the properties of the BZN
(or BZT)-type systems.128
Ba(Mg1/3Nb2/3)O3 (BMN) is known as a good dielectric material for a microwave
resonator9 since has εr of 32, Q value of 5600 at 10.5 GHz, and τf of 33 ppmoC-1.
Although this resonator has a high dielectric constant, its Q value is relatively lower
than those of other complex perovskites. In order to enhance the Q value, additives were
often used.129 However, the additives form a second phase, which is sometimes
detrimental to microwave dielectric properties. One way of improving the Q value is to
enhance the sinterability of the specimen through the change of the materials
stoichiometry.130 Paik et al. investigated130 the sinterability and microwave dielectric
properties of Mg-deficient BMN specimens. Mg deficiency has a significant effect on
the microwave dielectric properties. An improvement in both density and the dielectric
Q value when x = 0.02 in Ba(Mg1/3−xNb2/3)O3 has been reported. A 1:2 ordering of B-
site ions in a complex perovskite compound is formed due to the differences in size and
charge valencies of the two B-site ions.24 Furthermore, a 1:2 ordering produces the
deviation of c/a ratio from the ideal value for hexagonal structure (√3/√2 = 1.2247),
which introduces the lattice distortion in the hexagonal unit cell. The lattice distortion
can cause the splitting in (422) and (226) reflections.130 It has been found that 1:2
ordering and lattice distortion are sensitive to the Mg deficiency.
SEM results (Fig. 1.26) indicate that grain growth is not completed in the specimens
with x>0.02. Therefore, it is suggested that a small number of Mg vacancies enhances
Chapter 1Chapter 1Chapter 1Chapter 1
49
the sinterability, which leads to the increase of relative density of the specimen.
However, a large number of Mg vacancies inhibits the sintering and thus decreases the
relative density of the specimen.
Fig. 1.26: SEM micrographs130 of Ba(Mg1/3-xNb2/3)O3 at: (a) x = 0, (b) x = 0.01, (c) x=0.02, (d) x= 0.03, (e) x= 0.04.
The variation of dielectric constant with Mg deficiency is not significant. For all the
specimens, it is about 32. On the other hand, the temperature coefficient of resonant
frequency decreases from 32 to 22 as x varies from 0 to 0.03. However, for the
specimen with x =0.04, it increases to 45. The variation of τf with x is not clear. The Q
values are normalized values to a frequency of 8 GHz. The variation of Q value with
Chapter 1Chapter 1Chapter 1Chapter 1
50
Mg deficiency is very similar to that of relative density and lattice distortion. Therefore,
the improvement of Q value with Mg deficiency could be related to the increase of both
relative density and 1:2 ordering. The highest Q value achieved in this investigation was
about 12 000 at 8 GHz for Ba(Mg1/3-xNb2/3)O3.
Young-Woong Kim et al.,19 investigated the effect of heat treatment on the cationic
ordering behaviour in Ba(Mg1/3Nb2/3)O3 (BMN) complex perovskite microwave
ceramics. The effect of sintering temperature and time on the cationic ordering in the
BMN ceramics was studied by using transmission electron microscopy (TEM) and
energy-dispersive spectroscopy (EDS). The relationship between the local structure and
chemical composition, and the effect of the structural variation on microwave dielectric
properties of the ceramics were studied.
Table 1.6. Microwave dielectric properties of BMN ceramics sintered with various sintering conditions.19
Sintering condition
Dielectric constant
Quality factor (Q x f: GHz)
τf
(ppm/oC) Sintered density
1350oC - 4h 31.2 46,000 +18 96.9
1500oC - 4h 31.5 15,300 +20 96.0
1350oC - 40h 31.0 28,600 +18 96.0
The BMN ceramics show very diverse local cationic ordering behaviour according to
the sintering conditions, such as the development of domain twinning, “core-shell”
structured grains and local disordered structure, though having 1:2 cationic ordering
structure basically. When the BMN ceramics were sintered at 1350oC for 40 h and
1500oC for 4 h, a local order-disorder transition was observed in Mg-excess regions,
especially where the ratio of Nb/Mg is less than 1.8. It is postulated that the formation
of BaNb2O6-like second phase at grain boundary triple points resulted in local
compositional change in the 1:2 ordered matrix grain and thus caused the local
disordering. The microwave dielectric quality factor of the ceramics decreases greatly
with the increase of the structural and chemical inhomogeneity.
Chapter 1Chapter 1Chapter 1Chapter 1
51
Aim of the Thesis
The objective of the thesis is to understand how the dielectric properties in the
microwave dielectrics are affected by the processing conditions, and by synthesizing
methods and to synthesize and characterize the phase that forms during the syntheisis
and sintering of cubic perovskites.
Chapter 2
This chapter deals exclusively on the various methods of initial calcinations condition
in the synthesis of Ba(Zn1/3Ta2/3)O3 and we had compared between various methods.
• Comparison between Agate grinding and ball milling
• Comparison between various solvents in ball milling
• Comparison between various calcinations time in Agate grinding
• Comparison between various calcinations time in Ball-milling
Chapter 3
Flux method is one of the well known method in the literature, we explored this method
for synthesizing Ba(Zn1/3Nb2/3)O3 and Ba(Mg1/3Nb2/3) and studied the effect of
temperature and soaking time on the 1:2 ordering of these systems.
Chapter 4
In this chapter we studied the solid solution and composites of hexagonal perovskites,
which forms during the sintering of cubic perovskites, the various system studied are:
• (1-x) (Ba8ZnTa6O24) – x (Sr8ZnTa6O24)
• (1-x) [Sr(Zn1/3Ta2/3)O3] – x[Sr(Zn1/8Ta3/4)O3]
• (1-x) [Ba(Co1/3Ta2/3)O3] – x[Ba(Co1/8Ta3/4)O3]
Chapter 1Chapter 1Chapter 1Chapter 1
52
References
1. Kingery, W.D.; Bowen, H.K.; Uhlmann, D.R. Introduction to Ceramics, John
Wiley and sons, New York, 1975; 2nd Edition, pp182-184.
2. Scott, AW. Understanding Microwaves, New York, John Wiley & Sons, Inc
1993.
3. Agrawal, D. Mater. Res. Innovations 2010, 14, 3-8.
4. Takahashi, T. Jpn. J. Apl. Phys. 2000, 39, 5637-5641.
5. Li, M. Ferroelecrrrcs 1997, 195, 87-91.
6. Richtmyer, R.D. Dielectric Resonators; J. Appl. Phys. 1939, 10, 391–398.
7. Reaney, I.M.; Iddles, D. J. Am. Ceram. Soc. 2006, 89, 2063–2072.
8. Higuchi, Y.; Tamura, H. J. Eur. Ceram. Soc. 2003, 23, 2683–2688.
9. Nomura, S. Ferroelectrics 1983, 49, 61-70.
10. Moulson, A.J.; Herbert, J.M. Electroceramics; Chapman & Hall, London, 1990.
11. Lufaso, M.W. Chem. Mater. 2004, 16, 2148-2156.
12. Nomura, S.; Toyama, K.; Kaneta, K. Jpn. J. Appl. Phys. 1982, 21, L624-
L626.
13. Kawashima, S.; Nishida, M.; Ueda, I.; Ouchi, H. J. Am. Ceram. Soc. 1983, 66,
421-423.
14. Desu, S.B.; O′Bryan, H.M. J. Am. Ceram. Soc. 1985, 68, 546–551.
15. Davies, P.K.; Tong, J.; Negas, T. J. Am. Ceram. Soc.1997, 80, 1727.
16. Yang, J-I.; Nahm, S.; Choi, C-H.; Lee, H-J.; Park, H-M. J. Am. Ceram. Soc.
2002, 85, 165-68.
17. Galasso, F.; Katz, L.; Ward, R. J. Am. Chem. Soc. 1959, 81, 820.
18. Thirumal, M.; Ganguli, A.K. Prog. Cryst. Growth and Charact. Mater. 2002,
44, 147-154.
19. Kim, Y.W.; Park, J.H.; Park, J.G. J. Eur. Ceram. Soc. 2004, 24, 1775–1779.
Chapter 1Chapter 1Chapter 1Chapter 1
53
20. Barbers, D. J.; Moulding, K.M.; Zhou, J. J. Mater. Sci. 1997, 32, 1531-1544.
21. Jacobson, A.J.; Collins, B.M.; Fender, B.E.F. Acta Crystallogr.,B 1976, 32, 1083.
22. Galasso, F.; Pyle, J. J. Phys. Chem. 1963, 67, 1561-1562.
23. Galasso, F.; Darby, W. J. Phys. Chem. 1962, 66, 131.
24. Galasso, F.; Pyle, J. Inorg. Chem. 1963, 2, 482–484.
25. Kim, B.K.; Himaguchi, H.; Kim, I.T.; Hong, K. S. J. Am. Ceram. Soc. 1995,
78, 3117–3120.
26. Chen, J.; Chan, H.M.; Harmer, M.P. J. Am. Ceram. Soc. 1989, 72, 593-98.
27. Reaney, I.M.; Petzelt, J.; Voitsekhovskii, V.V.; Chu, F.; Setter, N. J. Appl.
Phys.1994, 76, 2086-92.
28. Kim, I.T.; Kim, Y.H.; Chung, S. J. J. Mater. Res. 1997, 12, 518- 525.
29. Ra, S-H.; Phule, P.P. J. Mater. Res. 1999, 14, 4259 - 4265.
30. Cullity, B.D. Elements of X-ray Diffraction 1978, 2nd ed.; pp 383-95.
31. Chen, J.; Chan, H.M.; Harmer, M.P. J. Am. Ceram. Soc. 1989, 72, 593-98.
32. Hench, L. L.; West, J. K. Principles of Electronic Ceramics, John Wiley &
Sons, Inc. 1990; pp 244-247.
33. Reaney, I.M.; Ubic, R. Int. Ceram. 2000, 1, 48-52.
34. Vanderah, T.A. Science 2002, 298, 1182-1184.
35. Mallinson, P.; Claridge, J.B.; Iddles, D.; Price, T.; Ibberson, R.M.; Allix, M.;
Rosseinsky, M.J. Chem. Mater. 2006, 18, 6227-6238.
36. Ra, S-H.; Phule, P.P. J. Mater. Res. 1999, 14, 4259 - 4265
37. Jiang, J.F.; Blair, W.D. IEEE Microwave Theory Techniques 1998, 46, 2493.
38. Tamura, H.; Sagala, D.A.; Wakino, K. Jpn. J. Appl. Phys. 1986, 25, 787–791.
39. Djuniadi, A. S.; Nambu, S. J. Am. Ceram. Soc. 1992, 75, 2573-75.
40. Penn, S.J.; Alford, N.M.; Templeton, A.; Wang, X.; Xu, M.; Reece, M.;
Scrapel, K. J. Am. Ceram. Soc. 1997, 80, 1885-1888.
Chapter 1Chapter 1Chapter 1Chapter 1
54
41. Lin, I-N.; Chia, C-T.; Liu, H-L.; Chen, Y-C.; Cheng, H-F.; Chi, C-C. J. Eur.
Ceram. Soc. 2003, 23, 2633–2637.
42. Petzelt, J.; Pačesová, S.; Fousek, J.; Kamba, S.; železný, V.; Koukal, C.;
Schwarzbach, J.; Gorshunov, B. P.; Kozlov, G.V.; Volkov, A. A. Ferroelectrics
1989, 93, 77.
43. Rupprecht, G. O.; Bell, R. O. Phys. Rev. 1962, 125, 1915.
44. Silverman, B. D. Phys. Rev. 1962, 125, 1921.
45. Gurevich, V. L.; Tagantsev, A. K. Adv. Phys.1991, 40, 719.
46. Colla, E. L.; Reaney, I.M.; Setter, N. J. Appl. Phys. 1993, 74, 3414-3425.
47. Harrop, P. J. J. Mater. Sci. 1969, 4, 370–4.
48. Reaney, I.M.; Colla, E.L.; Setter, N. Jpn. J. Appl. Phys.1994, 33, 3984-3990.
49. Goldschmidt, V.M.; Naturwissenschaften 1926, 14, 477-485.
50. McCormack, R.; Burton, B.P. Comput. Mater. Sci. 1997, 8, 153-160.
51. Graham, K.; Patrick M. W. J. Mater. Chem. 2010, 20, 5785-5796.
52. http://cst-www.nrl.navy.mil/lattice/struk/perovskite.html
(Accessed Feb.15, 2012).
53. Bernard Jaffe; William R. Cook; Hans Jaffe, Jr.; Piezoelectric Ceramics,
Academic Press Limited, 1971, pp49-51.
54. Honle, W.; Miller, G.; Simon, A. J. Solid State Chem. 1988, 75, 147-155.
55. Luana, V.; Costales, A.; Pendas, A.M. Phys. Rev.B 1997, 55, 4285-4297.
56. Clearfield, A. Acta Crystallogr. 1963, 16, 134-142.
57. Gingl, F.; Vogt, T.; Akiba, E.; Yvon, K. J. Alloys Compd. 1999, 282, 125-129.
58. Peschel, S.; Ziegler, B.; Schwarten, M.; Babel, D. Z. Anorg. Allg. Chem., 2000,
626, 1561-1566.
59. Malecki, G.; Ratuszna, A. Powder Diff. 1999, 14, 25-30.
60. Carlson, S.; Larsson, A.-K.; Rohrer, F.E.; Acta Crystallogr. 2000, B56, 189-196.
Chapter 1Chapter 1Chapter 1Chapter 1
55
61. Marchand, R.; Laurent, Y.; Guyader, J.; L’Haridon, P.; Verdier, P. J. Eur.
Ceram. Soc. 1991, 8, 197-213.
62. Ganguli, A. K.; Grover, V.; Thirumal, M. Mater. Res. Bull. 2001, 36, 1967–
1975.
63. Roy, R. J. Am. Ceram. Soc. 1954, 37, 581-588.
64. Megaw, H.D. Proc. Phys. Soc. 1946, 58, 133.
65. Sasaki, S.; Prewitt, C. T.; Bass, J. D. Acta Crystallogr., Sect.C 1987, 43, 1668-
1674.
66. Brown, I. D. Acta Crystallogr. 1992, B48, 553-572.
67. Glazer, A. M. Acta Crystallogr. Sect.B 1972, 28, 3384.
68. Arendt, R.H.; Rosolowski, J.H.; Szmaszek, J.W. Mater. Res. Bull. 1979, 14, 703.
69. Yoon, K.H.; Cho, Y. S.; Kang, D. H. J. Mater. Sci. 1998, 33, 2977.
70. Mao, Y.; Park, T-J.; Zhang, F.; Zhou, H.; Wong, S.S. small 2007, 3, 1122 – 1139.
71. Riedel, R.; Chen, I-W. Ceramics Science & Technology, Wiley-VCH Verlag
GmbH & Co. KGaA,Weinheim , 2008.
72. Liu, C.-L.; Wu, T.B. J. Am., Ceram. Soc. 2001, 84, 1291-1295.
73. Kikuo,W.; Toshio, N.; Youhei, I.; Hiroshi, T. Br. Ceram. Trans. 1990, 89, 39-43.
74. Youn, H.J.; Kim, K.Y.; Kim, H. Jpn. J. Appl. Phys. 1996, 35, 3947-3953.
75. Lu, C.H.; Tsai, C.C. J. Mater. Res. 1996, 11, 1219- 1227.
76. Kolodiazhnyi, T.; Petric, A.; Belous, A.; V’yunov, O.; Yanchevskij, O. J.
Mater. Res., 2002, 17, 3182–3189.
77. Galasso, F. Structure Properties and Preparation of Perovskite Type
Compounds; Pergamon Press, Oxford, London, 1969.
78. Reaney, I.M.; Wise, P.L.; Qazi, I.; Miller, C.A.; Price, T.J.; Cannell, D.S.;
Iddles, D.M.; Rosseinsky, M.J.; Moussa, S.M.; Bieringer, M.; Noailles, L.D.;
Ibberson, R.M. J. Eur. Ceram. Soc. 2003, 23, 3021–3034.
79. Galasso, F.; Pyle, J. J. Phys. Chem. 1963, 67, 482–484.
Chapter 1Chapter 1Chapter 1Chapter 1
56
80. Vincent, H.; Perrier, C.; Lheritier, P; Labeyrie, M. Mater. Res. Bull. 1993, 28, 951.
81. Kang, S.H.; Kim, D.J.; Lee, C.J.; Pezzotti, G. Mater. Res. Bull. 2008, 43,
2929–2936.
82. Moussa, S.M.; Ibberson, R. M.; Bieringer, M. ; Fitch, A. N.; Rosseinsky, M. R.
Chem. Mater. 2003, 15, 2527–33.
83. Bieringer, M.; Moussa, S. M.; Noailles, L. D.; Burrows, A.; Kiely, C.J.;
Rosseinsky, M.J.; Ibberson, R.M. Chem. Matter. 2003, 15, 586-597.
84. Koga, E.; Moriwake, H. J. Ceram. Soc. Jpn. 2003, 111, 767.
85. Koga, E.; Yamagishi, Y.; Moriwake, H.; Kakimoto, K.; Ohsato, H.
J. Eur. Ceram. Soc. 2006, 26, 1961–1964.
86. Kim, M-H.; Nahm, S.; Lee, W-S.; Yoo, M-J.; Park, J-C.; Lee, H-J. Jpn. J.
Appl. Phys. 2004, 43, 1438-1441.
87. Thirumal, M.; Davies, P. K. J. Am. Ceram. Soc. 2005, 88, 2126–2128.
88. Roth, R.S. Chemistry of electronic ceramic materials, Proceedings of the
International Conference, Jackson, Wyoming, August 17-22, 1990.
89. Tamura, H.; Konoike, T.; Sakabe, Y.; Wakino, K J. Am. Ceram. Soc. 1984, 67,
59–61.
90. Kim, J.S.; Kim, J.-W.; Cheon, C.I.; Kim, Y.-S.; Nahm, S.; Byun, J.D. J. Eur.
Ceram. Soc. 2001, 21, 2599-2604.
91. Kageyama, K. J. Am. Ceram. Soc. 1992, 75, 1767–1771.
92. Varma, M. R.; Nisha, P.; Varma, P.C.R. J. Alloys Compd. 2008, 457, 422–428.
93. Reaney, I.M.; Qazi, I.; Lee, W.E. J. Appl. Phys. 2000, 88, 6708.
94. Qazi, I.; Reaney, I.M.; Lee, W.E. J. Eur. Ceram. Soc. 2001, 21, 2613–2616.
95. Ioachim, A.; Toacsan, M.I.; Banciu, M.G.; Nedelcu, L.; Dutu, C.A.; Feder, M.;
Plapcianu, C.; Lifei, F.; Nita, P. J. Eur. Ceram. Soc. 2007, 27 1117–1122.
96. Ioachim, A.; Toacsan, M.I.; Banciu, M.G.; Nedelcu, L.; Dutu, C.A.; Alexandru,
H.V.; Antohe, S.; Andronescu, E.; Jinga, S.; Nita P. Thin Solid Films 2008, 516,
1558–1562.
Chapter 1Chapter 1Chapter 1Chapter 1
57
97. Roulland, F.; Terras, R.; Marinel, S. Mater. Sci. Eng., B 2003, 104, 156–162.
98. Tzou, W.C.; Chan, C.C.; Cheng, P.S.; Yang, C.F. J. Mater. Sci. Lett. 2005, 40,
4711-4714.
99. Chaouchi, A.; D’Astrong, S.; Marinel, S.; Aliouat, M. Bol. Soc. Espa. Ceram.
Vidr. 2011, 50, 79-84.
100. Matsumoto, H.; Tamura, H.; Wakino, K. Jpn. J. Appl. Phys. 1991, 30, 2347-2349.
101. Sugiyama, M.; Nagai, T. Jpn. J. Appl. Phys. 1994, 759, 4704-4708.
102. Liang, M.-H.; Wu, S.Y.; Hu, C.-T.; Lin., I.-N. Mater. Chem. Phys. 2003, 79,
276-278.
103. Renoult, O.; Boilot, J.P.; Chaput, F.; Pappienik, R.; Hubert- Pfalzgraf, L.G.,
Lejeune, M. J. Am. Cera. Soc. 1992, 75, 3337-3340.
104. Katayama, S.; Dekine, M. J. Mater. Chem. 1998, 2, 889-890.
105. Vincezini, P. Ceramics today- tomorrow ceramics, Ed.; Elsevier: Amsterdam,
1991.
106. Shannon, R. D. Acta Crystallogr., Sect. A 1976, 32, 751.
107. Shi, F.; Dong, H. Dalton Trans. 2011, 40, 6659.
108. You, C. C., Huang, C. L.; Wei, C.C.; Huang, J.W. Jpn. J. Appl. Phys. 1995, 34,
1911–1915.
109. Sebastian MT. Dielectric materials for wireless communication, 1st Edition,
Oxford, Elsevier: 2008.
110. Dai, Y.; Zhao, G.; Liu, H.; J. Appl. Phys. 2009, 105, 034111.
111. Liu, D.; X.M. Chen, X.M.; Hu, X. Mater. Lett. 2007, 61, 1166–1169.
112. Sert, D.; Mergen, A. J. Alloys Compd. 2009, 482, 396-399.
113. Khemakhema, L.; Kabadoua, A.; Maaleja, A.; Ben Salah, A.; Simonc, A.;
Maglione, M. J. Alloys Comp. 2008, 452, 451-455.
114. Barwick, M.; Azough, F.; Freer, R. J. Eur. Ceram. Soc. 2006, 26, 1767.
115. Varma, M. R.; Sebastian, M.T. J. Eur. Ceram. Soc. 2007, 27, 2827–2833.
Chapter 1Chapter 1Chapter 1Chapter 1
58
116. Mergen, A.; Korkmaz, E. J. Eur. Ceram., Soc., 2011, 31, 2649-2655.
117. Lee, C.T.; Lin, Y.C.; Huang, C. Y. J. Am. Ceram. Soc. 2007, 90, 483–489.
118. Lin, I.N.; Chia, C. T.; Liu, H. L.; Cheng, H. F.; Freer, R.; Barwick, M.;
Azough, F. J. Appl. Phys. 2007, 102, 044112.
119. Scott, R.I.; Thomas, M.; Hampson, C. J. Eur. Ceram. Soc. 2003, 23, 2467–
2471.
120. Millet, J. M.; Roth, R. S.; Ettlinger, L. D.; Parker, H. S. J. Solid State Chem.
1987, 67, 259–70.
121. Akbas, M. A.; Davies, P. K. J. Am. Ceram, Soc. 1998, 81, 1061–4.
122. Hong, K.S.; Kim, I-T.; Kim, C. D. J. Am. Ceram. Soc. 1996, 79, 3218–24.
123. Takeshi,T.; Eric, J.W.; Gerbrand, C.; J. Mater.Res. 2000, 15, 2061–2064.
124. Noh, S.Y.; Yoo, M.J.; Nahm, S.; Choi, C.H.; Park, H.M.; Lee, H. J. Jpn. J.
Appl. Phys. 2002, 41, 2978-81.
125. Koga, E.; Yamagishi, Y.; Moriwake, H.; Kakimoto, K.; Ohsato, H.
J. Electroceram. 2006, 17, 375–379.
126. Chaouchi, A.; Marinel, S.; D’Astrong, S.; Aliouat, M. Ceram. Int. 2009, 35,
939–942.
127. Sun, J.S.; Yu, J.J.; You, J.C. J. Mater. Sci. 1993, 28, 2163.
128. Borisevich, A.; Davies, P. K. Appl. Phys. Lett. 2004, 84, 23.
129. Nomura, S.; Tomaya K.; Kaneta, K. Jpn. J. Appl. Phys. 1983, 22, 1125.
130. Paik, J. H.; Nahm, S.; Bylin, J. D.; Kim, H. M.; Lee, H. J. J. Mater. Sci. Lett.
1998, 17, 1777–1780.