Microstructure, Property and Processing of Functional Ceramics Springer

Prévia do material em texto

Qingrui Yin 
Binghe Zhu 
Huarong Zeng 
Microstructure, Property and Processing of 
Functional Ceramics 
Qingrui Yin 
Binghe Zhu 
Huarong Zeng
Microstructure, Property 
and Processing of 
Functional Ceramics 
With 212 figures 
AUTHORS 
Prof. Qingrui Yin 
Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, China 
E-mail: qryin@sunm.shcnc.ac.cn 
Prof. Binghe Zhu 
Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, China 
E-mail: red-1688@163.com 
Associate Prof. Huarong Zeng 
Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, China 
E-mail: huarongzeng@mail.sic.ac.cn 
Based on an original Chinese edition: 
��������	
���
�����(Gongneng Taoci De Xianwei Jiegou, Xingneng Yu 
Zhibei Jishu), Metallurgical Industry Press, 2003 
ISBN 978-7-5024-4571-3 Metallurgical Industry Press, Beijing 
ISBN 978-3-642-01693-6 Springer Dordrecht Heidelberg London New York 
e ISBN 978-3-642-01694-3 Springer Dordrecht Heidelberg London New York 
Library of Congress Control Number:2009930028
This work is subject to copyright. All rights are reserved, whether the whole or part of the material is 
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, 
reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or 
parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its 
current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable 
to prosecution under the German Copyright Law. 
�� 2009 Metallurgical Industry Press, Beijing and Springer-Verlag GmbH Berlin Heidelberg 
Co-published by Metallurgical Industry Press, Beijing and Springer-Verlag GmbH Berlin Heidelberg
Springer is a part of Springer Science+Business Media 
Springer.com 
The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, 
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and 
regulations and therefore free for general use. 
Cover design: Frido Steinen-Broo, Estudio Calamar, Spain 
Printed on acid-free paper
Preface 
The functional ceramic materials (FCM) are potential for use in many electronic 
devices such as optical waveguides, non-volatile dynamic random access 
memories, micromotors, microactuators, thin film capacitors, and pyroelectric 
infrared detectors. FCM possesses unique properties like piezoelectricity, 
pyroelectricity, photoelectricity, photo-acoustic effect, photorefractive behavior, 
and non-linear optical activity that are closely depends closely on the common 
theme of composition-preparation-structure-property relationships in the solid state, 
especially microstructures (grain, grain boundary and domain structures, etc.) and 
their dynamic response to mechanical, electrical and optical loads at nanometer 
scale. Thus it is very important to understand the physical phenomenological 
behavior of ferroelectric structures and their dynamic evolution in nanoscale 
volumes. This is the context that motivated the publication of this book. 
The aim of this book is to present recent advances in the fabrication process of 
functional ceramic materials and their property study, particularly, in-depth 
observation/analysis of microstructures using the custom-built scanning electron 
acoustic microscopy (SEAM), acoustic and piezoresponse mode scanning probe 
microscopy based on atomic force microscopy. Along with the generally accepted 
concepts and experimental results there are numerous applications of functional 
ceramics and devices in industry. We hope that this book will make the readers 
aware of tremendous developments in the field of microstructure characterization 
and functional ceramic preparations. 
The first two chapters address fundamentals of microstructures in the functional 
ceramics. Chapter 1 presents the formation mechanism of microstructures 
including grains, grain boundaries, pores, domain structures, and their correlations 
with properties and processing for some typical ceramics like PLZT (lead 
lanthanum zirconate titanate) ceramics, PTC (positive temperature coefficient) 
ceramics, piezoelectric ceramics, ferroelectric ceramics, and so on. Chapter 2 
discusses grain boundary phenomena such as grain boundary segregation and 
migration in the functional ceramics. 
The next two chapters focus on near-field microscopy characterizing ferroelectric 
domains of functional ceramics. Chapter 3 describes two custom-built acoustic 
microcopies used for ferroelectric domain imaging, scanning electron acoustic 
microcopy (SEAM) and scanning probe acoustic microscopy (SPAM), presenting 
� �� Preface 
their operation principle, analyzing domain contrast formation mechanism, 
showing their high-resolution observations of domain configurations 
non-destructively, especially their unique capability of imaging sub-surface 
structures, and their dynamic response to the electric field as well as their 
applications to a variety of materials including structure ceramics, metals, single 
crystals, composites and coatings, etc. Chapter 4 presents piezoresponse force 
microscopy (PFM) of nanoscale domains in functional ceramics, involving 
interpreting the electromechanical vs electrostatic contributions to PFM imaging 
contrast in the ambient and high vacuum environment, observing domain 
arrangements, investigating their evolution under the inhomogeneous tip fields 
and local domain switching behavior in ferroelectric thin film, lead-free 
piezoelectric ceramics and relaxor-type single crystals. 
Finally, the last two chapters cover the fabrication processing and the prospect 
of functional ceramics. Chapter 5 describes science and technology of fabrication 
processing for different kinds of functional ceramics including capacitor ceramics, 
ferrite ceramics, corundum ceramics, PTC ceramics, and superconductor 
ceramics,etc. Chapter 6 summarizes the future development of functional ceramics, 
starting with an overview of the ceramic evolution and emphasis of ceramic 
processing. 
This book is intended for advanced undergraduate and postgraduate students in 
the field of materials science and microscopy techniques. Scientists and industrial 
engineers working on the functional ceramics, microelectronics, optroelectronics, 
and sensors may also find this book useful. 
���������
���	�
���
��
�
Qingrui Yin, 
Binghe Zhu, 
Huarong Zeng 
� �� Preface 
Acknowledgements 
We would like to thank our colleagues, Prof. Ji Zhou, Prof. Xiangming Chen, Prof. 
Haosu Luo, Prof. Yongxiang Li, and Prof. Guorong Li who contributed to this book. 
Also, we express gratitude to Yunshu Liu, Fukang Fan, Yingquan Zhu, Guangrui Ye, 
Taijun Deng, and Fengye Tao for their kind assistance. 
The authors are particularly indebted to Dr. Rubin Ye, Dr. Junkun Ma, Dr. Shiqun 
Xiao, Dr. Xiaohua Deng, Dr. Deji Fu, Dr. CV. Kannan and Dr. V. Vaithianathan for 
their appreciable support in proof-reading and revising the chapters. 
The authors greatly acknowledge National Science Foundation of China, Major 
State Basic Research Development Program of China, National High Technology 
Research and Development Program of China, and Shanghai Institute of Ceramics, 
Chinese Academy of Sciences for sponsoring various projects. 
Contents 
1 Microstructure and Properties of Functional Ceramics .............................1 
1. 1 General Description ...............................................................................1 
1. 2 Grain .....................................................................................................5 
1. 2. 1� Graincategory................................................................................5 
1. 2. 2� Grain properties..............................................................................9 
1. 3 Grain Boundary Structures ...................................................................13 
1. 3. 1� Concepts of grain boundary structures ..........................................13 
1. 3. 2� Properties of grain boundary structures .........................................14 
1. 3. 3� Nano grain boundary structures ....................................................15 
1. 4 Pore Phases..........................................................................................16 
1. 5 Domain Structure.................................................................................18 
1. 6 Mechanical Properties of Ferroelectric Ceramics ..................................27 
1. 6. 1� General ........................................................................................27 
1. 6. 2� Electric domain and internal stress................................................28 
1. 6. 3� PLZT ceramics and internal stress.................................................33 
1. 6. 4� PTC ceramics and internal stress...................................................39 
1. 6. 5� Aging...........................................................................................40 
1. 7 Capacitor Ceramics..............................................................................41 
1. 7. 1� Ordinary dielectric materials for capacitor.....................................41 
1. 7. 2� Relaxor ferroelectric materials ......................................................47 
1. 7. 3� Microwave dielectric materials .....................................................48 
1. 8 Piezoelectric Ceramics .........................................................................49 
1. 8. 1� Microstructures of piezoelectric ceramics......................................50 
1. 8. 2� Properties of piezoelectric ceramics ..............................................50 
1. 9 Transparent Ferroelectric Ceramics ......................................................53 
1. 9. 1� Microstructures of transparent ferroelectric ceramics.....................53 
1. 9. 2� Experimental method and two phases of PLZT ceramics...............55 
1. 9. 3� Domain switching properties of PLZT ceramics............................57 
1. 9. 4� Grain boundaries in PLZT ceramics..............................................67 
1. 9. 5� Summary......................................................................................77 
1. 10 Thermistor Materials..........................................................................77 
� Contents 
1. 10. 1� Microstructures and properties of PTC materials .........................78 
1. 10. 2� NTC materials and segregation at grain boundaries .....................82 
1. 11 Varistor Materials...............................................................................86 
1. 12 Ceramics for Humidity Sensitive Resistor...........................................91
1. 13 Magnetic Ceramics ............................................................................92
1. 14 Biologically Functional Ceramics.......................................................94
1. 15 Functional Ceramic Films ..................................................................98
1. 16 Alumina Ceramics ...........................................................................104
1. 17 Summary .........................................................................................105
References ..................................................................................................106
2� Grain Boundary Phenomena of Functional Ceramics............................112 
2. 1 Introduction .......................................................................................112 
2. 2 Generalization of Grain Boundary......................................................115 
2. 2. 1� Grain boundary structure ............................................................116 
2. 2. 2� Grain boundary properties ..........................................................118 
2. 3 Grain Boundary Segregation ..............................................................119 
2. 3. 1� Generalization ............................................................................119 
2. 3. 2� Boundary layer capacitors...........................................................122
2. 3. 3� PTC materials.............................................................................124
2. 3. 4� Magnetic ceramics......................................................................127
2. 3. 5� ZnO varistor materials................................................................128
2. 3. 6� Other examples of segregation....................................................129
2. 4� Grain Boundary Region .....................................................................134
2. 4. 1� General description about grain boundary region.........................134
2. 4. 2� Grain boundary region of BaTiO3 ceramics.................................134
2. 4 .3� Grain boundary region of PLZT ceramics ...................................135
2. 4. 4� Grain boundary region and stress ................................................139
2. 4. 5� “Core-shell” structure.................................................................141
2. 5� Grain Boundary Migration .................................................................143
2. 5. 1� Generalization ............................................................................143
2. 5. 2� Centripetal and acentric grain boundary migration ......................144
2. 5. 3� Liquid phase and abnormal grain growth during sintering ...........152
2. 6� Relation between Grain Boundary and Properties ...............................154
2. 6. 1� Influence on mechanical properties.............................................155
2. 6. 2� Influence on electric properties ...................................................162
2. 7� Summary ...........................................................................................166
References ..................................................................................................168
3� Near-field Acoustic Microscopy of Functional Ceramics........................176
3. 1 Introduction .......................................................................................176
Contents� �
3. 2 History and Development of Scanning Electron Acoustic 
Microscopy........................................................................................177
3. 3 Physical Principle of SEAM Imaging .................................................178
3. 4 Scanning Electron Acoustic Microscopy Image Processing System.....180
3. 5 Theory Studies of Electron-acoustic Imaging......................................182
3. 6 SEAM Imaging of Ferroic and Other Materials ..................................184
3. 6. 1� SEAM imaging features of ferroelectric domains ........................184
3. 6. 2� Electron-acoustic imaging of ferroelectric materials ....................185
3. 6. 3� Ferroelectric Bi4Ti3O12 single crystal ..........................................190
3. 6. 4� Ferroelasitc NdP5O6 single crystal ..............................................190
3. 7 Magnetic Domains in Austenitic Steel ................................................191
3. 8 Modulation Frequency Dependence of SEAM Imaging Domain 
Structures...........................................................................................193
3. 9 Electric Field Dependence of SEAM Imaging Domains......................195
3. 10 Temperature Dependence of Ferroelastic Domains in PMN- PT 
Single Crystals.................................................................................1963. 11 SEAM imaging of Other Materials ...................................................200
3. 11. 1� Residual stress distribution in Ti3N4 coatings.............................200
3. 11. 2� Stress distribution in ferroelectric composites............................202
3. 11. 3� Stress distribution in Si3N4 and ZrSiO4 ceramics .......................203
3. 11. 4� Stress distribution of Al metal...................................................205
3. 11. 5� Surface structures and internal defects in lead-free 
piezoelectric ceramics ..............................................................206
3. 11. 6� Phase transitions in superconductor ceramics ............................208
3. 11. 7� SEAM imaging of MEMS devices ............................................209
3. 12 Scanning Probe Acoustic Microscopy...............................................209
3. 12. 1� Tip-vibration mode scanning probe acoustic microscope ...........210
3. 12. 2� Sample-vibration mode scanning probe acoustic microscopy..........214
3. 13 Comparisons of SEAM with SPAM..................................................225
References ..................................................................................................225
4� Piezoresponse Force Microscopy of Functional Ceramics......................229
4. 1 Introduction .......................................................................................229
4. 2 History and Development of Scanning Probe Microcopy ....................230
4. 3 Piezoresponse Force Microscopy........................................................231
4. 3. 1� Operation principle.....................................................................231
4. 3. 2� PFM imaging features ................................................................234
4. 4 PFM Imaging of Ferroelectric Domains..............................................236
4. 4. 1� Ferroelectric thin films ...............................................................236
4. 4. 2� Ferroelectric ceramics.................................................................241
4. 4. 3� Ferroelectric single crystals ........................................................247
Contents���
� Contents 
4. 5 Dynamic Behavior of Nanoscale Domain Structure ............................261
4. 5. 1� Domain writing ..........................................................................261
4. 5. 2� Domain nucleation and reversal ..................................................262
4. 6 PFM and SPAM Characterization of Ferroelectric Materials ...............273
4. 6. 1� Bi4Ti3O12 lead-free ceramics .......................................................273
4. 6. 2� PMN-PT single crystal ...............................................................275
4. 7 Summary ...........................................................................................279
References ..................................................................................................279
5� Fabrication Processes for Functional Ceramics......................................283
5. 1� Introduction .......................................................................................283
5. 1. 1� Capacitor ceramics .....................................................................287
5. 1. 2� Ferrite ceramics..........................................................................288
5. 1. 3� Corundum ceramics....................................................................289
5. 1. 4� Piezoelectric ceramics ................................................................290
5. 1. 5� PTC ceramics.............................................................................290
5. 1. 6� Varistor ceramics........................................................................291
5. 1. 7� Superconductor ceramics............................................................291
5. 2� Raw Material and Powder Preparation................................................292
5. 2. 1� Ball mill mixing and grinding .....................................................293
5. 2. 2� Powder preparation by oxide methods.........................................294
5. 2. 3� Powder preparation by co-precipitation.......................................298
5. 2. 4� Powder preparation by sol-gel method ........................................299
5. 2. 5� Powder preparation by hydrothermal method ..............................300
5. 2. 6� Powder preparation by spray pyrolysis........................................301
5. 3� Shaping and Forming of Functional Ceramics ....................................301
5. 3. 1� Processing of thin films ..............................................................302
5. 3. 2� Processing of thick films.............................................................305
5. 3. 3� Dry pressing...............................................................................307
5. 3. 4� Iso-static pressing.......................................................................311
5. 3. 5� Hot injection moulding...............................................................312
5. 3. 6� Slip casting.................................................................................313
5. 4� Sintering............................................................................................314
5. 4. 1� Sintering mechanisms.................................................................314
5. 4. 2� Sintering process ........................................................................317
5. 4. 3� Grain growth ..............................................................................321
5. 4. 4� Abnormal grain growth...............................................................322
5. 4. 5� The effects of pressure and atmosphere on sintering....................323
5. 4. 6� Pressure sintering .......................................................................324
5. 4. 7� Micro-porosity sintering .............................................................325
5. 4. 8� Microwave sintering...................................................................326
��Contents�
Contents� �
5. 5� Mechanical Finishing.........................................................................327
5. 6� Electroding ........................................................................................329
5. 6. 1� Electroding from silver paste ......................................................330
5. 6. 2� Electroding from nickel plating...................................................332
5. 6. 3� Other electroding methods..........................................................334
References ..................................................................................................335
6� Review and Prospect of Functional Ceramics.........................................337
6. 1 Evolution of Ceramics........................................................................337
6. 2 Development of Functional Ceramics and Relation with 
Other Factors......................................................................................338
6. 3 Importance and Complexity of Understanding Functional 
Ceramic Effects and Mechanism ........................................................342
6. 4 Emphasis of Ceramic Processing........................................................344
6. 5 Future Development of Functional Ceramics ......................................345
6. 5. 1� Dielectric ceramics and devices ..................................................346
6. 5. 2� Chip type ceramic devices ..........................................................347
6. 5. 3� High performance, high temperature piezoelectric ceramics ........348
6. 5. 4� Lead-free piezoelectric ceramics.................................................349
6. 5. 5� Thermoelectric ceramics.............................................................3506. 5. 6� Functional ceramic films ............................................................352
6. 5. 7� Functional crystals......................................................................356
6. 5. 8� Battery materials ........................................................................358
6. 5. 9� High temperature superconductive ceramics................................360
6. 5. 10� Fabrication of ceramic micro-components.................................360
References ..................................................................................................362
Index..............................................................................................................364
Appendix .......................................................................................................366
��
Microstructure and Properties of 
Functional Ceramics 
�
 
This chapter mainly deals with the microstructure and property of functional 
ceramics. Following the description of grain, grain boundary, nano-grain boundary, 
pores, domain structure, and mechanical behavior, the emphasis is placed on the 
relationships among chemical, and phase composition, microstructure, and 
properties of various functional ceramics including capacitive ceramics, piezoelectric 
ceramics, transparent ceramics, thermistor ceramics, magnetic ceramics, bioceramics, 
thin film ceramics and corundum ceramics. 
About 90 photographs of representative microstructures are presented, which 
are useful for comparison and examination of the related structure characteristics, 
and are also taken as the criterion to judge the constitutive phase or defects and to 
evaluate the ceramic properties. 
1.1 General Description 
Modern advanced ceramics, have become key materials in the development of 
modern technology although ceramics are ancient materials. According to 
different applications, advanced ceramics could be roughly divided into structural 
ceramics and functional ceramics while functional ceramics account for a large 
�
�
�� Pictures in Chapter 1 and 2 without specific reference are works of former fourth Lab in Shanghai 
Institute of Ceramics CAS. The authors are grateful to Zujun Gu, Ruifu Huang, Xiangyun Song, Jing 
Sun, Yujun Zhang for their great assistance in the composition of the book. The authors appreciate great 
contribution from Haikuan Ao, Yao Yao, Rongming Sun, Zhili Chen, Weiping Yin, Xinsen Zheng and 
other colleagues from former 3rd group of fourth Lab. Thanks also go to Zhiwen Yin, Pingchu Wang 
and Xiangting Li for the enlightenment from discussion with them. 
� �� 1� Microstructure and Properties of Functional Ceramics 
percentage of advanced ceramics. Generally, polycrystalline inorganic materials 
possessing electric, magnetic, elastic, biologic, superconductive, or other chemical 
functions are called functional ceramics while those ceramics possessing mechanical, 
thermal or some chemical functions are called structural ceramics. At present, the 
production value of functional ceramics and structural ceramics are 3 to 1. The 
annual production value of functional ceramics around the world has reached 7 
billion USD, with 21% of capacitor, 18% of magnetic ceramics, 15%~16% of 
integrated package, 11% of piezoelectric ceramics, 5.6% of thermistor, 5.1 % of 
transducer, 2.4% of substrate, and 1.9% of varistor. All these functional ceramics 
have been widely used in the fields of computers, tele-communication, television, 
home appliances, space technology, automation, automobile and medical care, etc. 
Ceramic microstructures include various structural images obtained from 
different kinds of microscopes (Fig.1.1). 
Single crystal Ceramics Polyphase Pore (empty space) Crack 
 (poly-crystalline) Q�quartz grain G�grain � � � � C�crack 
 GB�grain boundary M�mullite P�pore 
 G�grain GP�glass phase 
(a) 
Fig.1.1 Microstructures of ceramics 
(a)Illustrative diagram for the microstructure of ceramics; (b) Microstructure of PTC BaTiO3; 
(c) Microstructure of PLZT ceramics; (d) Quartzitic sandstone�a natural hot pressed ceramics 
1.1 General Description� ��
According to Petzow, all phase regions and flaws contained in structures would 
be reflected in microstructures, which determine many properties of materials. 
According to Pask(1984), microstructures should include sizes and distribution of 
grains and pores, phase composition and distribution, nature of grain boundary 
and its defects and flaws, composition homogeneity as well as domain structures. 
Ceramics are materials derived from powdery raw materials through various 
processing, and possess specific microstructures and properties. Thus 
microstructures comprehensively reflect previous processing, and bring specific 
properties to materials. Microstructural analysis is also important for determining 
phase diagrams, providing bases for property analysis, instructing modification on 
formulation, processing improvement, production rationalization, and failure 
analysis. The following are several examples which further explain the importance 
of microstructure analysis. 
Example 1: There was a newly built transformer substation in Shanghai. In a 
very hot summer the elevated temperature caused a dramatical rise of the oil 
pressure with a ceramic container, and gave a blast on it. Luckily, it happened 
during the trial run, otherwise it would probably have caused life threat and power 
shut down for a massive area. The microstructural analysis afterwards on that 
ceramic debris showed that the silica particle had sharp boundaries in the 
high-tension insulator ceramics, which provided evidences that silica particles did 
not fully melt and react with feldspar and other glass frits during sintering while 
the boundaries of silica particle of normal insulating ceramics are corroded with 
glass phases. The microstructure demonstrated that the ceramic body had not been 
fully sintered, thus it had low tensile strength and couldn’t survival under high oil 
pressure. 
Example 2: At a PTC heater manufacturer in Cixi city of Zhejiang province, 
the ceramic pieces were not broken after voltage test, but cracked in a large 
amount after packing and transportation, which caused a loss of hundreds of 
thousands of ceramic pieces (0.65 Yuan/piece at that time). In the analysis of 
microstructure of PTC ceramics, abnormally grown grains of large sizes were 
found. During the puncture testing, large grain would expand or contract along 
axis, which produced large residual stress and micro cracks. Thus the as sintered 
ceramic plates had normal strength, but became fragile and brittle after puncture 
testing because of micro cracks. After discovering the cause of problem, some 
additional additives were introduced to the composition to restrain the abnormal 
grain growth , and the problem was solved (Zhu, Yao, Zhao, et al, 2001). 
Example 3: Ferroelectric thin films have wide applications in the field of 
nonvolatile ferroelectric memory and micro electromechanical system. The unit 
dimension of high-density ferroelectric memory has been reduced to 30~100 nm, 
thus physical properties, including formation of ferroelectric domain, polarization 
reversion mechanism, polarization fatigue, retaining, degradation and aging of 
� �� 1� Microstructure and Properties of Functional Ceramics 
polarization, need to be investigated in sub-micron scale. Scanning Probe 
Microscopy based on Atomic Force Microscopes could satisfy the requirements 
with resolution of 30 nm, and can be used to exhibit details of polarization 
reversion, which provides an observation foundation for improvement of 
anti-degradation. In addition, these kinds of microscopes are very useful for 
investigation on microstructures of all other nanoceramics too. 
Properties of materials can be divided into two categories. One kind is intrinsic 
or inherent properties, which mainly depend on internal characteristics of 
compounds and crystal structure and also depend on attributes of ferroelectrics, 
ferromagnetism, semi-conductivity and super-conductivity. the other is 
non-intrinsic properties, which often have relationship with microstructures. It is 
ceramic researchers’ task to investigate relationship among processing, structures 
and properties, in which microstructures play a significant role. During many 
global symposiums in last thirty years, it has been widely accepted that 
microstructure plays a significant role in material sciences with a special emphasis 
on interface and grain boundaries. When grain dimensions decrease to a nano level, 
proportion of grain boundaries abruptly increases, and materials turn into nano 
ceramics. Control of microstructures is an important approach to obtain materials 
with desirable properties. 
In early stage, optical microscopes were main tools for observing microstructures, 
while observation instruments currently including optical microscopy, 
Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), 
Scanning Electron Acoustic Microscopy (SEAM), Scanning Nonlinear Dielectrics 
Microscopy (SNDM), Atomic Force Microscopy (SFM), Infra-red or Laser 
Microscopy, are now widely used, with resolution from 1μm to 30nm and even 
to 0.5 nm, so as to enable comprehensive and precise investigation on material 
properties, such as optics, electronics, mechanics and thermology etc., and to 
widen applications of material functions. At present, it has been evolved from 
simply understanding materials’ microstructures to purposely controlling and 
preparing materials with specific microstructures. 
Many properties of ceramics conform to the combination principle. For 
example, mass per unit of volume decreased as porosity increases, which 
decreases permittivity. Density or thermal capacity is generally the proportional 
combination of all phases in materials, and thermal conductivity conforms to 
similar rules only with some exceptions. However, some other properties of 
ceramics do not conform to the combination principle. For example, small 
quantity of second-phase additive can dramatically change electric conductivity of 
materials. Some properties of materials, especially those related with functional 
processes and energy transformation can be affected by interaction between 
different grains and phases, and also fall into the second category. Many attributes 
or phenomena belong to interactive properties, such as alignment of electric 
1.2 Grain� ��
domains in ferroelectric ceramic under electric field, the function of high-energy 
grain boundary zone during the process, and the effect of space charge and 
induced field in grain boundaries. 
Key factors of functional ceramic microstructures will be discussed as below. 
1.2 Grain 
Grain category and grain property are very important to understand functional 
ceramic microstructures and physical property, here we introduce their concepts in 
details. 
1. 2. 1� Grain category 
Grains (or main crystal phase) are main body in microstructures, with three kinds 
of textures, i.e. idiomorphic, semi-idiomorphic and xenomorphic granular texture 
(as shown in Fig.1.2) (Shi, 1982). 
� �� 1� Microstructure and Properties of Functional Ceramics 
Fig.1.2 Crystalline morphology of ceramics (Shi, 1982)
(a) Idiomorphic texture (mullite porcelain); (b) Semi-idiomorphic texture (95% alumina porcelain); 
(c) Xenomorphic granular texture (smelted quartz crystals in porcelain) 
Grains grown in a sound sintering condition will exhibit the characteristic shape 
of crystal structure itself, which forms idiomorphic textures. Grains grown in an 
unsound or restraining condition will just partially exhibit its characteristic shape, 
which forms semi-idiomorphic textures. In addition, grains of anomalous shape with 
various sizes forming xenomorphic granular texture are quite common, especially in 
functional ceramics. According to different requirement on functions, different main 
crystal phase will be selected. For example, materials with large permittivity ε, (such 
as BaTiO3 and rutile) will be selected as main crystal phase. Materials with low ionic 
conductivity or covalent bond (such as α-Al2O3 or BeO) will be choosed. 
ferroelectric materials without symmetric center in crystal structure are excellent 
electro-mechanical materials due to its high spontaneous polarization Ps. Phase 
transformation induced under fields (electric or mechanic) will introduce changes of 
optical properties in some materials, which can be used as electric-optical functional 
(optical switch) or storage materials. Dimensions of some materials can be 
repeatedly and steadily changed under applied field, which can be used for 
electrostriction or actuator. Some transition metals (rare earth elements) retain their 
intrinsic spin angular distance after bonded into compound with magnetism (such as 
Mn-Zn ferrite), which can be used as magnetic materials. The properties of grains or 
main crystal phases have greatly determined properties of materials. For example, 
during the development of high voltage porcelain in past fifty years, the main crystal 
phases have changed from mullite, silica to corundum, with an increase of strength, 
from 30~70MPa (hard porcelain), 95~125MPa (silica porcelain) to 175~200MPa 
(alumina porcelain). Besides main crystal, grain sizes can also affect properties. 
Since grain have been formed from original powder particles after the processes of 
diffusion, gas elimination and grain boundary migration during sintering, grain sizes 
1.2 Grain� ��
depend on powdery raw materials, composition, additive phase and sintering process. 
The perfectness and idiomorphism of grain growth, as well as the interlocking grade 
would affect functional properties of materials. Generally, homogeneity of grains is 
preferred for microstructures, so dispersion or agglomeration, narrow or wide size 
distribution of powdery raw materials, appropriate second phase would affect 
microstructures. Narrow size distribution raw materials, appropriate second phase 
and forming processing are in favor of homogenous microstructures. In some cases, 
among fine grains emerge few coarse grains (abnormal grain growth), which possess 
greatly different thermal expansion or contraction along crystal axis and anisotropy. 
Stress concentration is generally formed along large coarse grain boundaries, which 
are electrically and mechanically weak and where micro cracks originate (as shown 
in Fig.1.3) (Burke, 1963). 
� �� 1� Microstructure and Properties of Functional Ceramics 
Fig.1.3 Microcracks appearing in the grain and grain boundary of ceramic material 
(a) Cracks in the grain of PTCR microstructure TC=240�; (b) Cracks in grain of 
KNaNb2O3 ceramic; (c) Large grain in high voltage porcelain (BaTiO3); 
(d) Cracks in grains of high tension porcelain insulator 
Rationality of sintering processes can be judged from grain shape and 
crystalline morphology (such as over-firing and under-firing as shown in 
Fig.1.4). 
For materials with domain structure, grain size smaller than a critical dimension 
will affect development of electric or magnetic domain, which will finally affect 
its dielectric or ferroelectric properties. 
1.2 Grain� ��
Fig.1.4 Crystalline morphology of the quartz grains in porcelain 
(a) Quartz grains with some smelted edges in porcelain (Chu, Weng, Wang, 1994); 
(b) Quartz grains with more smelted edges in porcelain 
1. 2. 2� Grain properties 
Grains possess similar properties to single crystals of the same composition. 
Ideally all micro zones in a grain possess the same properties,however 
differentiation in composition and properties could be found between boundary 
zone (shell) near boundaries and central zone (core) (Rawal, 1981). Two kinds of 
crystal forms could be even found in one grain (for example, cubic near the 
boundary and tetragonal at the core, as shown in Fig.1.5). 
� ��� 1� Microstructure and Properties of Functional Ceramics 
Fig.1.5 Core and shell structure in PLZT ceramics (optical, crossed Nicol) 
Micro zones with little differences in optic axial angle have been found in one 
grain of ferrite ceramics. Ununiformity in nano-scale has been observed within 
grains of solid solution ceramic of PbInNbO3-PbMgNbO3 due to defects of 
processing (Babakishi, Tai, Choy, 2002). Since grains are formed through 
diffusion and boundary migration (growth) during sintering, uniformity within 
grains can not be obtained if the process is out of equilibrium, so trace and residua 
of mixture of raw materials, boundary migration and diffusion will remain in 
non-uniform grains, which is quite common in many solid-solution functional 
ceramics such as PZT, BaTiO3, and ferrite. According to Laurent et al (2001), 
when preparing La-PZT from oxides, pores from chemical etching due to 
composition gradation of Zr, Ti within grains since the diffusion speed of Zr, Ti, 
and La are different; when preparing La-PZT from precursors, neither pores nor 
composition gradation was found in grains. In addition, grain sizes could also 
affect properties. For example, in piezoelectric ceramics, coercive field Ec
increases with decrease of grain sizes. Besides, the diffuseness of phase-transition 
is more significant in finer grain materials. 
Tartaj (2001) has proved that the tetragonality c/a of PbTiO3 ceramics decreases 
with the decrease of grain sizes, and obviously the internal stress status will be 
different. 
As for microstructures of ferroelectric ceramics, the relationship between grain 
sizes and properties is quite complicated. BaTiO3 will be taken as an example to 
further discuss the question (as indicated in section 1.15, Huarong Zeng has 
obtained the critical dimension for ferroelectric thin films, which indicated that 
ferroelectric vanishes with grain size under the critical dimension). 
Ceramics capacitor is one of the electronic devices with large amount of market 
volume. BaTiO3 capacitor widely used�dielectric coefficient of normally sintered 
BaTiO3 ceramics is around 1200~1500, while that of fine-grained BaTiO3 ceramics 
1.2 Grain� ���
could reach 4000~5000, or even 6000. However in single crystalline BaTiO3, ε
along c axis is 4000, and ε along a axis is only 170, i.e. the average ε is around 
950~1200, on which there is still no convinced explanation after around 50 years 
of researches. At present, there are two explanations: 
W R Buessem et al (1966), L Metoseriu et al (1999) and other scientists think 
that high value of ε is caused by internal stress; while G Arlt, D Hennings (1985) 
and others thinks that high ε is induced by a large number of domain walls in fine 
grain materials. 
According to the first theory, when cooling down from high temperature to TC,
phase transition from cubic to tetragonality takes place in BaTiO3 single crystal, 
and the deformation caused by phase transition is allowed since its surfaces are 
free of restriction (for example, surfaces perpendicular to c axis move outwards 
and surfaces parallel to c axis move inwards). However in BaTiO3 ceramics, 
surfaces of grains could not move freely since they are surrounded by other grains 
so as to form internal stress under TC. Many 90�domains (twining) would be 
induced to reduce internal stress in coarse-grain ceramics (5~50) μm, but very few 
90�domains are seen in fine-grain ceramics (around 1μm), which brings large 
internal stress to fine-grain ceramics. ε could reach to 3000 with internal stress of 
64 MPa, and even 6000 with internal stress of 80 MPa. Thus concerning the 
decrease of tetragonality with decrease of grain sizes, high value of ε in fine-grain 
BaTiO3 ceramics could be explained by the combination of large internal stress 
under TC and lack of 90�domains. 
According to the second theory, few 90�domains in fine-grain BaTiO3
ceramics as mentioned in the first theory is actually caused by defects of the 
chemical etching method. G Arlt indicated that 90�domains are visible after 
improving corrosion preparation. He also indicated that the width of ferroelectric 
domain d decreases with decrease of grain sizes α, following the relation of 
α�d1/2 (in the grain size range of 1~10μm) (Arlt, Hennings, 1985). 
G Arlt (1985) has proved that a large number of domain walls would make 
more contribution to dielectric coefficient of ε, which may reach 5000 with 
average grain diameter(a) of 0.8~1.0μm below temperature TC,When α decrease
further (for example, less than 0.7μm), ε would dramatically decrease, and the 
crystal form at room temperature would change from tetragonal to trigonal or 
orthorhombic system (Arlt, Hennings, 1985). 
G Arlt et al do not eliminate the contribution from internal stress for the 
increase of ε and attribute it to the combination of the two facts. But which fact is 
more significant needs further study. 
L Metoseriu et al (1999) have shown that transition temperature and 
tetragonality decreased when the grain size decreases, but the internal stress will 
be increase at this time. 
It is different from BaTiO3 ceramics, K Okazaki et al (1973) has proved that TC
� ��� 1� Microstructure and Properties of Functional Ceramics 
increases with decrease of grain sizes in piezoelectric and ferroelectric PZT and 
PLZT ceramics, which means that piezoelectric properties enhance and cohesive 
field declines with the decrease of grain sizes. Space charge theory could be 
utilized to explain the phenomena. It is reasonable to think that with the decrease 
of grain sizes, restriction enhances and it would restrain dielectric and polarized 
properties. Finer grains also mean more grain boundaries, which would affect 
transfer of some properties. As for the dependence of grain sizes on phase 
transition, the increase or decrease of transition temperature depends on expansion 
or contraction of grain during phase transition. 
With the decrease of grain sizes, space charge field effect would increased, 
which would restrict movement of domain walls so as to enhance the stability of 
resonant frequency in piezoelectric filter materials. PZT materials of fine grains 
often have higher fracture toughness than that of coarse grains, e.g. 40% higher. 
Grain sizes in superconductor also affect its toughness, while magnetic 
materials of fine grains have a higher saturation magnetization, a lower coercive 
field and lower resonant frequency. 
Generally optical axis of each grain in ceramics is randomized, so the properties 
of ceramics are average value of all grains. Nowadays it is practical to align grain 
along certain direction so as to provide ceramics with oriented properties, like that in 
single crystals. Following methods could be adopted to obtain this microstructure: 
1. Hot forging or hot pressing sintering. 
2. Eutectic solidification. 
3. Templated grain growth sintering. 
Hot forging method could be used to produce considerable directional 
alignment. For example, piezoelectricity in some layered ferroelectric system 
containing Bi is not quite strong, only with a Kp of 0.2. However, its Kt could be 
enhanced to 0.35~0.42 by doping MnO, NiO, Cr2O3. Due to advantages including 
desirable temperature stability of resonance frequency ((0~20)�10 – �4%), low 
dielectric constant ε (100~200), high Curie Point TC (above 550○C), good 
resistance against degradation, these materials could be effectively utilized at high 
frequency. Dielectric constant ε perpendicular to hot forging direction is highly 
differentfrom that parallel to hot forging direction as shown in Table 1.1. 
Table 1.1 Dielectric constants of some typical materials by hot forging method 
Materials ε parallel to hot forging direction 
ε perpendicular to hot 
forging direction 
Bi4Ti3O12 270 1170 
Na0.5Bi4.5Ti4O15 280 3030 
PbBi4Ti4O15 1350 5300 
M M Seabaugh (1997) prepared highly oriented corundum ceramics by using 
templated grain growth sintering, with materials as below: 
1.3 Grain Boundary Structures� ���
1. Colloid gibbsite ( 15 mass% ) was selected as alumina precursor. 
2. α-Al2O3 fine particles ( <0.1μm ) were used as seeds for phase transition (0.2 
mass%). 
3. Ca, Si ( 1:1 ) glass phase was added to promote grain growth ( 5 mass % ). 
4. Plated Al2O3 particles of hexagonal shape were used as templated grains (10 
mass %), and their oriented growth were promoted by tape casting. 
Corresponding preparing steps included: with homogeneously mixed slurry of 
above-mentioned materials, tape casting were performed on glass substrate to 
obtain films of about 500μm in thickness, which were dried and then calcined at 
600� for 1h. The sintering was performed at 1100~1600� for 6s�6h. As 
sintered specimen surfaces were corroded with hydrofluoric acid to remove glass 
phase for SEM observation. Microstructures of ordinary ceramics are randomly 
arranged, while the ceramics prepared by templated grain growth sintering has 
complexly oriented structures. 
By templated grain growth sintering, Sabolsky obtained ceramics with strong 
anisotropy in Sr0.53Ba0.47Nb2O6, PMN-PT, and (Sr0.9La0.1) Nb2O7 systems. 
J S Patwardhan (2002) prepared Bi4Ti3O12 materials with templated grain hot 
forging and obtained ceramics with strong orientation. In addition, slightly 
changing in composition, such as excessive Bi or Ti by 1 mass%, could 
dramatically change its microstructures. 
1.3 Grain Boundary Structures 
Grain boundary, as an important interface in functional ceramic microstructure, 
plays a key role in governing the grain growth,domain nucleation.Grain boundary 
structures,their properties and nanograin boundary are presented here. 
1. 3. 1� Concepts of grain boundary structures 
Raw materials for ceramics are powders with fine particle size. Many nucleus of 
crystal will form and grow during sintering. When nucleus grow into grains and 
join each other, grain boundaries are then formed. Atoms are out of order in grain 
boundary and phase boundary (boundary between different phases), which are also 
zones with higher energy and will affect many functional processes in functional 
ceramics. Grain boundaries are regarded as the most active part in microstructures 
since they provide main diffusion pathway in sintering, driving force for 
densification and grain growth, nucleation location for domain formation, and 
sources for many processes related to energy in materials. 
Since ceramics are sintered at high temperatures and solubility usually 
decreases as temperature goes down, some elements will be precipitated or 
� ��� 1� Microstructure and Properties of Functional Ceramics 
separated out during cooling (called segregation), which often occurs at grain 
boundaries. The segregation content at grain boundaries could be 100~1000 times 
higher than that within grains. By adjusting segregation, material properties could 
be improved, even new materials could be developed. In addition, the second 
phase in grain boundaries could be glass or crystalline phase, and proper second 
phase could help to reduce sintering temperature, prevent crystal from transition 
(as shown in Fig.1.6), or restrain grain 
growth to obtain fine-grain structures. 
Under high temperature diffusion in 
boundaries is much greater than that 
within grains due to its out-of-order and 
exoteric structures as well as liquid like 
phases. Paterson has indicated that at 
sintering temperature (at about 2/3Tm,
with Tm as the melting point of materials) 
atomic leap in defect planes (grain 
boundaries) is millions faster than that in 
ordered lattice. The transportation of mass 
by grain boundaries during sintering is just 
like transportation in a city. In ABO3
materials including BaTiO3 and PbZrTiO3,
oxygen could be easily removed by 
diffusion through oxygen vacancy in 
ABO3 structure, but nitrogen could not, 
so sintering in oxygen helps to obtain transparent materials with high density, but 
N2 would be difficult to be remove by diffusion when sintering in air or nitrogen 
atmosphere. Sintering techniques for transparent ceramics could be designed by 
controlling segregation and diffusion. 
Potential barriers caused by segregation of oxides and impurities could lead to 
PTC effect, which could be utilized for developing self-regulating heating 
materials (Zhu, Yao, Zhao, et al, 2001). 
1. 3. 2� Properties of grain boundary structures 
Usually ceramics have been through with sintering and cooling, which causes 
stresses in ceramics. For asymmetric crystal it would produce grain boundaries 
with tensile or compressive stress, and property differentiation between grains and 
boundaries including capacity, vacancy concentration, conductivity, and stress 
state, which are often origin of material degradation. 
The disorder and free space in grain boundary zones provide viscoelasticity, 
Fig. 1.6 Microstructures of talc porcelain 
(a) Grain; (b) Glass phase 
1.3 Grain Boundary Structures� ���
which accommodate stress and strain, i.e. boundary zones could not only serve as 
source and sink for vacancy but also accommodation for stress and strain. High 
energy of grain boundary makes itself as a source and nucleation for domain 
formation, which influence many processes related to energy including diffusion, 
segregation, phase transition, domain formation, degradation and fracture. 
Grain boundaries often act as captive centers, where space charge is 
accumulated (as shown in Fig.1.44(b)). When electric field applied, properties of 
grain boundaries perpendicular to the applied field may differ from that of those 
parallel to the applied field, or even composition and properties of one side of 
grain boundary may differ from that of the other side. 
During sintering grain boundaries would migrate towards curvature center so as to 
decrease boundary area and reduce energy of system. However migrating outwards 
may also take place to expand volume of low-energy phases. Regions swept by 
boundary migration are often with no or few pores. Migration speed would be 
affected by impurities and pores, which could produce differences of 104 times. 
Grain boundary migration would directly affect sintering and abnormal grain growth. 
Thermal etching could be used to observe the boundary migration �including 
acentric and centripetal migration, or restrained or pinned boundary migration). 
Most sintered ceramics have not reached equilibrium, and above-mentioned 
“shell” or “core” may occur. Grain boundaries are only several nanometers 
(several layers of atoms), but boundary zones may span hundreds of nanometers. 
Since properties in boundary zones differ from those within grains, properties of 
ceramics would be greatly affected by proportion of shell to core, width of 
boundary zones, and volume percentage of boundary zones. With decrease of 
grain sizes, proportion of boundary zones increases and turn into nano materials 
after certain dimension. Normally grain sizes for nano ceramics are defined 
between 1 to 100 nm. 
1. 3. 3� Nano grain boundary structures 
Nano materials possess extraordinary properties that are different from traditional 
materials, which are related with its special microstructure. Since their grain sizes 
reach nanometers, atoms at boundaries are in a state of newly solid state structures, 
which differs from that of crystal or amorphous states. The proportionof boundary 
atoms are quite large, even as many as those within grains. For example, if grain 
sizes are 6 nm and boundary layers are 1 nm in width, volume of boundaries could 
account for 50%. Grains are micro crystals with long-range order while grain 
boundaries have various inter-atomic distances with an average atomic density 
10%~30% lower than that of grains. High concentration of grain boundaries in 
nano ceramics gives rise to small size effect including quantum effects and 
� ��� 1� Microstructure and Properties of Functional Ceramics 
macroscopic quantum tunnel effects. Since having high surface energy, large 
specific surface, high concentration of defects, and high activity, nanomaterials 
have very important applications in medical treatment, as catalyst and filtering 
materials. Research on atomic alignment in grain boundary zones and effects of 
triple junctions in nano ceramics have received broad attention for better 
understanding of their properties (Yuan, Gao, 2002). 
Recently, preparations of nano powders have been numerous. However, 
difficulties have been encountered when preparing nano ceramics from nano 
powders since grain growth could not be restrained during long-time heating and 
soaking through ordinary sintering process. Therefore many techniques including 
microwave sintering, spark plasma sintering and hot-press sintering have been 
used to prepare nano ceramics. 
Ritzhaupt-Kleissl (1999) has successfully sintered nano ceramics of PZT, 
ZrO2, TiO2, and Al2O3 with grain sizes from 100~150 nm. The sintering 
temperature has been reduced by 100� and the soaking time was almost zero, 
so as to effectively avoid grain growth during sintering. For example, PZT 
ceramics with a density of 97%~98% theoretic density (TD) has been obtained 
and sintering temperature has been reduced from 1200� to 1000~1050�. By 
using impulse discharges of high current in air, metal wires have been turned 
into nano powders (24~40 nm) as raw materials for nano ceramics. Compressed 
with magnetic impulse compression of 2.8 GPa, green bodies of 83% TD were 
produced from the obtained powers. Microwave sintering has been finally used 
to prepare nano ceramics. 
Z Surowiak et al (2001) also have prepared nano PZT ceramics from nano 
powders. Sol-gel method was used to produce powders with 30 nm, which were 
then hot-pressed at 1127� to obtain nano ceramics with grain sizes of 0.5 μm and 
density of 99% TD, each grain consists of many nano crystals. 
Lian Gao (2002) from Shanghai Institute of Ceramics, Chinese Academy of 
Sciences has successfully prepared nano ZnO ceramics with grain sizes of 100 nm 
by using spark plasma sintering at 550�.
H Varma (2002) have prepared transparent nano hydroxyapatite ceramics with 
grain size of 250 nm. It is worthwhile to notice that human teeth are live nano 
ceramics with grain size of 50~200nm. 
1.4 Pore Phases 
Porosity of green bodies may reach 25%~35%, and is also called original pores. 
During sintering, mass is driven by surface energy to fill cavities so as to reduce 
pore volume. With increase of green body density, connected and open pores 
would be reduced to form separated and closed pores. Some pores may be 
1.4 Pore Phases� ���
eliminated while some would be finally left in grain boundaries or within grains 
if migration of boundaries is too fast. Usually sintered ceramics may contain 5% 
porosity. The pores are a little bit far from grain boundaries and difficult to be 
completely eliminate due to long distance for diffusion. Pores in ceramics are 
also location for stress concentration, which could affect strength and many 
other properties of materials. For example, pores may decrease magnetic 
induction, elastic modulus, bending strength, magnetic permeability, 
piezoelectric coefficient, and electric strength. In thermal shock tests, crack 
often originate from pores (Clarke, 1962). Pores could also affect ferroelectrics 
and ferromagnetism since pores act as pins against domain formation and 
movement. For example, when porosity decreases with density increased from 
7.45 g/cm3 to 7.65 g/cm3, (as shown in Table 1.4 later) electromechanical 
coupling coefficient Kp could increase from 0.65 to 0.75 because pores could 
interdict transfer or connectivity to some properties. In addition, solid-gas 
interfaces (or internal surface) of pores tend to catch space charge, and pores are 
sometimes called reservoir for space charge. When applied with Alternating 
Current, ferroelectric ceramics tend to present electric fatigue, i.e. ferroelectric 
degradation. For ferroelectric ceramics with high porosity, electric fatigue is 
even more substantial. For example, normally sintered PLZT ceramics (7/65/35) 
with density of 93% TD show fatigue after 105 cycles while hot-press sintered 
PLZT ceramics with density above 99% TD present no degradation even after 
109 cycles, owing to very few pores in hot-press sintered ceramics. 
Pores play a significant role on transparency of transparent ceramics. When the 
porosity decreased from 3% to 0%, the transparency increases from dramatically 
0.0x% to 100%. Meanwhile the material strength and dielectric withstanding 
voltage also increase remarkably. Following measures could be taken to reduce 
porosity: 
1. Introduction of additives to prevent re-crystallization so as to remove pores through 
grain boundaries. 
2. Sintering in oxygen to remove O2 in closed pores through amalgamation and 
diffusion. 
3. Improving particle packing and fluidity of granule in die-pressing to increase 
density of green bodies. 
4. Improving wettability between main crystal phase and glass phase etc.. 
However pores are necessary instead of baneful in some other functional 
materials and manufacture of porous ceramics with specific pore size and 
proportion has been developed into a subject, which is very important for 
functions including filtering, purifying water, purifying gas, humidity sensor and 
gas sensor. It is reported (Beppu, 2002) that porous ceramics with 99% porosity 
have been prepared from fine saponite powdery (0.1μm) with a high specific 
surface of 100 m2/g. 
� ��� 1� Microstructure and Properties of Functional Ceramics 
1.5 Domain Structure 
There are electric and magnetic domains within grains in some materials. For 
some crystals, there’s a distance between positive and negative electric charge 
centers, and it gives rise to spontaneous polarization Ps, whose direction could 
be changed according to applied field. These crystals are called ferroelectric 
materials characterized by Electric Hysteresis Loop. Inside these crystals, there 
are many micro zones similar to twins, i.e. domains. Let’s take BaTiO3 as an 
example. When cooling down to below phase transition temperature TC, cubic 
BaTiO3 will turn into tetragonal phase, and spontaneous polarization Ps, will 
then be produced in neighboring grains along certain crystal axis. Electric 
domains consisting of many crystal cells then emerge in tetragonal BaTiO3.
Above TC BaTiO3 exhibit cubic symmetry; below TC any one of the three a axes 
could be slightly prolonged to become polarization axis, i.e. c axis. In 
single-crystallized tetragonal BaTiO3, Ps of neighboring domains could only 
form an angle of 90�or 180�. Fig.1.7 shows domain structure in tetragonal 
BaTiO3 crystal (Liu, Xu, 1990; Xu, et al, 1978). As shown in the figure, every 
square stands for a crystal cell with an arrow as its spontaneous polarization 
direction. Zones consisting of crystal cells with the same spontaneous 
polarization Ps direction are called electric domains. In Fig.1.7, the area 
A1A2A3A4 or A1A2B1B2 are electric domains, and the boundary between two 
domains, such as A1A2, B1B2, are called domain walls. Domains with opposite Ps
direction are called 180�domains and the boundaries are called 180�domains 
walls, suchas B1B2. If Ps directions of two domains form a right angle, it is 
called 90�domains walls, such as A1A2 and A3A4.
Fig.1.7 Domain structure in tetragonal BaTiO3 crystal 
(Liu, Xu, 1990; Xu, et al, 1978) 
Fig.1.8(a) shows a cubic BaTiO3 crystal. At temperature under TC, electric 
domains emerge. Let’s suppose there are only two domains in a crystal. They 
could be either 180�domains (as shown in Fig.1.8(b)) with the crystal cell 
1.5 Domain Structure� ���
elongated along c axis and retracted along a axis, or 90�domains (as shown in 
Fig.1.8(c)) with the crystal cell elongated along their own Ps direction and 
retracted along a axis. Lattice parameters of BaTiO3 crystal at 20� are: a = b =
0.39920 nm, c = 0.40361 nm, and c/a = 1.011, 90�. Domains walls are 101 twin 
planes, and the measured angle between Ps of two adjacent 90�domains is 
actually 88�30�due to c>a. Fig.1.9 shows domain structure of tetragonal BaTiO3
(macroscopic model). 
Fig.1.8 The distortion of domain structure in BaTiO3 single crystal (Liu, Xu, 1990) 
(a) The topography image of cubic BaTiO3 crystal; 
 (b) The 180�twin domain structure in tetragonal BaTiO3 crystal (solid line); 
(c) The 90�twin domain structure in tetragonal BaTiO3 crystal (solid line) 
Fig.1.9 Domain structure of tetragonal BaTiO3
(macroscopic model) (M. D. Liu, Y. C. Xu, 1990) 
Electric domains with polarization axis parallel to crystal surface are called 
a-domains; while domains with polarization axis perpendicular to crystal surface 
are called c-domains. The difference of refractive indices along a axis and c axis 
at room temperature is: na�nc=0.055, thus domains could be observed under 
polarized light due to its characteristics of birefringence. With orthogonal 
polarized light perpendicular to the crystal surface, c domains (as dark area) and 
a domains (as bright area) could be observed. In BaTiO3 crystal of orthorhombic 
system, Ps is along 011 direction of formerly cubic system, thus there are also 
60�and 120�domains besides 90�and 180�domains; In BaTiO3 crystal of 
trigonal system, Ps is along <111> direction of formerly cubic system, thus there 
are also 60�and 109�domains besides 180�domains. Since positive sides of 
domain corrode faster than negative sides during chemical etching, domains could 
be observed from etched surface with “up and down” like morphology. Recently 
developed scanning probe acoustic microscope (SPAM) could be used to observe 
� ��� 1� Microstructure and Properties of Functional Ceramics 
domain structures without any special treatment. Actually domain structures of 
ferroelectric crystals or ceramics are quite complicated and randomly oriented, 
containing 90�, 180�domains and other interactive structures, as shown in 
Fig.1.10 (a), (b), (c), (d). 
An external field needs to be applied to compel randomly oriented domains to 
realign coincidently to the polarized direction, and to expand domain volume to 
the same direction as the applied field, so as to produce an overall Ps. Fig.1.11 (a) 
shows the schematic poling process in piezoelectric ceramics PZT, and (b) shows 
domain orientation after poling and chemical etching. 
1.5 Domain Structure� ���
� ��� 1� Microstructure and Properties of Functional Ceramics 
Fig.1.10 Domain and domain configuration 
(a) 90�and 180�domains in piezoelectric ceramics Z-4; (b) Domain configuration (PLZT, 
chemical etching, Phase contrast microscopy, direction of electric field E: shown as arrow); 
(c) Ribbed domain, replica, TEM; (d) Intersection of domains 
(PLZT, 8/67/33, after chemical etching, TEM) 
Fig.1.12 shows domain configuration under different poling conditions. If Ps
inverses for 180�, no additional stress would be introduced; if non-180�domain 
wall motion occurs, deformation would be induced (for example, elongation along 
poling direction) to produce mechanical stress. Ceramic plates may fracture if the 
internal stress is too high. For poled ceramics, internal stress would be gradually 
released in deposition for a period of time, which is called aging. 
T Ogawa (2002) has estimated switching and reorientation of electric domains 
by measuring ferroelectricity and piezoelectricity under applied field and their 
relationship. For example, switching of 180�domains could be evaluated from 
1.5 Domain Structure� ���
the poling fields for minimum Kp and K33; switching of 90�, 71�, 109�
domains could be evaluated from the poling fields for εmax or minimum value of 
frequency constant. 
Fig.1.11 Poling process in piezoelectric ceramics (sketch) (Xu, et al, 1978) (a) 
and domain orientation after poling piezoelectric ceramics (PZT, chemical etching) (b)
Electric domains often nucleate at locations where defects (such as grain 
boundaries) or stress concentrate. Nucleation of 90�domains would eliminate 
substantial internal stress. The possibilities of nucleation at grain boundaries are 
as follows: intersection points of four gains > intersection lines of three grains > 
intersection plane of two grains. Several twin-crystals often share a point along 
grain boundaries, i.e. nucleation of domain may cross grain boundaries: when a 
grain turns into tetragonal system from cubic system, domains show band 
structures and stress would occur along grain boundaries, which makes the 
domain movement in neighboring grains move easier along specific direction. 
During phase transition, the front side of band domains tends to cross grain 
boundaries to release stress, so as to minimize system energy, as shown in 
following section (Fig.1.43 (c)). Band width of 90�domains may vary with 
strength of internal stress, i.e. larger band width under smaller stress and vice 
versa. ε is often larger with more domain walls, and domain width is proportional 
to grain size: 
� ��� 1� Microstructure and Properties of Functional Ceramics 
Fig.1.12 Domain configuration under different poling conditions 
(Li0.11Na0.89NbO3,Ec=2400V/mm, crossed nicols) 
(a) 1900 V/mm, < Ec, 120�/20min; (b) 2500V/mm, around Ec, 120�/20min; 
(c) 7000 V/mm, >Ec, 120�/ 20min 
90�domain width � (grain size)1/2.
90�domains play a very important role in adjusting and relaxing stress. There 
are a large amount of 90�domains within grains of ferroelectric ceramics. 
However, very few domain walls would be left if a grain is separated out since 
these domain walls are created from twining to release stress produced during 
phase transition. When a grain is separated from other grains, volume change from 
1.5 Domain Structure� ���
phase transition could be easily accommodated by free surface movement, so 
there’s no need for domains to adjust strain. Poly-crystals always have a high 
stress status relative to single crystals, thus there’s always a driving force to reduce 
residual stress, which causes aging. Aging of properties is often related to 
following processes: 
1. Change of domain configuration with time. 
2. Trend of domain configuration towards equilibrium. 
3. Migration of impurities and vacancy towards domain walls or grain boundaries. 
4. Regularization of impurities and vacancy along poling axis etc. 
For example, back rotation of 90�domains increases ε, while back rotation of 
180�domains decreases ε. The internal friction of ferroelectric ceramics also lies 
on domain configuration and the interaction between domain walls and lattice 
defects. 
N G Zhang et al (2001) have investigated poling fatigue of PLZT (2/70/30) 
ceramics, and discovered that after certain number of switching cycles fatigue 
occurred, and piezoelectricity degraded with a decreased Pr. In addition, fatigue 
occurred earlier at lower frequency. For example: 
Degradation of ferroelectricity occurred after 103 cycles at 10Hz; 
Degradation of ferroelectricity occurred after 104 cycles at 50Hz; 
No degradation observed after 1010 cyclesat 100Hz; 
Little degradation observed after 1010 cycles at 500Hz; 
M L Eng (1999) has investigated domain switching properties of nano BaTiO3
ceramics. It has been concluded that in a specific field, domain switching would 
occur only when critical switching time τc is satisfied, i.e. if frequency of applied 
field is higher than 1/τc, electric domains could not be switched and no aging or 
fatigue could occur even under applied field. 
It is commonly accepted that fatigue originate from (1) micro cracks caused 
by mechanical stress, and (2) migration of defects under electric field. A 
Levstik et al (1997) has performed test of cycled electric load (15 kV/cm, 
50Hz) on PLZT (8/65/35) ceramic and measured its piezoelectric coefficient 
d33 and quality factor Q. It has been discovered that d33 started to decrease 
after 4�105 cycles, Q started to decrease after 106 cycles, and micro cracks 
occurred after 3�106 cycles. The earlier degradation of d33 may be caused by 
domain switching. 
Cross research group from Pennsylvania University of the US has conducted 
systematic research on electric fatigue of ferroelectric ceramics(Jiang, Cross, 
1993). If materials degrade dramatically under cycled loading of alternative 
field, it would limit their applications in high-strain actuators and non-volatile 
memory devices. Fatigue is related to many factors including surface status, 
preparation of electrodes applied, microstructures, composition and working 
temperatures etc., in which pores in microstructures play an important role. 
� ��� 1� Microstructure and Properties of Functional Ceramics 
The number of switching cycles to produce 70% decrease of remnant 
polarization has been taken as the criteria to estimate electric fatigue of 
materials with different density. 
It is well known that there’s space charge on surfaces of poled ceramic to 
neutralize polarization, and pore sizes in ceramics could reach 1μm or even 
dozens of micron. When high voltage field applied, space charge could enter 
grains, grain boundaries to interact with domain walls, which would stabilize 
domain configuration and make domain switching more difficult under applied 
fields, so as to decline the polarization. Under applied field, domain may extend 
from one surface to the opposite side in as-received BaTiO3 single crystal. 
However after several cycles of switching, domain would be confined within 
crystal body instead of extending from one side to the other. Space charge often 
accumulates around pores and thick grain boundaries which behave like pins on 
domain, and cause electric fatigue. In addition, the ceramic layer next to 
electrodes would undergo some electrochemical reactions under applied field 
and change its color obviously, which is un-restorable; while above mentioned 
fatigue due to domain pining could be recovered to as-received status by heating 
up to phase transition temperature T
εmax.
There’s also space charge within materials to neutralize the polarization: 
positive and negative space charge are accumulated on positive and negative sides 
of domain respectively and establish a space charge field. Space charge also 
accumulates at grain boundaries and could affect domain motion. Analysis on 
domain configuration of ferroelectric ceramics is an important method to 
understand relationship between poling condition and domain orientation, aging 
property and switching behavior. A large amount of electric domains and domain 
walls are produced to minimize internal stress during phase transition, so as to 
reduce strain energy and overall system energy, same as magnetic domains. 
Recently, strong piezoelectricity has been found in a single crystal of 
Pb(Zn1/3Nb2/3Ti0.09)O3 solid solution: K31=80%, K33=95%, d33=2500pC/N 
(K31=40%, K33=70% for ordinary soft PZT ceramics). To explain the strong 
piezoelectricity, Ogawa et al ( 2002) has conducted research. They found that 
under proper poling conditions, a single domain could be produced cross over a 
single crystal, which brings along strong piezoelectricity. However, if the poling 
voltage is unfavorably increased, multiple domains will be formed and 
piezoelectric property will be degraded. Thus proper poling conditions are crucial 
to produce ideal domain configuration, so as to enhance piezoelectric properties. 
However, there are also other theoretical explanations to the phenomenon of 
strong piezoelectricity in these materials. 
1.6 Mechanical Properties of Ferroelectric Ceramics� ���
1.6 Mechanical Properties of Ferroelectric Ceramics 
Ferroelectric ceramics are operated at high electric fields and at mechanical loads 
to use their full electromechanical potential.The intrinsic coupling of electrical and 
mechanical effects leads to the development of fracture mechanical experiments 
and concepts in ferroelectric ceramics. 
1. 6. 1� General 
Electric properties of ferroelectric ceramics have been comprehensively studied 
while only a few researches have been focused on their mechanical properties and 
internal stress which play a very important role on materials functions. Here is an 
example to demonstrate the effects of internal stresses on working lifetime of 
ceramics. When preparing transparent ferroelectric ceramics through hot-press 
sintering, 99% alumina ceramics were used to make hot-pressed dies which had to 
sustain pressure of 20~30MPa under 1150��for dozens of hours including multiple 
heating and cooling. They should also be reusable without damage. Casting process 
was initially used to produce hot pressed dies with 65mm in diameter and 85mm in 
height, and the dies were sintered at 1780�. However the obtained dies consistently 
fractured after being used only once. Isostatic pressing process was introduced then 
to produce the dies, and they were still intact after being used over 50 times. The 
dies produced with isostatic pressing had homogeneous structures and low internal 
stresses, and retained their high sustaining strength after repeated uses; while dies 
produced with casing process contained substantial internal stresses which 
significantly decreased the strength. Thus internal stresses play significant roles on 
working lifetime of ceramics. Similarly, once refilled with boiling water, vacuum 
flask made without annealing would fracture immediately. 
Internal stresses in ceramics play a significant role on material properties, and 
methods have been developed to measure and control internal stresses. 
Recently, more interests have been focused on mechanical properties than 
electric properties for piezoelectric and ferroelectric ceramics (Pferner, 1999), 
since these materials would be exposed to severe mechanical loading in the 
applications such as ultrasonic transducers, actuators and piezoelectric 
transformers. In some newly applications such as of piezoelectric powered 
pressure heads and fuel injection system of automotive technology, the 
expectation on mechanical properties for ceramics are even highly. 
The research on mechanical properties also helps to understand electric 
properties. For example: 
� ��� 1� Microstructure and Properties of Functional Ceramics 
BaTiO3 ceramics with fine grains are frequently used as capacitors, and their 
dielectric coefficient is even higher than that of average value of single crystals, 
which was quite elusive for almost fifty years. However the researches on 
mechanical properties and internal stresses have uncovered this puzzle. Details are 
discussed as below. 
Ceramics or polycrystals are prepared through sintering and cooling. Following 
factors could bring various deformations to grains during cooling: 
1. Differences in expansion and contraction among grains. 
2. Anisotropy of non-cubic crystal system. 
3. Variety in grain sizes. 
4. Composition diversity among grains. 
5. Ionic doping. 
Since grains