Zuo-Guang. Ye Advanced Dielectric Piezoelectric and Ferroelectric Materials: Synthesis, Characterisation and Applications

Подождите немного. Документ загружается.

Handbook of dielectric, piezoelectric and ferroelectric materials850

ferroelectric polarization of Bi

single crystals’, Appl. Phys. Lett. 10 (1),

14–16.

12. Cummins S E and Cross L E (1968), ‘Electrical and optical properties of ferroelectric

single crystals’, J. Appl. Phys., 39 (5), 2268–2274.

13. Aurivillius B (1949) ‘Mixed bismuth oxides with layer lattices-I. The structure type

of CaNb

’, Arkiv Kemi 1, 463–480; ibid. (1949), ‘Mixed bismuth oxides with

layer lattices-II. Structure of Bi

’, 1, 499–512; ibid. (1950), ‘Mixed bismuth

oxides with layer lattices-III. Structure of BaBi

’, 2, 519–527.

14. Smolenskii G A, Isupov V A and Agranoskaya A I (1959), ‘A new group of ferroelectrics

(with layered structure)’, Sov. Phys. Solid State 1, 149–150.

15. Subbarao E C (1961), ‘Ferroelectricity in Bi

and its solid solutions’, Phys.

Rev. 122 (3), 804–807.

16. Fang P H, Robbins C R and Aurivillius B (1962), ‘Ferroelectricity in the compound

’, Phys. Rev. 126 (3), 892–896.

17. Aurivillius B and Fang P H (1962), ‘Ferroelectricity in the compound Ba

’,

Phys. Rev. 126, 893–896.

18. Subbarao E C (1962), ‘A family of ferroelectric bismuth compounds’, J. Phys.

Chem. Solids 23, 665–676.

19. Newnham R E, Wolfe R W and Dorrian J F (1971), ‘Structural basis of ferroelectricity

in the bismuth titanate family’, Mater. Res, Bull. 6, 1029–1039.

20. Armstrong R A and Newnham R E (1972), ‘Bismuth titanate solid solutions’, Mater.

Res. Bull. 7 (10), 1025–1034.

21. Takenaka T and Sakata K (1986), ‘Piezoelectric properties of grain-oriented bismuth

layer-structured ferroelectric ceramics’, Proc. 6th Inter. Symp. Appl. Ferro, IEEE,

414–417.

22. Takenaka T and Sakata K (1988), ‘Grain-oriented and Mn-doped (NaBi)

(1–x)/2

ceramics for piezo- and pyrosensor materials’, Sensors and Materials,

1, 35–46.

23. Takenaka T, Sakata K and Toda K (1985), ‘Piezoelectric properties of bismuth layer-

structured ferroelectric Na

0.5

4.5

Ceramic’, Jpn. J. Appl. Phys. 24 (Suppl.

24–2), 730–732.

24. Nagata H, Takahashi T, Chikushi N and Takenaka T (2000), ‘A series of bismuth

layer-structured ferroelectrics’, Ferroelectrics 241, 309–316.

25. Nagata H, Takahashi T, Yano Y and Takenaka T (2001), ‘Electrical properties of

bismuth layer-structured ferroelectrics Sr

m–3+x

4–x

m–x

3m+3

(m = 2, x = 1~2;

m = 3, x = 0~2)’, Ferroelectrics 261, 219–224.

26. Yamashita Y, Yoshida S and Takahashi T (1983), ‘Effects of MnO additive on

piezoelectric properties in modified (PbCa)TiO

ferroelectric ceramics’, Jpn. J. Appl.

Phys. 22 (Suppl. 22–2), 40–42.

27. Lotgering F K (1959), ‘Topotactical reactions with ferrimagnetic oxides having

hexagonal crystal structures – I’, J. Inorg. Nucl. Chem. 9 (2), 113–123.

28. Cross L E and Pohanka R C (1971), ‘Ferroelectricity in bismuth oxides type layer

structure compounds’, Mater. Res. Bull. 6, 939–949.

29. Takenaka T and Sakata K (1981), ‘Electrical properties of grain-orientated piezoelectric

ceramics in layer-structure oxide family’, Jpn. J. Appl. Phys. 20 (Suppl. 20–3), 161–

163.

30. Fouskova A and Cross L E (1970), ‘Dielectric properties of bismuth titanate’, J.

Appl. Phys. 41 (7), 2834–2838.

31. Schulgasser K: (1976) ‘Relationship between single-crystal and polycrystal electrical

conductivity’, J. Appl. Phys. 47 (5), 1880–1886.

WPNL2204

Grain orientation and electrical properties of bismuth 851

32. Wu S Y, Takei W J, Francombe M H and Cumuins S E (1972), ‘Domain structure

and polarization reversal in films of ferroelectric bismuth titanate’, Ferroelectrics 3,

217–224.

33. Baerwald H G (1957), ‘Thermodynamic theory of ferroelectric ceramics’, Phys.

Rev. 105 (2), 480–486.

34. Marutake M (1956), ‘A calculation of physical constants of ceramic barium titanium’

J. Phys. Soc. Jpn. 11, (8), 807–814.

35. Nanao M, Hirose M and Tsukada T (2001), ‘Piezoelectric properties of Bi

TiNbO

–

BaBi

ceramics’, Jpn. J. Appl. Phys., 40 (9B), 5727–5730.

36. Shibata K, Shoji K and Sakata K (2001), ‘Sr

1–x

piezoelectric ceramics

with high mechanical quality factor’, Jpn. J. Appl. Phys., 40 (9B), 5719–5721.

37. Byer R L and Roundy C B (1972), ‘Pyroelectric coefficient direct measurement

technique and application to a NSEC response time detector’, Ferroelectrics 3, 333–

338.

38. Dougherty J P and Seymour R J (1980), ‘Automated simultaneous measurement of

electrical properties of pyroelectric materials’, Rev. Sci. Instrum. 51 (2), 229–233.

39. Smolenskii G A, Isupov V A and Agranovskaya A I (1961), ‘Ferroelectrics of the

oxygen-octahedral type with layered structure’, Sov. Phys. Solid State, 3 (3), 651–

655.

40. Takeuchi T, Tani T and Saito Y (1999), ‘Piezoelectric properties of bismuth layer-

structured ferroelectric ceramics with a preferred orientation processed by the reactive

templated grain growth method’, Jpn. J. Appl. Phys., 38 (9B), 5553–5556.

41. Nagata H, Takahashi T and Takenaka T (2000), ‘Piezoelectric anisotropies of bismuth

layer-structured ferroelectrics’, Trans. Mater. Res. Jpn, 25 (1), 273–276.

42. Dorrian J F, Newnham R E, Smith D K and Kay M I (1971), ‘Crystal structure of

’, Ferroelectrics, 3, 17–27.

43. Villegas M, Caballero A C, Moure C, Duran P and Fernandez J F (1999), ‘Factors

affecting the electrical conductivity of donor-doped Bi

piezoelectric ceramics’,

J. Am. Ceram. Soc., 82 (9), 2411–2416.

44. Shulman H S, Testorf M, Damjanovic D and Setter N (1996), ‘Microstructure,

electrical conductivity, and piezoelectric properties of bismuth titanate’, J. Am. Ceram.

Soc., 79 (12), 3124–3128.

45. Noguchi Y and Miyayama M (2001), ‘Large remanent polarization of vanadium-

doped Bi

’, Appl. Phys. Lett., 78 (13), 1903–1905.

46. Shannon R D (1976), ‘Revised effective ionic radii and systematic studies of interatomic

distances in halides and chalcogenides’, Acta Cryst. A 32 (5), 751–767.

47. Suzuki M, Nagata H, Ohara J, Funakubo H and Takenaka T (2003), ‘Bi

3–x

TiTaO

(M = La or Nd) ceramics with high mechanical quality factor Q

’, Jpn. J. Appl.

Phys. 42 (9B), 6090–6093.

48. Nagata H, Fujita Y, Enosawa H and Takenaka T (2004), ‘Piezoelectric properties of

Nb and V substituted Bi

ferroelectric ceramics’, in Ceramic Materials and

Multilayer Electronic Devices, Ceramic Transactions, Vol. 150, The American Ceramic

Society, 253–263.

WPNL2204

852

28.1 Introduction

Conventional processing of perovskite and perovskite-related materials is

usually carried out by solid-state reactions of metal carbonates, hydroxides

and oxides. These reactions require a long grinding time and high-temperature

treatment so that the starting ingredients are provided with sufficient energy

to come together and react. This process may lead to the formation of multiple

phases in the final product because of incomplete reaction. The final product,

if not homogeneous, usually suffers a decrease in its physical properties. For

this reason, soft chemistry has forged its way into the synthesis of mixed

oxides. Soft chemical routes to mixed oxide compounds are based on

transformations that take place in solution. These methods allow the starting

materials to homogeneously mix in solution and, therefore, react at the atomic

or molecular level to give rise to precursors that can be converted to the final

oxide under milder conditions than those required by solid-state synthesis.

The sol–gel process and the co-precipitation method are two commonly used

soft chemical routes for the synthesis of a variety of mixed oxides including

perovskite and perovskite-related materials.

This chapter will focus on two soft chemical methods, namely, the sol–gel

process and the co-precipitation method, and their roles in the synthesis of

perovskite compounds exhibiting ferroelectric and relaxor ferroelectric

properties, useful for a wide range of technological applications. Sections

28.2.1 and 28.2.2 will describe the principles of the sol-gel and co-precipitation

methods, respectively, highlighting the advantages they offer to the synthesis

of mixed oxide compounds. In Section 28.3, a new room temperature sol–

gel method for the synthesis of the relaxor ferroelectric-based lead magnesium

niobate–lead titanate solid solution system, (1 – x)Pb(Mg

1/3

2/3

–xPbTiO

(with x = 0.10 and 0.35) is described. Section 28.4 will deal with two new

soft chemical processes, namely a sol–gel and a co-precipitation method for

the synthesis of ferroelectric strontium bismuth tantalite, SrBi

(SBT),

of layered perovskite structure. The sol–gel method developed for the synthesis

Novel solution routes to ferroelectrics

and relaxors

K BABOORAM and

Z-G YE, Simon Fraser University, Canada

WPNL2204

Novel solution routes to ferroelectrics and relaxors 853

of SBT has been extended to the synthesis of ferroelectric bismuth titanate,

(BT), and the lanthanum-doped system, bismuth lanthanum titanate,

4–x

(BLT). The details of the synthesis and characterization of

BT and a series of BLT compounds ranging from x = 0.2 to 0.75 will be

described in Section 28.5. Finally, Section 28.6 will be dedicated to the

future development in this exciting field of soft chemical approaches to

ferroelectric and relaxor ferroelectric materials.

28.2 Soft chemical methods for the synthesis of

mixed metal oxides

28.2.1 Sol–gel process

The sol–gel process has gained considerable attention and popularity in

recent years. The ability to homogeneously mix (and dissolve) the starting

materials, commonly metal alkoxides, in solution is one of the major benefits

of the sol–gel processing method [1]. Developed in 1845 by Ebelman who

made the first silica gels, this technique has extensively been employed in

the synthesis of inorganic mixed oxides because of the numerous advantages

it offers over the conventional solid state reactions [2]. Sol–gel processing

has highlighted the importance of chemistry in the fabrication of materials

starting from chemical precursors. It was not until 1950 that the first non-

silicate ceramics, the perovskites, were synthesized using the sol–gel process.

In the last two decades, this soft chemical technique has attracted considerable

interest and has been used for a major development in the synthesis of mixed

oxides in the form of glass, ceramics, and thin films. Using the sol–gel

method, the syntheses of electronic and ionic conductors, and ferroelectric

and magnetic materials in the form of high-purity submicrometer powders,

have been accomplished [2,3].

The basic requirement of this type of processing is the formation of a

homogeneous solution (sol) containing all the necessary cationic ingredients

in the desired stoichiometry. Therefore, it is very important to choose the

right solvent, one that is capable of effectively dissolving all the starting

chemicals in question. This allows for a good mixing of the reactants at the

molecular level and, once this has been achieved, the next step of progressively

forming a network of oxide in solution through polymerization of the dispersed

constituents becomes a lot easier. Broadly speaking, the sol–gel method

consists of polymerization reactions which are based on the hydrolysis and

condensation of metal alkoxides M(OR)

, where M is a metal having an

oxidation state of n

and R is an alkyl group. Briefly, in the first step of the

reaction, the metal alkoxide is hydrolyzed by a water molecule. Once the

hydrolysis step has been initiated, the next reaction that follows is condensation

of the hydrolyzed molecules to produce M–O–M linkages or networks. Further

WPNL2204

Handbook of dielectric, piezoelectric and ferroelectric materials854

condensation reactions can then take place, leading to the formation of longer

chains or oligomers. The processes can be summarized in the reaction sequence

in Fig. 28.1. The formation of more extended networks in the sol through

further polymerization of the chains and oligomers causes gelation to occur.

The homogeneous network of M–O–M units then decomposes at a much

lower temperature to give rise to the desired inorganic oxide that is required

in solid-state reactions. One of the key features of the sol–gel method is the

ability to produce small particles, as small as nanometer size. The precursor

powder obtained by the decomposition of the homogeneous gel is actually

much finer than powders obtained by the solid-state reactions and this explains

the increased reactivity of the former [3].

In summary, the sol–gel process offers a better control of stoichiometry

and homogeneity, a higher purity and reactivity of the precursor powders,

and a significantly reduced temperature of ceramic densification when

compared with solid state synthesis. By allowing the preparation of stable

sols, the sol–gel method also opens opportunities to the fabrication of thin

and thick films through, e.g., spin-coating and other methods. The main

disadvantages of the sol–gel process are the cost of the metal alkoxide precursors

and the complexity of the processing.

28.2.2 Co-precipitation method

The co-precipitation method, often referred to as the solid-state analog or a

side-branch of the sol–gel process, is another efficient soft chemical route to

mixed oxide materials [4]. Briefly, this method, also starting from solution,

involves dissolving the starting materials in a common solvent and then

Hydrolysis

M(OR)

O M(OR)

n–x

(OH)

(a)

(b)

Dehydration

RO M OH

HO M OR

RO M O

M OR

Dealcoholation

RO M OH

RO M O

M OH

ROH

28.1

(a) Hydrolysis of metal alkoxides and (b) condensation of the

hydrolyzed species through dehydration or dealcoholation steps.

WPNL2204

Novel solution routes to ferroelectrics and relaxors 855

adding a precipitating agent so as to form a homogeneous and single phase

inorganic solid. The precipitate can then be decomposed at a high temperature

to produce the target mixed oxide material. Co-precipitation helps to hold

the required cations close together in the reaction medium and lowers the

temperature of decomposition, just as in the case of the sol–gel process. The

starting materials can be simple metal salts that can easily be dissolved in

water or other appropriate solvents.

Figure 28.2 shows, as an example, the synthesis of the complex metal

oxide, FeCr

, by the co-precipitation method. The first step involves

dissolving an iron(III) salt and a chromate in water to obtain Fe

and

CrO

2–

ions in solution. These ions are then precipitated in a complex form by an

ammonium solution

(N )

, and the resulting precipitate can be decomposed

at high temperature into the target FeCr

28.3 Polyethylene glycol-based new sol–gel route to

relaxor ferroelectric solid solution

(1 –

)Pb(Mg

1/3

2/3

–

PbTiO

(

= 0.1 and 0.35)

28.3.1 Background

The relaxor ferroelectric-based solid solution system (1 – x)Pb(Mg

1/3

2/3

–xPbTiO

((1 – x)PMN–xPT) has attracted considerable interest

because of its remarkable dielectric and electromechanical properties in the

form of ceramics and single crystals [5–7]. Pb(Mg

1/3

2/3

(PMN) is well

known to exhibit typical relaxor ferroelectric behavior characterized by a

broad and frequency-dependent dielectric maximum at T

max.

(around –15 °C

at 1 kHz) [8]. Addition of the normal ferroelectric PbTiO

(PT) to PMN

induces a long-range ferroelectric order, shifts the transition temperature

upwards and enhances the dielectric, piezoelectric and ferroelectric properties

of the solid solution [9, 10]. A morphotropic phase boundary (MPB) appears

in the composition region of 30–37 mol% PT with the presence of a

rhombohedral, a tetragonal, and/or a monoclinic phase [11]. The (1 – x)PMN–

xPT system embraces a wide range of compositions which find important

applications in multilayer dielectric capacitors, electromechanical sensors

and actuators [12]. Compositions within the MPB (e.g. 0.65PMN–0.35PT)

+ 2CrO

+ NH

2–

Fe(CrO

)

Precursor to FeCr

2NH

Fe(CrO

)

1150°C

2Fe

+ 2NH

+ H

O + 7/2O

28.2

Synthesis of FeCr

by the co-precipitation method.

WPNL2204

Handbook of dielectric, piezoelectric and ferroelectric materials856

exhibit typical ferroelectric hysteretic character, with a large spontaneous

polarization and a low coercive field, showing great potential for advanced

electromechanical applications both in the forms of electroceramics and thin

or thick films [13]. Recently, single crystals of (1 – x)PMN–xPT solid solutions

have been found to exhibit electromechanical performance superior to that

of the (1 – x)PbZrO

–xPbTiO

(PZT) ceramics [14,15] which have been

used for decades as piezoelectric materials for transducers and actuators

[16]. On the other hand, being more on the PMN side of the solid solution

system, 0.90PMN−0.10PT is a typical relaxor ferroelectric material with a

weak rhombohedral distortion [10]. It exhibits a diffuse dielectric constant

maximum at about 40°C (T

max

) and its dielectric constant at room temperature

is very high (ε′ ~ 10 000), making it a potential candidate for multilayer

high-power density capacitor applications [16,17].

The synthesis of pure perovskite 0.90PMN−0.10PT and 0.65PMN–0.35PT

by conventional solid-state reactions can be challenging because of the

formation of a more stable but undesired impurity phase of pyrochlore structure

which considerably impairs the dielectric properties of the materials [18].

For this reason, soft chemical methods have emerged as the best alternatives

for the synthesis of the (1 – x)PMNT–xPT ceramics.

Among the various routes to homogeneous multi-component oxides that

have recently been developed, the polymeric precursor method has drawn a

lot of attention [19]. Pioneered by Pechini in the 1960s, this method provides

a simple way to oxide powders by first making a solution containing all the

cations in the form of polymer–cation complexes in the desired stoichiometry,

and then firing this solution to drive off any organic moeties and to form the

target oxides [20]. Chelation and entrapment of the cations by the polymer

chains give rise to crystalline powders with better chemical homogeneity

and smaller particle size. This is explained from the fact that chelation

considerably lowers the mobility of the cations, thereby stabilizing the cations

against precipitation [21], which is the key requirement to obtaining a stable

precursor solution that can be used for making thin films.

Several researchers have reported the synthesis of multi-component ceramic

oxides via the oxidation of polyethylene glycol– or poly(vinyl alcohol)–

cation complexes [19,21–24]. The Pechini method, which uses α-

hydroxycarboxylic acids such as citric and lactic acids together with

polyhydroxyl alcohols such as ethylene glycol to make a polymer resin, has

been modified to some extent to the use of water-soluble sources such as

metal nitrates and chlorides. This has now become a very common method

because these inorganic salts are more stable than the corresponding alkoxides

and, hence, relatively easy to handle. Polyethylene glycol (PEG) has proved

to be a very suitable polymer for use in the polymeric precursor method in

the preparation of mixed oxide powders and ceramics. The reason for this is

the coordination of the ether oxygen atoms in PEG to the metal cations,

WPNL2204

Novel solution routes to ferroelectrics and relaxors 857

which allows it to dissolve a number of inorganic salts as well as metal

alkoxides. Commonly, sol–gel processes using alkoxides require refluxing

at temperatures usually above 100°C to allow for hydrolysis and condensation

reactions to take place and to form the cross-linked structures.

Besides polymer-binding, cross-linking can also greatly help in obtaining

a very homogeneous distribution of cations in solutions. This is achieved by

forming extended structures in solution by crosslinking the cations using a

molecule with several binding sites. Sriprang and coworkers [25] demonstrated

this through a triol-based route to PZT thin films in which 1,1,1-

tris(hydroxy)methylethane (THOME) was used as the complexing agent.

They demonstrated that, with its three hydroxyl groups, THOME could bind

to up to three metal ions at a time, forming oligomeric species that were then

oxidized to form the final mixed oxide compound. However, to the best of

our knowledge, no application of such technique was made to the synthesis

of the (1 – x)PMN–xPT system.

28.3.2 Results and discussion

We have developed a novel soft chemical route for the synthesis of 0.90PMN–

0.10PT and 0.65PMN–0.35PT ceramics [26,27]. Using a 1:2 volume mixture

of polyethylene glycol-200 (PEG200) and methanol as solvent for the starting

materials, and THOME as a complexing agent, we were able to perform a

room temperature reaction to produce the precursor solution, as well as to

obtain the pure (pyrochlore-free) perovskite phase for the PMN–PT ceramics.

The flowchart shown in Fig. 28.3 details the steps involved in the new sol–

gel method. An interesting feature of this method is that the precursor solutions

to the target PMN–PT ceramics were formed at room temperature under

simple mixing and stirring conditions, thus avoiding the reflux and distillation

steps usually required in sol–gel reactions. Moreover, the sol is not moisture

sensitive and shows very good stability against precipitation, which makes it

a very promising stock solution for depositing thin films. Furthermore, this

new route to PMN–PT ceramics does not require the use of any excess of

lead starting material, which is necessary in all conventional and most of soft

chemical syntheses in order to compensate for lead oxide loss during sintering

at high temperatures.

In the case of the 0.65PMN–0.35PT system, ceramics of pure perovskite

phase, high density, high dielectric constant and excellent ferroelectric

properties were obtained from a stoichiometric amount of the lead starting

material in the preparation of the sol. A decrease in the ceramic quality in

terms of the perovskite phase purity and dielectric properties was observed

in those ceramics prepared without using the triol molecule, THOME, in the

reaction. Figure 28.4 shows the positive effect of THOME on the perovskite

phase formation by comparing the 0.65PMN–0.35PT powders prepared with

WPNL2204

Handbook of dielectric, piezoelectric and ferroelectric materials858

(PEG(S)T) and without (PEG(S)) the triol molecule and calcined at different

temperatures.

The presence of THOME in the system during the synthesis of precursor

solution has not only led to the formation of a pure perovskite phase, but also

greatly improved the dielectric and ferroelectric properties of the 0.65PMN–

0.35PT ceramics with a maximum dielectric constant,

′

max

of ~ 27000 and

a remnant polarization, P

= 21µC/cm

, as compared with

′

max

= 9000 and

= 7.5 µC/cm

for the ceramics prepared without THOME. The temperature

and frequency dependences of the real part of dielectric permittivity (ε′) of

the ceramic prepared with THOME (PEG(S)T) and the one prepared without

THOME ([PEG(S)), both sintered at 1140 °C, are shown in Fig. 28.5. The

arrow (MPB) in Fig. 28.5(a) indicates the morphotropic phase transition

from the rhombohedral or monoclinic to the tetragonal phase. The higher

maximum dielectric constant with a sharp peak in the ε′ versus temperature

curve is associated with a larger grain size (>2µm) in the ceramic prepared

Soln. containing all

the cationic species

Precursor soln.

Precursor powder

PMN-PT

ceramics

Pb-acetate

Mg-pentanedionate

THOME

(PEG200/MeOH)

Nb-ethoxide

TIAA

THOME

(PEG200/MeOH)

Pb-Mg

soln.

Nb-Ti

soln.

3h stirring

at room temperature

3h stirring

at room temperature

5h stirring

at room temperature

Rotary evaporator

at ~ 75°C

Heat at 500°C

Calcination and

sintering

28.3

Flowchart for the preparation of precursor solutions and

ceramics of (1 –

)PMN–

PT (

= 0.10 and 0.35). (Reproduced with

permission from

Chem. Mater.

2004, 16(25), 5365–5371.)

WPNL2204

Novel solution routes to ferroelectrics and relaxors 859

10 20 30 40 50 60 70 80

2θ (Deg.)

(a)

Intensity (a.u.)

1040 °C

950 °C

850 °C

750 °C

650 °C

500 °C

10 20 30 40 50 60 70 80

2θ (Deg.)

(b)

Intensity (a.u.)

1040 °C

950 °C

850 °C

750 °C

650 °C

500 °C

28.4

X-ray diffraction patterns of (a) PEG(S)T and (b) PEG(S)

precursor powders of 0.65PMN–0.35PT calcined at different

temperatures for a period of 8h. (Reproduced with permission from

Chem. Mater.

2004, 16(25), 5365-5371.)

WPNL2204