Luận văn effect of zr and la based co doping on electrical properties of lead free barium titanate batio3 thin films

and energy-storage efficiency 1 on applied electric field for BZT thin films al various ammealing temperatures.. Dependence of volumetric energy-storage density ae os recoverable energy-

TLANOI UNIVERSITY OF SCIENCE AND TECIINOLOGY

MASTER’S THESIS

Effect of Zr and La based co-doping on

electrical properties of lead-free barium

titanate BaTiO; thin films

TRAN THI DOAN

Doan.‘t'l202748M @sis.hust.eduyn

Major: Materials Science

Supervisor: Prof Vu Ngoc [lung

Department: Micro-Blectro-Mechanical-System Laboratory

Tnstitute: International Training Institute for Materials Science

HANOI, 05/2022

HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY

MASTER’S THESIS

Effect of Zr and La based co-doping on

electrical properties of lead-free barium

titanate BaTiO; thin films

TRAN THI DOAN Doan TT202748M @sis.husteduyn

Major: Material Science

Supervisor (Sign and write fall name)

HANOI - 05/2022

Acknowledgment

Fustly, I would like to cxpress my deepest appreciation to my supervisor,

Prof, Vu Ngoc Hung, who direcfly instructed me throughout this tescarch project

I am also grateful to Dr Nguyen Duc Minh of MDSA+ Institute for

Nanotechnology at University of ‘Iwente, Netherlands { am extremely thankful

and indebled to him for sharing his expertise as well aa the valuable guidance and encouragement extended to me

T would like to extend my sincers thanks lo MSc Dang Thi Ha, Dr Vu Thu,

Lien, Dr Ngo Due Quan, and all of members in MUMS laboratory for their

enthusiasm to help me through the process of research

Ltake this opportunity to express gratitude to all professors, lecturers, and

employees al TTTIMS for their kindness to support me durmg a period T have

already studied and worked there

Finally T want to give special thanks to my family for providing me with

their unfailing support and encouragement during my years of study and

research

Abstract

BaTliO;-based materials with a perovskite structure have attracted mterest

because some of them are potentially valuable materials due lo their

environment-(riendly properties In this study, lead-free Ba(79 25Tip.7)05 (BZT)

and La-doped Ba(Z1925'ip 75)03 (BLZ‘L) thin films were grown on Pt/l'i/Si0,/Si

substrates via a sol-gel spin-coating method

‘The effects of various annealing temperatures (450 ?00 °C) for BZ thin

Ons on microstructure, dielectric, and energy storage performances were

systematically investigated As XRD result, it was found that the degree of crystallization of the films increased with the increasing amnealing temperature (2) This result indicated a pure polycrystalline perovskite phase of BZT thin

films achieved at 700 °C At the same time, the dielectric constant of the BZT

film also increases as the armealing temperature increases In particular, the

optimal energy-storage density of 30.9 J/cm’ and a large energy-storage

efficioncy of 67.8% could be obtained in the film anncaled at 500 °C, which not

only achieved a large breakdown strength (up to 7000 kV/cm) but also exhibited

a great temperature-dependent energy storage performance stability in a wide

temperature (from 30 °C to 200 °C) a good frequency-dependent energy storage

performance stability in ranging from 100 ta 10000 Hz and an excellent charge-

discharge cycling life with fatigue-free performance up to 10° cycles

Moreover, the effects of La doping of BZT thin films (from 0 — 8 mol.%)

on microstructure, dielectric, and energy storage performances were also

invesligaled As XRD result, if was shown that, La doping enhanced the ability of crystallization of the films with the perovskite B7.T phase achieved at 650 °C

La-doped BZT thin films indicated prominently increasing relaxor behavior with

increasing La-doping concentration In particular, the films with 5 mol.% La-

doping simultaneously exhibit a quite high recoverable energy-storage density

(~7.0 vem") and a large enerpy-storage efficiency (~ 60.7%) under an gp of

1650 kV/em Moreover, dieleewie coustant of La-doped BZT thin films was found to be significantly improved, reaching the wiaximum value of 164 for 3 mol.% La-doped BZT thin film

These results indicated thal lead-free BalZr92sTigz3)O3 and Ta-doped Ba(21/asTia;;)O; thín films were expected to become a candidate materials for

cnergy-storage capacitors

Master student

(Sign and write fall name)

1.2.1 Basic knowledge on dielectric capacitor

Measuring methods of cncrgy-storage deusity for dielectric

11 1.23 Potential dielectrics for high energy-storage application 14

1.3 Overview of barium titanate-based materials - - 15

1.3.2 Liffeels of doping on BaliO; properlie "¬¬

21 Fabrication of BZI and BLZT thin films by sol-gel spin coating

21.1 Overview of sol-gel spin coating method 20

2.2 Methods to investigate the structure and properties of BZT and BLZT

3.1.2 — Ferroelectric properties and breakdown strength (Epp) 34

3.1.4 Dieleetrieproperiies Xueesssusoou 38

3.2 Hfeots of La-doping on properties of BZ/1 thn ñlme 38

Chemical Vapor Deposition

Differential scanning calorimetry

Barium titanate, BaTiO, Barium zirconate titanate, Ba(4r,Ti}O;

Barium lanthanum zirconate titanate, (Ba,La)(Zr.Ti)O;

Pulsed laser deposition

Lead zirconate, PhZrO;

Lead zirconate titanate, Pb(Z1,Ti)O;

Lead lanthanum zirconate titanate, (Pb,La}{(Zr,TijO3

Polar nanoregions

Lead magnesium niobate, Po(Mg,gNbia)Os

Lead zirconate titanate, Pb(Zx,Ti)O3

Relaxor ferroelectric

Thermogravimetric analysis X-Ray diffraction

LIST OF FIGURES

Figure 4.1 (a) Cubic perovskite unit call ABOs, (b) Perovskite lettice structure

Figure 1.2 Polarization as a function of temperature in (a) first and (b) second

Figure 1.3 Frequency dependence of polarization I9 - - 4

Figure 1.5 Orientation of dipoles in the ferroelectric materials (a) absence of electric field (b} under electric field and (c) after removal of electric field [23] 6 Iigure 1.6 Temperature evolution of dielectric constant showing the characteristic temperatures in RFE Representative hysteresis loops for each

Figure 1.7 Hysteresis behavior in (a) ferroelectric and (b) relaxor materials [27]

igure 1.10 The diagram of charge separation in parallel-plate capacitor under

Figure 1.11 The diagram of measurement, circuit for the energy-slorage density

Figure 1.12 (Color online) The typical dependence of (a) polarization and (b)

permittivity on electric field of ferroelectrics in the first quarter [30], L3 Vigure 1.13 Diagram of hysteresis and energy storage density for (a) linear

dielectrics, (b) lerroclectries, (c) relaxer [erroclevtrics, and (d) anli-[erreeicetrics

The green area in the first quadrant is the recoverable energy densily Uyeco, and

the red area is the energy loss U,,, [32] - - 14

Figure 1.14 Schematic of the perovskite structure of BaTiO; ‘@ Cubic lattice

(above Curie temperature, 120°C), (b) ‘Tetragonal lattice (below Curie

Figure 2.3 Example of processing routes to obtain sol-gel spin coatings [16] 23 Figure 2.4, Flow diagram for producing BZT and BLZT sols 24

Figure 2.6 Schematic of spin-coating and heat treatment process for producing

Figure 2.7 Spin-coaling machine al, Inlemalional Traming Tnstilule for Matcrials

Figure 2.8 Schematic representation of the Drage’ s law for diffraction BỊ 28 Figure 2.9 The working principle diagram of X-ray diffractometer and the PANalytical X’Pert PRO system [48] khien "— ˆ Figure 2.10 The schematic drawing of a Sawyer-Tower circuit used for hysteresis measurement al ferrocleciric thin fil [49] 30

Figure 3.3 Polarization-electric field (P-2) hysteresis loops and th values of

Pree, Py and Pyyey P, for BET thin films at various annealing temperatures The

measurements were performed at 1000 kV/cm and 1 kllz 34

Figure 5.4 Electric field dependence of P,,., and, values for BZT thin films at

various annealing temperatwes, measured until their conesponding electric

breakdown strength (Egy) The dala were calculated from the corresporiding P-F

Tigure 3.5 ‘Dependence of vohmett ¢ enerpy- storage density Cv, recoverable

energy-storage density U,, and energy-storage efficiency (1) on applied electric

field for BZT thin films al various ammealing temperatures The dala were

calculated from the corresponding P-E loops 37

Figure 3.6, (4) Fnergy-storage densilies, recoverable energy-slorage and (b) energy-storage efficiency were measured at the corresponding Lp, values, for BZT thin films at various annealing temperatures 38 Figure 3.7 (a) Dielectric constant — electric field (e- 0 curves and by dielectric loss curves of BZY thin films at various annealing temperatures, measurement at

Figure 3.10 Polarivation-electric field 104 FR) liysteresis loops and (b) values of Prey, Py and Pag, - P, for BET thin films with various La doping contents (0-8 mol.%), The measurements were performed at 1000 kV/om and 1 kHz 41

Figure 3.1L Hlectric field dependence of 2,4, and P,, values for La-doped BZ‘

thin films at various La doping contents measured until their corresponding

electric breakdown strength (2p) The data were calculated from the

corresponding #-# LOOPS ccsssscsessssssiesssssseisssssseeseessseeeseesssineeiess + 42

Figure 3.12 Dependence of volumetric energy-storage density (ae os recoverable energy-slorage đensity (17.„„), and cnergy-storage efficiency (17) on

applied electric field (or BZ.T thin fibns with various Ta doping conLents (a) 0%,

(b) 3%, (c) 5%, (d) 8% The data were calculated from the corresponding P-E

Figure 3.13 (a) Energy: storage © densities, recoverable energy-storage and (b)

energy-storage e[liciency were measured at the comresponding gp values, for

BZT thin films with various La doping contents 44

Figure 3.14 (a) Dielectric constant electric field (c-K) curves and (b) dielectric

loss curves of BZT thin films with various doping content, measurement at room

Figure 3.15 The operating-temperature dependence of (a) P-E keop, (©) Pinas Pr

and Py ?,-values for BZ'T thin film at an annealing temperature of 500 °C The measurements were performed al 4000 kV/em and 1000 Ay 45

Figure 3.16 The operating-lomperalure dependence of (a) cnorgy storage densily (O) and (b) energy storage efficiency (y) for BZT thin film at annealing temperature of 500 °C The measurements were performed at 4000 kV/cm and

Figure 3.17 The Sequencies temperature dependence of of PE ! loops for BAT thin

film at annealing temperature of 500 °C The measurements were performed al

Figure 3.18 The operating-frequencies dependence of @) Pa P, and Prax - Py

values, (b) E, values, (c) energy storage density (U) and (d) energy-storage efficiency (9) for BZT thin film BZT thin film al anmealing temperature of 500

°C The measurements were performed at 4000 kV/em and room temperature 47

Figure 3.19 a) Comparison of P-E hysteresis loops measured at different

charge-discharge oycles, (b) Pix and P, values as a function of number charge- discharge cveles under an applied electric field of 4000 kV/em and 1 kHz, for the

BAT thin film at annealing temperature of 500 °C The fatigue testing was

performed by applying a tpolar electric field of pulse height 200 KV/em and at

pulse width 100 KLIz (07 5 U8) ccccsssscessseeseesunssseeee sees .49

Figure 3.20 Dependence of (a) energy storage density and q@) energy storage efficiency () on cycling for BZT thin film at an annealing temperature of 500°C

The dala were calculated from the corresponding P-E hysleresis loops performed

at 1000 kVvem, 1 kHz and room temperature - - 49

LIST OF TABLES Table 2.1 Parameters of chemical componenls used to synthesize BZT and

‘Table 3.3, Electric breakdown strength (#5p) of BZT thin films with various La

Table 3.4 The measured dielectric constant, dicieviric loss for BZT thin fis

INTRODUCTION

Nowadays, onergy storage is realized mainly based on traditional devices

such as accumulators and electrochemical cells Electrochemical cells typically

have a fairly high energy density but their power density takes on a relatively low value, Therefore, current studies are focusing on developing capacitors with high

energy density and small charge/discharge times (~10° seconds) to be used in

storage devices Although these devices have a lower energy density than electrochemical cells, their power density is much higher As a result, capacitors are offen usually beg used to gorrate a pulsed voltage or current in devices

such as shock absorbers and medical pacemakers To enhance the features of

cunrent energy storage devices the dielectric properties of a capacitor need to be optimized and for that finding a suitable dielectric material is key for industrial

applications Dicleotric materials with high power densily, fast charge-discharge process, high discharge efficiency, and low cost have gained the potential

application in pulsed capacitors technology recently, such as radar transmitters,

lasers, comrol devices, and pacemakers ‘hese materials need to meet criteria

such as high spontaneous polarization (P,), small remanent polarization (P,), and

high breakdown strength (Zpp) Thus, ferroelectric materials, relaxor-ferroelectric

materials, and anti-ferroelectric materials are suitable to satisfy the above criteria

‘The current research in the field of energy storage by capacitors is carried oul based on the application of ferrucleviric materials and Jead-containing anti- ferroelectric materials such as Ph(Zx,TH}Os, (PhiL.a)(7x,Ti)Os, and PhZzO; Many studies have shown that these materials have outstanding advantages such as high charge/discharge energy density and high energy storage efficiency (Pbạp

Lag y(Zt9.52'8i9.49)03 (PLZT) thin films): Ujeco ~ 23.2 Jom? and 9~ 91.6) [1]

However, for most of the components of PAT materials, the lead element

accounts for quite a large proportion, more than 60% by weight Meanwhile, the

lead element is highly toxic and can evaporate quickly during the heat treatment

process, it is considered environmental pollution and is harmful to human health

Thus, there is a necessity to find novel environmentally friendly malerials as an alternative to traditional lead-containing materials for high-power capacitor applications

Recently, BaliO;-based materials such as Ba(⁄r,Tu„}O; (H⁄I) with a

perovskite structure have attracted interest duc to its fcrroclcetric proporties

dependent on the substitution concentration of Ti ions (ionic radius of 0.745 A) with Zr‘ ions (ionic radius of 0.860 A) [2,3] In this casc, the isovalent

substitution of Ti” ions by Zr" ions can transform ferroelectric nmuoro-domains

into high dynamic polar nano-regions (PRNs), exhibiting normal ferroelectric transition to relaxor behavior with changing Zr concentration [4,5] Due to the larger ionic radius of Zr" ions, the lattice parameters of BZ.T are expanded while

the ferroelectric-paraelectric phase transition is decreased [6] Moreover, the Zr" ion is more chemically stable than ‘Ti ion, which is expected to depress the conduction caused by clectron hopping between TẾ ion and TỶ" ion |7], and

vi

therefore, the properties of BZI are improved Recently, there has been considerable interest in the development of BZT-based thin films for various applications due to their excellent ferroelectric properties at a low La-doping concentration, La-doping enhances the relaxor behavior in lead-free BZT thin filus by introducing a disorder al the A siles (Ba?') of BZT unit cells, which can

‘be achieved by donor substitution of La’ for Ba’ ioms The results achieved the optimum values of 72.2 J/cm? recoverable energy-storage density and 78.2%

energy-storage efficiency under a high 3.8 MV/em electric breakdown strength for 5 mol.% La-doping [8] This showed that proper La-doping concentration can enhance the relaxor behavior and, hence, significantly improve the energy- storage performance and breakdown strength of BZ thin-film capacitors According to the above analysis, the research topic was sclected: “Effect of Zr and La based co-daping on electrical properties of lead-free barium titanate BaTiO; thin films”

Research objectives of the thesis:

Fabrication of lead-free ceramic material for practical application in the pulsed-power capacitors industry

Lnvestigation of electrical properties of fabricated materials

Investigation of energy-storage properties of capacitors using fabricated xuaterials

Investigation of the slabilily of capacitors using fabricated materials toward

the practical application in pulse-power electronic systems

Investigate the effect of various annealing temperatures on microstructure,

dielectric, and energ

Investigate the thermal stability, frequency stability, and fatigue endurance

of capacitors using BZT materials

‘The layout of the thesis: includes 3 main chapters, 55 pages (no mention

of table of contents and references)

Introduction: In this section, the reasons for choosing topic, research

objectives, research subjects, research methods, and research contents of the

thesis are presented

Chapter 1: Literature reviews

Chapter 2: Experiments and methods

Chapter 3: Results and discussion

Conclusions: In this section, the main points and results in the thesis are briefly re-stated

viii

CHAPTER 1 LITERATURE REVIEWS

This chapter presents the theoretical basis of ferroclectric and relaxor

ferroelectric properties At the same time, principles for high energy-storage in

dielectric capacitors, as well as potential dielectrics, are also given In particular, this discussion will focus on giving an overview of barium titanate-basod

qmalenals, which have allacled considerable allention due io their excellent

energy storage characteristics Not only that, BaTiO,-based materials are environmentally friendly materials as an altemative to relaxor-ferroelectric PLZT thin films (or high-power capacitor applications

1.1 Overview of ferroclectric and relaxor ferroclectric propertics

111 Perovskite structure

Among all the structures, perovskites and tungsten branze are found to have

best piezoelectric and ferroelectric characteristics The perovskite structured selaxor materials are being widely studied ‘'he main advantage of this structure

1s that many different cations can be substituted on both A and B sites without

drastically changing the overall structure and also a complete solid solution can

be achieved between many cations over a range compositions [9] Anisotropy in piezoelectric properties is lage in perovskites compared to all other structures The perovskite walerials can readily undergo phase transitions [10] aud their

structure is that of mineral perovskite (CaTiOQ;), which is orthorhombic The

ideal perovskite is centrosymmetrie with general formula ABO; where ‘A’ site cation valence varying from +1 to +3 and ‘B” site is occupied by the cations of

valence 13, 4 or 05 The schematic represenation of ABO; lype perovskile is

shown in ligure 1.1 (a), where the 13-site ion is in the body center position, A-

site cations are in the cubic comer position and the oxygen atoms are at the face

centered position and form an octahedron around the B-site [11] The easiest way

to visualize the structure is in terms of the BO, octahedra which share corners

infinitely in all 3 dimensions, making for a very wice and symmetric slructure

(Figure 1.1(b)) The A cations occupy every hole which is created by eight BOs

octahedra, giving the A cation a 12-fold oxygen coordination, and the B-cation a 6-fold oxygen coordination

‘Yo get a stabilized structure there is a size constraints like charge neutrality

constraints The sive constraint is described by the tolerance factor £ For the

perovskite structure,

Tạ + Tạ

V2 0 +7,)

where 7 and ry are the radius of the A-site cation (in 12 coordination) and B-site

cation (in 6 coardination) respectively and rp is the oxygen ion radius For the

ideal perovskite system ¢ should be in the range of 0.95 to 1.04 for cubic symmetry and larger lor the distorted perovskite system:

Egil

() (b)

Figure 1.1 (a) Cubic perovskite unit cell ABOs, (b) Perovskite lattice structure BO;

[12]

11.2 Ferroelectrics

Ferroelectricity was discovered by Valasek in 1921 in Rochelle single

crystals (NaKCH,0¢.4H,0) [13] From then on interest in fabrication of new

ferroelectric material is increasing rapidly A compound exhibiting permanent

dipole moment is called as ferroelectrics Ferroelectricity can be defined as the material that contains one or more polar axes along which spontaneous polarization can be developed below the Curie temperature (7) Similar to

pyroelectric materials, ferroelectric materials have spontaneous polarization and

the direction can be reversed by the applied external electric field The arrangement of cations and anions within the ferroelectric materials gives rise to

dipole moments within each unit cell, and the resulting polarization can be measured via material surface current A distinctive feature of ferroelectric

material is hysteresis behavior in polarization vs electric field Spontaneous

polarization exists even after the removal of electric field and is called as

remnant polarization, P, At 7, transformation from ferroelectric to paraelectric is

happening Paraelectric phase materials will behave as a normal dielectric with

no hysteresis Detailed discussion about the hysteresis will be found in the next

section

1.1.2.1 Phase Transitions in Ferroelectric Materials

There are two types of ferroelectric phase transition, order-disorder and

displacive [14] In the order-disorder type of ferroelectrics, there is a dipole

moment in each unit cell At high temperature these dipoles are in random

directions and lowering the temperature there will be phase transition where the

dipoles will be orderly arranged and within a domain all the dipoles are pointing

towards the same direction This type of transition can be observed in hydrogen

bonded ferroelectric materials [15] The displacive transition can be understood

in terms of polarization catastrophe, in which, if an ion is displaced from

equilibrium position, the force from the local electric fields due to the ions in the crystal increases faster than the elastic-restoring forces This leads to an

asymmetrical shift in the equilibrium ion positions and hence to a permanent

dipole moment lonic crystals such as barium titanate (BYO) are displacive

ferroelectrics

Ferroelectric matenals will undergo sccorl order or first order transition The second order phase transition is characterized by gradual reduction of spontaneous polarization (P,), with increase in temperature and becoming zero at

T, and above Transition in tri glycine sulphate is sceand order type [16]

Whereas in Ihe first order transition there will be a chscomtimuous reduction of P,

to zero at 7, Barrum titanate undergoes first order transition [17] ‘The first and

second order transitions are explained schematically in Figure 1.2

Figure 1.2 Polarisation as a function of temperature in (a) first and (bj second order

phase transition [18]

1.1.2.2 Polarization

Spontancous polarization is duc to the ordering of dipoles under the

influence of intemal process in a dielectric material without the effect of extertal

factors Polarization occurs due to several atomic mechanisms The total

polarization (P, ;) can be written as,

Protat = Pz + Ppt Pot Poo Fig l2 where P,P, #, and P,, correspond to electronic, ionic, orientational and space charge polarization respoetively [19]

Electronic polarization (P2): The clectric ficld causcs deformation or

translation of the originally symmetrical distribution of the electron clouds of atoms or molecules ‘t'his is essentially the displacement of the outer electron clouds with respect to the inner positive atomic cores

Atomic or ionic polarization (P): In an ionic lattice, the positive ions are

displaced in the dircolion of an applied field while the negative ious are displaced

in the opposite direction, giving a resultant (apparent) dipole moment to the

whole body

Orientational polarization (2,): Polarization arises from the orientation of

anolecules that has permanent dipole moments and these dipole moments are due

to the asymmetric charge distribution Tis also known as dipole polarization,

Space charge polarisation (Py): Al ligher fields, carrier injcetion

‘becomes important For materials consisting of a high concentration of charge

carriers, polarization due to the migration of charge carriers to form space

charges at interfaces or grain boundaries becomes important This type of

polarization is called space charge polarization

‘The polarization is not constant rather it will vary concerning the measuring

frequency [19] ‘The variation of polarization for the frequency is given in Figure

nommal dielectric materials The observalion of the hysteresis loop is still

frequently used for the identification of ferroelectrics The rectangularity of the hysteresis loop is the main requirement for memory cells The linear relationship between the electric field and the polarization is given by,

where g and y are the vacuum dielectric permittivity (8.854 x 10™ Fim) and

susceptibility of the material, respectively A typical ferroelectric hysteresis loop

ix shown in Figure 1.4

As the electric ficld £ strength increases, the munber of domains with

dilTerent polarization dircotions will switch towards the ficld direction, producing

a rapid increase in polarization (AJ3) When all the domains are aligned in the field direction saturation is reached (BC) A† this saturation state, appropriately oriented crystals will be composed of a single domain The extrapolation of the

linear segment on the polarization axis represents (he saturation polarizalion, P,

(CBE) As the field strength decreases, the polarization will decrease but does not go back to zero (BD) When the field is reduced to zero, some of the domains

will remain aligned and the material will exhibit remnant polarization (P,) The

field required to remove the P, or to reduce the polarization back to zero is called the coercive field (Z,), Further increase of field in the negative direction will

cause dipole alignment in this direction and the cycle can be completed by

reversing the field direction

Electric field (E)

Figure 1.4 Ferroelectric hysteresis loop [19]

1

Ferroelectric materials are composed of domains which is a small regions

with uniform polarization, Usually, in ferroelectric material, there are many domains and the direction is different for neighboring domains In a single

domain, all the dipoles are aligned in the same direction and this direction can be

reversed by the external electric field The net polarization along one particular

direction will depend on the ratio of oppositely aligned domains along that

direction If they are in equal volume then the net polarization is zero The

domains are separated by the domain walls The change in dipole moment can be brought by a change in the temperature and by an external electric field This

results in domain wall movement, nucleation, and the growth of new domains [14] The ferroelectric domains were first demonstrated in the study of

spontaneous birefringence [20] Ferroelectric domain structure can be observed

by second harmonic generation [21], etching, a liquid crystal method [22] The

usefulness of each technique varies from one material to another with the shape,

size, and transparency of the material The structure of the domains depends on

the structure of the crystal In a single orystal, there is a variety of domain patterns and the number of types of domain walls depends on the number of

orientations of the dipole moment when the spontaneous polarization occurs Domain structure is strongly dependent on the symmetry of the ferroelectric

phase

4, Ferroelectric Domains

1.1.2.5 Poling

Single crystals or polycrystalline ferroelectric materials are having

multiple domains A single domain can be obtained by domain wall motion

which is possible by the application of an appropriate electric field A very strong

field that could reverse the polarization in the domain is called as poling or

polarization switching Simply, the process of applying an electric field to a ferroelectric material to orient the dipoles in the same direction is called poling

As already discussed when the electric field increases, the polarization is also

increasing, as the alignment of the dipole in the same direction is boosted, When all the dipoles are aligned in the same direction maximum polarization value is

attained and the material is said to be saturated then the electric field is reduced

to zero Though the applied electric field is decreased to zero the dipoles are still

aligned in the direction of the applied field with some relaxation due to the

remnant polarization The dipole behavior during poling is schematically

explained in Figure 1.5 The random orientation of dipoles is represented in

Figure 1,5 (a) At the maximum field, all the domains are aligned in the field direction and the material is said to be saturated (Figure 1.5 (b)) Even after the removal of the external field, the dipoles are aligned in the field direction but with some relaxed orientation depending on the material property (Figure 1.5 (c))

Figure 1.5, Orientation of dipoles in the ferroelectric materials (a) absence of electric

field (b) under electric field and (c) after removal of electric field [23]

1.13 Relaxor ferroelectric

Diffuse phase transition behavior is characteristic of disordered structures

In perovskite structures, random lattice disorder introduces dipolar impurities and

defects that influence the static and dynamic properties of materials The

presence of the dipolar entities on a lattice site of the highly polarizable FE

structure, induced dipoles in a region determined by the correlation length (7,)

The correlation length is a measure of the extent of dipoles that respond in a

correlated manner In normal FE, r, is larger than the lattice parameter, a and it is strongly temperature-dependent, On decreasing the temperature, a faster increase

of r, promotes the growing of polar domains yielding a static cooperative long-

range ordered FE state at T < T, This is not the case for RFE where a small

correlation length of dipoles leads to the formation of polar nanodomains

frustrating the establishment of a long-range FE state Therefore, the dipolar

nanoregions form a dipolar-glass-like or relaxor state at low temperatures with some correlation among nanodomains.

sizes of the nano-domains The temperature dependence of dielectric constant as

shown in Figure 1.6 identifies the main temperatures associated with relaxor

ferroelectric behavior The temperature that corresponds to the transformation is called Burn’s temperature (7';) At temperatures close to 7’; polar nano regions

(PNRs) are mobile and their behavior is ergodic On cooling, their dynamics

slow down enormously and at low temperature, 7; (freezing temperature) the

PNRs become frozen Freezing of the dipole dynamics is associated with a large

and wide peak in the temperature dependence of the dielectric constant (€) with

characteristic dispersion observed at all frequencies The definition of Ty follow

from the fit of the frequency dispersion of 7, with the Vogel-Fulcher law [24]

The dynamics of polar nano regions do not follow Arrhenius type temperature

dependence; instead, the nice fit of the frequency dispersion for each relaxor system is obtained with the Vogel-Fulcher law:

—E„

f =foexp (Lưng T= To) E.q 14

where fy is the attempt frequency which is related to the cut-off frequency of the

distribution of relaxation times, E, is the activation barrier to dipole

reorientation, 7, is the dielectric maximum temperature, and 7; is the freezing

temperature

The temperature dependence of the dielectric constant below 7 and in the

vicinity of Z,, is normally fitted by the empirical power law [24];

1 _ 1

£(0,T) - Emox(@,1)

where ¥ and are parameters desoribing the degree of relaxation and diffuseness

of the transition respectively The parameter y vanes belween 1 and 2, where

values closer to 1 indicate normal ferroelectric behavior whereas values close to

2 indicate good relaxor behavior Above 7's the inverse of dicleetric permittivity

is (tiled by the Curie-Weiss law:

differentiated from the normal ferroelectrics by the following properties [26]

Firsily, the P-E hysteresis loop is the signature of a FE in the low-

temperature FE phase The large renmant polarization, P,, is a marifestalion of

the cooperative nature of the I'l} phenomenon A relaxor, on the other hand,

exhibits a so-called slim loop For sutficitently high electric fields the

nanodomains of the relaxor can be onerled with the eld leading to large

polarization, however, on removing the field most of these domains reacquire

their random orientations resulting in a slim hysteresis loop Figure 1.7 represents

the hysteresis behavior of RFE and FE materials

Secondly, the saluration and remmant polarization of a FE decreases with

increasing temperature and vanishes at the Curie temperature (T,) The vanishing

of polarization at T, is continuous for a second-order phase transition and discontinuous for a first-order transition which implies the absence of polar

domains above T, By contrast, the field-mduced polarization of a relaxor

decreases smoothly through the dynamic transition temperature Tq and retains finite values to rather high temperatures and it is shown in Kigure 1.8

Figure 1.7, Hysteresis behavior in (a) ferroelectric and (6) relaxor materials [27]

‘Thirdly, the static dielectric susceptibility or dielectric constant of a Itt exhibits a sharp, narrow peak at 7 The KH response is frequency independent in the audio frequeney range By contrast, a relaxor exhibils a very broad dielectric

8

peak and strong frequency dispersion m the peak temperature (7„) and the magnitude below T, The broad peak is also referred to as a ‘diffuse phase transition’ and is associated with compositional fluctuations leading to many inicro FE regions with different compositions ‘he sharp transition in barium titanale (RTO) and diffuse transition m1 PMN single crystals are shown im Figure

Finally, the FE lransition can be thormodynamtically first er second order

and involves a macroscopic symmetry change al 7; Transparent, FE extubils a

strong optical anisotropy across T By contrast, there is no structural phase transition across J’, in a relaxor, ‘his was evidenced by X-ray and neutron dillraction studies The poak is simply a manifestation of the slowing down of

the dipolar motion below 7+ For transparent relaxors, there is no optical

anisotropy across Ty In addition, the relaxor ferroelectric materials are having

ferroelectric to antiferroelectric transition at Tz, called depolarization

Substitution of ions in the A and B sites of the ABO; perovskite with

different polarizabilities, valence states, and sizes will sufficiently produce

dipolar defecis Ti can lead to a high degree of disorder to break the translational

symmetry thereby preventing the formation of long-range order This supports the fonnation of PNRs ‘the difference in ionic radius, electronegativity, and valence state in the A and B sites can induce enough charge fluctuation,

vacancies, and lacal ordering to introduce the relaxor praperty

Very high response coefficient and an enhanced width of the high response regime around the ordering temperature T,, (Curie range) make relaxors as

popular systems for application in piezoelectric/ electrostrictive actuators &

sensors (eg scaring probe microscopy, inkjel pritder, adaptive optics,

micramotors, vibration sensors/atteruators, Hubble telescope correction, etc.)

alectro or electro-optic and photorefractive elements (segmental displays, amodulators, image storage, holograpluc data storage, etc.) [29]

1.2 Principles for Ligh Energy-Storage in Diclectric Capacitors

12.1 Basic knowledge on dielectric capacitor

A capacitor typically consists of two conductor plates filled with cerlaim

dielectric materials, and is commonly in the parallel-plate form, as shown in igure 1.10 ‘the electric energy storage is the function bases of capacitors in electronic devices Ihe energy-stored ability of a capacitor is the so-called

capacitance, which is only determined by the physical dimension (geometry) of

the conductors and the permittivity of the dielectrics It is independent of the potential difference between the conductors and the total charge on them For example, the capacitance of a parallel-plate capacitor constructed of two parallel plates fulfilled with certain dielectrics is approximately equal to the following

[30]

A

where C is the capacitance, A is the area of overlap of the two plates, ¢, is the

relative permittivity, sạ is the electric constant ( 8.85 x 107!Fm~1), and d is

the distance between the plates Obviously, the capacitance is directly proportional to the overlap arca of the conductor plates and the relalive

permittivity of the dielectrics, while inversely proportional to the separation

distance between the plates

As shown in Figure 1.10, if an external voltage V is applied to the

conductor plates, the electric polarization has happened This will result in positive and negative charges with equal content accumulating on the two plates,

respectively, which is the so-called charge process of the capacitor The charge

process will be finished when the electrical potential caused by the accumulated

charge +Q on both plates is equal to the external applied voltage V.Q/V is equal

to the capacitance C of the capacitor Sometimes, the relative permittivity of the dielectrics is changed by the external bias, causing the capacitance to vary In this

case, capacitance is defined in terms of incremental change:

Figure 1.10 The diagram of charge separation in parallel-plate capacitor under the

function of electric field [30]

During the charging process, the charges are moved between the

conductor plates by the function of external bias, indicating that work must be

done and that the electric energy is stored in the dielectrics at the same time

Hence, the amount of the stored energy W could be obtained from the following

research Generally, U values could be obtained in two ways: static method and

dynamic method

Figure 1.11 gives the circuit for measuring energy-storage density in a static

way [31] In this case, the sample capacitor is first charged by an external bias,

thus the electric energy is stored in the dielectric Then, the capacitor is

connected to the load R to complete a circuit by MOSFET switching So part of the stored energy is discharged, accompanied by the transient current formed in the closed circuit, According to the I (t) - t curve, the discharged energy could be obtained by the following formula:

where R is the resistance of the load, and ¢ is the discharged time Finally, the

energy density U could be calculated by using the ratio between W and the volume of capacitor It should be noted here that the obtained U value in this way

is the recoverable energy-storage density, because some stored energy is lost during the charge and discharge process

Figure 1.11 The diagram of measurement circuit for the energy-storage density [30]

As for the dynamic method, the energy-storage density could be induced from the above formula (1.9) From the physics base, it could be known that the charge density (Q/4) on the conductor plate of capacitor is equal to the electrical displacement D (D = &9¢,E) in the dielectrics Thus, combined with formula

(1.9), the energy-storage density U could be expressed as follows:

wf 9 Vaq Emax

where E is the external applied electrical field and equal to /d, other letters are defined just as before For the dielectrics with high permittivity, the electrical

displacement D is very close to their electrical polarization P As a result, formula (1.11) can be rewritten as:

Emax

0

Evidently, based on formula (1.9), the U value of the dielectrics can be

easily obtained by numerical integration of the area between the polarization and

the curves of the electric field-polarization (P - E) loops As shown in Figure 1.12 (a), when the electric field increases from zero to the maximum E,, , the

polarization also increases to its maximum P,,,,, and the electrical energy is

stored in the capacitor as Urs, illustrated by the green and red area; during the

discharge process from Ey to zero, the recoverable electrical energy density Uyeco i8 then released, represented by the green area in the figure, This means that

part of the stored energy (the red area enclosed by the loops) is exhausted during

the depolarization process because of the hysteresis loss Based on these results, the energy-storage efficiency can be defined as:

Figure 1.12 (Color online) The typical dependence of (a) polarization and (b)

permittivity on electric field of ferroelectrics in the first quarter [30]

Since the permittivity is defined as dP/dE, as shown in Figure 1.12 (b), the formula (1.9) can be expressed as

Emax

họ For the linear dielectric materials, whose permittivity is independent of the external applied field, the formula (1.11) could be simply expressed as follows:

ọ This result indicates that the energy-storage density for the linear dielectric materials is directly proportional to the relative permittivity of the dielectrics and the square of the operating field It should be noted here that the U value obtained from the dynamic way is usually larger than that from the static way.

1.2.3 Potential dielectrics for high energy-storage application

According to the above analysis, to design a proper dielectric material with

high recoverable energy-storage density and high efficiency (small energy loss)

for practical application, three requirements have to be satisfied simultaneously

at least: high electric breakdown field, large saturated polarization, and small remnant polarization, Figure 1.13 shows the typical P-E loops and the energy-

storage illustration of four kinds of dielectrics [32]: (a) linear dielectric with

constant permittivity (example, Al,Os, glass), (b) ferroelectric with spontaneous

polarization (example, BaTiO3, PbT103, (c) relaxor ferroelectric with nanosized

domains (example, (Pb,La)(Zr,Ti)O3, (Ba,La)(Zr,Ti)O;) and (d) anti-ferroelectric

with zero net remnant polarization (example, PbZrO;)

in the first quadrant is the recoverable energy density and the red area is the

energy loss Use: [32]

Although linear dielectrics usually possess a higher breakdown field and

lower energy loss, their smaller polarization value (permittivity) makes them not

suitable for high energy-storage applications Ferroelectrics often have larger

saturated polarization and moderate electric-field endurance, but their larger remnant polarization leads to a smaller recoverable energy-storage density and

lower efficiency Comparatively, as shown in Figure 1.13, relaxor ferroelectrics

and anti-ferroelectrics are more likely to be used for high energy storage because

of their larger saturated polarization, smaller remnant polarization, and moderate

breakdown field) Meanwhile, with the development of new manufacturing

processes of materials, such as glass-crystallization technique and composite

14

technology, other two kinds of materials, glass-ceramic and polymer-based

ferroelectrics, are also been found to have the potential for application in this

area, which combine with the higher breakdown field of linear dielectric and larger polarization of ferroelectrics

Thus, in general, above mentioned four kinds of dielectrics: relaxor-

ferroelectric is believed to be the most promising candidates for high energy storage application, and quite a few works have been reported on these materials 1.3 Overview of barium titanate-based materials

In recent years, a large number of lead-free BaTiO; - based ceramics have been studied, because some of them are potentially valuable materials due to

their environment-friendly properties Some of the ceramics display a

composition-dependent relaxor behavior This relaxor behavior varies with

composition, substitution of different types of ions, and the rate of substitution

When the overall composition of ceramics is far away from the base composition

of BaTiO; and when the ions substitution is heterovalent in the 6-coordination

number crystallographic site, the relaxor behaviour is more apparent

13.1 Barium titanate (BaTiO;) structure

Barium titanate (BaTiO; or BTO) was synthesized to become the first and the most widely studied ceramic material, due to its excellent dielectric, improved ferroelectric and piezoelectric properties [33] The high dielectric

constant of BaTiO; ceramics results from its crystal structure BaTiO; has the perovskite structure as shown in Figure (1.14)

Figure 1.14 Schematic of the perovskite structure of BaTiO; (a) Cubic lattice (above

Curie temperature, 120°C), (b) Tetragonal lattice (below Curie temperature, 120°C)

134

In Figure (1.13), each barium ion is surrounded by 12 oxygen ions The

oxygen ions plus the barium ions form a face-centered cubic lattice The titanium

atoms reside in octahedral interstitial positions surrounded by six oxygen ions Because of the large size of the Ba ions, the octahedral interstitial position in

BaTiO; is quite large compared to the size of the Ti ions The Ti ions are too

small to be stable in this octahedral position There are minimum-energy

15

positions off-center in the direction of each of the six oxygen ions surrounding the Tỉ ion Since each Ti ion has a 14 charge, the degree of polarization is very high When an electric field is applied, the Ti ions can shift from random to aligned positions and result in high bulk polarization and a high dielectric

constant [35]

Barium litaate has three crystalline forms: cubic, tetragonal, and hexagonal The tetragonal polymorph is the most, widely used because of ils excellent ferroelectric, piezoelectric, and thermoelectric properties [36] The temperature has a strong effect on the crystal structure and polarization characteristics of BaTiOs Above 120°C (and up to 1400°C), BaTiOg is cubic and the BaTiO, has a spontaneous random polarization as described above In this

temperature range the ‘Ii ion lies in the center of an octahedron of oxygen ions,

ag shown in Figure 1.14 (a) The thermal vibration is high crough to result in the

Tandom onentalion of the Glarium ions im its octahedral inlerstigal posilion in

BaTiO; The Ti ion does shift position, resulting in polarization when an clectric ficld is applicd, but it returns to its stable central position as soon as the ficld is removed Thus, there is no retained polarization, no hysteresis loop, and

no ferroelectric behavior As the temperature of BaTiO; is lowered slightly below 120°C (Curie temperature), a displacive transformation occurs in which the structure of the BaTiO; changes from cubic lo tetragonal (Figure 1.14 (b)) One crystallographic axis increases in length (unit cell goes from 4.010 lo 4.022 A) and the other two decrease in length (fram 4.010 to 4.004 A) The Ti** ion moves off-center toward one of the two oxygen ions of the long axis, resulting in a

sponlancous inercasc

posilive charge in this dircetion This is illustrated 1m

Figure 1.15 (a) Application of an electric field opposite to the polarity of this

original dipole will cause the Ti’* ion to move through the center of the

octahedral site and to an equivalent off-center position, ‘Ihis is shown in Figure 1.15 (b) This results in a reversal polarization, hysteresis in Lhe E versus P curve,

Figure 1.15 Reversal in the direction of spontaneous polarization in Bal iO; by

reversal af the direction of the applied field [35]

The dielectric properties of BaTiO; were found to be dependent on the grain size and temperature At the Curie point, large-grained HaliO; (10 um) has a high dielectric constant because of the formation of multiple domains ina

16

single grain, the motion of whose walls increases the dielectric constant at Curie

point Far a BaTiO, with fine-grains (~1um), a single domain forms inside each

grain The movement of domain walls is restricted by the grain boundaries, thus leading to a low dielectric constant at the Curie point compared to coarse-grained

BaTiO At room temperature, the dielectnc constant of coarse-grained BaTiO;

ceramics was found to be in the range of 1500-2000 On the other hand, fine-

grained La'fiO; exhibits a room temperature dielectric constant between 3500-

6000 ‘Lhis is because the internal stresses in fine-grained BaTiO, are greater than

in the coarse-grained material, which leads to a higher permittivity at room

temperature [37]

1.3.2 Effects of doping on BaTiO; properties

According to the abave analysis, barium Lilanale (BaTiO,) ceramic iaterial has a relatively high dielectric and has been used in piezoelectric applications such as ceramic-sonar-transducers However, this ceramic exhibits typical ferrocleetric features with a sharp dicleetic response near the Curie temperature about 120°C, which is no! suitable for cnery storage usage because of its high F, and inferior thermostability It is well-known that doping is an effective way to snodify the electrical characteristics in electroceramies In the ABO; perovskites, the substitution of A and/or B cation sites by the proper acceplor, donor, or

isovalent impunity ions alters the electric behavior of these materials

Recently, doping barium titanate with isovalent Zr ions brings about Ta(Zz,Tì„)Q; (in short form barium zirconium titanate (BZT)), which has been found effective in decreasing Curie temperature (‘I,) In addition, the substitution concentration of Ti’ (avomiv weight of 47.9, ioniv radius of 0.0605 mm) with

Zr" (atomic weight of 91.2, ionic radius of 0.072 nm) unveiled several

interesting features in the dielectric behavior of Bal'iQ; ceramics such as reducing leakage current and dielectric loss in the system The Curie temperature

(T,) of pure barium titanate is 120 ° C [38] This temperature decreases by introducing zirconium to the structure Also, other phase transitions in barium titanate are affected In this case, the isovalent substitution of Ti** ions by Za” ions can transform Tezoelecuie micro-demains inte high dynamic polar nano- regions (PRNs), exlubiling normal ferroelectric transition to relaxor behavior with changing Zr concentration [4,5] When the Zr concentration is less than 10 smol.%, BZT ceramics show normal ferroelectric behavior, and BZT ceramics

exhibit relaxor behavior wher the Zr substitution is higher than about 20 mol.%

Due to the larger ionic radius of Zr" ions, the lattice parameters of BZT are

expanded while the ferrcefectric-paraelectrie phase transition is decreased [2]

Moreover, the Zr*' ion is more chemically stable than Ti*' ion, which is expected

to depress the conduction caused by electron hopping between Ti ion and TẾ” ion, and therefore, the properties of BZT are improved, Sun et al reported a large energy storage density of 30.4 J/ em’ and high energy efficiency of 81.7% under anc cirical field of 3 MV/ern was achieved al room temperature by the

fabrication of environmentally friendly lead-free BaZry,Tiegs epitaxial thin

17

films on Nb-doped SrTiO; (O01) substrates by using a radio-frequency

magnetron sputtering system Moreover, the BZT film capacitors exhibit great

thermal stability of the energy storage density from 16.8 J/om’ to 14.0 J/em* with

the efficiency of beyond 67.4% and high fatigue endurance (up to 10° cyeles) ina

wide femperalure range from room lemperalire lo 125 °C [39] Tiang el al

indicated an ultrahigh recoverable energy density (Urecs) (78.7 I/em”) and

efficioney (9) (80.5%) in RaZry3sTiessQy film capacitors by enhancing the breakdown electric field strength at room temperature lurthermore, the BaZ1o3sTiggsO3 film capacitor exhibits great energy storage properties when measured from - 150 °C to 200 °C Ujeco and z can reach the value of 41.9 Jem" and 66.4% under an electric strength of 4.0 MV/em even at 200 °C, respectively

[40] Compared to other RaTiQ;-hased energy storage capacitor materials and

aven Pb-based systems, BaZro2TigQ3 thin film capacitors show either high energy storage density or great energy efficiency Although BZ capacitors have

huge potential in the appheation of modern clectronies due to ibs high dicleetric

constant, low dielectric loss, and large tunability However, it is found that the dielectric properties are dependent on the temperature and frequency

‘To improve the electrical and other properties of barium titanate (3'T'O) and barium zirconate titanate (BZT) materials, doping is a common method Some

dopants such as Zn™!, Ca™!, S$’, Sia?! or La? , ele, which can replace the A siles

of ABO, perovskite structure and act as electron acceptors have been used to

improve the dielectric properties

Liu et al indicated that the crystal structures, surface morphology, and dielectric properties of Zn-doped BZT films were investigated as a function of

Zn content In addition, the dielectric canstant decreases at first and then

increases with the increase of Zn content, but the dielectric loss decreases with the increasing of Zn content at room temperature It also can be found that the Curic temperature of the Zn-doped BZT films is lower than that of the pure BZT films [41]

Chen et al showed a low leakage current density of 7.65 x 107 A/c al 60

V, and large breakdown strength of 4 MV/em in Sr-doping BZ thin films Ln addition, it not only exhibits an almost linear and acceptable change (AC/C -~13.6%) of capacitance from room temperature to 180°Cbut also a large

capacitance density of 1.7 nl'/mm’ at 100 k[lz, which shows great potential for

coupling and decoupling applications [42]

Amrit P, Sharma et al reported the ferroclectric phase transition of BZT/BCT thin fils has been prebed above room lomperalure with relaxor behavior These nanostructures show high discharge and charge energy densities

of 9.74 Jem? and 26.55 Jem’, respectively These heterostructures show high

Minh D Nguyen et al showed that La-doping enhanced the relaxor behavior in lead-free BZT thin films by introducing a disorder at the A sites

(Ba) of BZT unit cell, which can be achieved by donor substitution of La** for Ba’ ions The results achieved the optimum values of 72.2 J/cm’ recoverable

energy-storage density and 78.2% energy-storage efficiency under a high 3.8 MV/cm electric breakdown strength for 5 mol % La-doping [8] This showed that proper La-doping concentration can enhance the relaxor behavior and, hence, significantly improve the energy-storage performance and breakdown strength of

BZT thin-film capacitors

Conclusions

Tn conclusion, many studies have shown that doping based on barium tiatante (BTO) materials has significantly enhanced the energy-storage density, especially energy-storage efficiency and other properties for pulse-power systems, In particular, La-doping improved properties in BZT thin films, which

is a promising environmentally friendly candidate for the next generation of advanced energy-storage capacitor applications According to the above analysis, the research topic was selected: “Effect of Zr and La based co-doping on electrical properties of lead-free barium titanate BaTiO; thin films”

CHAPTER 2 EXPERIMENTS AND METHODS

Currently, there are two groups of methods to fabricate thin films: physical methods: sputtering, laser pulse deposition (PLD), molecular beam epitaxy, and chemical methods: vapor phase deposition, and sol-gel spin coating In this chapter, the author presents the technology to fabricate BZT and BLZT thin films

by sol-gel spin coating method

To investigale the propertics of the fabricated filns, modern measurements, and analysis techniques were used such as the crystallographic properties of the thin films were analyzed by X-+ay scans (XRD) using a PANalytical X-+ay diffractometer (Malvern PANalytical) with Cu-Ka radiation (wavelength: 1.5405 A) The microstructure of the BZT thin films was invesligaled by cross-scotional SEM measurements using Iligh Resolution Scanning Llectron Microscopy equipment (HRSEM: Zeiss-1530, Carl Zeiss Microscopy GmbH) ‘the ferroelectric properties of the materials were invesligated by polarization hysteresis loop measurements using aixACCT TF2000 Analyzer (aixACCT Systems Gmbll)

2.1 Fabrication of BZT and BLZT thin films by sol-gel spin coating

method, the molecular prectssor (usually metal alkoxide) is

alcohol and converted to gel by heating and stirring by hydrolysis/alcoholysis

Since the gel obtained from the hydrolysis/alcoholysis process is wet or damp, it

should be dried using appropriate methods depending on the desired properties

and application of the gel

lisgolyed in water or

Unlike the more conventional methods, the sol-gel technique allows [or preparing porous materials in a “one-pot” with a homogeneous distribution of components on the atomic scale through a technology of low-temperature synthesis and with full control of the finite product microstruclure The sol-gel

chemistry involves two distinc phases: solution and gelation: a sol is a colloidal

suspension of solid particles, whereas a gel is an interconnected network of solid- phase particles that form a continuous entity throughout a secondary, usually liquid, phase The advantages of sol-gel molhods include high yicld, low

operation temperatures, and low production costs Even more, the sol-gel

synthesis resulted in possessing unique features, namely the possibility of control over the physico-chemical properties of the resulting compounds through a

careful variation of the parameters alfecling the various synthesis steps [44]

The reaction mechanisms of sol-gel methed consist of two main reactions:

(1) hydrolysis of precursors in acidic or basic media and (2) condensation of hydrolysis products

20

Hydrolysis reaction: in this reaction, a nucleophilic substitution mechanism is hypothesized which results in the replacement of an alkoxy group with a hydroxyl

Condensation reaction: in parallel with the hydrolysis reaction, the condensation reaction ovcurs: the partially hydrolyzed alkoxy molecules may either react with another OII-bearing species by removing water or react with the alkoxy group producing an alcohol molecule:

MOR + MOH ¬ M — 0 — M + ROH B 22

Several parameters influence the hydrolysis and condensation reactions (sol-gel process), including the activity of the metal alkoxide, the water/alkoxide

ratio, solution pH, temperalure, nature of the solvent, and additive used Another

consideration is that catalysts are frequently added to control the rate and the

extent of hydrolysis and condensation reactions By varying these processing parameters, materials with different microstructures and surface chemistry can be

Figure 2.1 An overview of the various stages of the sal-gel process

2.1.1.2 Techniques for fabricating films from sol-gel solutions

The sol-gel process is casily coupled to a deposilion technique, such as dip-

coating, spin-coaling, or spray-coaling (Figure 2.2) For the coating production,

after the preparation of the sol with a composition suitable to the coating

application, its deposition on the substrate is carried out using a deposition

technique

21

Figure 2.2 Schematic representation of (a) dip coating; (b) spin coating; and (c) sprav

coating [45]

* Dip-coating techniques

Dip coating is a simple, reliable, and reproducible method that involves the deposition of a wet liquid film by immersion of the substrate into a solution containing hydrolysable metal compounds (or readily formed particles) and its

withdrawal at constant speed into an atmosphere containing water vapor After

the removal of the substrate from the solution, a homogeneous liquid film is

formed on the substrate's surface Dip-coating technique is almost used to

fabricate transparent layers of oxides on a transparent substrate with a high

degree of planarity and surface quality

“ Spin-coating techniques

Spin coating is a procedure used to apply uniform thin films to flat substrates A typical process involves depositing a small puddle of fluid resin onto the center of a substrate and then spinning the substrate at high speed

Centrifugal force will cause the resin to spread to, and eventually off, the edge of

the substrate leaving a thin film of resin on the surface Final film thickness and

other properties will depend on the nature of the resin (viscosity, drying rate,

percent solids, surface tension, etc.) and the parameters chosen for the spin

process Factors such as final rotational speed, acceleration, and fume exhaust

contribute to how the properties of coated films are defined

s* Spray-coating techniques