ANODIC ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION OF
NANOCRYSTALLINE CERIUM OXIDE AND CERIUM OXIDE / MONTMORILLONITE
NANOCOMPOSITES
Qi Wang, BS., MS.
Dissertation Prepared for the Degree of
DOCTOR OF PHILOSOPHY
UNIVERSITY OF NORTH TEXAS
August 2003
APPROVED:
Teresa D. Golden, Major Professor Jeffrey A. Kelber, Committee Member Michael G. Richmond, Committee Member Nandika A. D’Souza, Committee Member, Department of Materials Science and Engineering Zhibing Hu, Committee Member Ruthanne D. Thomas, Chair of the Department of Chemistry Sandra L. Terrell, Dean of the Robert B. Toulouse School of Graduate Studies
Wang, Qi, Electrochemical synthesis of CeO2 and CeO2/montmorillonite nanocomposites. Doctor of Philosophy (Chemistry), December 2003, 326 pp, 24 tables, 95 illustrations, 419 references.
Nanocrystalline cerium oxide thin films on metal and semiconductor substrates have been fabricated with a novel electrodeposition approach – anodic oxidation. X-ray diffraction analysis indicated that as-produced cerium oxide films are characteristic face-centered cubic fluorite
structure with 5 ~ 20 nm crystal sizes. X-ray photoelectron spectroscopy study probes the non-
stoichiometry property of as-produced films. Raman spectroscopy and Scanning Electron
Microscopy have been applied to analyze the films as well. Deposition mode, current density,
reaction temperature and pH have also been investigated and the deposition condition has been
optimized for preferred oriented film formation: galvanostatic deposition with current density of
–0.06 mA/cm2, T > 50oC and 7 < pH < 10. Generally, potentiostatic deposition results in random
structured cerium oxide films. Sintering of potentiostatic deposited cerium oxide films leads to
crystal growth and reach nearly full density at 1100oC. It is demonstrated that in-air heating
favors the 1:2 stoichiometry of CeO2.
Nanocrystalline cerium oxide powders (4 ~ 10 nm) have been produced with anodic
electrochemical synthesis. X-ray diffraction and Raman spectroscopy were employed to
investigate lattice expansion phenomenon related to the nanoscale cerium oxide particles. The
pH of reaction solution plays an important role in electrochemical synthesis of cerium oxide
films and powder.
Cyclic voltammetry and rotation disk electrode voltammetry have been used to study the
reaction mechanisms. The results indicate that the film deposition and powder formation follow
different reaction schemes. Ce(III)-L complexation is a reversible process, Ce3+ at medium basic 2+ pH region (7~10) is electrochemically oxidized to Ce(OH ) 2 and then CeO2 film is deposited on the substrate. CE mechanism is suggested to be involved in the formation of films, free Ce3+
- species is coordinated with OH at high basic pH region (>10) to Ce2O3 immediately prior to electrochemically oxidation Ce2O3 to CeO2.
CeO2 / montmorillonite nanocomposites were electrochemically produced. X-ray diffraction and Raman spectroscopy illustrate the retaining of FCC structure for cerium oxide.
Fourier Transform Infrared Spectroscopy and Differential Scanning Calorimetry of composites indicate the insertion of montmorillonite platelets into the structural matrix of cerium oxide.
Sintering study of the nanocomposites demonstrates that low concentration of montmorillonite platelet coordination into cerium oxide matrix increases crystal growth rate whereas high concentration of montmoillonite in nanocomposites retards the increase of crystallite size during the densification process. ACKNOWLEDGMENTS
I sincerely express my special gratitude to my academic advisor, Dr. Teresa D. Golden, for her kind guidance and great inspiration in research, also her continuous encouragement and wonderful support in my knowledge pursuit and working spirit development.
I appreciate Dr. Nandika, D’Souza for her expertise and advice in the composite project.
I wish to extend my acknowledgments to my committee members for their time and effort devoted to me and my research.
I would like to thank Dr. Paul Braterman, Dr. Seifollah Nasrazadani (Engineering
Department), Dr. Richard F. Reidy (Material Science Department), Dr. Robert M. Wallace
(Material Science Department), Dr. Michael Lance (Oak Ridge National Laboratory), Ms. Laura
Riester (Oak Ridge National Laboratory) and Mr. David Garrett (Biology Department) for their support and help in the material characterization. I am indebted to Dr. Oliver Chyan for his kindly providing silicon wafer for the research. Thank Mr. Williams A. Smith for making and adjusting glassware.
I would like to thank Dr. Zhiping Xu, Hanjiang Dong, Dr. Prakaipetch Punchaipetch and
Haritha Namduri for their helpful assistance.
I truly appreciate my big sister-like colleague, Charoendee Pingsuthiwong, for her invaluable help in the research and wonderful friendship during daily life. And help from other colleagues is also appreciative.
Financial support from Robert A. Welch Foundation and UNT Faculty Research Grant are highly grateful. Thanks also go to Chemistry Department of UNT for financial support as a
Teaching Assistant in my first four-year graduate study and research here.
ii Last but not least, I deeply appreciate my dear parents and brother for their never-ending love, support and encouragement.
iii TABLE OF CONTENTS
Page
LIST OF TABLES ……………………………………………………………………………..viii
LIST OF ILLUSTRATIONS……………………………………………………………………..xi
Chapter
1. INTRODUCTION AND LITERATURE REVIEW………………………………….1
1.1 Cerium Oxide Properties, Applications, Synthesis and Characterizations………..1
1.1.1 Structure and Properties of Cerium Oxide………………………………...1
1.1.2 Practical Applications of Cerium Oxide…………………………………..7
1.1.3 Production of Cerium Oxide……………………………………………..21
1.1.4 Characterization of Cerium Oxide
1.2 Cerium Oxide / Layered Silicate Nanocomposites ……………………………...32
1.2.1 Montmorillonite Nanoclays……………………………………………...32
1.2.2 Secondary Phase Reinforcement in Composites…………………………34
1.2.3 Layered Silicate Composites …………………………………………….42
1.2.4 Synthesis of Layered Silicate Composites ………………………………44
1.2.5 Characterization of Layered Silicate and Composites …………………..46
1.3 Chapter Summaries and Dissertation Introduction ……………………………...46
1.4 Chapter References………………………………………………………………48
2. ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION
METHODS…………………………………………………………………………..64
2.1 Electrochemical Synthesis and Other Electrochemical Techniques……………..64
2.1.1 Electrochemical Synthesis……………………………………………….64
iv 2.1.2. Cyclic Votammetry (CV)………………………………………………...79
2.1.3. Rotating Disk Electrode (RDE) Method…………………………………83
2.1.4 Corrosion Electrochemistry and Corrosion Measurement……………….85
2.2 Characterization and Analysis Methods…………………………………………88
2.2.1 X-ray Diffraction (XRD)………………………………………………...88
2.2.2 X-ray Photoelectron Spectroscopy (XPS)……………………………….95
2.2.3 Infrared Spectroscopy (FTIR) and Raman Spectroscopy……………….98
2.2.4 Scanning Electron Microscopy (SEM)………………………………...103
2.2.5 Brunauer-Emmett-Teller (BET) ……………………………………….104
2.2.6 Nanoindentation………………………………………………………..106
2.2.7 Sintering Technique……………………………………………………108
2.3 Chapter Conclusions……………………………………………………………110
2.4 Chapter References……………………………………………………………..112
3. ELECTROCHEMICAL DEPOSITION OF CERIUM OXIDE FILMS…………….118
3.1 Introduction……………………………………………………………………..118
3.2 Experimental……………………………………………………………………121
3.2.1 Electrochemical Deposition of CeO2 Films…………………………….121
3.2.2 Characterization of CeO2 Films………………………………………...122
3.2.3 Property and Application Studies ……………………………………...123
3.3 Results and Discussions………………………………………………………..124
3.3.1 Deposition Conditions Investigations ………………………………….124
3.3.2 Evaluation of Particle Sizes and Lattice Strains………………………..137
3.3.3 Determination of Oxidation State with XPS……………………………141
v 3.3.4 Morphology……………………………………………………………..149
3.3.5 Sintering Studies ……………………………………………………….152
3.3.6 Corrosion Test Results………………………………………………….165
3.4 Chapter Conclusions……………………………………………………………168
3.5 Chapter References……………………………………………………………..170
4. ELECTROCHEMICAL SYNTHESIS OF NANOCRYSTALLINE CERIUM OXIDE
POWDERS…………………………………………………………………………175
4.1 Introduction …………………………………………………………………….175
4.2 Experimental …………………………………………………………………..178
4.3 Results and Discussions………………………………………………………..179
4.3.1 Electrochemical Synthesis of Cerium Oxide Powders…………………179
4.3.2 Nanosized Cerium Oxide Powder Sintering……………………………184
4.3.3 Lattice Relaxation of Nanocrystalline Cerium Oxide…………………..193
4.4 Chapter Conclusions…………………………………………………………...209
4.5 Chapter References…………………………………………………………….210
5. MECHANISM STUDIES ON ELECTRODEPOSITION OF CERIUM OXIDE
FILMS AND ELECTROSYNTHESIS OF CERIUM OXIDE POWDER IN
SOLUTION OF CE (III) COMPLEXES…………………………………………...214
5.1 Introduction……………………………………………………………………..214
5.2 Experimental …………………………………………………………………...216
5.2.1 Electrochemical Techniques……………………………………………216
5.2.2 FTIR…………………………………………………………………….216
5.3 Results and Discussions………………………………………………………...217
vi 5.3.1 Ligand-pH Effect on Film or Powder Formation………………………217
5.3.2 Mass Transfer Study using Rotation Disk Electrode…………………...224
5.3.3 Coordination Investigation Through Varying Ligand Concentration…..228
5.4 Chapter Conclusions……………………………………………………………236
5.5 Chapter References……………………………………………………………..238
6. ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION OF CeO2 /
MONTMORILLONITE NANOCOMPOSITES…………………………………...240
6.1 Introduction ……………………………………………………………………240
6.2 Experimental …………………………………………………………………...241
6.2.1 Materials and Electrosynthesis of Composites…………………………241
6.2.2 Composite Property Analysis…………………………………………..241
6.3 Results and Discussions………………………………………………………..243
6.3.1 CeO2 / Na-cloisite Composites…………………………………………243
6.3.2 Nanocomposites from CeO2 and Other Nanoclays…………………….267
6.3.3 Structural Model for CeO2 / Montmorillonite Nanocomposites………..269
6.4 Chapter Conclusions……………………………………………………………270
6.5 Chapter References……………………………………………………………..271
7. ELECTROCHEMICAL FORMATION OF CeO2 / LAYERED SILICATE
NANOCOMPOSITES……………………………………………………………...274
7.1 Introduction……………………………………………………………………..274
7.2 Experimental……………………………………………………………………275
7.2.1 Electrodeposition of Nanocomposte Films……………………………..275
7.2.2 Characterization of Nanocomposite Films……………………………...275
vii 7.3 Results and Discussions…………………………………………………………….276
7.3.1 Treated and Untreated Montmorillonites as Second Phases……………276
7.3.2 Electrodeposited CeO2 / Na-Cloisite Nanocomposite Films…………...289
7.4 Chapter Conclusions………………………………………………………………..293
7.5 Chapter References…………………………………………………………………298
BIBLIOGRAPHY………………………………………………………………………301
viii LIST OF TABLES
Table Page
CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW
1.1 Physicochemical properties of pure stoichiometric CeO2.………………………………3
CHAPTER 2 ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION
METHODS
2.1 Three regions for infrared radiation.……………………………………………………..99
CHAPTER 3 ELECTROCHEMICAL DEPOSITION OF CERIUM OXIDE FILMS
3.1 Comparison between experimental (correspond Figure 3.4 A) and PDF data………...127
3.2 Comparison between experimental (correspond Figure 3.4 B) and PDF data…………127
3.3 Comparison between experimental (correspond Figure 3.4 C) and PDF data…………128
3.4 Cerium-ligand complexation constants.………………………………………………...128
3.5 The employed substrates for deposition of CeO2 film..……………..………………….130
o 3.6 The preferred orientation of CeO2 films at temperature 25, 40, 50, 60, 70, 80 C .……135
3.7 Peak list of cerium oxide XPS and calculation for x in CeOx …………………………155
CHAPTER 4 ELECTROCHEMICAL SYNTHESIS OF NANOCRYSTALLINE CERIUM
OXIDE POWDERS
4.1 Sintering temperature and particle sizes .………………………………………………194
3+ 4.2 Estimation of Ce percentage of oxygen vacancies in nanosize CeOy…………………203
4.3 Comparison of peak Raman breath at different sintering temperatures ……………….207
4.4 Comparison of experimental and calculated Raman shifts..……………………………208
ix CHAPTER 5 MECHANISM STUDIES ON ELECTRODEPOSITION OF CERIUM OXIDE
FILMS AND ELECTROSYNTHESIS OF CERIUM OXIDE POWDER IN
SOLUTION OF CE (III) COMPLEXES
5.1 IR assignment for Ce(III)-acetate complex and calculation frequency
difference.………………………………………………………………………………234
CHAPTER 6 ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION OF CeO2 /
MONTMORILLONITE NANOCOMPOSITES
6.1 Experimental and calculated XRD pattern position and d-spacing for Na-cloisite sample
in Figure 6.1.……………………………………………………………………………246
6.2 Particle size vs. content of Na-cloisite in composites.………………………………….248
+ + 6.3 OH vibration band shifts versus concentration of Na -cloisite in CeO2/Na -cloisite
nanocomposites.………………………………………………………………………..256
6.4. Crystallization of pristine CeO2, CeO2 / Na-cloisite nanocomposites with different
percentages of Na-cloisite at different sintering temperatures…………………………265
CHAPTER 7 ELECTROCHEMICAL FORMATION OF CeO2 / LAYERED SILICATE
NANOCOMPOSITES
7.1 Experimental and calculated XRD pattern positions and d-spacings for treated
montmorillonite sample in Figure 7.1(A).……………………………………………...278
7.2 Experimental and calculated XRD pattern positions and d-spacings for untreated
montmorillonite (UTC) sample in Figure 7.2 (B)………………………………………279
7.3 Particle sizes of CeO2 / montmorillonite nanocomposite films calculated from XRD
patterns in Figure 7.3 and Figure 7.5 using Scheerr equation on CeO2 (111)………….286
7.4 Particle sizes of CeO2 / Na-cloisite composite films.…………………………………..290
x 7.5 Particle sizes of composites at different unltrasonication time…………………………297
xi LIST OF ILLUSTRATIONS
Figure Page
CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW
1.1 FCC structure of CeO2 ………………………………………………………………….2
1.2 Illustration of montmorillonite unit structure.…………………………………………33.
1.3 Intercalation and exfoliation of montmorillonite………………………………………34
CHAPTER 2 ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION
METHODS
2.1 Electrode-solution interface with potential profile and electrochemical reaction.………70
2.2 Cathodic base generation electrochemical synthesis…………………………………….73
2.3 Anodic oxidation electrochemical synthesis…………………………….……………….78
2.4 Scheme of cyclic voltammetry …………………………………………………………..82
2.5 Schematic representation of rotation disk electrode and hydrodynamic flow pattern at
RDE………………………………………………………………………………………84
2.6 Corrosion measurement technique.………………………………………………………88
2.7 Diffraction of X-ray by crystal.………………………………………………………….89
2.8 Mathematic exploration to obtain true lattice parameters.………………………………93
2.9 Depicts of Raylegh scattering and Raman scattering.………………………………….101
2.10 Schematic representation of load-displacement curve in nanoindentation……………..107
CHAPTER 3 ELECTROCHEMICAL DEPOSITION OF CERIUM OXIDE FILMS
3.1 Simplified dourbaix diagram for cerium system……………………………………….120
3.2 Setup of electrochemical synthesis……………………………………………………..122
3.3 Corrosion measurement cell……………………………………………………………124
xii 3.4 X-ray diffraction patterns for CeO2 films formed from cathodic and anodic methods...126
3.5 X-ray diffraction patterns for CeO2 films on different substrates………………………129
3.6 X-ray diffraction patterns for CeO2 films deposited with galvaonstatic mode at different
current densities.………………………………………………………………………..132
3.7 Cyclic voltamgraphs for Ce(NO3)3-acetate system at different temperatures………….133
3.8 X-ray diffraction patterns for CeO2 film produced with galvanostatic mode (J = -0.06
mA/cm-2) at different temperatures……………………………………………………..134
3.9 Cyclic voltamgraphs for Ce(NO3)3-acetate system at different pHs…………………...136
3.10 X-ray diffraction patterns of CeO2 films deposited at different pHs…………………...136
3.11 Raman spectra of electrodeposited CeO2 films………………………………………...138
3.12 Williams-Hall plots βcosθ versus sinθ for XRD patterns of galvanostatic and
potentiostatic depositions……………………………………………………………….140
3.13 Grain size of CeO2 versus deposition temperature for galvanostatic deposition……….141
3.14 X-ray photoelectron spectra of Ce 3d in potentiostatic deposited CeO2 film…………..144
3.15 X-ray photoelectron spectra of O1s in potentiostatic deposited CeO2 film.……………146
3.16 X-ray photoelectron spectra of Ce 4d in potentiostatic deposited CeO2 film.………….148
3.17 X-ray photoelectron spectra of valance band in potentiostatic deposited CeO2 film…..150
3.18 SEM morphologies of CeO2 films deposited with potentiostatic and galvanostatic
modes…………………………………………………………………………………...151
3.19 X-ray photoelectron spectra and curve fitting of Ce 3d for CeO2 film after sintering at
different temperatures…………………………………………………………………..153
3.20 X-ray photoelectron spectra and curve fitting of O1s for CeO2 film after sintering at
different temperatures…………………………………………………………………..154
xiii 3.21 X-ray photoelectron spectra and curve fitting of Ce 4d for CeO2 film after sintering at
different temperatures…………………………………………………………………..156
3.22 X-ray diffraction patterns for potentiostatic deposited CeO2 film before and after
sintering at different temperatures……………………………………………………...158
3.23 Crystallite size versus sintering temperature calculated from XRD patterns in Figure
3.22……………………………………………………………………………………...159
3.24 SEM morphologies of potentiostatic deposited CeO2 films before and after sintering at
different temperatures…………………………………………………………………..160
3.25 X-ray diffraction patterns for galvanostatic deposited CeO2 film before and after sintering
at different temperatures………………………………………………………………..163
3.26 Crystallite size versus sintering temperature calculated from XRD patterns in Figure
3.25……………………………………………………………………………………...164
3.27 SEM morphologies of galvanostatic deposited CeO2 films before and after sintering at
different temperatures…………………………………………………………………..165
3.28 Corrosion testing of CeO2 film using 0.1M NaCl solution……………………………..167
3.29 Illustrations of corrosion of stainless steel in chloride solution………………………...168
CHAPTER 4 ELECTROCHEMICAL SYNTHESIS OF NANOCRYSTALLINE CERIUM
OXIDE POWDERS
4.1 X-ray diffractions of CeO2 powder produced from anodic electrodeposition and particle
size calculation………………………………………………………………………….181
4.2 Raman spectra of CeO2 powder produced from anodic electrodeposition……………..182
4.3 Adsorption-desorption isothermal plot for CeO2 powder………………………………183
4.4 BJH and BET plots deferred from adsorption-desorption for CeO2 powder…………...185
xiv 4.5 X-ray diffraction patterns and particle size variations for CeO2 powder submitted to
isothermal sintering……………………………………………………………………..187
4.6 Specific area and average pore size variation with sintering temperatures…………….188
4.7 Sintering curve for isothermal sintering of CeO2 powder……………………………...189
4.8 X-ray diffraction patterns and particle size variation for CeO2 powder submitted to ramp
sintering…………………………………………………………………………………192
4.9 Nilson-Riley exploration and Bradley-Jay exploration to obtain lattice parameters…...195
4.10 X-ray diffraction peak shifts and broadening of reflection (422) and reflection (400) for
nanosized CeO2 powder………………………………………………………………...197
4.11 Lattice parameter versus particle size for nanosized CeO2 …………………………….198
4.12 C-type sesquioxide structure illustration……………………………………………….199
4.13 (A) Linear relationship between stoichiometry (CeO2 & CeO1.5) and lattice parameters
according to Tsunekawa’s method; (B) Linear relationship between Ce3+ doping and
lattice parameters according to McBride’s theory……………………………………...202
4.14 Full range Raman spectra of nanosized CeO2 powder at different sintering
temperatures…………………………………………………………………………….205
4.15 Raman peak at ~ 465 cm-1 of nanosized CeO2 at different sintering temperatures……206
CHAPTER 5 MECHANISM STUDIES ON ELECTRODEPOSITION OF CERIUM OXIDE
FILMS AND ELECTRSYNTHESIS OF CERIUM OXIDE POWDER IN
SOLUTION OF CE (III) COMPLEXES
5.1 Setup of rotation disk electrode experiment……………………………………………217
5.2 Cyclic voltammgraphs of Ce(NO3)3-acetate complex at different conditions………….219
5.3 Cyclic voltammgraphs of Ce(NO3)3-EDTA complex at pH ~ 8.0……………………...220
xv 5.4 Cyclic voltammgraphs of Ce(NO3)3-acetate complex at pH ~ 12.0 on substrate stainless
steel and platinum………………………………………………………………………221
5.5 Cyclic voltammgraph of Ce(NO3)3-acetate complex at pH ~ 8.0………………………223
5.6 Linear sweep voltammographs of Ce(NO3)3-acetate on CeO2 prelayer………………..225
5.7 Levich plots for Ce(III)/Ce(IV) system at different reaction temperatures…………….225
5.8 Activation plot for linear sweep voltammetry data at rotation rate of 1000 rpm………226
5.9 Coordination schemes for acetate as a ligand…………………………………………..229
5.10 Coordination modes of acetate with metals…………………………………………….230
5.11 Cyclic voltammgraphs for cerium-acetate complex at different molar ratios………….232
5.12 FTIR spectra for acetate blank and cerium-acetate complex at different molar ratios…235
5.13 FTIR spectra of cerium-acetate complex at ratio of 1:1, 1:6; and 1:10………………...236
CHAPTER 6 ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION OF CeO2 /
MONTMORILLONITE NANOCOMPOSITES
6.1 X-ray diffraction patterns for air-dried Na-cloisite and ethylene glycol-solvated Na-
cloisite in comparison to CeO2 powder………………………………………………...245
6.2 X-ray diffraction patterns for CeO2 / Na-cloisite composites at different content of Na-
cloisite…………………………………………………………………………………..249
6.3 IR spectra of Na-cloisite in the regions of 3850 ~ 3350 cm-1 and 1350 ~ 400 cm-1……255
-1 6.4 IR spectra of CeO2 / Na-cloisite nanocomposite in the regions of 3850 ~ 3350 cm and
1350 ~ 400 cm-1………………………………………………………………………...256
6.5 Zoom-in IR spectra in the region of 1060 ~ 1010 cm-1 for CeO2 / Na-cloisite
nanocomposite at 0.5% (w/w) of Na-cloisite…………………………………………..257
6.6 Raman spectra of CeO2 / Na-cloisite composites………………………………………258
xvi 6.7 Differential scanning calorimetry curves of pure Na-cloisite, cerium oxide and their
Nanocomposites………………………………………………………………………...260
6.8 XRD patterns for Na-cloisite subjected to different heating temperatures……………..261
6.9 X-ray diffraction patterns of CeO2 / Na-cloisite composites at 1% and 20% before
sintering after sintering at different temperatures…………………………………262-263
6.10 Si-O stretching IR spectra of composites obtained from transmission mode and ATR
mode…………………………………………………………………………………….267
6.11 Transmission IR and ATR-IR spectra in region of 1300 ~ 400 cm-1 for 3% (w/w)
composite pellet before and after sintering at different temperatures…………………..268
6.12 Proposed structural model for formation of CeO2 / Na-cloisite nanocomposites………270
CHAPTER 7 ELECTROCHEMICAL FORMATION OF CeO2 / LAYERED SILICATE
NANOCOMPOSITES
7.1 X-ray diffraction patterns for air-dried and ethylene glycol-solvated octadecylammonium
montmorillonite and untreated montmorillonite in comparison to CeO2 powder……...277
7.2 X-ray diffraction patterns for physical mixtures of (A) CeO2-untreated montmorillonite
and (B) CeO2 – treated montmorillonite ……………………………………………….280
7.3 X-ray diffraction patterns for (A) CeO2 / treated montmorillonite composites and (B)
CeO2 / untreated montmorillonite composites at different clay concentrations………..283
7.4 SEM of CeO2 / treated montmorillonite composite films (20%, 10%) and CeO2 /
untreated montmorillonite composite films (20%, 5%)………………………………...284
7.5 X-ray diffraction patterns for treated montmorillonite-composite films and untreated
montmorillonite-composite films produced with potentiostatic and galvanostatic
depositions……………………………………………………………………………...285
xvii 7.6 Optical micrographs of CeO2 / untreated montmorillonite composites films produced with
potentiostatic and galvanostatic depositions……………………………………………285
7.7 Hardness versus loading displacement for nanocomposite films with different content of
untreated montmorillonite and 5% of treated montmorillonite………………………...287
7.8 Young’s modulus versus loading displacement for nanocomposite filmss with different
content of untreated montmorillonite and 5% of treated montmorillonite……………..288
7.9 X-ray diffraction patterns for CeO2 / Na-cloisite composite films at different
concentration of Na-cloisite with potentiostatic deposition and 5% with galvanostatic
deposition……………………………………………………………………………….290
7.10 SEM images of 10% NTC, (A) film surface X 5,000; (B) film surface, X 10,000; (C) film
cross-section, X75; (B) film cross-section, X2,000…………………………………….291
7.11 CeO2 / Na-cloisite composite films at different clay ultrasonication time (A) XRD
pattern; (B) FTIR spectra; (C) Raman spectra………………………………………….295
7.12 Optical micrographs of CeO2 / NTC nanocomposite filmes X800 (A) 0.2%, 24 h; (B)
0.2%, 216 h (C) 0.2%, 432 h (D) 0.5%, 24h; (E) 0.5%, 216h (F) 0.5%, 432 h………...296
7.13 SEM image of CeO2/ Na-cloisite (NTC) nanocomposite films after clay was
ultraconicated for 216 h. (A) 0.2% X1000; (B) 0.5%, X1000; (C) 0.2%, X5000; (D)
0.5%, X5000……………………………………………………………………………297
xviii CHAPTER 1
INTRODUCTION AND LITERATURE REVIEW
1.1. Cerium Oxide Properties, Applications, Synthesis and Characterization
1.1.1 Structure and Properties of Cerium Oxide
1.1.1.1 Electronic Structure
Cerium dioxide (CeO2) is the most stable oxide of cerium. This compound is also called
ceria or ceric oxide. It is necessary to discuss cerium chemistry in general before a detail review
of CeO2. Cerium is the second member in the lanthanide series and ranked as the second most
reactive element in the series. It is very electro-positive, and like other lanthanides, it has a
Ce(III) oxidation state due to a low ionization potential for removal of the three most weakly-
bound 4f electrons. However, the electronic structure for Ce(IV) is [Xe] 4f0, as compared with
[Xe]4f1 for Ce (III), is the most stable oxidation state in cerium due to the empty 4f0. From
valence state theory, cerium is supposed to have two types of oxides, cerium dioxide (CeO2) and
cerium sesquioxide (Ce2O3). Since CeO2 is the most stable oxide, when the cerium oxide is
mentioned, it usually refers to the tetravalent oxide. Cerium Dioxide can also be formed if
cerium salt is calcined in oxygen rich environment.1
1.1.1.2 Crystal Structure
Cerium sesquioxide (Ce2O3) has two structural forms, hexagonal (A-type) and cubic (C-
type). Cerium dioxide (CeO2) has a fluorite (CaF2) structure (fcc) with space group Fm3m.
Figure 1.1 illustrates the structure of the stoichiometric CeO2 with the oxygens (represented by
circles) four coordinated and the cerium (represented by a solid ball) eight coordinated. The cerium is at the center of tetrahedron and the tetrahedral corners are occupied by oxygens.2
1 Usually, cerium oxide exists as a non-stoichiometric oxide that is a mixture of Ce(III) oxide and
Ce(IV) oxide, while still retaining the fluorite cubic structure.
Figure 1.1 FCC structure for CeO2 (circle represents oxygen, solid ball represents cerium)
1.1.1.3 Physical Properties
Stoichiometric CeO2 is pale yellow due to Ce(IV)-O charge transfer. However, since
CeO2 can be reduced in a reducing environment and Ce2O3 can be partially converted to CeO2, it
is usual that cerium oxide is an oxygen-deficient, non-stoichiometric oxide (CeO2-x with
0 3,4 properties for CeO2 are listed in Table 1.1. 2 Table 1.1. Physicochemical properties of pure stoichiometric CeO2 Property Value (unit) Color Yellow-White Density 7.22 g cm-1 Surface Area ~9.5 m2 g-1 Melting Point Ca. 2750 K Formation Heat -246 kcal mol-1 Specific Heat Ca. 460 J kg-1 K-1 Acidity Weak base Conductivity 1,2 ~ 2 x 10-8 Ω-1 cm-1 Thermal Conductivity Ca. 12 W m-1K-1 Ca. 2.1 visible Refractive index Ca. 2.2 infrared Absorption Edge ~ 420 nm Relative Dielectric Constant 11 Young’s Modulus Ca. 165 x 109 N m-2 Poisson’s Ratio Ca. 0.3 Hardness 5 ~ 6 (Mohs) ~ 2.95 eV (UV), Bandgap 5.5 eV (electronic cal.) 3 1.1.1.4 Redox Chemistry As mentioned above, the CeO2 structure switches between stoichiometric CeO2 and non- stoichiometric CeO2-x since pure Ce(IV) is prone to reduce to Ce(III) in a reducing environment. The reduction of cerium dioxide can be written as follows:3 1 Ce + O = O (g) +V ⋅⋅ + 2Ce' (Equation 1.1) Ce o 2 2 o Ce ' where CeCe , according to Kröger-Vink-notation, represents the oxidizing Ce(IV) and CeCe ⋅⋅ represents the reducing Ce(III) and Vo represents the oxide vacancies. Reduction introduces ⋅⋅ oxygen vacancies Vo that can be imported from lower valence cations into the CeO2 structure. Therefore, CeO2 is a good oxidizing agent, for instance, CeO2 acts as an oxidant for the conversion of iron (II) to iron (III). On the other hand, defective cerium oxide can be oxidized by higher valance compounds and be converted back to stoichiometric CeO2. With the continuous transformation between oxygen-rich CeO2 and oxygen-deficient CeO2-x (0 CeO2 has the ability to collect and release oxygen, which has much application in catalysis. 1.1.1.5 Electrical Conductivity It was recognized a long time ago that cerium oxide is an electrical conductor.5 Estel6 reviewed the electrical properties of ceria material. Indaba et.al. discussed the electrical conductivity, diffusivity and transference number properties of ceria-based solid electrolytes 7 from both theoretical and experimental views. It was revealed that pure CeO2-x is a mixed conductor made up of almost same partial conductivities from oxygen ions, electron and hole conductivities. -26 Hole conductivity of CeO2 is negligible in the region of oxygen pressure of 10 ~1atm. Therefore electrical conductivity usually refers to the summation of electronic conductivity (σe) 4 8 and ionic conductivity (σi). Pure cerium oxide is basically an electronic conductor with n-type semiconductor characteristics and this property occurs through the transport of polarons that are ' oxide vacancies or CeCe in the structure of CeO2-x with thermally activated electron mobility. The electronic conductivity of CeO2-x can be expressed as ' σ e = [CeCe ]eµ (Equation 1.2) ' 3+ in which [CeCe ] is the concentration of Ce ions, e is the charge of electron and µ is the electron mobility. Applying the thermal factor of electron mobility, the final equation for the electronic conductivity is: ' σ e = [CeCe ]e(C /T )exp(−EH / kT) (Equation 1.3) in which EH represents the activation energy of electron mobility and C is a constant. This equation shows that electronic conductivity of cerium oxide is related to temperature as well as concentration of Ce3+. 3 The total electrical conductivity is the addition of the electronic and ionic conductivity (σ = σ e +σ i ). The percentage of ionic conductivity increases significantly with the oxides of two- or three- valent metals doping, since doping introduces more oxide ion vacancies. Doped ceria acts as an ionic electrolyte in which the conductivity is derived mainly from the transport of oxide ions to the vacant sites. The ionic conductivity can be expressed as σ i = (σ 0 /T )exp(−EH / kT) (Equation 1.4) in which EH is activation energy for hopping of oxide ions. Temperature is also is a controlling factor. Combining ionic and electronic conductivity, it is inferred that electrical conductivity increases with decrease of temperature. At higher concentrations of Ce3+, electronic conductivity 5 contributes more than at lower concentration of Ce3+. It is recognized that by controlling temperature, oxygen activity, chemical structure and composition, it is possible to vary the composition of the different conductivities.9 1.1.1.6 Optical Properties 10 Calculations about electronic structure of CeO2 result in a band gap of 5.5 eV. A 1eV 4f band that is around 3eV above the valence band, separates the conduction band (Ce5d) and valence band. No observation of absorption in visible or IR region is existent for pure stoichemetric CeO2. The optical properties of cerium oxide are usually investigated with spectral absorption and reflectance in blue/violet region. Absorptions are observed at photon energies of 3eV, 90~ 190 eV. 11 The 3eV sharp absorption is attributed to transition between 2p oxygen valence band and cerium 4f band that lies above Fermi level of the material.12 1.1.1.7 Catalytic Properties Due to the unique fluorite structure and easy redox switch between Ce3+ and Ce4+, cerium oxide has been widely studied as a catalyst or catalytic promoter.13-15 There is an extensive list of reactions that cerium oxide can provide direct or indirect catalytic functions.2 (1) Oxidation Reactions: CeO2-containing materials have been employed as oxidation catalysts. Pure CeO2, CeO2 mixed with the metal oxide (CuO, TiO2, Al2O3, La2O3, etc), CeO2 added into precious metals (Pt, Pd, Rh, Au, Ag, etc) and even CeO2 supported metal oxides (SiO2, Al2O3, etc) catalyze or help catalyze oxidation of CO, conversions of hydrocarbons to yield CO2, ammoxidation, wet oxidation of organics and oxidative coupling reactions of hydrocarbons. For the oxidation reactions above, CeO2 acts as an oxygen provider, and sometimes the active site for the reactants to be adsorbed on the catalyst surface as well. 6 (2) Hydrogenation Reactions: 3+ It is easy for Ce to be formed in a reduced environment. CeO2 has often been used as the catalyst or catalyst support for hydrogenation of CO, CO2, hydrocarbons, and other organic compounds. Doping of rare earth oxides (ZrO2, Y2O3, La2O3, MgO, Cs2O, CaO, etc) assists the formation of Ce3+ and the increase of active sites for hydrogenation reaction promotes the reaction to a certain degree. Metal (Rh, Pd, Ni, etc.) involvement strongly enhances activity and selectivity of ceria in the hydrogenation reaction because of the interaction between ceria and the metal. (3) Reactions Involving SOx and NOx: SOx and NOx are environmental pollution gases like CO. It is found that CeO2 has an effect on the conversion of these two types of compounds to less toxic compounds. Combining MgAl2O4 and ceria is a very efficient approach to convert SOx to H2S that can then be easily removed. In this catalyst system, MgAl2O4 stores the SO3 after CeO2 oxidizes SO2 to SO3. It is said that CeO2 could also play a role in the storage portion. Similar catalysts have been suggested for the removal of NOx by converting it to N2. Since it is possible for CeO2 to undergo the Ce(IV)-Ce(III) and Ce(III)-Ce(IV) redox cycling, and SO2 needs CeO2 to be oxidized and NOx needs to reduced, the elimination of noxious gases in one cycling process with help of cerium oxide has also been explored recently. 1.1.2 Practical Applications of Cerium Oxides 1.1.2.1 Cerium Oxide Powder Applications 1.1.2.1.1 Applications of Catalytic Properties There are several industrial processes where cerium oxide functions significantly as a catalyst or catalytic promoter. These processes include Three-Way Catalysis (TWC) for removal 7 of exhaust in automotive, Fluid Catalytic Cracking (FCC) for elimination of sulfur oxide, and Catalytic Wet Oxidation (CWO) for removal of organics in wastewaters. (1) Three-Way Catalysis (TWC) Cerium oxide has played a very important role in three-way catalysis (TWC). TWC is designed to simultaneously control and remove three emitting pollutions from automotive exhaust, carbon monoxide (CO), nitrogen oxide (NOx) and hydrocarbons (HC). Usually the catalyst formulation for auto exhaust treatment consists primarily of noble metals and metal oxides dispersed either on an aluminum surface anchored to a ceramic substrate or stand alone.2,16,17 The functions of ceria in the TWC is shown in following equations: CeO2 + xCO → CeO2−x + xCO2 (Equation 1.5) CeO2 + HC → CeO2−x + (H 2O,CO2 ,CO, H 2 ) (Equation 1.6) CeO2 + xH 2 → CeO2−x = xH 2O (Equation 1.7) CeO2−x + xNO → CeO2 + 0.5xN 2 (Equation 1.8) CeO2−x + xH 2O → CeO2 + xH 2 (Equation 1.9) CeO2−x + 0.5xO2 → CeO2 (Equation 1.10) Ceria provides oxygen buffer capacity that can be used to store and release oxygen during the whole process. At rich fuel situations, the oxygen-deficient portion of the cycle, ceria donates its oxygen to help the oxygenation of carbon monoxide (CO) and hydrocarbon compounds (HC) (equations 1.5-1.7), while for lean fuel conditions, it has the capability to adsorb oxygen from oxygen (O2), nitrogen monoxide (NO) and water (H2O), then stores the adsorbed oxygen in its own structure (equations 1.8-1.10). These processes cycle throughout the system. A variety of interactions are involved with ceria besides as an oxygen bank in TWC. The interaction between the noble metal catalysts such as Pd, Pt, Rh, and cerium oxide can thermally 8 stabilize catalyst so as to increase the activity of the catalyst at high temperature.13 Cerium oxide was also reported to mechanically increase the strength of Al2O3 and prevent Al2O3 surface area loss. Cerium dioxide also functions significantly in the water gas shift reactions (WGSR), which simultaneously remove carbon monoxide (CO) and provide hydrogen. The working mechanism is similar to that in TWC. (2) Fluid Catalytic Cracking (FCC) In FCC units, most of the sulfur remains in the reactor or converts to H2S to be eliminated downstream. But around 10% sulfur remains trapped in the catalyst and then is oxidized to SO2 and SO3 in the catalyst regeneration step. This regenerated SO2/SO3 mixture needs to be treated before being released to the atmosphere. CeO2/MgAl2O4·MgO is a perfect catalyst for this situation. The functions of cerium oxide and magnesium aluminate spinel in this system are indicated as followings: CeO2 + xSO2 → xSO3 + CeO2−x (Equation 1.11) SO3 + MO → MSO4 (Equation 1.12) MSO4 + red ⇔ MO + H 2 S + H 2O (Equation 1.13) CeO2−x + 0.5xO2 ⇔ CeO2 (Equation 1.14) In the above equation: (red = CeO2−x , H 2 ) And M refers to sulfur storage component CeO2 as an oxidizing agent first converts SO2 to SO3, where itself is reduced to non- stoichiometric CeO2-x (Equation 1.11). Then SO3 are adsorbed by the storage components Mg2Al2O5 (Equation 1.12). CeO2 also participates as storage for sulfur oxides. Reduced cerium oxide has the potential to facilitate the next step involving the release of sulfate to H2S (Equation 9 1.13). The reductive cerium oxide can be easily regenerated by oxygen in the system (Equation 1.14). In summary, cerium oxide plays a significant role because it participates in almost every step in the treatment of sulfur oxides in the FCC process. (3) Catalytic Wet Oxidation (CWO) CWO is a method that is utilized to remove pollutants from wastewater or to transform these toxic components in the wastewater to less toxic compounds. Usually the CWO is operated at high temperature and high pressure and oxidizes organic compounds in aqueous solution. Ceria is a material that can be used to oxidize organics and can withstand high temperature and high pressure. The following radical mechanism is a possible illustration for the function of cerium oxide in the catalytic wet oxidation process. RH+ Me(n+1)+ → R ⋅ +M n+ + H + (Equation 1.15) ROOH + Me n+ → RO ⋅ +Me (n+1)+ + OH − (Equation 1.16) ROOH + Me(n+1)+ → ROO ⋅ +Men+ + H + (Equation 1.17) The formation of radicals in the reactions is most critical steps and from Equation 1.15- 1.17, it is obvious that the oxidizing and corresponding reducing ability of Me(n+1)/Men+ is the primary channel to produce these actively oxidative intermediates. CeO2/Ce2O3 is a perfect redox couple that suits this situation. Therefore, the catalytic function of cerium oxide in this system embodies the redox property of cerium oxide that is distinct from its oxygen vacancy performance for most other catalytic situations. In addition to these three well-established systems that employ cerium oxide and cerium oxide containing materials as catalyst, co-catalyst, and catalyst promoter, it has been utilized in the removal of soot from diesel engine exhaust, combustion catalysis and other organic conversion reactions. The applications benefit from the rigid fcc structure, high oxygen storage 10 property of cerium oxide and fast conversion between Ce(IV) and Ce(III). It is believed that cerium oxide will offer catalytic assistance to more and more industrial systems. There are several issues that affect the application of cerium oxide catalysis. The most severe is aging problem, i.e. the loss of oxygen storage property of the catalyst system over operating time. It is observed that this problem increases with the increase of cerium oxide crystallite size, accompanied with the decrease effect for surface area of the catalyst. This hurdle can be overcome by developing new synthesis approaches to produce nanocrystalline CeO2 powder. 1.1.2.1.2 Application of Electrical Conductivity Electrical conductivity is another big issue for CeO2 and CeO2-based materials. Electronic conductivity is characteristic of pure cerium oxide, which is one of the reasons that CeO2 can be used as a component in electrodes for solid oxide fuel cells (SOFC). However, relatively more applications are derived from the doped CeO2, which shows competitive oxygen ionic conductivity. The possible potential application of oxygen ion conducting solid electrolytes includes gas separations, power generation (SOFC), oxygen sensors (automotive pollution control), catalysis (catalytic membrane reactor), and electrocatalysis.18 Sensor applications will be addressed in the film application section (1.1.2.2.4). Here we will focus on electrolyte application in SOFC. (1) Electrolytes in Solid Oxide Fuel Cells Solid oxide fuel cells (SOFCs) are ceramic electrochemical energy conversion reactors, which have the capability of converting oxidizable gaseous fuel directly into electricity by the electrochemical process but without combustion.19-21 Cerium oxide-based materials have multiple applications in this system: as electrolytes, component material in anode design, solid- 11 electrolyte oxygen pumps and mixed-conductive membranes for oxygen-separation and partial oxidation of hydrocarbons.22 Solid electrolytes in SOFC systems are required to be materials that must have the following qualities: high and pure oxygen ionic conductivity and low electronic conductivity; stable at both reducing and oxidizing atmospheres; compatible with electrodes; resistant to high temperature; capable to form thin layer of films. Compared with traditional YSZ, doped ceria materials can provide the higher ionic conductivity and stability at lower temperature. The lower operation temperature is preferable to the utilization of natural gas as fuel in SOFC without carbon deposition on the anode.23 The most often applied doped ceria are: gadolinia-doped cerium oxide (GDC (Ce0.90Gd0.10)O1.95), samaria-doped cerium oxide (SDC (Ce0.85Sm0.15)O1.925), yttium-doped cerium oxide (YDC (Ce0.85Y0.15)O1.925) and calcium–doped cerium oxide 20 3,6,24 (Ce0.88Ca0.12)O1.88. The electrical properties of these ceria-based materials are reviewed. The problem with ceria is that this material is prone to be reduced from Ce4+ to Ce3+, introducing electronic conductivity,25,26 increasing grain boundary impedance27 and inducing expansion of ceria and thus mechanical stresses, which delaminate the materials.3,28 The result is that SOFC efficiency is reduced. To solve these reducing ionic conductivity problems, a protective layer of YSZ are deposited on the outside of the ceria-based electrolytes, which can suppress the reduction of the ceria.3 Ceria based materials can act as an anode for SOFC with high electrical conductivity, stability at both oxidizing and reducing atmospheres, and inertness to electrolytes.29 Ceria based materials not only provide conductivity but also can suppress the carbon deposition when direct utilization of hydrocarbons as fuel in SOFC so that Y-based ceria and La-based ceria were used as anode as early as 1960s.3 Investigation by Park30 indicates conventional Ni-YSZ cermet 12 anode, on the contrary, catalyzes the carbon deposition. Ceria plays multiple roles in the as- assembled anode for SOFC: electron transfer, increasing the surface area and catalyzes oxidation of the fuels.31 1.1.2.1.3 Cerium Dioxide as a Photocatalyst Cerium dioxide, to a certain degree, is a semiconductor that can adsorb light between UV and visible region. And because of the variety of valence states of cerium, the cations of cerium are well-known as electron-acceptors. Bamwenda et.al. investigated the possibility of cerium oxide as a photocatalyst to convert water to oxygen.4 The results indicated that cerium dioxide demonstrated photostability and activity as a photocatalyst, and further exploration of oxygen production in this photocatalysized reaction strongly depends on the concentration of cerium dioxide. All the details proved that cerium dioxide is a very promising material for photocatalysis. 1.1.2.1.4 Cerium Dioxide as Polishing Powder Cerium oxide was used as a main component for polishing powders as early as 1940s because cerium oxide crystalline structure can offer fair mechanical strength and wear resistance which confer excellent polishing ability. In Kosynkin’s report,32 not only a detailed process of producing polishing powder based on cerium dioxide is described, but the development and history of cerium dioxide based polishing powder are listed as well. Before 1992, mainly two types of cerium dioxide based polishing materials (cerite and photopol) were involved in the industry. These two polishing powders had as high as 90% of cerium dioxide in their contents. Relatively new types of polishing powder were released in 1997 with improvement of the properties of the polishing powder. 1.1.2.1.5 Cerium Dioxide Fluorite Type Pigments 13 Cerium dioxide can act as a pigment in dye engineering. In Šulcova’s report,33 the preparation process of a CeO2 based pigment was described, in which CeO2 and Pr6O11 were mixed in an agate mortar and calcinated at a higher temperature. As-produced pigment demonstrated fluorite structure and gave pink-orange ceramic glaze. According to the research, increasing content of praseodymium changed the color hue more reddish. When different types of mineralizers are present, the color hue of the pigment will shift from pink-orange to red- brown, i.e. effect of the mineralizers increases red hue. 1.1.2.2 Applications of Cerium Oxide Thin Film 1.1.2.2.1 Corrosion Protection Generally, surface modification has been recognized as a very important approach to enhance the corrosion resistance of many materials. Surface modification only covers or alters the surface layers of the materials but does not interfere with the physical, chemical, and mechanical properties of bulk materials underneath. Cerium compounds not only can confer corrosion resistance comparable to chromates but also have no toxicity and carcinogenic properties that are a problem for chromates.34,35 It is found that the cerium species can be applied to protect zinc,36,37 aluminum and aluminum containing alloys,38 stainless steel,35,39,40 magnesium containing alloys,41 tin-containing alloys,34,42 even SiC/Al metal-matrix composites from corrosion. It can be used to reduce the rate of general corrosion, pitting, crevice corrosion and stress corrosion.35,39,40 It is recognized that cerium oxide/hydroxide formation is the main reason for the corrosion protection property for cerium compounds. Basically, both anodic and cathodic electrochemical processes are involved in aqueous corrosion. At anodic sites, the possible corrosion reactions are:38 14 Metal dissolution M → M n+ + ne− (Equation 1.18) − Anodic passivation M + H 2O → M 2On + ne (Equation 1.19) While at cathodic sites, − − Oxygen reduction O2 + 2H 2O + 4e → 4OH (Equation 1.20) + − Hydrogen evolution 2H + 2e → H 2 (Equation 1.21) In general, metal dissolution and reduction of oxygen are the main corrosion reactions for metal surface. Hence, to hinder the corrosion means to find a way to block or diminish one or more of these reactions. Cerium species have been used in two ways for metal corrosion protection. The first one is low concentrations of cerium ions to be added into an aqueous solution in which the metal is immersed as an electrode.35,38 Another one is that the cerium species will be incorporated directly into passive oxide film on metal surfaces or electroplated on the surface of the metal to provide a protective coating.40,43,44 The immersion approach has been applied to aluminum and 34,35 stainless steel. The employed cerium species for stainless steel were not limited to CeCl3 and 45,46 Ce(NO3)3, but extends to Ce(CH3COO)2 as well. According to investigations for cerium inhibition properties, it is found that most of the mechanisms for the inhibition involve the decrease in the cathodic reduction rate. This has been shown in the corrosion inhibition of cerium species for aluminum, stainless steel and zinc. The corrosion inhibition is explained as follows: local pH at cathodic site increases so that the cerium ions in solution can form cerium oxide or hydroxide on the surface. The layer blocks cathodic reduction of oxygen.38,39 Cathodic inhibition can also be observed by the observation of negative shifts of the corrosion potential in cerium-containing solution.35 15 Cerium species also can be an anodic inhibitor to prevent metals from corrosion. Crossland et.al. reported that when aluminum is alloyed with cerium, the alloy was anodically oxidized to a layer of oxide. Ce3+ in this inner oxide layer has a higher migration rate to the surface than that of Al3+, which results in cerium-rich oxide but alumina-free outmost layer. The as formed anodic film can suppress ejection of Al3+ to the hydroxide electrolyte to form a corrosion layer.43 This protection can be kept as efficient as possible until pH 12 as the layer is stable in alkaline conditions. Cerium combined with yttrium was also used on the corrosive wear of stainless steel, aluminum alloy as well and laser surface cladding of ceria is used for nickel- based alloy.47,48 It is demonstrated that cerium species markedly reduced synergic effect of corrosion and wear in both systems. 1.1.2.2.2 Optical and Electrochromic Applications It is evidenced that stoicheometric CeO2 film has no absorption property in visible and infrared regions, and has a very sharp absorption edge and high refractive index. It has been employed for optical coatings.49,50 In contrast, cerium oxide is a relatively very efficient UV radiation absorber, it can be used as a protection film for some UV sensitive materials. The most impressive application of CeO2 film in this area is the utilization of cerium oxide as the counter electrode in electrochromic devices.51,52 Electrochromic devices are usually composed of conducting lithium as electrolytes and amorphous tungsten oxide (WO3) as the electrochromic 53 layer. CeO2 meets the requirement for the counter electrode because of its high optical transparency property in the visible region combined with excellent ability of charge delivery and storage in both reduced and oxidized state. It is reported that the intercalation / de- intercalation of lithium in CeO2 indicates reasonably good reversibility. However, since the size of insertion site in CeO2, which is around 1.02 Å, is much larger than the radius of lithium (0.6 16 Å), the structure limits the diffusion of the lithium, affecting cell performance. The cation of smaller size replacing the cerium ion in the cubic structure is favorable for the insertion/release of lithium from the counter electrode. Titanium,54,55 zirconium,56 tin,57 tungsten58 and vanadium59 have been doped with cerium oxide for this purpose. It shows that the counter electrode made by mixed oxide films of cerium and titanium or tin have shown good transmission in both reduced and oxidized states and better performance in electrochemical 54,57 56 properties. ZrO2-CeO2 shows good promise in this area, and due to large charge capacity 59 and good optical property of V2O5, CeO2-V2O5 is being of explored. 1.1.2.2.3 Applications in Electronic Devices Cerium dioxide is an attractive buffer layer material for silicon technology because it exhibits high insulating resistance and great chemical stability even at high temperatures. It is demonstrated that the epitaxial grown CeO2 can be used in silicon on insulator (SOI), and the silicon-based devices of superconduction, colossal-magnetoresistance and ferroeletrics.60 The silicon-on-insulator (SOI) structure is composed of a thin layer of low-defect epitaxial silicon and an insulating layer. This structure has potential application in the high-speed device and Very Ultra-Large-Scale Integrated (VULSI) circuits.61 Choosing insulator layer material is a critical stage. Two types of problems exist in the previously explored material: the mismatch of lattice between silicon and the insulator (zirconia, spinel, sapphire, etc.) and the silicon-insulator (CaF2) interface reaction induced contamination to the silicon. Cerium oxide has the fcc symmetrical structure with lattice parameter of 0.5411nm, which matches well with the lattice of silicon (0.5431nm) so that it is possible to deposit epitaxial CeO2 thin layers on silicon surface without twinning and any other imperfection between silicon and the insulating layers. Simultaneously, cerium oxide is inert to silicon so contamination in silicon is also not a problem 17 for the SOI structure. Inoue reported the first successful cerium oxide epilayer deposition on silicon surface in 1990,62 and from then on, a lot of researchers have focused on the growth of cerium oxide epilayer on different type (e.g. p-type Si and n-type Si63) or different orientations (Si (110) ,Si (111) or Si (100)60,64-66) of silicon surfaces. Techniques that have been applied to the deposition of this structure include electron beam assisted evaporation,67 ion beam deposition,61 laser ablation,63,64 spray prolysis,65 and atomic layer deposition.68 More work has been focused on the characterization of the grown films and the Si/CeO2 interface to control and improve the 12,61,64,65 quality of the SOI structure. Oxidation of silicon at the interface of Si/CeO2 is the biggest problem during the growth of CeO2 on silicon and production of electronic device. It is suggested that Ce played a role as oxygen reservoir in the process.64 It is predictable that better understanding the silicon oxide formation mechanism will shed light on the suitable control and use of oxidation of silicon in the production of electronic devices. Cerium oxide has also been fabricated as the buffer layer for the superconductor-related structures. Conventionally, the epitaxial deposition of the high quality and high-temperature superconducting YbaCu3O7-δ thin films are obtained at the substrates of YSZ, SrTiO3, LaAlO3, MgO.69-71 These substrates have stable chemical properties to endure high-temperature superconductor growth and the resulting superconductor thin film is epitaxial, uniform and defect-free over the whole film surface.72 However, they have difficulty in forming large single crystals, which is required for electronic devices, and the processing cost is expensive. In contrast, silicon is a well-established material with the advantage of commercialized production of large single crystal pieces and uniform epitaxial clean surface, and it has been widely applied in different electronic devices. It is a perfect substrate for YBCO used in the electronic application. The epitaxial deposition of YBCO on the surface has been reported as well.62 18 However, it is found that there was diffusion of barium from YBCO to Si surface and further reaction between Si substrate and the outer-diffused barium, which severely affects properties of the silicon surface and weakens the superconductivity of YBCO.72 To prevent the inter-diffusion, the buffer layer construction was suggested and studied. CeO2 has all the merit of the conventional oxide substrates. Compatibility between silicon and CeO2, together with possible epitaxial growth CeO2 thin film on silicon surface encourages using CeO2 as the intermediate layer between Si and superconductor.62 According to the investigations, there is a large lattice o mismatch between CeO2 and cuprate superconductors but by rotating 45 in the CeO2 basal plane, the lattice mismatch will be less than 0.16% and 1.7% along a and b axial of YBCO, 68,69 respectively. The inertness of CeO2 to either Si or YBCO further illustrates the advantage of 72 CeO2 as a buffer layer for YBCO on Si surface. Growth of a-axis oriented YBCO on (110) oriented CeO2 buffered silicon surface was reported and thus-produced YBCO structure is uniquely favorable to the fabrication of Josephson tunnel junctions because the electrical transport is along Cu-O planes and superconducting coherence lengths in a,b-directions are much larger than that along c-direction, and the longer coherence length facilitates the construction of Josephson tunnel junctions.70,72 In addition to the superconductor-on-buffered-silicon system, the similar structure based 71,73 on CeO2 as buffer layer has also been constructed with sapphire (Al2O3) as the substrate. And more investigations are focusing on this topic since some question the inertness of CeO2 related 74,75 to superconductor or substrate lately. Furthermore, the CeO2 intermediate layer has been used in the fabrication of superconductor-insulator-superconductor (SIS) multiplayer structure as well.72 It is recognized recently that the buffer layer not only plays the role in suppression of the inter-diffusion but also functions as the template for the further epitaxial growth of YBCO.76 19 CeO2 was also deposited on the outer-surface of YBCO as protective thin film to prevent corrosion of YBCO against the environmental humidity and severe processing conditions.77 The new development of CeO2 in the superconductor industry is the inclusion of CeO2-SnO2 combination in YBCO to improve the critical current density of the superconductor.78 Due to the high dielectric constant of cerium oxide of around 26, the epitaxial deposited film is a promising material to be applied for the capacitor structure.62 1.1.2.2.4 Applications in pH Sensors and Gas Sensors Oxygen vacancy is a well-understood property for nonstoichiometric cerium oxide. Oxygen vacancy diffusion coefficient of cerium oxide is reported as 10-5cm2 s-1 at 970 oC.79 In addition, cerium oxide has high chemical stability at high temperature. Therefore cerium oxide or ceria based films have been used to act as oxygen gas sensors at high temperature. The application examples include sputtered CeO2 film and mist pyrolysized CeO2 powder based thick film, both presents high sensitivity and fast response time to the oxygen partial pressure 79,80 change. As a n-type semiconductor, CeO2 based sensors can be applied to detect other reductive gases because the oxygen vacancies can react with these gases, such as CO, NO, and acetone.81,82 To evaluate the sensitivity of ceria-based gas sensors, Stefanik studied the relationship between the electrical conductivity of the Pr-Ce-O system with oxygen partial pressure and temperature.83 The pH sensing application of cerium oxide is based on oxygen vacancy induced ionic conductivity of cerium oxide. The potentiometric metal/ metal oxide pH sensor has advantage of working at high temperature and some special situations.84 The samarium doped ceria shows near Nernstian behavior in the pH range of 3 ~ 9.85 This proton related reaction leads to the interfacial potential between sensor surface and the solution.86 20 1.1.3 Production of Cerium Oxide The use of cerium oxide in a number of broad applications has encouraged much research for the production of cerium oxide. The production of cerium oxide films or powders has attracted the attention of various researchers. The following will review the often-employed approaches of cerium oxide synthesis in detail. This section starts with cerium oxide film deposition. 1.1.3.1 Cerium Oxide Film Deposition A variety of techniques can be utilized and has already been applied to prepare cerium oxide films. The techniques can be categorized as vapor phase deposition and liquid-based methods. Vapor phase methods include chemical vapor deposition (CVD) and physical vapor deposition (PVD) (e.g. sputtering, thermal evaporation, pulsed laser ablation/deposition (PLA/PLD), ion (assisted) beam deposition (IBD), electron beam deposition (EBD), molecular/atomic beam epitaxial (MBE)). Liquid based methods cover the deposition by chemical reactions from solution (e.g. homogeneous precipitation, spray pyrolysis or electrochemical reactions), deposition from solutions or suspension (e.g. sol-gel methods), screen printing and liquid phase epitaxy.87 The electrochemical methods will be presented in detail in Chapter 2. 1.1.3.1.1 Sol-Gel Methods Sol-gel synthesis is a solution-based method that can be applied in both ceramic powder formation and film deposition. As the term implies, the starting material in the sol-gel process is in a sol-like phase that is composed of colloidal ceramics particles dispersed in a liquid or polymerizable metalorganic precursors. The final product is a gel-like phase in which the solid network is interspersed with continuous liquid phase. The sol-gel process has been discussed in 21 detail in literature.88-90 A large variety of molecular precursors have been developed for sol-gel oxide ceramics, which are divided in terms of solvents: metal alkoxides in organic solvent and metal salts in aqueous solution. The ceramic gel can be formed through hydrolysis and condensation:91 Hydrolysis: M − OR + H 2O = M − OH + ROH (Equation 1.22) Condensation: M − OH + RO − M = M − O − M + ROH (Equation 1.23) The commonly employed coating techniques used in sol-gel film formation are dip coating, spin coating, spray coating and electrophoresis.90,92 Compared with other techniques, sol-gel can provide a large versatility for compositions and structures. Direct control of ion ratio in making precursors can decide the composition.93 This deposition method is a relatively inexpensive technique for film deposition. Drawbacks exist for this technique in several aspects. First of all, since the precursor is a solid particle dispersed in solution, the solution is vulnerable to aggregation before and during the reaction, the composition and properties of the solution are susceptible to change. The second limitation is that the film usually cracks due to drying and thermal treatment, and high temperature may induce malfunctional transformation of the product. Despite the limitations of the technique, sol- gel deposition has been and still will be widely employed in ceramic synthesis. Sol-gel techniques have been extensively used in cerium oxide thin film 52,65,89,93,94 deposition. Czerwinski employed CeO2 in HNO3 as the precursor to dipping for deposition of cerium oxide on substrates. As produced CeO2 film was utilized to prevent the oxidation of substrates (Ni, Cr, Ni-based alloys) at high temperature.89 In the papers of Özer, Ce(NH4)2(NO3)6 dissolved in a mix of ethyl alcohol, HNO3, and diethanolamine produced 52,95 optically sensitive CeO2 films by spin coating from the aged precursor solution. 22 93 Ce(NH4)2(NO3)6 was also chosen to form the CeO2 film on glass plates. Sol-gel method is also employed for the doped or mixed ceria formation. The sol-gel deposited CeO2-TiO2 film has been studied in detail from the view of formation process and conditions such as precursors employed, coating mode applied by a variety of research groups.55 The characterization of the film revealed the distinguished optical and electrochemical properties and predicts 55,96 electrochromic application. SiO2-CeO2 and CeO2-PZT films were also deposited on glass and sapphire respectively with sol-gel method.94,97 Most of the film depositions employed cerium salt (Ce(NH4)2(NO3)6) or cerium alkloxides as the component for the precursor. Following formation of the precursor solution, dipping or spinning were used to form gel on the substrate, seldom spray deposition has been applied in this step. 1.1.3.1.2 Sputtering Sputtering98-100 is a technique usually operated in vacuum. In the reactor, a large negative potential is applied to the cathode, and the sputtering gas ions are accelerated to a high energy and bombard the target placed over the cathode. Upon collision, the atoms and ions from the target are ejected, and some reach the substrate to form a thin film. This physical sputtering dissociates the target surface and emits secondary electrons, which affects the quality of the deposited film by local heating or local biasing.99 Magnetron sputtering has been adopted to trap the secondary electrons. Usually, the rare gases such as He, Ne, Ar and N2 are employed in the sputtering system for ionization.98 Among three sputtering modes, dc-sputtering is the most commonly used to prepare pure metals and alloys. Post-oxidation after dc-sputtering can give a ceramic oxide, whereas rf-sputtering can directly deposit ceramic oxide films if oxide target is used.98,100 Reactive sputtering was developed to support the deposition of a film composed of more than 23 one element.100 The most important application of reactive sputtering is the growth of ceramic nanostructures and cermets.99 Sputtering is a versatile technique that can be utilized in the film formation of a variety of materials. It has no constraints in the shape and configuration of the substrates and the resulting films show uniform surface and homogeneous properties. The stoichiometry of the final film can be adjusted according to requirement. However, the need for a vacuum and the careful handling of various types of gases, complicate the operation of the technique. All three types of sputtering have been adopted in the growth of CeO2 thin films. Gerblinger used this technique to grow CeO2 films for high temperature applications of the gas 79 sensors. Chin and his colleagues extensively worked on the r.f. sputtering of CeO2 and observed that the stoichiometry of the CeO2 film varied with the sputtering temperature and 101,102 O2/Ar ratio in the sputtering gas and deposition pressure. r.f sputtering of epitaxial CeO2 thin films on Al2O3, LaAlO3 and MgO2 with CeO2 as the target was described in Jacobsen’s dissertation. And the relationship between the properties of the thin films (morphology, orientation, microstructure, and chemical states, etc.) and deposition conditions (substrate type, on/off axis substrate/magnetron configuration, sputtering pressure, etc.) was also inspected in detail.100 In addition, reactive dc-sputtering was employed with cerium metals as the target and Ar/O2 mixture to deposit CeO2. The investigation of epitaxy evolution indicates that high 103 temperature is necessary for epitaxial CeO2/MgO, CeO2/Al2O3, CeO2 /YSZ thin films. Since sputtering has the ability for epitaxial growth of the cerium oxide thin film at high temperature, it is often applied in CeO2 buffer layer deposition in superconductors and semiconductors. 1.1.3.1.3 Ion Beam Deposition (IBD) and Ion Beam Assisted Deposition (IBAD) 24 Ion beam deposition (IBD) is a well-established physical vapor deposition method and is capable of forming a variety of metallic and ceramic deposits.104 It employs an energetic, broad ion beam source to impinge onto the target. The sputtered materials then deposit on the substrate. For Ion Beam Assisted Deposition (IBAD) process, reactive or non-reactive ions are employed to form thin films or assist in the deposition of thin films. This concurrent combination of deposition and ion bombardment results in films with improved characteristics in adhesion, density, intrinsic stress, morphology, orientation, durability and resistance to moisture.105 Both IBD and IBAD have been applied in the deposition of single crystal cerium oxide thin films. Huang et.al. reported the deposition of CeO2 on Si (111) by low energy dual ion beam 106 deposition. Wu et.al. epitaxially grew the CeO2 on single crystal Si via the same method as Huang and his colleagues studied the interface of Si/CeO2 hetero-structure and suggested that compared with other PVD methods, the benefit of the IBD method for the deposition of CeO2 on 61 Si is that no SiO2 is involved. The application of IBAD in CeO2 thin film formation is indicated by Gnanarajan’s report, in which ion beam assisted magnetron sputtering was combined to 104 deposit the polycrystalline CeO2 films for buffer layer usage in superconducting systems. 1.1.3.1.4 Pulsed Laser Deposition/Ablation (PLD/PLA) Pulsed laser deposition/ablation as a film deposition technique has attained wide attraction since the successful growth of high temperature superconducting films in 1987.70 It has been used on a large scale for the epitaxial deposition of thin films including metallic multilayers, cermics oxides, and other various crystalline lattices. Basically, the excimer laser beam is impinged onto the target and the elements in the target are ablated out of the target surface by the high-energy short pulse laser. Then the evaporated materials diffuse to the substrate and further nucleate and grow on the substrate to form a film. 25 The stoichiometry of the deposited film is controlled by the content ratio of different elements in the target. Film thickness can be manipulated with pulse characteristics. Because of the high energy of laser and high heating rate of the ablated materials, the deposition of thin films with PLD can be handled at lower temperatures. Compared with other physical vapor deposition methods, it is a much more promising deposition technique used for semiconductor and underlying integrated circuit system. The problems can be summarized as follows: the angular distribution of ablated species is narrow because of the small spot size of the laser beam, so there is difficulty in large area substrate deposition. The laser splashes large size of particulates, affecting the uniformity of the thin films. In addition, the high-energy species may collide and damage the deposited films.107 PLD/PLA has been identified as an important deposition approach for the epitaxial deposition of cerium oxide thin films on semiconductor or insulator substrates.60,63,75 Amirhaghi 108 et.al.. presented the deposition of CeO2 film on single crystal silicon wafers via PLD. The study of deposition parameters suggested high temperature and low oxygen partial pressure can produce better crystalline CeO2 films. Kang et.al.. deposited oriented CeO2 on p-Si (100) -3 substrate using PLD in the chamber of base pressure below 10 Pa, in which N2 and O2 were introduced to prevent substrate oxidation and help CeO2 formation respectively. The deposition result indicates that controlling oxygen and nitrogen pressure can adjust the epitixial properties 60 of the film. Shi’s report indicates that the H2 involved atmosphere is more beneficial to lower temperature growth of pure epitaxial CeO2 film on YSZ than O2. Hydrogen-assisted PLD was 109 also applied in epitaxial deposition of CeO2 films on InP (001). 26 1.1.3.1.5 Chemical Vapor Deposition (CVD) Chemical vapor deposition is a technique where gaseous sources can be used to deposit thin films by an appropriate chemical reaction.110,111 Most common CVD methods include hydride CVD, trichloride CVD and metal-organic CVD. The gas phase in the CVD reactor is the mixture of carrier gas and reacting gases. The reacting gases act as the chemical precursors individually or react with each other to supply reaction precursors. The carrier gas functions as the dilution agent for the reacting gas at the time of reaction and deposition process, and also helps to carry the reactants when the reactants are not in the gas phase before going into the CVD supply system. Basically, the solid or liquid reactants are transferred by carrier gas to CVD reactor. The diluted precursors in the reactor are continuously flowing over the substrate surfaces and so forms a boundary layer to further reactions. The precursors close to the surface are adsorbed on the surface and through surface diffusion and surface reactions grow as the film. After deposition, gas effluents flow downstream to the tail system for toxic elimination and discharge to the outside environment. The most attractive advantage of CVD compared to PVD is conformal deposition (deposit is formed all over the exposed parts of a three-dimensional structure). CVD can be operated at a relatively low temperature. The morphology and stoichiometry of the grown film are controlled by engineering the deposition conditions and parameters. The first problem involved in CVD is the toxic or flammable gases used in the reaction and the complicated design for precaution chemical safety hazards. Compared with most wet chemical methods, CVD operation temperature is still high and it may induce high-temperature by-products. Since the advantages of CVD outweigh its limitations, it is widely applied in the deposition of metals, 27 semiconductors, and ceramics and the deposited materials can be amorphous, single-crystalline, and polycrystalline.112 CVD is a well-established method for the preparation of cerium oxide films. Usually the metal-organic CVD is employed for the formation of the CeO2 films. Lu et.al.. and McAleese et.al.. used Ce(thd)4 solid source as the precursor to deposit the CeO2 film on sapphire substrate 41,113 and the silicon wafer (100), respectively. Powders of Ce-β-diketonate, Ce(hfd)4, Ce((hfac)3(glyme) and Ce(hfc)3(diglyme) were also employed in the CVD production of CeO2 114-117 and doped CeO2 thin films. The decomposition of metal-organic precursors was involved in the aforementioned MOCVD processes. As-produced CeO2 thin films showed dense and high adherent properties and have potential application as the buffer layer for high temperature superconducting systems. In addition to MOCVD, Carter and his colleagues used Ce(III)-2EH and Ce-(TMHD)4 as chemical precursors to deposit CeO2 films on sapphire with combustion CVD. They studied the deposition process in details to establish the relationship between 118 morphology and orientation of the CeO2 film with the deposition conditions. 1.1.3.1.6 Other Techniques The application of spray pyrolysis in the deposition of cerium oxide has been demonstrated by several references.50,65,119 In this approach, a solution of cerium precursor (e. g. cerium nitrate, cerium chloride, cerium acetylacetonate) is neutralized and sprayed on the substrate and then thermally decomposed to form the CeO2 film. Electron beam bias evaporation, electron-beam assisted evaporation and dual plasma deposition were also employed for the 120 epitaxial deposition of CeO2 on silicon substrate to lower the deposition temperature. In 121 122 addition, flash evaporation and photoresist-free lithography were applied in the CeO2 formation as well. 28 1.1.3.2 Cerium Oxide Powder Production There are also two categories of approaches involved in the formation of cerium oxide powders. Wet-chemical processing techniques include sol-gel methods, chemical homogeneous precipitation, hydrothermal reaction of aqueous metal-salt solutions and electrochemical synthesis. Another type is direct thermal treatment, such as thermal decomposition. 1.1.3.2.1 Sol-Gel Method The sol-gel method has been discussed in detail for thin film deposition. As one of the wet-chemical methods, it is also used in cerium oxide or cerium oxide based powder 97,123-125 formation. It is reported that CeO2-ZrO2 mixed oxide prepared by the sol-gel method has 125 good redox properties and sol-gel based CuO-CeO2 has been used in the catalytic wet oxidation of phenol.124 1.1.3.2.2 Homogeneous Precipitation and Hydrothermal Reaction Chemical precipitation is a wet-chemical method to produce ceramic powder, in which a precipitating agent is added to the solution containing metal cations so that the metal species can be precipitated because of non- or low- solubility of the precipitant-metal-species-reacted- product. Sometimes, hydrothermal treatment is necessary to form the final product. Directly pouring the precipitant into the mother solution results in non-homogeneous and agglomerated precipitate by an instant reaction. And the resulting precipitate is vulnerable to inclusion of the solvent and contamination molecules in the crystal. When the precipitating agent is released into the mother solution slowly and uniformly, the thermal-kinetic properties for precipitation formation are well controlled. This will eliminate inclusion and agglomeration and lead to homogeneous precipitation. Precipitation produces hydroxide followed by hydrolysis to obtain 29 the oxide. Hydrolysis can be achieved by increasing solution pH, increasing solution temperature, and hydrothermal treatment with higher temperature at controlled pressure.126 Chen et.al. employed hexamethylenetetramine decomposition to prepare cerium oxide powders, in which the pH increase is the decisive factor to form oxide powders.127 Djuričić and Pickering employed hydrogen peroxide to oxidize cerium species in solution and then added ammonium hydroxide in drops to increase pH of the mixture. The method is a slow precipitation process compared with direct introduction of ammonium salt, which diminished the agglomeration of the nano-particles.87,126 Hydrothermal treatment was usually preformed on precipitated powder to obtain the final pure powder.87,126,127 Pure thermal hydrolysis was utilized 128 by Hirano et.al.. to prepare CeO2 and sulfate is found to assist in spherical agglomeration. Organic agents are involved in hydrothermal precipitation of CeO2. Wang and his colleagues prepared nanocrystalline CeO2 using ethanol as a solvent to dissolve cerium (III) nitrate.129 Konishi et.al.. precipitated cerium oxide powder directly through hydrolyzing tertiary carboxylate solution of cerium (III) at 130-170 oC and 0.27-0.78 Mpa. In this type of preparation method, the contamination of organic materials can be eliminated with more severe thermal treatment.130 1.1.3.2.3 Thermal Decomposition Thermal decomposition is another approach to produce CeO2 powder. Salts are employed as precursors, as temperature is applied, precursors proceed via some reactions such as dehydration, dehydrogenation etc, which release the volatile species (H2O or CO2, small organic species etc.) to form the oxide powder. Thermal decomposition can occur in both air131 and vacuum.132 Often used precursors include oxalate, carbonate, acetate, and some organic acid ligated metal complexes. 30 Cerium oxide powder resulting from thermal decomposition is usually a mixture of all stoichiometries which include fluorite CeO2, hexagonal Ce2O3 and other non-stoichiometric oxides. The composition and phases of the product will depend on the reaction precursor and reaction conditions. The earliest decomposition of cerium salt can be dated back to 1950s, when Wendlandt utilized Ce(III) oxalates as the precursor to produce CeO2 at a temperature of 360oC.131 Oxalate thermal decomposition have been well examined for mechanism, process, and the decomposition condition (in air or in vacuum).132,133 It is reported that Ce(III) basic carbonate hydrate, hydrated cerium benzene-1,4-dioxygacetat, cerium mallonate, and ammonium cerium 134 oxalate were studied for the decomposition to form CeO2 as well. Thermal decomposing different precursors leads to cerium oxide through different reaction paths. 1.1.3.2.4 Other Synthesis Methods Combustion synthesis is a direct powder preparation process in which a saturated aqueous solution with desired metal ions and a suitable organic fuel is ignited. The combustion leads to a dry oxide powder. In this process, no intermediate decomposition is involved and no further thermal treatment needed and thus the produced powder is homogeneous and unagglomerated. This method has been employed to produce Ga-doped ceria.135 Microemulsion method was explored to prepare cerium oxide nanocrystals with controlled size and surface area. In this method, the cerium (III) nitrate and ammonium hydroxide were dispersed in a water-in-oil system to form so-called microemulsion and the microemulsion was evaporated to obtain cerium oxide nanocrystalline powder.136 In addition, some vapor related deposition methods also can be applied to powder formation, for example, sputtering.98 1.1.4 Characterization of Cerium Oxide 31 A long list of techniques has been applied to study cerium oxide powders and thin films. X-ray diffraction (XRD),137 Neutron diffraction,138 Fourier-transform infrared (FT-IR),139,140 Raman spectroscopy,141,142 Nuclear Mass Resonance (NMR),143 and UV-visible spectroscopy144 can be used to probe and determine the structure. X-ray absorption spectroscopy (e.g.XANE),45,145,146 X-ray photoelectron spectroscopy (XPS),147,148 Auger electron spectroscopy (AES), low energy electron diffraction (LEED) and scanning probe microscopy (SPM)149,150 study the cerium oxide surface variation and provide information of oxidation state and stoichiometry information. BET technique characterizes the microstructure of particles in the form of surface area, pore size and distribution.151 Scanning electron microscopy (SEM)152 and transmission electron microscopy (TEM)153 illustrate the morphologies. Thermal gravimetric analysis (TGA) and sintering/annealing examine the thermal properties.154 Impedance spectroscopy is used to study the conductivities of cerium oxide.155 In addition, the magneticity, semiconductivity, corrosion protection property, catalytic properties have also been investigated with various techniques.2 1.2 Cerium Oxide/ Layered Silicate Nanocomposites 1.2.1 Montmorillonite Nanoclays Montmorillonte layered silicate with the formulas (Na,Ca)(Al,Mg)6(Si4O10)3(OH)6·nH2O) has a 2:1 structure in which the single aluminum octahedral layer is sandwiched between two layers of a silicon tetrahedral (Figure 1.2). Each layer extends infinitely in the x-y dimension. In the silicon tetrahedral, the silicon is bonded with four oxygen atoms at four corners. Each of three oxygen atoms in one plane is shared by three other silicon tetrahedrons forming a two- dimensional plane, whist the fourth oxygen is extruded from the plane and participates in another layer. In the aluminum octahedral layer, six oxygen atoms or hydroxyls surround one aluminum 32 atom to form an octahedron, the oxygen-oxygen edges are shared between the octahedrons. This octahedral structure is repeated and extended to infinity in the plane dimension. The tetrahedral layers and octahedral layer are connected together by the corner oxygen atoms aforementioned. In this structure, the distance of the Si-O bond is around 1.62Å, O-O around 2.64Å and Al-O 1.77Å. The Si4+ in the tetrahedral layer and the Al3+ in the octahedral layer are sometimes substituted by lower valent cations (e. g. Al3+ replaces Si4+ and Mg2+ or Fe2+ replace Al3+, etc.), which introduces negative charges into the layered structure. The negative charges are counterbalanced by the loosely held alkaline or alkali cations from the associated water. For example, sodium montmorillonite indicates the associated cation is sodium.156-158 Figure 1.2 Illustration of montmorillonite unit structure Horch, A. et.al. Chem. Mater. 14(8), 2002, P3531, Copyright permitted by ACS The individual montmorillonite platelet exists as 2:1 coordinated layers measuring about 1 nm in thickness and several microns in width. This unit is stacked in the z dimension with a regular gap of nanometer regime in between and the forces that bond the units together are van der Waals force. This internal gap is coined the gallery spacing and is a very significant parameter in characterizing the properties of layered silicates.157,159 In addition to cations, it is 33 possible for organic species to be inserted into this spacing to make these materials more hydrophobic for some applications. Therefore, the basal spacing of the layered silicate not only depends on the exchanged cations and associated water but relies on the size of the inserted organic species as well. The swollen form of the layered silicate is an intercalated structure (Figure 1.3A). In this structure, the intercalant species overcome van der Waals force and separate silicate layers but leave the units intact.160 Another structure is the exfoliated structure (Figure 1.3B) in which the individual silicate platelet not only separates but delaminates as well.161-163 This is an attractive property for this material because the individual platelets can be incorporated into a matrix of other crystals so as to change the properties of the crystal. Figure 1.3 (A) Intercalation (left) and (B) Exfoliation (right) of montmorillonite Horch, A. et.al. Chem. Mater. 14(8), 2002, P3532, Copyright permitted by ACS 1.2.2 Secondary Phase Reinforcement in Composites There are two ways that can be used to improve the mechanical strength of the ceramics. The first method is thermal treatment such as sintering and annealing of the ceramics, which will change the properties of the ceramics. And the second method is to introduce a second phase into the matrix of the materials to form composites, tuning the properties of the ceramic materials. 34 A composite is defined as the combination of two or more solid phases containing different compositions or structure. Nanocomposite refers to the composite in which one phase is in the nanoscale regime. Usually, the composite performs differently from the pristine materials that compose it but to a degree its performance greatly depends on the properties of compositions.164 The second phases to form the ceramics composite and improve toughness and strength of the ceramic materials are in the form of particles, fibers, whiskers, and platelets.165 The employed materials can be metals, ceramics, and naturally or artificially formed compounds. The reinforcing mechanisms are dependent upon the properties of the secondary phase that form the composites. 1.2.2.1 Particle Reinforced Ceramic Nanocomposites Particles are the most often used phase to improve the strength of ceramic materials. According to the morphological properties of secondary phase, the particle can be defined by the aspect ratio (length/diameter) as around 1.0.165 It is reported that nickel, iron, chromium, molybdenum and stainless steel metal particles have been dispersed in the matrix of zirconia to form reinforced ceramic matrix composites for solid electrolyte or anode in solid oxide fuel cells 166 (SOFC). Other examples for particle reinforced ceramic materials include Al2O3 and MgO 167,168 composites toughened with materials such as TiC, TiB2, SiC, (W, Ti)C and Ti(C,N). The reinforcement mechanisms of the ceramics with particles as the second phase have been studied for several decades. The most recent opinion was proposed in 1990s by Niihara and his coworkers. They suggested that the nanosized particles are dispersed in the matrix grains and grain boundaries. Thermal expansion mismatches between the matrix material and the second phase bring up dislocations in close vicinity of the second-phase particles. The dislocations from 35 the dispersed particles are connected as networks to reinforce the structure of the matrix materials.169 This proposal was supported by the study of Awaji et.al..170 Three types of toughening mechanisms related to crack development are identified by Tan et.al.:171 the switch of the intergranular cracking to the transgranular crack by nanoparticle distributed along grain boundaries; nanocomposite residual stresses perturbing original crack path; the transgranular and intergraular crack paths blocking crack tip development. Crack-tip bridging was regarded as the primary toughening scheme for the particle reinforced ceramic nanocomposites.172 Crack-tip bridging refers to the particle bridging a propagating crack behind the crack tip. The extension path of the main crack is blocked by the particle and deviates in another direction. The bridging particles constrain the crack-opening displacement and finally prevent the crack development. The deep theoretical analysis of this toughening mechanism can be found in the review paper by Jeong and his coworkers.168 Microcrack is another important mechanism that has been proposed and extensively speculated for the particle-reinforced ceramics.173,174 In the composites, microcracks develop at the vicinity of the macrocrack tip as a result of the external loads and local stresses. The emergence of the microcracks has two opposite effects: on one hand, microcracks or microdefects redistributes and further relieves local stress near the macrocrack tip, preventing the further propagation of the microcrack. On the other hand, the microcracks have the tendency to coalesce and to combine with the existent macrocrack, which promotes the growth of macrocrack. The former crack-shielding function and the latter crack accumulation compete and lead to the ultimate enhancing or degrading in the toughness of the materials. It is revealed from the microcrack models that the toughening depends on density, distribution, and residual opening 36 of the microcracks.175 The detail theoretical study of microcrack toughening with mathematical models can be found in the papers by Huchinson and Kotoul.176,177 Transformation toughening theory was initially proposed in the 80s.178 Establishment and perfection of the theory have been underway for two decades, the detailed description of this 179 toughening mechanism can be retrieved in Green’s book. Transformation appears in the ZrO2 toughened composites.63,180,181 Transformation occurs in the high stress field at the vicinity of crack tip and toughening arises as the result of the volume expansion during the phase transformation of zirconia from tetragonal to the monoclinc phase. This volume expansion induces the compressive stresses that prevent the further development of the macrocrack so that the toughness of the composites is highly improved.182,183 Pull-out has been mentioned as the mechanism for the particle toughening as well. However, it is more related with the interface between the phases so it contributes more in the fiber, whisker or platelet toughening and will be covered in the following sections. Although the particle secondary phase has the advantage of easy fabrication, the toughening efficiency of anisotropically morphological materials such as the lengthy fibers and whiskers and the disk-like platelets is predicted comparably higher.184 1.2.2.2 Fiber Reinforced Ceramic Composites Compared with particle, the aspect ratio for fibers is much higher than 10, which corresponds to the one-dimensional phase.165 Inserting the continuous fibers into the matrix of the crystal is another approach to toughen materials. Continuous SiC unidirectional fiber reinforced lithium-α-sialone matrix composites were fabricated and the effect of different sintering additives was studied.185 Mullite fiber was coated with zirconia and then used in the toughening of the mullite ceramic composites.186 SiC monofilament and the multifilament 37 Nicalon and HPZ fibers have been used to reinforce the celsian BaAl2Si2O8 (BAS) and 187 SrAl2Si2O8 (SAS). Metal fibers have also been used to improve the mechanical properties of the ceramic materials: nickel fiber added into the matrix of alumina led to stable crack propagation and the energy dissipation during fracture has been improved from 15 J/m2 to 867 J/m2.187 More and more materials can be fabricated as fibers and employed to form the composites with ceramics materials to improve the strength and toughness of this ceramic material (carbon, oxide and even polymer materials). Fibers can be categorized as two types according to length: one is continuous fibers that usually refer to the naturally formed long fibers whereas another is the short fibers or chopped fibers. The reinforcement scheme for continuous fibers comes from the distribution of the load over length to support the matrix at a site of local stress.163 The toughing mechanisms of chopped fibers are much closer to particle toughening. Therefore, crack bridging, transformation and microcrack are also used to describe the toughening and strengthening function of chopped fibers in composites. It is identified that the properties of the fiber/matrix interface plays a very significant role in the development of the fiber reinforced ceramic materials.188 Simulation and various models have been suggested for the short-fiber reinforced ceramics.188-190 Crack deflection followed with fiber pullout and debonding is suggested as an important mechanism for ceramic composites. It is well known that if the interaction and bonding between the fiber and matrix are strong, the material will be so brittle that it cannot bear damage. A weak interface between the matrix and fiber phases is helpful. When the external force induced crack is propagated, the crack can be deflected along the matrix-fiber interfaces, which results in the debonding and the pull-out of the intact fiber away from the matrix instead of crack through the fiber. The pull-out and debonding at the interface dissipate energy, which lead to the insufficient 38 energy for the crack development and thus toughens the material.188,191 It is possible as well that the unbroken strong fibers behind the crack front bridge the broken crack surface, which assists the reinforcement to a certain degree.189 1.2.2.3 Whisker Reinforced Ceramic Composites Similar to fibers, whiskers have the aspect ratio of larger than 10 and can be incorporated into the matrix of ceramic materials to form the reinforced materials. The distinguished example 192,193 is the SiC whiskers, and matrix ceramics can be Al2O3, ZrO2, Y2O3. Si3N4 whiskers also act 194 195 as reinforcing phases, which have been applied in the strengthening of BAS, MgO/CeO2, and SiC.196 Besides inorganic and ceramic materials, whiskers composed of some other types of materials have been seen recently, such as starch, cellulose and chitin whiskers to function as reinforcement phase in dental composites.197,198 Since the elongated whiskers are easy to be entangled and agglomerated and reluctant to be dispersed the system matrix, the reinforcement effect is retarded. Therefore, researcher suggested adding and fusing silica particles with whiskers, which hinders the entanglement of whiskers and benefits the distribution of whiskers in the matrix.199 Since the morphological properties (e. g. aspect ratio) of whiskers are close to that of fibers, the toughening mechanisms of fiber-ceramic composites are applicable for whisker- ceramic composites as well. 1.2.2.4 Platelet Reinforced Ceramic Composites Platelets that have an aspect ratio of less than 0.1 (ratio between the thickness and width) are a different phase from the previously mentioned fibers and whiskers, since this phase has a similar dimension for length and width so that it is regarded as a two-dimensional phase in comparison to the one-dimensional definition for fibers and whiskers. Two-dimensional phase, 39 in the terms of load distribution, has an advantage over the one-dimensional phase. It was suggested that the platelet reinforcement is much more desirable and practical when 200 environmental and health are concerned. Therefore various platelets (e.g. SiC, Al2O3, ZrB2) have been widely applied in the toughening and strengthening of ceramic materials with obvious improved mechanical properties. The two famous reinforcing ceramic materials Al2O3 and SiC were studied from experimental and theoretical views by Janssen et.al. for the possibility of effective toughening in various types of ceramics.201 It was found that less than 20% of SiC- platelets embedded RBSN showed a 60% increase in fracture toughness while the Al2O3- platelets addition resulted in a decrease in the strength of yttrium-stabilized zirconia (3Y-TZP). SiC-platelets were added to Barium alunimosilicate (BAS) to form composites, the fracture toughness and flexural strength increased 213% and 182%, respectively.202 Mica platelets were incorporated into the matrix of barium aluminosilicates (BAS), which increased the fracture toughness by 110% and Young’s Modulus by 124%.203 The investigation on mechanical properties of SiC platelet embedded Al2O3 composites concluded that both Young’s modulus and fracture toughness of the composites with small platelets increased with the increase of SiC 204 volume fraction. And SiC platelets have also been used in the reinforcement of Al2O3/SiC- 205,206 particle hybrid and Al2O3/Y-TZP composites. The platelet-reinforced Ti-B-C composites also showed higher hardness, flexural strength and fracture toughness.207 Parallel to the experimental investigations of mechanical strength and fracture toughening, the study of platelet reinforcement mechanisms has been carried out. The recognized toughening schemes related to platelets can be listed as followings: platelet pulling-out (slipping), crack deflection, crack bridging, platelet/matrix debonding, platelet cleavage, and load transfer.201 The crack bridging mechanism for platelets is similar to that of particle 40 reinforcement and the working schemes of pulling-out, crack defection, debonding for platelets demonstrate the close properties of those for fibers and whiskers, in which the interface between the matrix phase and platelet phase play a significant role. However, the anisotropy properties, which have been mentioned previously, have made this two-dimensional toughening distinct from one-dimensional toughening. Platelets give reinforcement in two-dimension whereas one-dimensional secondary phase only functions at one direction.165 Therefore for platelets as two-dimensional secondary phase, an extra point needs to be considered when mechanisms are applied: the orientation of the platelet. Pulling out of platelets from the matrix followed with crack bridging mechanisms is usually only applied to the situation that platelets oriented parallel to the tensile stress plane.201 Crack deflection demonstrates different toughening between the aligned orientation and random distribution of platelets in the matrix. The orientation of platelet reinforcement has been explored in detail from the experimental and theoretical views in Chou and coworker’s study.208 Another distinguished strengthening mechanism for platelets is load transfer. It is established that load distribution/transfer is the decisive function for the strength of the materials when the matrix has polymer-like structure. Fibers and whiskers have a pretty small diameter, so the load over cross-sectional area of one-dimensional phase basically relies on length to transfer to the environment. However, because the width dimensions for platelets are much larger, much less length is needed to achieve the same load distribution.163 The thickness of the platelet is fixed, the smaller the aspect ratio (thickness/width), the higher fraction of platelets oriented parallel to the tensile stress plane, and stronger adherence between interfaces will achieve better load distribution.201 41 Due to the distinguished toughening capabilities of platelets, more and more platelets have been proposed and applied in the research and industry. The relative large size of platelets may be the defect in the matrix weakening the material, hence platelet size is critical.201 Various types of reinforcement mechanisms have been recommended and adopted in explaining the secondary phase reinforcement. However, sometimes, one particular scheme cannot describe the phenomenon while the synergic effects of various mechanisms demonstrate the advantages. 1.2.3 Layered Silicate Involved Composites Layered silicates have been combined with various structures to form composites. Due to easy intercalation and exfoliation accompanied with nanoscale thickness of crystal unit, layered silicates can be utilized to form nanocomposites. The most popular layered silicate related nanocomposites are the polymer/clay nanocomposites. This type of compounds have been studied extensively from preparation, characterization, properties to various applications,209,210 since nylon/layered silicate composites, the first composite was reported by the Toyota’ researchers in 1980s.211 These nanocomposites show remarkable enhancement in properties compared with the corresponding original polymers: increased mechanical strength and moduli, improved heat resistance, reduced gas permeability and good flame retardancy.209,212 The unusual properties of layered silicate related nanocomposites are basically derived from the uniqueness of the layered silicates: platelet reinforcement. The chemical force between the layers of the layered silicate is inter-molecular bonding that is weak and can be broken easily. Therefore, when the layered silicates are incorporated with other materials, any external force to make the incorporation can induce the intercalation of the layered silicates. The dispersed silicate units in the matrix are the high aspect ratio-silicate platelets rather than the intact layered 42 structure. The toughness improvement of as-produced composites follows the platelet- reinforcement schemes. If polymers are involved, load transfer dominates, otherwise, depression of crack propagation type mechanisms dominate. The success of layered silicate in enhancement of polymer materials enlightens the researchers. It is conceived that these attractive properties of layered silicates are possible to be applied to the ceramics industry mainly to toughen the ceramic materials. The combination of silicates with ceramics has been studied for several decades. Most of the composites combined with the layered silicate and ceramic materials can be categorized as Pillared Interlayer Clays (PILC) or Cross-Linked Compounds (CLC). The PILC represents the clay materials that have been intercalated with other materials so that the basal spacing of the clay will increase.213 Such intercalated clay can be used as catalysts, adsorbents and molecular sieves.214 To improve the thermal stability, more and more inorganic species, metals, oxides or hydroxides have been 215,216 216 217 218 employed to intercalate the layered smecitites, such as Pd, CdS, Fe3O4, SiO2, 219 220 221 222 223 214,224 TiO2, ZrO2, Al2O3, Fe2O3, Cr2O3 and mixed oxide clusters. Cerium compounds have also been employed to form these types of the composites.225 To fulfill the need for the fine catalyst, the investigations of the transition from the microporous to nano-phase pillared layered structures have been attempted.215-217,226 Even though the oxide pillared interlayer clays can be looked as the layered silicate (mainly montmorillonite)-oxide composites, montmorillonite is the primary content in them while the metal oxide only acts as the additives for property modification. This can be regarded as the fundamental distinction between the pillared interlayer clay materials from the layered silicate reinforced composites in which the silicate is the additive and the second material will be the composite matrix,159,227 even sometimes, the percentage of silicate is as low as one out of thousands in the composites. 43 1.2.4 Synthesis of Layered Silicate Composites A variety of approaches have been employed to form layered silicate based composites where silicates function either as matrix or additives. Some of the synthesis methods will be listed in the following section. 1.2.4.1 Ionic Exchanging Pillared layered clays are formed through the ionic exchanging of interlayer cations with the sol or colloidal cations and then thermal treatment converts the intercalated inorganic species to pillared structure. This method is valid for the composites with silicates as the additives. 1.2.4.2 Layer-by-Layer Assembly Layer-by-layer assembly is usually used in the formation of the polymer composites in which the two precursors have opposite charges. The assembly will be accomplished by the adsorption of the charged components onto the oppositely charged polyelectrolyte-coated surface so that the stratified layered system can be formed. Magnetite and montmorillonite were two negatively charged components that need to be adsorbed to the substrate. The precursors for the layers were the magnetite nanoparticles and exfoliated montmorillonite by vigorous agitation. The positively charged polymer was first adsorbed onto the substrate and then one layer of magnetite or montmorillonite dependent on the requirement for the stack structure. Then the process was repeated until the adsorption saturation or the synthesis requirement was met.217 1.2.4.3 Colloid Synthesis Colloid synthesis has been applied to form the metal or semiconductor nanoparticle- intercalated layered silicate nanocomposites. Layered silicates are dispersed in a binary liquid mixture system. One component of the binary liquid mixture is a good solvent for the reactants and it is readily adsorbed in the interlayer of layered silicates while another cannot be adsorbed 44 onto the silicate and the solubility of the reactants in it is nearly zero. At suitable conditions, the reaction occurs and forms the required nanoparticles in the interlayer of silicates.215,216 1.2.4.4 Sintering of Powder Sintering of metal-coated layered silicate powders was adopted by Ohtsuka and his coworkers to produce the nickel-layered silicate composites. In this method, homogeneous precipitation combined with ion-exchange forms a nickel hydroxide-silicate complex. This complex was treated with H2 at high temperature to give nickel-layered silicate composites. The silicate phases were uniformly aligned and dispersed in the nickel matrix. The composites show anisotropic property because of the incorporation of the platelet-like layered silicate in the metal nickel.228 1.2.4.5 Organometallic host and layered silicate guest interaction Organometallic host and layered silicate guest interaction has been used to synthesize Cu- metallated form of tetrakis-(1-methyl-4-pyridyl) porphyrin (CuTMPyP) in the galleries of Na hectorite and Li fluorohectrorite. It was reported that this synthesis method can provide a way to ameliorate the structure and orientation of intercalatants and further tailor the properties of the composites.229 1.2.4.6 Electrochemical Synthesis During the synthesis of layered silicate composites, it is found that ultrasonification assists the delamination of the layered silicate.230,231 Addition of 20-35% of montmorillonite in the alumina composite improved the mechanical strength of the alumina.227 Therefore it is proposed by our research group to incorporate lower percentage of delaminated layered silicate into metal or ceramics to improve the mechanical properties of the green materials. A novel synthesis technique – electrochemical synthesis for the layered silicate composites has been 45 adopted. We successfully produced the nickel-montmorillonite nanocomposite thin films with electrodeposition and the material demonstrated 60% increase in the hardness over the pure nickel films.159 No literature has been observed for ceramic / layered silicate composites. The synthesis and preliminary property study of cerium oxide / layered silicate nanocompososites has been done and the result will be presented in Chapter 6 & 7. 1.2.5 Characterization of Layered Silicate and Composites XRD,159,232 Small-angle X-ray scattering (SAXS),233 FT-IR,234 FT-Raman,235 Si-NMR,236 TGA-DTA,237 TEM162 can be used to analyze layered silicate incorporation to composites. 1.3 Chapter Summaries and Dissertation Introduction The properties, applications, synthesis and characterization techniques are reviewed extensively in the first part of this chapter. And layered silicate composite related topics are covered in the second half: structure of montmorillonite, toughening effect in composites, composite synthesis and analysis. Chapter 2 will detail electrochemical techniques related to cerium oxide film deposition, nanocrystalline powder synthesis, electrochemical reaction mechanism study and corrosion studies (Tafel and linear polarization). In addition, a variety of characterization techniques employed in this research work will be discussed: X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), Scanning electron microscopy (SEM), BET technique, FT-IR, Raman spectroscopy, Nanoindentation and Sintering procedure. Chapter 3 focuses on anodic electrodeposition of cerium oxide thin films on stainless steel. The deposition conditions and optimization are to be discussed. 46 Chapter 4 presents electrochemical synthesis of nanocrystalline cerium oxide powders. The sintering properties and lattice variation of nano-powder will be illustrated. Chapter 5 studies the electrochemical formation of cerium oxide based on previous two chapters and proposes the corresponding mechanisms. Chapter 6 reviews electrochemical synthesis of cerium oxide/ layered silicate nanocomposite powders. Chapter 7 summarizes the preliminary results of cerium oxide / layered silicate nanocomposite film deposition and mechanical properties. 47 Chapter References (1) Molycorp Cerium: A Guide to its Role in Chemical Technology; Molycorp, Inc.: Mountain Pass, CA U.S.A, 1992. (2) Trovareli, A. Catalysis Reviews: Science and Engineering. 1996, 38, 439-520. (3) Mogensen, M.; Sammes, S. M.; Tompsett, G. A. Solid State Ionics 2000, 129, 63-94. (4) Bamwenda, G. R.; Arakawa, H. Journal of Molecular Catalysis A: Chemical 2000, 161, 105-113. (5) Lavik, E. B.; Chiang, Y.-M. Material Research Society Symposium Proceedings 1997, 457, 63-68. (6) Etsel, T. H.; Flengas, N. Chemical Reviews 1970, 70, 339-376. (7) Ihaba, H.; Tagawa, H. Solid State Ionics 1996, 83, 1-16. (8) Suzuki, T.; Kosacki, I.; Anderson, H. U. Journal of the American Ceramic Society 2001, 84, 2007-2014. (9) Chiang, Y.-M.; Lavik, E. B.; Blom, D. A. Nanostructured Materials 1997, 9, 633-642. (10) Koelling, D. D.; Boring, A. M.; Wood, J. H. Solid State Communications 1983, 47, 227- 232. (11) Haensel, R.; Rabe, P.; Sonntag, B. Solid State Communications 1970, 8, 1845-1848. (12) Morshed, A. H.; Moussa, M. E.; Leonard, R.; Liu, S. X.; El-Masry, N. Applied Physics Letters 1997, 70, 1647-1649. (13) Shen, W.; Matsumura, Y. Journal of Molecular Catalysis A: Chemical 2000, 153, 165- 168. (14) Xavier, K. O.; Sreekala, R.; Rashid, K. K. A.; Yusuff, K. K. M.; Sen, B. Catalysis Today 1999, 49, 17-22. 48 (15) Yao, C.; Weng, H. Industrial & Engineering Chemistry Research 1998, 37, 2647-2653. (16) Trovareli, A.; De Leitenbutrg, C.; Boaro, M.; Dolcetti, G. Catalysis Today 1999, 50, 353- 367. (17) Trovareli, A.; Boaro, M.; Rocchini, E.; DeLeitenburg, C.; Dolcetti, G. Journal of Alloys and Compounds 2001, 323-324, 584-591. (18) Ziehfreunt, A.; Maier, W. F. Chemistry of Materials 1996, 8, 2721-2729. (19) Yamamoto, O. Electrochima Acta 2000, 45, 2423-2435. (20) Stambouli, A. B.; Traversa, E. Renewable and Sustainable Energy Reviews 2002, 6, 433- 455. (21) Dokiya, M. Solid State Ionics 2002, 152-153, 383-392. (22) Kharton, V. V.; Figueiredo, F. M.; Navarro, L. Journal of Materials Science 2001, 36, 1105-1117. (23) Livermore, S. J. A.; Cotton, J. W.; Ormerod, R. M.; et.al. Journal of Power Sources 2000, 86, 411-416. (24) Steele, B. C. H. Solid State Ionics 2000, 129, 95-110. (25) Ralph, J. M.; Scheler, A. C.; Krumpelt, M. Journal of Materials Science 2001, 36, 1161- 1172. (26) Huijsmans, J. P. P. Current Opinion in Solid State and Materials Science, 2001, 5, 317- 323. (27) Boivin, J. C.; Mairesse, G. Chemistry of Materials 1998, 10, 2870-2888. (28) Arkinson, A.; Ramos, T. M. G. M. Solid State Ionics 2000, 129, 259-269. (29) Marina, O. A.; Bagger, C.; Primdahl, S.; Mogensen, M.; Sammes, S. M.; Tompsett, G. A. Solid State Ionics 1999, 123, 199-208. 49 (30) Park, A.; Gorte, R. J.; Vohs, J. M. Applied Catalysis A: General 2000, 200, 55-61. (31) Putna, E. S.; Stubenrauch, J.; Vohs, J. M.; Corte, R. J. Langmuir 1995, 11, 4832-4837. (32) Kosynkin, V. D.; Arzgatkina, A. A.; Ivanov, E. N.; Chtoutsa, M. G.; Grabko, A. I. Journal of Alloys and Compounds 2000, 303-304, 421-425. (33) Šulcova, P.; Trojan, M.; Šolc, Z. Dyes and Pigments 1998, 37, 65-70. (34) Arenas, M. A.; Conde, A.; de Damborenea, J. J. Corrosion Science 2002, 44, 511-520. (35) Breslin, C. B.; Chen, C.; Mansfeld, F. Corrosion Science 1997, 39, 1061-1073. (36) Aramaki, K. Corrosion Science 2001, 43, 2201-2215. (37) Aramaki, K. Corrosion Science 2002, 44, 2621-2634. (38) Arnott, D. R.; Hinton, B. R. W.; Ryan, N. E. Corrosion Science 1989, 45, 12-18. (39) Lu, Y. C.; Ives, M. B. Corrosion Science 1993, 34, 1773-1785. (40) Lu, Y. C.; Ives, M. B. Corrosion Science 1995, 37, 145-155. (41) Lu, Z.; Hiskes, R.; DiCarolis, S. A.; Nel, A.; et.al. Journal of Crystal Growth 1995, 156, 227-234. (42) Liu, H.; Yang, J.; Liang, H.-H.; Zhuang, J.-H.; Zhou, W.-F. Journal of Power Sources 2001, 93, 230-233. (43) Crossland, A. C.; Thompson, G. E.; Skeldon, P.; Smith, G. C.; et.al. Corrosion Science 1998, 40, 871-885. (44) Fahrenholtz, W. G.; O'Keefe, M. J.; Zhou, H.; Grant, J. T. Surface and Coatings Technology 2002, 155, 208-213. (45) Davenport, A. J.; Isaacs, H. S.; Kendig, M. W. Corrosion Science 1991, 32, 653-663. (46) Forsyth, M.; Forsyth, C. M.; Wilsom, K.; Behrsing, T.; Deacon, G. B. Corrosion Science 2002, 44, 2651-2656. 50 (47) Liu, R.; Li, D. Y. Journal of Materials Science 2000, 35, 633-641. (48) Wang, K. L.; Zhang, Q. B.; Sun, M. L.; Zhu, Y. M. Surface and Coatings Technology 1997, 96, 267-271. (49) Al-Robaee, M. S.; Rao, K. N.; Mohan, S. Journal of Applied Physics 1992, 71, 2380- 2386. (50) Elidrissi, B.; Addou, M.; Regragui, M.; Monty, C.; Bougrine, A.; Kachouane, A. Thin Solid Films 2000, 379, 23-27. (51) Krishna, M. G.; Hartridge, A.; Bhattacharya, A. K. Materials Science Engineering B 1998, 55, 14-20. (52) Özer, N. Solar Energy Materials & Solar Cells 2001, 68, 391-400. (53) Monk, P. M. S.; Mortimer, R. J.; Rosseinsky, D. R. Electrochromism: Fundamentals and Applications; VCH: New York, 1995. (54) Camino, D.; Deroo, D.; Salardenne, J.; Treuil, N. Solar Energy Materials & Solar Cells 1995, 39, 349-366. (55) Ghodsi, F. E.; Tepehan, F. Z.; Tepehan, G. G. Solar Energy Materials & Solar Cells 2001, 68, 355-364. (56) Varsano, F.; Decker, F.; Masetti, E.; Cardellini, F.; Licciulli, A. Electrochimica Acta 1999, 44, 3149-3156. (57) Ovel, Z. C.; Ovel, B. J. Journal of Materials Science 1995, 30, 2284-2290. (58) Azen, A.; Kullman, L.; Ragan, D. D.; Granqvist, C. G. Solar Energy Materials & Solar Cells 1998, 54, 85-91. (59) Masetti, E.; Varsano, F.; Decker, F.; Krasilnikova, A. Electrochimica Acta 2001, 46, 2085-2090. 51 (60) Kang, J.; Liu, X.; Lian, G.; Zhang, Z.; et.al. Microelectronic Engineering 2001, 56, 191- 194. (61) Wu, Z.; Huang, D.; Yang, X. Vacuum 1998, 51, 397-401. (62) Inoue, T.; Yamamoto, Y.; Koyama, S.; Suzuki, S.; Ueda, Y. Applied Physics Letters 1990, 56, 1332-1333. (63) Li, M.; Wang, Z.; Fan, S.; Zhao, Q.; Xiong, G. Nuclear Instruments and Methods in Physics Research B 1998, 135, 535-539. (64) Hirschauer, B.; Göthelid, M.; Janin, E.; Lu, H.; Karlsson, U. O. Applied Surface Science 1999, 148, 164-170. (65) Wang, S.; Wang, W.; Liu, Q.; Zhang, M.; Qian, Y. Solid State Ionics 2000, 133, 211-215. (66) Chikyow, T.; Bedair, S. M.; Tye, L.; El-Masry, N. A. Applied Physics Letters 1994, 65, 1030-1032. (67) Inoue, T.; Yamamoto, Y.; Satoh, M. Thin Solid Films 1999, 343/344, 594-597. (68) Mölsä, H.; Niionstö, L. Materials Research Society Symposium Proceedings 1994, 335, 341-350. (69) Wu, X. D.; Dye, R. C.; Muenshasusen, R. E.; Foltyn, S. R.; et.al. Applied Physics Letters 1991, 58, 2165-2167. (70) Luo, L.; Wu, X. D.; Dye, R. C.; Muenchausen, R. E.; Foltyn, S. R.; et.al. Applied Physics Letters 1991, 59, 2043-2045. (71) Denhoff, M. W.; McCaffrey, P. Journal of Applied Physics 1991, 70, 3986-3988. (72) Qiao, J.; Yang, C. Y. Materials Science and Engineering R 1995, 14, 157-202. (73) Maul, M.; Schulte, B.; Häussler, P.; Frank, G.; Seinborn, T.; Fues, H.; Adrian, H. Journal of Applied Physics 1993, 74, 2942-2944. 52 (74) Develos, K. D.; Yamasaki, H.; Sawa, A.; Nakagawa, Y. Physica C 2001, 361, 121-129. (75) Develos, K. D.; Yamasaki, H.; Nakagawa, Y.; Sawa, A. Physica C 2002, 372-376, 642- 648. (76) Tian, Y. J.; Linzen, S.; Schmidl, F.; Matthes, A.; et.al. Thin Solid Films 1999, 338, 224- 230. (77) Aguiar, R.; Sánchez, F.; Ferrter, C.; Varela, M. Thin Solid Films 1997, 306, 74-77. (78) Harnoic, C.; Hervieu, M.; Monot-Laffez, I.; Desgardin, G. Journal of European Ceramic Society 2001, 21, 1909-1912. (79) Gerbinger, J.; Lohwasser, W.; Lampe, U.; Meixner, H. Sensors and Actuators B 1995, 26-27, 93-96. (80) Izu, N.; Shin, W.; Murayama, N.; Kanzaki, S. Sensors and Actuators B 2002, 87, 95-98. (81) Bene, R.; Perczel, I. V.; Réti, F.; Meyer, F. A.; Fleiseher, M.; Meixner, H. Sensors and Actuators B 2000, 71, 36-41. (82) Khodadadi, A.; Mohajerzadeh, S. S.; Mortazavi, Y.; Miri, A. M. Sensors and Actuators B 2001, 80, 267-271. (83) Stefanik, T. S.; Tuller, H. L. Journal of European Ceramic Society 2001, 21, 1967-1970. (84) Hamagami, J.; Inda, Y.; Umegaki, T.; Yamashita, K. Solid State Ionics 1998, 113-115, 235-239. (85) Shuk, P.; Ramanujachary, K. V.; Greenblatt, M. Solid State Ionics 1996, 85, 257-263. (86) Shuk, P.; Ramaswamy, A. V.; Greenblatt, M. Solid State Ionics 1996, 86-88, 1115-1120. (87) Yang, X.; Rahaman, M. N. Journal of European Ceramic Society 1997, 17, 525-535. (88) Roy, R. Science 1987, 238, 1664-1669. (89) Czerwinski, F.; Szpunar Thin Solid Films 1996, 289, 213-219. 53 (90) Francids, L. F. Sol-Gel Methods for Oxide Coatings; Marcel Dekker,Inc: New York, Basel, 1999. (91) Livage, J.; Ganguli, D. Solar Energy Materials & Solar Cells 2001, 68, 365-381. (92) Laubversheimer, J.; Ritzhaupt-Kleissl, H.-J.; Haubelt, J.; Emit, G. Journal of European Ceramic Society 1998, 18, 255-260. (93) Orel, Z. C. The Internet Journal of Vibrational Spectroscopy 1999, 3, http://www.ijvs.com/volume3/edition4/section3.htm. (94) Majumder, S. B.; Mohapatra, Y. N.; Agrawal, D. C. Journal of Materials Science 1997, 32, 2141-2150. (95) Özer, N.; Cronin, J. P.; Akyuiz, S. SPIE 1999, 3788, 103-110. (96) Deroo, D.; Petit, J.-P.; Kéoman, D. Solar Energy Materials & Solar Cells 1995, 36, 397- 408. (97) Tulun, T.; Akinci, H. Ceramics International 1991, 23, 141-148. (98) Ying, J. Y.; Tschöpe, A. The Chemical Engineering Journal 1996, 64, 225-237. (99) Aita, C. R. In Intermetallic and Ceramic Coatings; Sudarshan, T. S., Ed.; Marcel Dekker. Inc.: New York, 1999, pp 181-231. (100) Jacobsen, S. N. In Lingköping Studies in Science and Technology; Lingköping University,: Lingköping, Sweden, 1998. (101) Chin, C. C.; Lin, R. J.; Yu, Y. C.; Wang, C. W.; et.al. Physica C 1996, 260, 86-92. (102) Chin, C. C.; Lin, R. J.; Yu, Y. C.; Wang, C. W.; et.al. IEEE Transactions on Applied Superconductivity 1997, 7, 1403-1406. (103) Savvides, N.; Thorley, A.; Gnanarajan, S.; Katssros, A. Thin Solid Films 2001, 338, 177- 182. 54 (104) Gnanarajan, S.; Savvides, N. Thin Solid Films 1999, 350, 124-129. (105) Cotell, C. M.; Hirvonen, J. K. Surface and Coatings Technology 1996, 81, 118-124. (106) Huang, D.; Qin, F.; Yao, Z.; Ren, Z.; et.al. Applied Physics Letters 1995, 67, 3725-3726. (107) http://ap.polyu.edu.hk/pld/PLD.html (108) Amirhaghi, S.; Li, Y. H.; Kilner, J. A.; Boyd, I. A. Materials Science and Engineering B 1995, 34, 192-198. (109) Shi, D. Q.; Ionescu, M.; McKinnon, J.; Dou, S. X. Physica C 2001, 356, 304-310. (110) Gates, S. M. Chemical Reviews 1996, 96, 1519-1532. (111) Kapat, J. S.; Kumar, A. In Intermetallic and Ceramic Coatings; Sudarshan, T. S., Ed.; Marcel Dekker. Inc.: New York, 1999, pp 441-484. (112) Choy, K. L. Progress in Materials Science 2003, 48, 57-170. (113) McAleese, J.; Plakatouras, J. C.; Steele, B. C. H. Thin Solid Films 1996, 280, 152-159. (114) Pan, M.; Meng, G. Y.; Xin, H. W.; Chen, C. S.; et.al. Thin Solid Films 1998, 324, 89-93. (115) Wang, A.; Belot, J. A.; Tobin, J. M.; Markworth, P. R. Physica C 1999, 320, 154-160. (116) Pollard, K. D.; Jenkins, H. A.; Puddephatt, R. J. Chemistry of Materials 2000, 12, 701- 710. (117) Nigro, R. L.; Malandrino, G.; Fragalà, I. L. Chemistry of Materials 2001, 13, 4402-4404. (118) Carter, W. B.; Book, G. W.; Pollery, T. A.; Stollberg, D. W.; Hampikian, J. M. Thin Solid Films 1999, 347, 25-30. (119) Wang, S.; Wang, W.; Zuo, J.; Qian, Y. Materials Chemistry and Physics 2001, 68, 246- 248. (120) Wang, L. P.; Tang, B. Y.; Huang, N.; Tian, X. B.; Chu, P. K. Materials Science and Engineering A 2001, 308, 176-179. 55 (121) Ramírez-Duverger, A.; Ruiz-Salvador, A.; Hernández-Sáchez, M. P.; García-Sánchez, M. F.; Rodríguez-Gattorno, G. Solid State Ionics 1997, 96, 89-93. (122) Law, W. L.; Hill, R. H. Materials Research Bulletin 1998, 33, 69-80. (123) Huang, W.; Shuk, P.; Creenblatt, M. Solid State Ionics 1997, 100, 23-27. (124) Ho evar, S.; Kra ovec, U. O.; Antonino, B. O.; et.al. Applied Catalysis B: Environmental 2000, 28, 113-125. (125) Luo, M.; Lu, G.; Zheng, X. Journal of Materials Science Letters 1998, 17, 1553-1557. (126) Djuricic, B.; Pickering, S. Journal of European Ceramic Society 1999, 19, 1925-1934. (127) Chen, P. L.; Chen, L. W. Journal of the American Ceramic Society 1993, 76, 1577-1583. (128) Hirano, M.; Fukuda, Y.; Iwata, H.; Horta, Y.; Inagaki, M. Journal of the American Ceramic Society 2000, 83, 1287-1289. (129) Wang, C.; Qian, Y.; Xie, Y.; Wang, C.; et.al. Materials Science and Engineering B 1996, 39, 160-162. (130) Konishi, Y.; Murai, T.; Asai, S. Industrial & Engineering Chemsitry Research 1997, 36, 2641-2645. (131) Wendlandt, W. W. Analytical Chemistry 1958, 30, 58-61. (132) Glasner, A.; Steinberg, M. Journal of Inorganic and Nuclear Chemistry 1961, 16, 279- 287. (133) Moosath, S. S.; Abraham, J.; Swaminathan, T. V. Zeitschrift Fuer Anorganische und Allgemeine Chemie 1963, 324, 103-105. (134) Hussein, G. A. M. Journal of Analytical and Applied Pyrolysis 1996, 37, 111-149. (135) Jurado, J. R. Journal of Materials Science 2001, 36, 1133-1139. 56 (136) Zhang, J.; Ju, X.; Wu, Z. Y.; Liu, T.; Du, T. D.; Xie, Y. N. Chemistry of Materials 2001, 13, 4192-4297. (137) Audebrand, N.; Auffrédic, J.-P.; Louër, D. Chemistry of Materials 2000, 12, 1791-1799. (138) Mamontov, E.; Egami, T. Journal of Physics and Chemistry of Solids 2000, 61, 1345- 1356. (139) Binet, C.; Daturi, M.; Lavalley, J.-C. Catalysis Today 1999, 50, 207-225. (140) Bassini, L.; Marchionna, M.; Aragno, A. The Journal of Physical Chemistry B 1992, 96, 9431-9441. (141) Long, R. Q.; Wan, H. L. Journal of the Chemical Society Faraday Transactions 1997, 93, 355-358. (142) Twu, J.; Chuang, C. J.; Chang, K. I.; Yang, C. H.; Chen, K. H. Applied Catalysis B Environment 1997, 12, 309-324. (143) Cunningham, J.; Cullinane, D.; Sanz, J.; Rojo, J. M.; Soria, X. A.; Fierro, J. L. G. Journal of the Chemical Society Faraday Transactions 1992, 88, 3233-3240. (144) Laachir, A.; Perrichon, V.; Badri, A.; Lamotte, J.; Catherine, E.; Lavalley, J. C.; El Fallah, J. H.; Hilaire, L. Journal of the Chemical Society Faraday Transactions 1991, 87, 1601-1609. (145) Nachimuthu, P.; Shih, W.-C.; Liu, R.-S.; Jang, L.-Y.; Chen, J.-M. Journal of Solid State Chemistry 2000, 149, 408-413. (146) Balasubramanian, M.; Melendres, C.; Mansour, A. N. Thin Solid Films 1999, 347, 178- 183. (147) Mullins, D. R.; Overbury, S. H.; Huntley, D. R. Surface Science 1998, 409, 307-319. 57 (148) Wu, Z.; Huang, D.; Yang, X.; Wang, J.; Qin, F.; Zhang, J.; Yang, Z. Vacuum 1998, 49, 133-137. (149) Berner, U.; Schierbaum, K. Thin Solid Films 2001, 400, 46-49. (150) Nörenberg, H.; Briggs, G. A. D. Surface Science 1999, 433-435, 127-130. (151) Francisco, M. S. P.; Mastelaro, V. P.; Nascente, P. A. P.; Florebtino, A. O. The Journal of Physical Chemistry B 2001, 105, 10515-10522. (152) Zhou, Y.; Phillips, R. J.; Switzer, J. A. Journal of the American Ceramic Society 1995, 78, 981-985. (153) Kepinski, L.; Wolcyrz, M. Catalysis Letters 1992, 15, 329-337. (154) Zhou, Y.; Rahaman, M. N. Acta Materialia 1997, 45, 3635-3639. (155) Tschöpe, A.; Ying, J. Y.; Tuller, H. L. Sensors and Actuators B 1996, 31, 111-114. (156) http://www.nanocor.com/nano_struct.asp (157) Moore, D. M.; Reynolds, R. C. J. X-ray Diffraction and Identification and Analysis of Clay Minerals; 2nd ed.; Oxford University Press, Inc.: New York, 1997. (158) Tan, K. H. In Principles of Soil Chemistry; Marcel Dekker, Inc.: New York, 1998, pp 177-259. (159) Horch, R. A.; Golden, T. D.; D'Sousa, N. A.; Riester, L. Chemistry of Materials 2002, 14, 3531-3538. (160) Pinnavaia, T. J. Science 1983, 220, 365-371. (161) Lewis, R.; Kuroda, H. Solid State Ionics 1989, 32/33, 373-377. (162) Ranade, A.; D'Sousa, N. A.; Gnade, B. Polymer 2002, 43, 3759-3766. (163) Butzloff, P. R. In Materials Science; University of North Texas: Denton, 2000, p 97. (164) Komarnenl, S. Journal of Materials Chemsitry 1992, 2, 1219-1230. 58 (165) Poorteman, M.; Descamps, P.; Cambier, F.; Poulet, A.; Descamps, J. C. Journal of European Ceramic society 1999, 19, 2375-2379. (166) Wildan, M.; Edrees, H. J.; Hendry, A. Materials Chemistry and Physics 2002, 75, 276- 283. (167) Xu, C.; Ai, X. International Journal of Refractory Metals & Hard Materials 2000, 19, 85-88. (168) Jeong, Y.-K.; Choa, Y.-H.; Niihara, K. Journal of the American Ceramic Society 1998, 81, 1453-1260. (169) Nihara, K. Journal of the Ceramic Society of Japan 1991, 99, 974-981. (170) Awaji, H.; Choi, S.-M.; Yagi, E. Mechanics of Materials 2002, 34, 411-422. (171) Tan, H.; Yang, W. Mechanics of Materials 1998, 30, 111-123. (172) Ohji, T. Ceramic Engineering and Science Proceedings. 1997, 18, 353-360. (173) Meguid, S. A. Journal of Materials Processing Technology 1996, 56, 978-989. (174) Sauma, V. E.; Natekar, D.; Sbaizero, O. Materials Science and Engineering 2002, 323, 129-137. (175) Rühle, M.; Evans, A. G.; McMeeking, R. M.; Charlambides, P. G. Acta Metallurgica 1987, 35, 2701-2710. (176) Hutchioson, J. W. Acta Metallurgica 1987, 35, 1605-1619. (177) Kotoul, M.; Profant, T. Mechanics of Materials 2000, 32, 203-234. (178) Evans, A. G.; Heure, A. H. Journal of the American Ceramic Society 1980, 63, 241-248. (179) Green, D. J.; Hannink, R. H. J.; Swain, M. V. Transformation Toughening of Ceramics; CRC press, Inc.: Florida, USA, 1989. 59 (180) Casellas, D.; Ràfols, I.; Llanes, L.; Anglada, M. International Journal of Refractory Metals & Hard Materials 1999, 17, 11-20. (181) Zhou, J.-J.; Zhang, H.-T.; Chen, L.-X.; Qi, P. Journal of Materials Science Letters 1994, 13, 846-848. (182) Reimanis, I. E. Materials Science and Engineering A 1997, 237, 159-167. (183) Thomsen, N. B.; Karihaloo, B. L. Journal of the Mechanics and Physics of Solids 1996, 44, 207-231. (184) Faber, K. T.; Evans, A. G. Acta Metallurgica 1983, 31, 565-576. (185) Yu, Z. B.; Thompson, D. P.; Bhatti, A. R. Composites A 2002, 33, 621-629. (186) Kaya, C.; Gu, X.; Al-Dawery, I.; Butler, E. G. Science and Technology of Advanced Materials 2002, 3, 35-44. (187) Bansal, N. P.; Setlock, J. A. Composites A 2001, 32, 1021-1029. (188) Chiang, Y.-C. Composites Science and Technology 2001, 61, 1743-1756. (189) Suemasu, H.; Kondo, A.; Itatani, K.; Nozue, A. Composites Science and Technology 2001, 61, 281-288. (190) Song, G. M.; Zhou, Y.; Sun, Y. Ceramics International 1999, 25, 257-260. (191) Ahn, B. K.; Curtin , W. A.; Parthasarathy, T. A.; Dutton, R. E. Composites Science and Technology 1998, 58, 1775-1784. (192) Lee, T.-C.; Deng, J. Ceramics International 2001, 27, 755-760. (193) Ye, F.; Lei, T. C.; Zhao, Y. C.; Zhou, Y.; Yang, J. M.; Zhang, L. T. Ceramics International 1998, 24, 181-188. (194) Chen, S.; Ye, F.; Zhou, Y. Ceramics International 2000, 28, 51-58. 60 (195) Gao, L.; Yang, H.; Yuan, R.; Huang, P.; et.al. Journal of Materials Processing Technology 2001, 115, 298-301. (196) Günay, V.; Hampshire, S. Journal of Materials Processing Technology 1995, 54, 348- 354. (197) Favier, V.; Chanzy, H.; Cavaillé, J. Y. Macromolecules 1995, 28, 6365-6367. (198) Paillet, M.; Dufresne, A. Macromolecules 2001, 34, 6527-6530. (199) Xu, H. H. K.; Quinn, J. B.; Smith, D. T.; Antonucci, J. M.; et.al. Biomaterials 2002, 23, 735-742. (200) Kaya, C.; Kaya, F.; Marsoglu, M. Journal of Materials Science Letters 1997, 892-896. (201) Janssen, R.; Heussner, K.-H. Advanced Ceramics 1995, 23, 241-245. (202) Gum, J. C.; Ye, F.; Zhou, Y.; Lei, T. C. Ceramics International 2000, 26, 855-863. (203) Grigge, J.; University of Florida: Gainsville, 1998. (204) Chou, Y.-S.; Green , D. Journal of the American Ceramic Society 1993, 76, 1452-1458. (205) Choa, Y.-H.; Nakahira, A.; Niihar, K. Journal of Materials Science 2000, 35, 3143-3149. (206) Kaya, C.; Kaya, F.; Trusty, P. A.; Boccaccini, A. R.; Marsoglu, M. Ceramics International 1999, 25, 359-366. (207) Li, S.-B.; Zhang, B.-S.; Wen, G.-W.; Xie, J.-X. Materials Letters 2002, 57, 1445-1452. (208) Chou, Y.-S.; Green, D. J. Journal of the American Ceramic Society 1993, 76, 1985-1992. (209) Alexandre, M.; Dubois, P. Materials Science and Engineering 2001, 28, 1-63. (210) LeBaron, P. C.; Wang, Z.; Pinnavaria, T. J. Applied Clay Science 1999, 15, 11-29. (211) Fukushima, Y.; Inagaki, S. Journal of Inclusion Phenomena 1987, 5, 473-482. (212) Kim, T. H.; Jang, L. W.; Lee, D. C.; Choi, H. J.; Jhon, M. S. Macromolecular Rapid Communications 2002, 23, 191-195. 61 (213) Hernando, M. J.; Pesquera, C.; Blanco, C.; González, F. Chemistry of Materials 2001, 13, 2154-2159. (214) Kiš, E. E.; Ranogajec, J. G.; Marinkovic-Neducin, R. P.; Vulic, T. J.; et.al. APTEFF 2001, 32, 71-78. (215) Szücs, A.; Berger, F.; Dekány, I. Colloids and Surface A: Physicochemical and Engineering Aspects 2000, 174, 387-402. (216) Dekány, I.; Turi, L.; Király, Z. Applied Clay Science 1999, 15, 221-239. (217) Mamedov, A.; Ostrander, J.; Aliev, F.; Kotov, N. A. Langmuir 2000, 16, 3941-3949. (218) Han, Y.-S.; Matsunoto, H.; Yamanaka, S. Chemistry of Materials 1997, 9, 2013-2018. (219) Ooka, C.; Akita, S.; Ohashi, Y.; Horiuchi, T.; et.al. Journal of Materials Chemistry 1999, 9, 2943-2952. (220) Yamanaka, S.; Brindley, G. W. Clays and Clays Minerals 1979, 27, 119-124. (221) Brindley, G. W.; Sempels, R. E. Clay Minerals 1977, 12, 229-237. (222) Rightor, E. G.; Tzou, M. S.; Pinnavaia, T. J. Journal of Catalysis 1991, 130, 29-40. (223) Pinnavaia, T. J.; Tzou, M. S.; Landau, S. D. Journal of the American Chemical Society 1985, 107, 4783-4785. (224) Mishra, T.; Parida, K. Applied Catalysis A; General, 1998, 174, 91-98. (225) Garcia-Rodriguez, A.; del Rey-Bueno, F.; del Rey-Perez-Caballero, F. J.; Ureña-Amate, M. D.; Mata-Arjiona, A. Materials Chemistry and Physics 1995, 39, 269-277. (226) Szostak, R.; Ingram, C. Studies in Surface Science and Catalysis 1995, 94, 13-38. (227) Drozdov, V. A.; Doronin, V. P. S.; Porokina, T. P.; Gulyaeva, T. I.; Duplyakin, V. K. Kinetics and Catalysis 2001, 42, 117-125. 62 (228) Ohtsuka, K.; Koga, J.; Tsunoda, M.; Suda, M.; Ono, M. Journal of the American Ceramic Society 1990, 73, 1719-1725. (229) Giannelis, E. Chemistry of Materials 1998, 2, 627-629. (230) Sivakumar, S.; Damodaran, A. D.; Warrier, K. G. K. Ceramics International 1995, 21, 85-88. (231) Katdare, S. P.; Ramaswamy, V.; Ramaswamy, A. V. Catalysis Today 1999, 49, 313-320. (232) Pospíšil, M.; Capkivá, P.; Weiss, Z.; Malác, Z.; Šimoník, J. Journal of Colloid and Interface Science 2002, 25, 126-132. (233) Carrado, K. A.; Xu, L.; Gregory, D. M.; Song, K.; Seifert, S.; Botto, R. E. Chemistry of Materials 2000, 12, 3052-3059. (234) Madejová, J.; Arvaiová, B.; Komadel, P. Spectrochima Acta A 1999, 55, 2467-2476. (235) Frost, R. L.; Rintoul, L. Applied Clay Science 1996, 11, 171-183. (236) Occelli, M. L.; Auroux, A.; Ray, G. J. Microporous and Mesoporous Materials 2000, 39, 43-56. (237) Paama, L.; Pitkãnen, I.; Perãmãki, P. Talanta 2000, 51, 349-357. 63 CHAPTER 2 ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION METHODS 2.1 Electrochemical Synthesis and Other Electrochemical Techniques 2.1.1. Electrochemical Synthesis1-6 Electrochemical synthesis is another synthesis technique that can be used in the production of ceramic materials. The product can be both ceramic powder and ceramic thin films on a substrate. An electrochemical synthesis is a synthesizing process in which oxidation or reduction is accomplished through electron transfer between two or more electrodes immersed in an electrolyte. Synthesis occurs at the interface of an electrode and the electrolyte. Electron transfer can be achieved via two ways: through an external power supply or by an electroless process in which the reagent in the solution functions as the electron source.4 The power-provided electrochemical process has been studied in our research and will be discussed in this chapter. 2.1.1.1 Electrochemical Synthesis Modes1,4 There are three operation modes for electrochemical synthesis: potentiostatic, galvanostatic and pulse period deposition. Current, potential, and charge are the most basic and important parameters that dominate the electrochemical synthesis. Two electrodes are involved in the setup of the typical galvanostatic electrosynthesis, the working electrode where the reactant is converted to the product and the counter electrode (or coined as accessory electrode) for the counter reaction occurrence. A constant current is applied to the electrochemical cell and the potential varies correspondingly during the reaction. Conversely, a constant potential is applied to the working electrode for the reaction to take place in potentiostatic electrochemical synthesis mode. A standard three-electrode assembly 64 is employed in the setup. A third electrode, the reference electrode must be provided, which is used as the standard electrode. The potential applied to the working electrode is with respect with the reference electrode. Cyclic voltammetry is usually employed to find the best potential window for the reaction. For aqueous solution, the limits for the potential window are the potentials for the evolution of oxygen and hydrogen respectively at the anode and cathode. Based on potentiostatic mode synthesis, a pulse period deposition method has been developed. In this method, the synthesis can be accomplished by repeating the cycle of deposition and stripping. During deposition, a desired potential is applied to the working electrode, while an opposite potential is applied to the working electrode for stripping. The stripping step is much shorter than the deposition. It is well known that there are four types of reactions that may be involved in electrodeposition: electron transfer, mass transport, chemical reaction and crystallization (Figure 2.1). For most electrochemical reaction cases, the electron transfer is the rate-determining step and in this case, the reaction rate is proportional to the current density. Therefore, delicate tuning of the current can properly control the reaction rate. In galvanostatic deposition process, the applied current is constant, and the reaction rate is constant. The resultant deposit will demonstrate the homogeneous morphology in surface and good adhesion to the substrate. However, the potential drifts as the reaction proceeds, and usually to a higher value with the reactant activity decreasing, which will lead to impure deposition. While in the potentiostatic deposition, the constant applied potential leads to a homogeneous, single-phased coating. The problem is the deposition rate varies, which affects the adhesion of the thin film onto the surface. Usually the reaction rate becomes slower with time for two reasons: concentration of reactant in solution decreases resulting in lower reactant diffusion; thicker insulating film formed on the 65 electrode-which lowers the conductivity of the electrode.1 Different from the previous two electrosynthesis, the applied potential or current in pulse period deposition alternates anodically and cathodically. The periods that the desired potential or current is applied to synthesize the product is called “on-times” and the lengths of time for the opposite process is represented by “off-times”.4 Usually the on-time is scheduled as much longer than the off-time. For instance, if the pulse deposition mode is used for synthesis of metal oxide at high oxidation state from solution with lower oxidation state metal ions, the working electrode is alternately polarized anodically for deposition and cathodically for deposit stripping. The cathodic stripping process, though assigned with an ultra-short time within each cycle, prevents the further nucleation and growth of the deposit. This is a novel but effective means to control the grain size of electrosynthesized materials. This deposition mode has been applied in the electrochemical fabrication of nanometer scale thin films and powders. 2.1.1.2 Electrochemical Synthesis Types1,2,4 Basically, there are two types of electrochemical synthesis: cathodic and anodic synthesis. For cathodic deposition, the reactant accepts electrons and this reduction process converts the reactant to the desired deposit on a working electrode. This type of electrochemical synthesis is most often used in the production of metals or lower valence of compounds. And in anodic electrochemical synthesis, the electrons of reactant species in solution will be transferred away to the counter species, which leads to the oxidation of reactant to higher valence deposit on the working electrode. 2.1.1.3 Electrochemical Synthesis Conditions1,4,5 In addition to applied current, potential, deposition type and mode, some other factors also play significant roles in the electrochemical synthesis. 66 (1) Electrolytes Electrolytes function as solvents, ligands, or buffers for the electrochemical solution. Electrolytes also adjust solution pH, balance the reaction system and enhance or weaken the reaction activity. Therefore, the electrolytes are critical in electrochemical reactions. (2) Substrates or Electrodes Different substrates demonstrate different redox properties for a particular electrolyte system. A linear or cyclic sweep voltammetry is necessary to determine the potential condition for synthesis or deposition before each deposition onto a particular substrate. The selection of substrate has a close relationship with the deposition mechanism for the electrochemical synthesis. Two main types of mechanisms are present for electrodeposition: layer growth and nucleation-coalescence growth. The latter is more popular in explanation of electrodeposition. The electrodeposition on different substrates involves different levels of competition of the mechanisms at different stages. Deposition on single crystal substrates tends toward epitaxial growth of the deposits at the beginning of the deposition due to uniformity of the substrate. Electrodeposition on polycrystalline and randomly oriented substrates can be explained by growth rate competition between various crystallinite orientations. Generally this will result in random textured deposits. When the growth rate of one particular orientation is greater than all other orientations in the nucleation stage, it is possible to form oriented films. For film growth on amorphous substrates, the orientation and texture are determined in the coalescence or post-coalescence stage. Structure matching of substrate and product is important for producing uniform films without cracking. (3) Temperature 67 Temperature is another factor critical for the structure and morphology of resultant films or powders in electrochemical synthesis. Generally, like chemical reaction, higher temperature results in larger synthesis current when the potential is constant. Different structure or different preferred orientation at different temperatures is another aspect to consider. The temperature of electrochemical synthesis is generally controlled with a water circulator and dry-ice bath for temperature lower than room temperatures. (4) Setup Electrochemical synthesis can take place in undivided cell or divided compartments. Undivided cell refers to the regular electrochemical synthesis setup in which all three or two required electrodes are situated in the same cell. Most electrochemical synthesis reactions adopt this setup. However, in some electrochemical synthesis, the reaction on the counter electrode will interfere with the reaction at the working electrode, such as evolvement of hydrogen or oxygen, resulting in bubbles around the working electrode. Turbulence makes the deposition film heterogeneous and of low quantity. A divided-compartment cell can be applied in this situation to protect the film deposition from the interference. In the divided cell, the working and reference electrode are located in one compartment and the counter electrode will be assigned in another compartment, between which a glass frit is employed to do the separation. However, there will be uncompensated resistance involved with this setup. It is suggested to use the undivided compartment-setup when possible. 2.1.1.4. Features about Electrochemical Synthesis1,7 Compared with other synthetic methods, electrochemical synthesis has some distinct features and advantages that make this method as a popular approach for the fabrication of ceramic materials. 68 (1) Synthesis at the Interface Generally chemical synthesis occurs in one phase or two liquid phases. However, for electrochemical system, the synthesis takes place at the interface of the solid-liquid (electrode- electrolyte). The interface has distinct properties different from the bulk solid, so that at the interface, an electric double layer exists, which has a high potential gradient of 105 V/cm (Figure 2.1). It is this specific characteristic of the interface that promotes electrochemical reactions at the electrode resulting in products particular to the synthesis. Since the solid-liquid interface is only a thin layer, it facilitates the conformal coatings or deposition of thin films on electrodes with different shapes. (1) Production Parameters Electrochemical synthesis can be either oxidation or reduction reaction, and the oxidation or reduction power can be continuously varied and suitably selected to obtain different products. By varying the synthesis duration, the film thickness can be controlled as well. Electrochemical synthesis can also give different film compositions by changing the composition of the bath. It is possible for electrochemical synthesis to obtain different orientation or structure for the same material at the different conditions. Another advantage of electrochemical synthesis is the formation of multilayers. These are all things that common chemical synthesis cannot accomplish as easily as electrochemical synthesis. 69 Chemical Mass Adsorption Reactions transfer O’ O bulk Os O a’ Desorption Metal Electron transfer Desorption R a’ Chemical Mass Reactions transfer R’ Rs R bulk Adsorption Metal Solution IHP OHP Solvent molecule Adsorbed ions Isolated ions Potential Profile lin e Figure 2.1 Electrode-solution interface with potential profile and electrochemical reaction (2) Instrument Availability Electrochemical synthesis can be set up and operated at normal temperature and normal pressure to obtain products usually obtained by special conditions such as high temperature, high pressure, ultra-high vacuum or other synthesis methods. The instrument used for electrochemical 70 synthesis is a potentiostat/galvanostat, which is easy to obtain and the price of which is low. The availability of the instrument is another asset for this technique. However, for this technique, there are also some features that pose trouble for the synthesis reactions. (1) Delicate Control of Conditions1,7 Electrochemical synthesis can lead to various products. It is possible for the electrochemical synthesis to produce impurities in the composition of the resultant product if the experimental condition varies. For current-controlled synthesis, the applied potential varies as the reaction proceeds, which will result in a mixture of multiple compositions. Therefore, delicate control of the synthesis condition is important for the electrochemical synthesis. To make uniform films by electrodeposition also requires careful control of the cell geometry conditions, since improper set-up can result in high density of deposition at edges or corners instead of in the middle.7 (2) Conductive Substrate A conductive electrode is required in electrochemical synthesis. This requirement confines the electrochemical synthesis to metals or some semi-conductive materials as the electrodes or substrates, which limits the application of the electrochemical synthesis with non- conductive substrates. 2.1.1.5 Electrochemical Synthesis of Cerium Oxides Electrochemical synthesis of ceramic materials especially oxides has received considerable attention in recent years because the listed advantages of the electrochemical synthesis outweighs the limitations of this approach. There are basically two types of 71 electrochemical methods that have been employed in the synthesis of cerium oxide: cathodic base generation synthesis and anodic oxidation deposition.1 2.1.1.5.1 Cathodic Base Generation Electrochemical Synthesis Cathodic deposition is a method that is used to plate metal coatings at the cathode.4 However, with selection of the pH and anion used in the plating solution, it is possible that the deposition will result in the hydroxide and further loss of water, which then leads to the formation of the oxide. Cathodic deposition used to produce the oxide and hydroxide has also been coined as the base generation electrochemical method due to the mechanism, where base is produced at the vicinity of the working electrode. Base generation deposition was first proposed by Switzer to make ceramic oxide films and powders8-10 and later proved to be an attractive 11,12 13 14 15 method for producing powders such as ZrO2, TiO2, Al2O3, and Nb2O5. In the base generation synthesis, the reactions to generate base and increase pH at the cathode site can involve the electrolysis of water (Equation 2.1), the reduction of dissolved oxygen at electrode (Equation 2.2), the reduction of the anions in electrolytes (Equation 2.3) and the reaction consuming H+ ions (Equation 2.4).1,16 − − 2H 2O + 2e → H 2 + 2OH (Equation 2.1) − − O2 + 2H 2O + 4e → 4OH (Equation 2.2) − − − − NO3 + H 2O + 2e → NO2 + 2OH (Equation 2.3) + − 2H + 2e → H 2 (Equation 2.4) The above reactions either decrease the concentration of hydrogen ions or increase the concentration of hydroxide ions. And since all these reactions occur at the electrode site, it causes the increase of local pH at the electrode surface. It is reported that in some particular anion-contained solutions, the potentials for the above reaction are more positive than the metal 72 formation from metal ions. Therefore, electron transfer reaction at the electrode generates base and the base immediately participates in the hydroxylation of metal ions or metal complex close to the working electrode (Figure 2.2), which usually results in the formation of metal hydroxide at the working electrode (Equation 2.5) instead of pure metal. The hydroxides then are dehydrated to form oxides. n+ − M + nOH → M (OH ) n (Equation 2.5) Electrode Molecules OH- Mn+ Figure 2.2 Cathodic base generation electrochemical synthesis In addition to hydroxide as the intermediate in electrosynthesizing ceramic oxide, peroxides are also used as precursors for oxide formation when water is used to generate base. Peroxocomplex is hydrolyzed at the cathode to form peroxide which will be dehydrated with (4-n)+ thermal treatment to form corresponding oxide. Peroxocomplex of titanium [Ti(O2)(OH)n-2] was used instead of titanium ions to cathodically deposit titania by Zhitomirsky et.al...17 And 16,18 later the same electrosynthesis method was adopted to form ZrO2, ZrTiO4, SnO2, and Ni2O5. 73 This mechanism has also been hypothesized for the successful formation of TiO2 and Ni2O5. It was pointed out that the basic properties of the cathodic region give rise to colloidal particles of (4-n)+ hydrated peroxide cationic ions (such as [Ti(O2)(OH)n-2] ) which will be accumulated in the vicinity of the cathode with the power of the electrical field. This explanation is of electrophoretic characteristic, which has been discussed by Zhitomirsky in detail.3 And the later coagulation of the colloidal particles forms the deposit on the electrode. The interactions and migration of the colloidal particles have been rationalized with DLVO theory.19 And the function of the additives during the deposition process and the influence of the additives in the deposits have been investigated in detail.16,17 It is revealed that the addition of hydrogen peroxide bears practical importance in the conversion of hydroxide deposits to hydrated oxides or peroxides and acceleration of oxide crystallization process. And since peroxide has lower basicity than hydroxide, it alleviates the possibility of adsorption of CO2 and the formation of carbonates. The cathodic electrosynthesis has been applied in the formation of the oxide composites, in which either hydroxide or peroxide precursors precipitates in the formation of the desired compounds.16,20 The films grown with cathodic base generation electrochemical synthesis can be oriented or polycrystalline, but due to the chemical reaction involved in the process, the formed films are often powdery and crack-like. Sintering has been used to condense the deposits from wet chemical methods and it is confirmed that the oxides produced with base generation approach a higher sintering activity compared with other chemical methods owing to the finer grain sizes obtained with this electrosynthesis.9 Alternated deposition and sintering method has been utilized to prevent the drying shrinkage crack during the film deposition process.3 74 The production of CeO2 powders and films with base generation electrosynthesis can date back to Switzer’s work in which the nitrate or chloride of cerium (III) was employed for synthesis and the hydroxide is the intermediate for oxide. The CeO2 powder exhibits nanocrystalline properties and higher sintering rates compared to CeO2 powder from other wet 9 chemical methods. The as-grown nano-structure CeO2 film demonstrates the crystallinity without further thermal treatment which is necessary for other wet methods.10 An in-depth study of the complete mechanism for CeO2 deposition using the base generation method was done by Aldykiewicz.21 Aldykiewicz proposed a mechanism involving oxygen to produce an oxidizing agent for Ce(III) to Ce(IV) formation. With an oxidant available in the system, cerium oxide film formation was accomplished through a four-electron or two-electron process to a hydroxide 21 intermediate. The final step was the precipitation of CeO2 onto the electrode surface. As previously mentioned, the reduction of oxygen can have two possible pathways: the first proceeds through a direct four-electron reduction: − − O2 + 2H 2O + 4e → 4OH (Equation 2.6) The second proceeds through a two-electron reduction: − − O2 + 2H 2O + 2e → H 2O2 + 2OH (Equation 2.7) And with oxidant available in system, cerium oxide film formation was accomplished in following two oxidation steps: For the four-electron process: 3+ − 2+ 4Ce + O2 + 4OH + 2H 2O → 4Ce(OH)2 (Equation 2.8) For the two-electron process: 3+ − 2+ 2Ce + 2OH + H 2O2 → 2Ce(OH)2 (Equation 2.9) 75 The final step is the precipitation of CeO2 with increasing pH in the vicinity of the electrode: 2+ − 2Ce(OH)2 + 4OH → 2CeO2 + 4H 2O (Equation 2.10) The mechanism was supported by rotation disk electrode experiments and XANES studies. Li studied the mechanism proposed by Aldykiewicz with in-situ atomic force microscopy technique (AFM) and concluded that a cerium hydroxide species is produced at the electrode surface with CeO2 forming nuclei out of this hydroxide “gel”. The rate-determining 22-24 step for this mechanism is then the nucleation and growth of the CeO2 crystals. Hydrogen peroxide plays a dominant role in two-electron reduction process in the above mechanism. Zhitomirsky used it as the additive in the cathodic electrodeposition of CeO2 and doped ceria films from aqueous and mixed alcohol-water solutions of cerium chloride or 19,25 nitrate. It is proposed that CeO2.nH2O or Ce(OH)3OOH deposit is formed via two-electron or four-electron reduction, owing to the participation of hydrogen peroxide in oxidation step. In addition, poly (diallyldimethylammonium chloride) as a binder was electrochemically intercalated into the deposit so that the resultant films exhibit better adherence and crack-proof properties.25 2.1.1.5.2. Anodic Oxidation Electrodeposition of Cerium Oxide Anodic deposition is the reaction where electron transfer induces the oxidation of the reactant to form product on the anode. There are two types of anodic depositions that can be used to form ceramic oxides1 (Figure 2.3). The lower oxidation state species in solution is anodically converted to higher oxidation state on the electrode in the conventional anodic deposition. The examples for this technique include the formation of hydrous iridium oxide films,26-29 manganese dioxide film,30 titanium 31 32 33 34 oxide films, cobalt oxide film, hydrated nickel oxide, PbO2/Co2O3 composites, and single 76 crystalline bismuth oxide films.35 In this technique, the metal ions in the deposition solution are different from the metal as the substrate. In addition, the anions and pH of solution are also adjusted so that the lower state of cations can exist stably in solution whereas the higher oxidation can be easily hydrolyzed to form the hydroxide or oxide. The possible reaction pathway can be represented as follows:1 M n+ → M (n+x)+ + xe− (Equation 2.11) M (n+x)+ + (n + x)OH − → M (OH ) → MO + (n + x) H O (Equation 2.12) n+x (n+x) / 2 2 2 Figure 2.3B illustrates the scenario at the working electrode of this type of anodic electrochemical syntheses. Usually, the solution needs to be basic at the vicinity of the working electrode to promote reaction 2.12, but sometimes, the lower oxidation state of the metal ions cannot be stabilized at the higher pH condition, in which case, the complexes replace the free ions in Equation 2.11 and Equation 2.12 such as the deposition of cobalt oxide reported by Casalla and Gatta.32 Practically, the mechanism of anodic deposition of cobalt oxide with cobalt-citrate complex in alkaline condition has been studied. The complex (Co (II) -citrate) is oxidized to the corresponding complex of the higher oxidation states (Co (III or IV)-citrate) via electron transfer reaction on the electrode (Equation 2.13). The deposited complex is hydrolyzed by the base at the vicinity of the electrode to form cobalt (IV) hydroxide or oxide (Equation 2.14). It is probable that oxygen evolves simultaneously with the formation of oxide and hydroxide deposit due to the reaction between the cobalt-hydroxide intermediate and hydroxyl ions accompanied with electron transfer at the electrode, but the competition of oxide deposition and oxygen evolution usually occurs at a high applied potential. The function of the complexing agent has been investigated as well, which indicates that the addition of complexing agent shifts the anodic 77 oxidation peak to a range more positive and the higher the concentration of complexing agent, the slower the deposition rate.32 M II − L → M III ,IV − L + e− (Equation 2.13) M IV − L + OH − → M IV (OH ) + L (Equation 2.14) A: M → M (n+x)+ + (n + x)e − B: M n+ → M (n+x)+ + xe − Electrode OH- Mn+ (n+x)+ M Figure 2.3 Anodic electrochemical synthesis The alternative anodic approach involves the oxidation of the metal electrode itself to form the metal oxide (Figure 2.3A), such as the oxidation of metallic zirconium to produce zirconium oxide films36 and the formation of iridium oxide films sacrifying iridium.27 This type 1 of anodic deposition can be applied in the electrosynthesis of composites (eg. BaWO4, SrWO4). Compared with cathodic base generation electrosynthesis, the films from anodic deposition exhibit better adherence to the substrate and less powdery properties. As-prepared ceramics possess homogeneity not only in the morphology but in composition as well. 78 Most of electrochemical synthesis of cerium oxide has been accomplished through the cathodic base generation method and there has been little study of the anodic deposition of cerium oxide films. Anodic oxidation has been applied to Ce-Al alloys to form anodic oxide film which was used to prevent the corrosion of alloys.37 Another instance is in Balasubramanian and co-workers' studies on electrochemically deposited cerium related thin films utilizing X-ray absorption fine structure technique (XAFS). It mentioned that the anodic deposition resulted in 38 films with stoichiometry close to CeO2. In this research, anodic electrodeposition and electrochemical synthesis will be major approach to produce cerium oxide thin films and nanocrystalline powders. 2.1.2. Cyclic Votammetry (CV)5,6,39,40 2.1.2.1. Fundamentals of CV Cyclic votammery (CV) is the most popular and effective electroanalytical technique that can be applied to examine the redox system at and on the electrode. Cyclic votammetry is a potential-controlled process in that current response is continuously observed as a cyclic sweep of potential is imposed on a working electrode in a stagnant solution. Three-electrode configuration of a working electrode, a reference electrode and a counter electrode is adopted for cyclic voltammetry. The potential, as excitation signal, is applied in triangular waveform (Figure 2.4A) and the variation of response signal - current, with the scan of the potential is recorded as cyclic voltammogram (Figure 2.4B) which is featured by current (vertical axis) versus potential (horizontal axis) curve. Oxidation or reduction of a species is presented as peaks in this curve. Figure 2.4 displays a typical potential waveform and cyclic voltammogram for the redox reaction of Fe(III)/Fe(II) at a glassy carbon working electrode. For the forward sweep, the potential scans negatively in this case, producing the reduced species (Fe(II)) when the potential is sufficiently 79 negative to reduce the Fe(III) to Fe(II). Cathodic current, the current for the reduction increases with the reaction till the concentration of Fe(III) on electrode surface decreases to zero, which is indicated with the cathodic peak. Then cathodic current decreases, during which the Fe(II) on the surface diffuses to the bulk solution. The sweep switches to positive for the reverse scan, during which the pre-produced Fe(II) will be oxidized to Fe(III) and a corresponding anodic peak forms. The potentials and currents for anodic and cathodic scan are labeled as Epa, Ipa and Epc, Ipc, respectively. 2.1.2.2. Electrochemical Information from CV In addition to potentials and currents, cyclic voltammogram can provide more important kinetic and thermodynamic information about the reaction. For an electrochemically reversible redox, both oxidized species (Fe(III)) and reduced species (Fe(II)) can exchange electrons with the electrode rapidly, which results in oxidation and reduction peaks. In this case, Ipc = Ipa, and the peak current is a function of square root of the scan rate, which can be expressed through Randles-Sevcik equation (Equation 2.17): 5 3 / 2 1/ 2 1/ 2 i p = 269x10 n AD Cv (Equation 2.17) in which ip is the peak current (A), n is number of electrons transferred, A is effective electrode area (cm2), D is diffusion coefficient (cm2/sec), C is bulk concentration of the electrochemically active species (mol/cm3) and v is the scan rate (V/sec). Therefore, the peak current varies 1/2 accordingly with the square root of scan rate. A plot of ip versus v yields a straight line with a slope of 269x105 n3 / 2 AD1/ 2C (Figure 2.4C). The transferred electrons can be calculated from the separation of cathodic and anodic peak potentials (Equation 2.18): 0.059 ∆E = E − E = (Equation 2.18) p pa pc n 80 If the area and bulk concentration is known, from the slope, the diffusion coefficient can be calculated. The peak current is proportional to the solution concentration, which can be used to figure out the concentration of the species if diffusion coefficient is known. The information from CV shape and ip of the electrochemical reaction trend is a wealth for the study of kinetics and mechanisms for this reaction. Furthermore, the formal potential that has close relationship with thermodynamic properties can be obtained from the peak potentials of reversible cyclic voltammograms (Equation 2.19): E + E E = pa pc (Equation 2.19) f 2 The absence of the reverse peak in CV is indicative that the formed species are not electrochemically active or is depleted by a followed reaction. And for an electrochemically irreversible redox couple, electron transfers for both or one reaction are so slow as to lead to a larger separation between the anodic and cathodic peaks. The relationships established for reversible reactions (Equation 2.17-2.19) are invalid for both cases. A separate and more complex analysis needs to be employed for interpretation of the CVs to produce valuable information for the reactions. 2.2.2.3. CV in this Research CV spectra of different reaction systems are studied, which illustrates different reaction schemes for cerium oxide film and powder formations. 81 -0.2 0.0 0.2 ial (V) t 0.4 ten Po 0.6 0.8 (A):Triangular potential waveform 1.0 0123456 Tim e (sec) 0.8 v = 10 mV/sec E v = 20 mV/sec pc 0.6 v = 50 mV/sec v = 100 mV/sec v = 200 mV/sec i 0.4 pc A) 0.2 -4 0 t (1 n 0.0 e rr u (B):Cyclic voltammograms C -0.2 i pa at different scan rates -0.4 Fe(III)/Fe(II) : 0.001M E pa -0.6 0.8 0.6 0.4 0.2 0.0 -0.2 Potential (V) 6 I = -1.92-6 + 1.31-4 v1/2 P 5 4 XA) 5 (10 3 p i 2 1/2 (C): Ip~v plot 1 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 V 1/2(V/sec)1/2 Figure 2.4. Cyclic voltammetry 82 2.1.3. Rotating Disk Electrode (RDE) Method 2.1.3.1. Fundamentals of RDE5,41 Generally, there are three modes of mass transfer related with an electrode reaction: migration, diffusion and convection. Migration represents the movement of charge species in an electrical field. Diffusion is the movement of chemical species because of the gradient (e.g. concentration gradient) between electrode interface and bulk solution. And convection refers to the movement of species due to the solution motion with respect to the electrode, such as liquid flow, stirring of solution and rotating of electrode. The techniques that involve convection mass transport are called hydrodynamic methods.5 The rotating disk electrode (RDE), as one of the convective electrode systems, has been widely applied in the study of electrode kinetics due to its stable and reproducible hydrodynamic conditions at the vicinity of the electrode.41 The RDE is constructed with a disk electrode imbedded in an insulating rod that is attached to a rotating driving motor. Fast rotation induces the solution convection at the surface of the electrode. The disk electrode can be connected to a power supply by an internal conductive wire or rod, while RDE voltammetry is performed. Figure 2.5 demonstrates the hydrodynamic flow pattern in which the liquid at the surface of the electrode moves horizontally away from the center due to the centrifuging force of the electrode rotation. The detail discussion and theory about concentration and velocity profiles at RDE has been documented.5 2.1.3.2. Electrochemical Information from RDE The RDE voltammogram furnishes us with information about the redox system such as redox potentials and reversibility. It is the steady state that more interests us. The steady state limiting current varies with the rotation speed of electrode (Equation 2.16), which can be used to 83 derive the kinetic information such as reaction mechanism or rate constants of electrode reaction 1/2 6,42,43 through a plot of il versus ω . o 2 / 3 −1/ 6 1/ 2 il = 0.62nFAC D v ω (Equation 2.16) where Co is solution concentration (mol/L) D is diffusion coefficient of reactant (cm2/sec) A is effective area of electrode (cm2) n is number of transferred electron il represents limiting current (A) v represents kinematic viscosity of the fluid ω is angular velocity of the electrode Solution Figure 2.5 Schematic representation of rotating disk electrode and hydrodynamic glow pattern at RDE 2.1.3.3 RDE for this Research RDE is used for mechanism studies of the film electrodeposition. 84 2.1.4 Corrosion Electrochemistry and Corrosion Measurement44-46 Metal corrosion refers to the metal oxidation and the dissolution as the metals ions and leaves the associated electrons behind. Since the corrosion involves electron generation and transfer, it is categorized as an electrochemical reaction and the properties of corrosion can be measured through electrochemical techniques.44 The corrosion reaction and mechanisms have been well documented in the introduction section and the utilized electrochemical techniques for corrosion will be described in this section. The corrosion testing techniques include electrochemical-related: linear polarization (LP), Tafel test, potentiodynamic scanning (PDS), cyclic polarization (CP), electrochemical impedance spectroscopy (EIS)44 and non-electrochemical weight loss test.46 LP and Tafel test are two commonly used corrosion test methods in laboratories due to the relatively rapid and convenient data acquisition. 2.1.4.1. Linear Polarization Method Linear polarization, compared with other electrochemical techniques, is a non-destructive test and therefore can be applied for long-term monitor of corrosion. A potential spectrum between – 20 mV from the open circuit potential (OCP) to + 20 mV from the OCP is applied to the testing specimen in the electrochemical setup. The current is plotted versus the potential through the scan range to give a linear plot (Figure 2.6A). The slope of the straight line ∆E represents corrosion resistance of the material: R = . This resistance can be used to calculate p ∆i the corrosion current density according to Stern-Geary equation (Equation 2.20):44,45 β a β c icorr = (Equation 2.20) 2.303(β a + β c )R p Where: 85 2 icorr is the corrosion current density (A/cm ) β a , β c are anodic and cathodic Tafel slopes (V/decades), respectively. The measurement of linear polarization is error-prone when there is uncompensated IR drop and localized corrosion dominates, which limits its application. 2.1.4.2. Tafel Plot Method Tafel plot method uses a wider range of potentials for the scan (OCP ± 200mV ) , therefore more information will be uncovered by this technique. The typical Tafel plot is shown in Figure 2.6B in which the potential is plotted as the vertical axis and logarithmic current density as the horizontal axis. Both anodic and cathodic branches in the Tafel plot are approximately linear and the slopes ( β a , β c ) are utilized to calculate corrosion current density (icorr ). The corrosion current density can be estimated from the intersection of the extrapolation of the linear portion of both andic and cathodic branches in Tafel plot. For the most common Tafel plot, the activation energy controls the corrosion reaction. Another type of Tafel plot derives from diffusion controlled corrosion. The interpretation and analysis of the activation type Tafel plot can be extended to other Tafel plots. Tafel plot measurements have a destructive effect on the testing specimen surface due to large polarization potentials imposed on the specimen. Hence, the specimen cannot be subject to other test after the Tafel test. Uncompensated IR drop, like in linear polarization test, can bring error to the Tafel test. Corrosion rate can be deduced from the corrosion current density according to the following equation (Equation 2.21):44 e R = 0.0033i (Equation 2.21) mm / y corr ρ 86 Where: Rmm / y is corrosion rate (millimeter/year) 2 icorr is corrosion current density (µA/cm ) e is equivalent weight of the metal, ρ is density of material (g/cm3) Techniques with wider spectrum of potential can be employed to test and gather more information about corrosion, such as PDS, CP and EIS. Weight loss is a non-electrochemical simple method to test the corrosion and it is determined when the corrosion product is removed with chromic-phosphoric acid solution. ASTM G1-1972 describes the procedure for the test.46 2.1.4.3. Corrosion Test in this Research Linear polarization and Tafel method are applied to examine the corrosion inhibition ability of cerium oxide film on stainless steel substrates. 87 Ecorr + 20mV Negative Ecorr + 10mV PotentialPotential Applied Current Applied Current (Cathodic) (Anodic) Ecorr - 10mV Potential Positive Ecorr - 20mV Negative Anodic Branch ial E nt corr e Cathodic Branch Pot Positive i corr log (Current Density) Figure 2.6 Corrosion measurement: (A) Linear polarization (B) Tafel technique 2.2. Characterization and Analysis Methods 2.2.1. X-ray Diffraction (XRD) 2.2.1.1 Fundamentals of XRD47-51 X-ray diffraction is a powerful analytical technique to investigate structural properties of crystalline solids and it is widely used in powder structure determination and thin film characterization. X-rays are electromagnetic radiation of wavelength about 1Ǻ, and the energy ranges 200 eV to 1 MeV, which is sandwiched between γ-rays and UV radiation. An x-tube is composed of 88 two metal electrodes enclosed in a vacuum chamber. The electrons formed by heating the tungsten cathode are accelerated toward the anode (Cu, Mo, Fe, etc.) with a very high velocity. The collision of electrons with the anode loses the energy of the electrons to the anode. A small portion of energy ejects the inner-shell electrons and leaves the anode atom at the excited state with a hole in the electron shell. When the outer shell electrons drop to fill the hole in the inner shell, the x-ray photon with the energy equal to the difference between these two energy levels is produced. The filtered monochromatic X-ray beam, incident on the sample, is diffracted in the specific direction. This direction depends on wavelength of the incident x-ray beam and the crystalline properties of the sample. The relationship between the incidence radiation wavelength and the spacing of the sample atomic planes is indicated with Bragg’s law: nλ = 2d sinθ (Equation 2.22) where λ represents wavelength of the incident radiation (Copper Kα is used in most modern X- ray diffraction instrument: λ = 0.154056 nm), n is the order of the reflection, d is interplanar spacing of the detected lattice planes, and θ is called as diffraction angle or Bragg angle between specimen plane and X-ray source or reflection direction (Figure 2.7). θ θ Figure 2.7 Diffraction of X-rays by a crystal 89 Since the diffraction patterns result from the interference of different atoms, X-ray diffraction is unique for different compounds and it is used to characterize the crystal structure of the specimen. Technically, there are three components for an X-ray diffractometer including the X-ray sources, specimen and X-ray detector. We know the Bragg angle is the angle between the specimen and X-ray source, then the angle between the projection of X-ray and the detector is 2θ. X-ray setup can be θ-2θ, θ-θ configurations, with θ-2θ the most often used mode. In this mode, the X-ray source is fixed, as the detecting sample is tilted at an angle of θ relative to the X-ray source, the detector will move 2θ simultaneously. The resulting pattern consisting of a series of reflections are recorded as the intensity verses 2θ. Each peak in the diffraction pattern relates to a specific set of planes in the specimen. The combination of peak positions in the x-ray diffraction pattern is the basic information carrier for a specific specimen because unique particle size and unit cell of each compound are embodied in the unique X-ray pattern (ignoring the effect of X-ray source). Different preferred orientation brings about different types of intensity patterns with the peak positions retained for the single-phase material. The width of an individual peak, which is always evaluated by the full width at half the maximum height (FWHM) can furnish us with the information of the crystallite size of the sample and lattice distortion in the structure. 2.2.1.2. Lattice Parameters 49,51 Equations 2.23 and 2.24 relate lattice parameters and the interplanar spacing for cubic structure and hexagonal structures, respectively. As Bragg’s Law (Equation 2.22) is combined with Equation 2.23 or 2.24, Equation 2.25 and 2.26 result respectively, which are used to determine lattice parameters (a, b, c) for crystal structures. 90 1 h 2 + k 2 + l 2 Cubic structure: = (Equation 2.23) (nd) 2 a 2 1 4 h 2 + k 2 + l 2 l 2 Hexagonal structure: = + (Equation 2.24) 2 2 2 (nd) 3 a c λ Cubic structure: a = h 2 + k 2 + l 2 (n=1) (Equation 2.25) 2sinθ λ 4 l 2 Hexagonal structure: a = (h 2 + k 2 + l 2 ) + (n=1) (Equation 2.26) 2sinθ 3 (c / a)2 In the equations, h, k, l represent the miller indice for a specific plane and a, b, c is the lattice parameters for the crystals. (c/a) is a constant for a defined hexagonal diffraction pattern. Precise determination of lattice parameters usually considers source wavelength accuracy and the Bragg angle measurement accuracy. Instrument selection defines the wavelength accuracy, and commonly applied CuKα has the accuracy of 1/200000. Therefore accurate measurement of the Bragg angle is extremely critical for the exact calculation of lattice parameters. Take a cubic structure as an example, it is recognized that the lattice parameter (“a” in equation) calculation gets closer to the ideal number as the Bragg angle approaches to 90o. With detail analysis of errors, a bulk of methods has been suggested to correct and achieve high precision of lattice parameters. Three types of methods have been developed to attain precise lattice parameters: standard calibration, experimental technique refinement and mathematical minimization of errors. (1) Mathematical Extrapolation: Mathematical extrapolation approaches are popular due to the extensive use of diffractometers instead of Debye-Scherrer cameras lately. Usually the lattice parameters at various Bragg angles are calculated with Equation 2.25 or 2.26 as the crystal structure is known 91 and miller indices are well defined. In Bradley-Jay extrapolation various calculated lattice parameters are plotted against cos2θ whereas in Nelson-Riley extrapolation the lattice parameters cos 2 θ cos 2 θ are plotted as function of + , resulting in a straight line. The intercept corresponds sinθ θ to the lattice parameter at Bragg angle equal to 90oC, which is the true value of lattice parameters (Figure 2.8). Comparing these two error minimization methods, the latter has advantage of broader range of linearity and it can be applied in the structure with fewer high angle reflections. However, the former method requires θ = 60o to 90o. (2) Cohen’s Least-Square Minimization: Cohen’s least-square minimization can be derived from combining Bragg equation and error equation. Taking logarithm of squared Bragg equation and then differentiating leads to: ∆sin 2 θ 2∆d = − (Equation 2.27) sin 2 θ d and assuming errors can be defined as: ∆d = K cos 2 θ (Equation 2.28) d where K is a constant. Combining these two equations result in: ∆sin 2 θ = −2K cos 2 θ sin 2 θ = Dsin 2 θ (Equation 2.29) in which ∆sin 2 θ refers to the error between true and observed value. The true value is 2 2 λ 2 2 2 sin θ (true) = 2 (h + k + l ) . (Equation 2.30) 4atrue Hence, 2 λ2 2 2 2 2 sin θ (observed) − 2 (h + k + l ) = Dsin 2θ (Equation 2.31) 4atrue 92 which can be written as sin 2 θ (observed) = Aα + Cδ (Equation 2.32) 2 λ 2 2 2 2 where A = 2 , α = h + k + l , C = D /10 and δ = 10sin 2θ . 4atrue Considering the least-squares principle: 2 ∑e 2 = ∑[Aα + Cδ − sin 2 θ (observed)] (Equation 2.33) Differentiating Equation 2.33 with respect to A and C and equating to zero, we have ∑∑α sin 2 θ = A α 2 + C∑αδ (Equation 2.34) 2 ∑∑δ sin 2 θ = A αδ + C∑δ (Equation 2.35) Solving A is used to determine atrue. The essential application prerequisite of Cohen’s method is the valid cos2θ extrapolation. a true ) (nm ter e ram a P e c tti a L 2 2 2 cos θ or cos θ/sinθ + cos θ/θ Extrapolation Function Figure 2.8 Mathematic extrapolation to obtain true lattice parameter 2.2.1.3 Crystallite Size and Lattice Strain51,52 93 Another well-established math model is the relationship between the peak broadening and the crystallite size and lattice strain, which is known as the Williamson and Hall method:52 kλ β cosθ = η sinθ + (Equation 2.36) corrected L where λ is the wavelength of the x-rays, θ is the diffraction angle, η is the lattice strain, L is the crystallite size, k is a constant dependent on the crystal structure and βcorrected is the corrected FWHM of the given peak. Plotting λcosθ against sinθ leads to a straight line in which the intercept is attributed to the broadening due to particle size (kλ/L) and the slope is corresponding to the broadening due to lattice strain (η). Therefore crystallite size of the sample can be calculated from L = kλ/intercept, in which the intercept is read directly from the plot and the strain can be obtained directly from the slope. If only crystallite size contributes to the line broadening, the Scherrer equation (Equation 2.37) will account for crystallite size estimation. kλ β = (Equation 2.37) particlesize L cosθ It is evidenced that lattice strain and particle size effects can result in the broadening of X-ray lines. The X-ray line becomes broader as the crystalline size gets smaller. Generally, there are three factors contributing to the X-ray line broadening. In addition to the above, instrumental broadening of the X-ray diffractometer is also in effect. A stress free material is employed as a standard to correct the measured line broadening. And the instrumental broadening is subtracted from the measured sample broadening using one of two methods below: Lorentzian Method β corrected = β measured − β s tan dard (Equation 2.38) 2 2 1/ 2 Cauchy Method β corrected = (β measured − β s tan dard ) (Equation 2.39) 94 According to Cullity, the maximum particle size measurable from the method of line broadening is 100 nm and at most 200 nm with most refined techniques.47 2.2.1.4. XRD in Cerium Oxide and Layered Silicates In terms of cerium oxide powders and thin films, XRD is used to evaluate the sample produced and to characterize the structure of the sample. It estimates the crystallite size of the CeO2. Pure cerium dioxide has the typical fcc structure with the first reflection (111) at 2θ = o 10 28.541 if CuKα is used as x-ray source. The importance of XRD to layered structures relies on the layer spacing measurement. The intensity, shape or position change of characteristic (001) reflection will reflect the intercalation or delamination of layered materials.50 2.2.2 X-ray Photoelectron Spectroscopy (XPS) 2.2.2.1. Fundamentals of XPS 53 X-ray Photoelectron Spectroscopy (XPS) is a photoemission technique that has broad application in the examination of elemental composition and chemical state distribution of species at solid surface. In XPS, the irradiation of the sample by characteristic x-rays in vacuum results in the generation of photoelectrons, which have characteristic kinetic energies. The relationship between such kinetic energy and the x-ray energy can be expressed by the following equation: Eb = hν − Ek + ∆φ (Equation 2.40) where Eb is the electron binding energy (BE) in the solid, hν is the energy of the incident x-ray source, Ek is the kinetic energy of the emitted photoelectron, and ∆φ refers to an instrument related work function difference between the sample and detector. With the fixed incident x-ray 95 source provided, electron binding energy (BE) for the examined sample is of element characteristic and can be observed from the energy of the photoelectron. XPS instruments in UHV are composed of an x-ray source, a controller, an electrostatic charged-particle energy analyzer and a sample mount system. High vacuum helps prevent the contamination of the surface during the operation of XPS. X-rays that are employed in the XPS experiment are generated by bombarding a metallic anode with high-energy electrons. The commonly used anode is made of Mg or Al with energies of 1253.6 eV and 1486.6 eV, respectively. X-rays strike the sample surface, then the produced photoelectrons pass through an energy analyzer at a certain applied potential. The signal will be collected and sent to the data processing system for further analysis. Typical spectrum of XPS is recorded as signal intensity against the different binding energies. Spectral peaks can be attributed to the valence-band structure, core levels and also possibly arise from Auger processes. The peak positions (BEs), peak shape and the relative intensities of every spectral peak carry chemical composition and chemical state information, since every element has fingerprint peaks in XPS except for H and He. It is revealed that valence-band spectrum also furnishes information about the chemical bonding. 2.2.2.2. Practical Aspects of XPS54 One element possesses several features at different BEs embracing more than one chemical state, however, some of feature peaks are closely spaced. To gather compositional and chemical state information, curving fitting techniques are needed to resolve the peaks. The rough XPS data are smoothed and the background is removed, the proper individual peak positions and relative intensities are assigned and adjusted in iteration until the best fit to the data are obtained. 96 The compositions of the materials can be estimated using the relative intensity (area or height) ratio combined with appropriated sensitivity factors. The valence state preference and stoichiometry of the material result from the relative intensities (area or height) of clearly separated peaks. Usually, the prominent spectra present for material are chosen to determine valence and stoichiometry. The signal from the analyzer is in the form of electron current as a function of kinetic energy of photoelectrons. C 1s or some inert metal with known core-level binding energy is used to align the Fermi level and calibrated electron energies so that all of the other features can be assigned in the XPS system correctly. X-rays in XPS penetrate 1-4 nm in the sample depth and XPS is a surface sensitive technique. The improvement of detection sensitivity of XPS for very thin films can be realized with angle-resolved XPS (ARXPS), in which the sample plane orientation can be tilted with respect to the detector so that the same quantity of information can be obtained with shorter detection depth. Surface charging of insulating materials is vulnerable to random peak shift and peak overlaps. Both an external calibrant and internal core peak have been used to effectively abate this trouble. 2.2.2.3 XPS in Cerium Oxides Cerium oxide is distinguished with characteristic Ce3d, Ce4d, Ce valence band and O1s in XPS. Since Ce(III) oxide and Ce(IV) oxide demonstrate different features in Ce and O spectra, XPS is primarily used to examine the oxidation state of cerium oxide and to quantitatively estimate the stoichiometry of this material according to the ratio of Ce (III) and Ce(IV). 97 Experimental and theoretical studies of cerium compounds have supply rationale for further detection and analysis.55,56 2.2.3. Infrared Spectroscopy and Raman Spectroscopy 2.2.3.1. Fundamentals of Infrared Spectroscopy57-59 Infrared spectroscopy belongs to molecular absorption spectroscopy. When a sample is irradiated with infrared light, the molecule absorbs certain frequencies. These frequencies are proportional to the energy difference between the vibrational or rotational ground states and excited states. Accordingly, the corresponding energy level transitions are excited. The absorption of the infrared energy weakens the transmission. Infrared spectra record the relationship between the transmission (absorption) percentage and wavelength (wavenumber). Since Infrared energy is low and can only induce the transition of vibration or rotation level, it is also termed as vibrational & rotational spectroscopy. Infrared region locates between the visible light and microwave and the wavelength ranges 0.75 ~ 1000 µm, and encompasses 13,158 ~ 10 cm-1 in terms of wavenumber. The infrared is usually divided into far-, mid-, and near- infrared radiation from the view of instrumentation and applications. The wavelength, wavenumber and frequency ranges for each is listed in Table 2.1. Mid-infrared is the most useful region and has wide applications in quantitative and qualitative analysis of materials. 98 Table 2.1 Three regions of infrared radiation Infrared Wavelength Wavenumber Frequency Energy Level Transition Region (λ), µm (v), cm-1 (ν),10-14 Hz Far- 0.75 ~ 2.5 13158 ~ 4000 4.2 ~ 1.2 OH, NH, CH Overtones Molecular vibration, Mid- 2.5 ~ 25 4000 ~ 400 1.2 ~ 0.12 accompanied vibration Near- 25 ~ 1000 400 ~ 10 0.12 ~ 0.003 Molecular rotation Two conditions are necessary to absorb infrared radiations. The photon radiation is required to have the equivalent energy as the energy for the vibration transition. Assuming the two-atom molecule, the vibration energy is expressed as 1 E = (V + )hν (Equation 2.41) v 2 in which ν is molecular vibration frequency, h is Planck’s constant and V is the vibrational quantum number. V = 0, 1, 2, 3 and so on. The energy difference between energy levels is ∆E = ∆Vhν , which should equal radiation energy ( hν a ). When molecules are promoted to first excited energy state (V = 1) from ground state (V = 0), hν a = hν . The absorption frequency that brings the transition is equal to that of vibrational frequency of the molecule. Net change of dipole moment as the consequence of vibration or rotation is another required factor for infrared activity. Dipole moment determines the polarity of a molecule, and it induces infrared transition since energy transfer is achieved through the interaction between dipole moment fluctuation and alternating radiation electromagnetic field. Therefore, not all the vibrations can bring infrared absorption, instead only the molecules with dipole moment change are infrared active. Four types of infrared instruments are available: dispersive grating spectrometers, non- dispersive photometers, reflectance photometers and flourier transform infrared spectrometers. Fourier transform infrared spectrometer consists of radiation source (mercury arc or silicon 99 carbide rod, etc.), Michelson interferometer and photoconductive detector. The Michelson interferometer splits beam by interfering frequency modulation and then digitally regenerates the spectrum. This approach avoids the radiation throughput limit of slits, so fourier transform instrument has the advantage of high throughput, high S/N, high wavelength accuracy and short detection time. Usually the reference spectrum is obtained before the sample detected. Infrared spectrum displays as the plot of transmission percentage (%) vs. wavenumber (wavelength) or absorption percentage vs. wavenumber (wavelength). The infrared absorption band position, peak number and intensity represent the molecular structure characteristics, which is used to identify the materials. And the infrared absorption intensity is proportional to the concentration of the functional group, which can provide the way to quantitative analysis and appraisal of purity. 2.2.3.2 Fundamentals of Raman Spectroscopy57-61 Raman spectroscopy is molecular scattering spectroscopy. Raman scattering is the consequence of inelastic collision between molecule and radiation photon. When a sample is irradiated with photon source, a portion of portion energy is transferred to molecule, exciting the molecule to higher energy level ( E0 + hν 0 ), ( hν 0 represents transferred energy from photon to molecule). The molecule jumps back to a lower energy level because the excited state is not stable. Rayleigh scattering is produced if there is no energy loss. However, if the molecule is excited to E0 + hν 0 but returns to E1, the molecule gains but photon loses the energy of (E1-E0). The result is that scattering frequency will be lower than the incident frequency and the scattering line is termed Stokes line. Shifts toward higher frequency are termed anti-Stokes. Figure 2.9.depicts the Rayleigh scattering and Stokes/anti-Stokes lines. The difference between Stokes/anti-Stokes line frequency and the incident frequency are termed Raman shifts. 100 E1+hν0 ∆E = hν0 E0+hν0 hν0 E1 ∆E = hν0 E0 Raman Stokes Anti- Stokes ν0 ν0+∆ν ν0 ν0-∆ν Figure 2.9 Depicts of Rayleigh scattering (left) and Raman scattering (right) The vibration energy difference is basis for both Raman shifts and infrared absorption, the spectra from these two techniques are similar for a given material, if both spectra are available. However, from the mechanism view, infrared absorption requires permanent dipole moment, only polar molecules can be infrared active. Raman scattering develops an induced dipole moment as the molecule is at momentary excited state. Hence the molecule can be Raman active with this temporary polarizability even the molecule has no permanent dipole moment. Therefore, Raman spectroscopy is a complementary technique to infrared spectroscopy. In practice, basic rules are followed to judge whether the molecule is infrared or Raman active. The molecules with symmetry center, infrared active and Raman active properties are exclusive to each other; otherwise, both infrared active and Raman active properties exist generally. Laser is usually used as an irradiation source. And the instrument is also composed of sample handling system, spectrometer/detector. Raman spectra are obtained by plotting the 101 intensity as the Y-axis and the shift of the observed radiation from the incident source as X-axis. The Raman shift is directly related with the energy difference of molecular vibration and rotation levels, which are unique for a certain material structure. Therefore, Raman spectrum also can be applied to characterize and analyze materials structure. 2.2.3.3. Applications of FTIR /Raman Spectroscopy in Crystals59 Crystals are formed with unlimited array of repeated structure in three dimensions. And the unlimited crystals can be subdivided into any smaller units. The infinity and periodicity properties of crystals are unique for crystals, which are different from the generally mentioned molecules. The vibration of a crystal constitutes a 3n-dimensional representation of the unit cell of the crystal. And the motion in which all unit cells vibrate in phase is the spectroscopically active. It is reducible to the direct sum of irreducible representations of the group symmetry. From the representation, 3N-3 optical (infrared and Raman modes) with non-zero frequencies and 3 modes of acoustic mode can be predicted. The acoustic or optical modes can be either longitudinal (L: vibration takes place along vibarional wave propagation direction) or transversal (T: vibration takes place perpendicular to vibrational wave direction). Considering optical modes, neglecting interaction between atoms, some of modes are degenerated; otherwise. LO- TO splitting usually is observed for polar vibrations. In principle, infrared absorption and Raman scattering techniques in crystals are similar to those used in organic materials. However, due to the specialty of the crystal structure and variety of vibrations, crystal spectra are more complicated with possible second-order spectra and the impurity spectra in comparison with the general molecules. Crystal data from other techniques have complementary information to analyze the IR and Raman spectra. 2.2.3.4. FTIR and Raman in Cerium Oxide and Layered Silicates 102 Si-O absorption is present in low frequencies and different layer arrangements influence the OH absorption, therefore FTIR transmission and reflection are applied to identify clay type. As used in composites, FTIR detect the existence of clay components. In addition to conventional FTIR, Diffusion reflectance (DRIFT) and Attenuated total reflection (ATR) has also been widely used to obtain information for clays.62,63 Cerium dioxide has no absorption in FTIR, but nonostoichiometric cerium oxide shows IR activity due to adsorbed oxygen species.64 Cerium oxide demonstrates activity in Raman spectroscopy. The first-order Raman shift and second-order Raman intensity change characterize lattice variation of cerium oxide.65-67 2.2.4. Scanning Electron Microscopy (SEM) 2.2.4.1. SEM Imaging68 Scanning Electron Microscopy (SEM) is one of most versatile instruments available to examine and analyze microstructural characteristics of solid-state materials. High-energy electrons are focused into a fine beam, which is scanned across the surface of the specimen. The interaction between the electron beam and the specimen produces a rich measure of radiations, which can be used to illustrate some useful information for the specimen. Two types of radiations are used most often in SEM for image generation: Secondary electrons generated because of interactions between energetic electrons of the source and weakly bound conduction- band electrons and backscattering electrons derived from the elastic scattering of specimen electrons as source beam strikes the specimen. The modulation of signals of these two types of radiations is displayed as an image in the CRT and reveals the morphology of the specimen surface. The primary use of the SEM is to obtain the information for bulk objects in the range of 2-5 µm. 103 2.2.4.2. SEM / EDX68 This technique has also been used in three-dimensional appearance of the specimen. Electron Dispersive X-ray Spectroscopy (EDX) is attached with SEM. The beam-specimen interaction-generated X-rays are employed to identify and quantify elemental composition of the specimen. SEM/EDX determines the elemental composition as well as obtains the morphology image of the sample. 2.2.4.3. SEM /EDX in Cerium Oxide and Nanocomposite Films In terms of cerium oxide and nanoclay materials research, SEM is usually employed to observe the surface morphology and cross-section structure for the thin film and used to determine film thickness. The attached EDX is capable of probing the Si and Al contents in CeO2/nanocomposites. Conductive coatings or conductive connection are used in the sample preparation to avoid surface charging in SEM application. Lower filament voltage is another acceptable approach that is effective for the prevention of charging. 2.2.5 Brunauer-Emmett-Teller (BET) 2.2.5.1. Adsorption and BET Model69 BET method is well established for characterization of materials through gas adsorption and desorption phenomena. Commonly, nitrogen adsorption at 77K is applied in the BET technique. The adsorbed amount of nitrogen varies with the applied pressure, which comprises the adsorption isotherm. The technique can supply information about the investigating materials in the following areas: total pore volume, pore size and pore size distribution and surface area. There are two types of processes for adsorption: inter-molecular force related physical adsorption and chemical bond-based chemisorption. BET method is a mathematical model based on the adsorption theory postulated by Langmuir to describe the adsorption isotherms.69 104 Langmuir theory states that the surface of the solid can be regarded as an array of adsorption sites. The adsorption of individual site is independent from the neighboring sites and the lower layer of the molecules can be regarded as the adsorption sites for the next layer, and between molecules, there is no interaction.70 2.2.5.2 Surface Area and Pore Size69 Based on this model and some other kinetic postulations, BET equation can be expressed as follows: p 1 (c −1) p p = + (Equation 2.42) V ( p0 − p) Vmc Vmc p0 where p0 is the saturation vapor pressure, Vm is the monolayer capacity and c is a constant which ∆H L − ∆H l approximately corresponds to c ≈ exp and ∆HL and ∆Hl referred to the heat of RT liquefaction and heat of adsorption. BET equation is suitable for the type II isotherm (See reference 69 for isotherm types) that is derived from the physical adsorption of gases by non-porous solids. And this equation can be reduced to Langmuir equation at lower pressures. The surface areas can be evaluated from the p p monolayer capacity which can be obtained from the BET plot of against . The V ( p0 − p) p0 (c −1) 1 1 slope of this straight line is and the intercept is , then Vm = . Vm c Vm slope + int ercept Pore size and pore size distribution can be obtained from adsorption isotherm using various type of equation according to the adsorption type: Kevin equation for type IV. The most popular mathematical model for pore distribution nowadays was developed by Barrett, Joyer and Halenda (BJH). 105 2.2.5.3. BET in Cerium Oxide BET technique has been used to assess surface area variation of cerium oxide as catalysts. 2.2.6. Nanoindentation71,72 2.2.6.1 Mechanical Properties The mechanical properties that we concerned most with nanocomposites are strength and stiffness. Basically, the so-called strength refers to yield strength, which is the highest stress that the material can withstand without undergoing significant yielding. Hardness is measured to determine the strength of materials; On the other hand, stiffness of a material is the ratio of stress to stain. Young’s modulus is the value used to represent the stiffness of materials in the linearly elastic region. 2.2.6.2 Nanoindenting and Mechanical Properties Nanoindenting is a novel method to characterize the mechanical properties of materials in small scale that is less than 100 nm across and less 5 nm thick.73 Owing to this specific qualification, it is most often employed to probe the mechanical properties of thin films. The indenting propagation is described with the load-displacement curve (Figure 2.10), and the two most frequently used mechanical properties: mechanical hardness (H) and elastic modulus (E) can be derived from this curve. 106 Loading N) m S ( L max L , ad Unloading Lo h contact h final h max Displacement, h (nm) Figure 2.10 Schematic representation of load-displacement curve Hardness is calculated as the applied load (L) divided by projected area (Aproejcted) of contact between sample surface and indenter tip (Equation 2.43). In real indentation experiment, peak load is used and the contact area can be obtained from indentation depth and shape function. L H = max (Equation 2.43) Aprojected Elastic modulus is measured from the unloading slope of load-displacement curve (Equation 2.44 & 2.45). dL 2 S = = Er Aprojected (Equation 2.44) dh π In this equation, S is called stiffness, corresponding to the slope of upper portion of unloading curve. Er reduced modulus, which is related to real elastic modulus of species (E) and indenter material (Ei) by Equation 2.45. 107 1 (1−ν 2 ) (1− v 2 ) = + i (Equation 2.45) Er E Ei in which ν and νi are Poisson’s ratio for specimen and indenter material, respectively. In addition to these two primary mechanical properties for the materials, the force displacement curve also can provide further information, such as maximum indentation depth which refers to the indent as the load is at maximum, and the residual depth that means the depth after the load is removed. Since the load-displace curve records the whole process of the loading and unloading, any change from normal loading-unloading curve due to the physical or chemical change of the materials or the effect attributed to the loading itself can be revealed directly. Besides hardness and elastic modulus testing, some nanoindentation instrument also can be applied to detect wear resistance of the materials and uncover the potential of spalling or delamination of thin films. 2.2.6.3 Nanoindention in this Research Nanoindentation is employed to study CeO2/layered silicate nanocomposite thin films, which will determine the mechanical property change as layered silicate is added in cerium oxide films. 2.2.7. Sintering Technique 2.2.7.1. Fundamentals of Sintering74 Sintering is defined as “a thermal treatment for bonding particles into a coherent, predominantly solid structure via mass transport events that often occur on the atomic scale. The bonding leads to an improved strength and lower system energy”. It has been demonstrated sintering can have a dramatic improvement in various properties such as: mechanical strength, ducitility, conductivity, magnetic permeability and corrosion resistance. Therefore, sintering has become an important processing technique in industrial ceramic production. 108 The microstructure change is the origin of the enhancement of so many properties. From the view of basic physical property, bond formation between the particles helps grain growth and reduces grain boundary density. Heating eliminates pores in the structure, and indirect consequences include decrease of surface area and densification of the sample. Therefore, grain size, pore size, pore volume, surface area and material density are significant parameters to measure the sintering level and quality. The changes of one or more of these properties are usually used to judge the efficiency of a sintering process. Densification and coarsening are two forms of practical sintering. Densification refers to pure increase of the sample density during sintering whereas coarsening is the sintering process to increase the grain size and strength without dimensional change. Various mass transport mechanisms have been proposed to contribute to sintering: viscous flow, surface diffusion, volume diffusion, grain boundary diffusion and evaporation- condensation. Viscous flow is accepted for glass or polymer materials. Evaporation-condensation is to explain the sintering of materials with high percentage of pores. Evaporation repositions the atoms via the transport of pore space, resulting in the condensation. Surface diffusion refers to the atoms reposition between different surface defects, such as vacancies, kinks, adatoms etc. Volume diffusion means that vacancies move in and between crystalline structures. And when grain boundary becomes the mass transport path, it is termed as grain boundary diffusion. 2.2.7.2. Solid State Sintering74,75 Solid state sintering is the most understandable process and the three diffusion mechanisms are important for this type of sintering. The commonly accepted sintering technique is to heat sample in air or inert environment. Generally, solid-state sintering progresses in three stages according to the microstructure variation. In the first stage, the neck bonds are formed 109 between contacting particles, the grain and pore structures are almost unchanged. Intermediate stage is relatively more significant stage for sintering, in which grains start growing and small pores are eliminated, therefore in this stage, densification degree increases dramatically. Final sintering continues eliminating pores and grain boundaries, but since the remaining pores are comparably larger, removal is relative difficult compared with smaller ones. And the material density is high, reducing the densification drive force. So in this stage, coarsening coexists with densification, which slows densification rate to some extent. To improve the sintering ability, different heat procedures have been assessed. Simultaneously, novel sintering approaches other than furnace heating have also been utilized. 2.2.7.3 Cerium Oxide Sintering Sintering is an inevitable production process for a number of materials, especially ceramic materials. Sintering is a necessary process when cerium oxide is used as electrolyte in SOFC.76 To investigate sintering property of a material is a necessary part to understand the material and to improve material sinterability is significant to extend the material application. Therefore, sintering properties of cerium oxide produced by different techniques have been documented in literatures. In this research, the sintered cerium oxide powder and films at different temperature will be studied with XRD, BET or XPS. 2.3 Chapter Conclusions Electrochemical synthesis technique is discussed in details from various aspects and the electrosynthesis of cerium oxide powders and films are well documented. Other electrochemical techniques, characterization and analysis methods that have been applied in this research work are also illustrated in this chapter. These electrochemical techniques include the cyclic 110 voltammetry, rotating disk electrode method for the electrodeposition mechanism study and the linear polarization and Tafel measurement for corrosion test. Characterization/analysis methods are XRD, XPS, SEM, FTIR, Raman Spectroscopy, BET, nanoindentation and sintering. Practical application of these techniques and methods in the research will be further elaborated in following chapters. 111 Chapter References (1) Therese, G. H. A.; Kamath, P. V. Chemistry of Materials 2000, 12, 1195-1204. (2) Zhitomirsky, I.; Gal-Or, L. In Intermetallic and Ceramic Coatings; Dahortre, N. B., Sudarshan, T. S., Eds.; Marcel Dekker Inc.: New York, 1999, pp 83-145. (3) Zhitomirsky, I. The American Ceramic Society Bulletin 2000, 9, 57-60. (4) Paunovic, M.; Schlesinger, M. Fundamentals of Electrochemical Deposition; John Wiley & Sons, Inc.: New York, 1998. (5) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; 1st ed.; John Wiley & Sons, Inc: New York, 1980. (6) Heineman, W. R.; Kissinger, P. T. In Laboratory Techniques in Electroanalytical Chemistry; 2nd ed.; Kissinger, P. T., Heineman, W. R., Eds.; Marcel Dekker, Inc.: New York, 1996, pp 51-125. (7) Tan, Y.-J.; Lim, K. Y. Surface and Coatings Technology 2003, 167, 255-262. (8) Switzer, J. A. American Ceramic Society Bulletin 1987, 66, 1521-1524. (9) Zhou, Y.; Phillips, R. J.; Switzer, J. A. Journal of American Ceramic Society 1995, 78, 981-985. (10) Zhou, Y.; Switzer, J. A. Journal of Alloys and Compounds 1996, 237, 1-5. (11) Switzer, J. A.; Phillips, R. J. Material Research Society Symposium Proceedings 1988, 121, 111-114. (12) Zhitomirsky, I.; Gal-Or, L. Journal of the Ceramic Society 1996, 16, 819-824. (13) Narayanan, R.; Seshadri, S. K. Metal Finishing 2001, 2, 84-89. (14) Gal-Or, L.; Sibermar, I.; Chaim, R. Journal of the Electrochemical Society 1991, 138, 1939-1942, 1942-1946. 112 (15) Lee, G. R. Journal of Materials Chemistry 1996, 6, 187-192. (16) Zhitomirsky, I. JOM 2001, 53, 48-50. (17) Zhitomirsky, I. Electrochemical Society Proceedings 2000, 4, 381-388. (18) Zhitomirsky, I. Electrochemical Society Proceedings 2000, 4, 338-345. (19) Zhitomirsky, I.; Petric, A. Materials Letters 1999, 40, 263-268. (20) Zhitomirsky, I. Materials Letters 1998, 33, 305-310. (21) Aldykiewicz, A. J. J.; Davenport, A. J.; Isaacs, H. S. Journal of the Electrochemical Society 1996, 143, 147-153. (22) Li, F.; Newman, R. C.; Thompson, G. E. Electrochima Acta 1997, 42, 2455-2464. (23) Li, F.; Thompson, G. E.; Newman, R. C. Applied Surface Science 1998, 126, 21-33. (24) Li, F.; Thompson, G. E. Journal of The Electrochemical Society 1999, 146, 1809-1815. (25) Zhitomirsky, I.; Petric, A. Ceramics International 2001, 27, 149-155. (26) Baur, J. F.; Spaine, T. W. Journal of Electroanalytical Chemistry 1998, 443, 208-216. (27) Birss, V. I.; Bock, C.; Eizanowska, H. Canadian Journal of Chemistry 1997, 75, 1687- 1693. (28) Marzouk, S. A. M.; Ufer, S.; Buck, R. P. Analytical Chemistry 1998, 70, 5054-5061. (29) Petit, M. A.; Plichon, V. Journal of Electroanalytical Chemistry 1998, 444, 247-252. (30) Hu, C.-C.; Tsou, T.-W. Electrochemistry Communications 2002, 2, 105-109. (31) Kavan, L.; O'Regan, B.; Kay, A.; Gratzel, M. Journal of Electroanalytical Chemistry 1993, 346, 291-307. (32) Casella, I. G.; Gatta, M. Journal of Electroanalytical Chemistry 2002, 534, 31-38. (33) Orel, Z. C.; Hutchins, M. G.; McMeeking, G. Solar Energy Materials & Solar Cells 1993, 30, 327-337. 113 (34) Musiani, M. Chemical Communications 1996, 21, 2403-2404. (35) Switzer, J. A.; Shumsky, M. G.; Bohannan, E. W. Science 1999, 284, 293-296. (36) Bensadon, E. O.; Nascente, P. A. P.; Olivi, P.; Bulhoes, L. O. S.; Pereira, E. C. Chemistry of Materials 1999, 11, 277-280. (37) Crossland, A. C.; Thompson, G. E.; Skeldon, P.; Wood, G. C.; Smith, C. J. E.; Habazaki, H. S.; Shimizu, K. Corrosion Science 1998, 40, 871-885. (38) Balasubramaian, M.; Melendres, C. A.; Mansour, A. N. Thin Solid Films 1999, 347, 178- 183. (39) Heineman, W. R.; Kissinger, P. T. Current Separations 1989, 9, 15-18. (40) Gosser, D. K. Cyclic Voltammetry: Simulation and Analysis of Reaction Mechanisms; VCH: New York, 1993. (41) Lovric, M.; Osteryoung, J. Journal of Electroanalytical Chemistry 1986, 197, 63-75. (42) Mariani, M. P.; Bartolozzi, M. Journal of Electroanalytical Chemistry 1996, 209, 275- 282. (43) Kiekeng, P.; Verplaetse, H.; Temmerman, E. Journal of Electroanalytical Chemistry 1980, 113, 233-238. (44) Tait, W. S. An Introduction to Electrochemical Corrosion Testing for Practicing Engineers and Scientists; Pair O Docs Publications: Racine, Wisonsin, 1994. (45) Sedriks, A. J. Corrosion of Stainless Steels; 2nd ed.; John Wiley & Sons, Inc.: New York, 1996. (46) ASTM: Philadelpha, Pennsyvivania. (47) Cullity, B. D.; Stock, R. S. Elements of X- Ray Diffraction; 3rd ed.; Prentice Hall, 2001. 114 (48) Jenkins, R.; Snyder, R. L. Introduction to X-ray Powder Diffractometry; John Wilery & Sons, Inc.: New York, 1996; Vol. 138. (49) Klug, H. P. A.; Alexander, L. E. X-ray Diffraction Procedures for Polycrystalline and Amorphous Materials; 2nd ed.; Wiley: New York, 1974. (50) Moore, D. M.; Reynolds, R. C. J. X-ray Diffraction and Identification and Analysis of Clay Minerals; 2nd ed.; Oxford University Press, Inc.: New York, 1997. (51) Suryanarayana, C.; Norton, M. G. X-ray Diffraction: A Practical Approach; Plenum Press: New York, 1998. (52) Williamson, G. K.; Hall, W. H. Acta Metallurgica 1952, 1, 22-31. (53) Hoflund, G. B. In Handbook of Surface and Interface Analysis; Riviere, J. C., Myhra, S., Eds.; Marcel Dekker Inc.: New York, 1998, pp 57-86. (54) Smith, G. C. In Handbook of Surface and Interface Analysis, Methods for Problem- Solving; Riviere, J. C., Myhra, S., Eds.; Marcel Deker, Inc.: New York, 1998, pp 159- 208. (55) Burroughs, P.; Hamnett, A.; Orchard, A. F.; et.al. Journal of the Chemical Society Dalton Transactions 1976, 17, 1686-1698. (56) Al-Dhhan, Z. T.; Hashen, T.; Hogarth, C. A. Spectrochimica Acta. 1989, 44B, 205-208. (57) Zhao, X.; et.al. Instrumental Analysis; Beijing University Publishing: Beijing, 1997. (58) Skoog, D. A.; Leary, J. L. Principles of Instrumental Analysis; 4th ed.; Saunders College Publishing,: Forth Worth, 1992. (59) Schrader, B. Infrared and Raman Spectroscopy: Methods and Applications; VCH Publishers, Inc.: New York, 1995. 115 (60) Lewis, I. R.; Edwarsds, H. G. M. Handbook of Raman Spectroscopy, from the Research Laboratory to the Process Line; Marcel Dekker, Inc.: New York, 2001; Vol. 28. (61) Turrel, G.; Corset, J. Raman Microscopy, Developments and Applications; Academic Press, Harcourt Brace & Company: San Diego, 1996. (62) Madejová, J. Vibrational Spectroscopy 2003, 31, 1-10. (63) Madejová, J.; Komadel, P. Clays and Clay Minerals 2001, 49, 410-432. (64) Li, C.; Domen, K.; Maruya, K.; Onishi, T. Journal of the American Chemical Society 1989, 111, 7683-7687. (65) Binet, C.; Daturi, M.; Lavalley, J.-C. Catalysis Today 1999, 50, 207-225. (66) Weber, W. H.; Hass, K. C.; McBride, J. R. Physical Review B 1993, 48, 178-185. (67) McBride, J. R.; Hass, K. C.; Poindexter, B. D.; Weber, W. H. Journal of Applied Physics 1994, 76, 2435-2441. (68) Goldenstein, J. I.; Newbury, D. E.; Echlin, P.; et.al. Scanning Electron Microscopy and X-ray Microanalysis: A Text for Biologists, Materials Scientists, and Geologists; 2nd ed.; Plenum publishing Corporation: New York, 1992. (69) Gregg, S. J.; Sing, K. S. W. Adsorption, Surface Area and Porosity; 2nd ed.; Academic Press: London & New York, 1982. (70) Sing, K. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2001, 187- 188, 3-9. (71) Doerner, M. F.; Nix, W. D. Journal of Materials Research 1986, 1, 601-609. (72) Oliver, W. C.; Pharr, G. M. Journal of Materials Research 1992, 7, 1564-1583. (73) Woirgard, J.; Tromas, C.; Giard, J. C.; Audurier, V. Journal of the European Society 1998, 18, 2297-2305. 116 (74) German, R. M. Sintering Theory and Practice; John Wiley & Sons, Inc.: New York, 1996. (75) Chen, I.-W.; Wang, X.-H. Nature 2000, 404, 168-171. (76) Kharton, V. V.; Figueiredo, F. M.; Navarro, L. Journal of Materials Science 2001, 36, 1105-1117. 117 CHAPTER 3 ELECTROCHEMICAL DEPOSITION OF CERIUM OXIDE FILMS 3.1. Introduction The fabrication of cerium oxide films has attracted attention due to the broad applications for cerium oxide.1-6 Functioning as the structure barrier for the insulator on silicon or insulator on superconductor materials is the basic spur for the deposition of cerium oxide films on semiconductors or even insulating materials.7-9 Whereas the ability of cerium oxide coating to 10-12 prevent corrosion on metal and the potential application of CeO2 coated materials as electrodes in SOFC or other systems5,13 have encouraged cerium oxide film deposition on metal substrates. Electrochemical deposition is an attractive method for the synthesis of thin films. It offers the advantages of low processing temperature, normal handling pressure, high purity of deposition and controlled thickness of the film. Cathodic electrodeposition (i.e. base generation electrochemical methods) was first introduced for the plating of cerium oxide films14 and has 15-21 been successfully used to prepare cerium oxide films. However, producing CeO2 films with the base generation method was limited since as-produced films tended to be powdery and loosely adherent for the precipitation reactions.20 The hydroxide ions at the vicinity of the working electrode combined with Ce(III) free species in the bulk solution to form Ce(III) hydroxide that contaminated Ce(IV) compounds, therefore the stoichiometry (O:Ce) of the formed film was less than 2.21 As the reaction of Ce(III)→Ce(IV) is an oxidation reaction, theoretically, it is possible to apply anodic deposition to obtain CeO2 thin films. Compared with the well-understood base generation electrodeposition method, the anodic electrodeposition of metal oxide is relatively 118 less documented in literature. Anodic electrodeposition has not been studied as a technique for cerium oxide films though a number of other metal oxide films have been obtained with this approach.22-24 Since anodic deposition is a direct oxidation of Ce(III) to cerium oxide, it is predicted this deposition mode will result in higher purity of CeO2. XANE studies on electrodeposited cerium related thin film revealed that the anodic deposition preferred the Ce(IV) compounds while cathodic base generation method led to the formation of high percentage of Ce(III) species in the composition. Poubaix diagrams’ (E vs pH) consist of redox species and solubility of products for an element in aqueous solutions. The Poubaix diagram of cerium species in solution is shown in Figure 3.1. Single solid lines separate species related by acid-base equilibria and solid double lines separate species related by redox equilibria. The dotted lines frame the boundaries of oxidation and reduction for water. The Poubaix diagram reveals that CeO2 exists at potentials higher than 1.2 V vs. SHE and pH higher 8.25 However, Ce(III) free species cannot be stable in high pH due to the cerium hydroxide formation when OH- exists. To electrochemically convert Ce(III) in high pH aqueous solution to solid CeO2 but not form Ce(OH)3, a complexing agent is employed to stabilize Ce(III) in the solution, then a suitable potential or corresponding current was applied to directly oxidize Ce(III) species to Ce(IV) without precipitation of Ce(OH)3. 119 04812 2 2 ++ Ce(OH) CeO 2 2 (orange) (yellow) 1 1 3+ Ce Ce(OH) 3 0 (Ce O hydr.) 0 2 3 ) (grey-green) (V E -1 -1 -2 -2 -3 Ce -3 04812 pH Figure 3.1 Simplified Pourbax diagram for cerium that illustrate the existence of cerium species at certain potential and pH. This chapter describes the anodic electrodeposition of the cerium oxide film on metal and semiconductor substrates. In this electrodeposition process, organic ligands are used for the stabilization of cerium (III) in high pH aqueous solutions. The crystal structure and surface texture of electrodeposited CeO2 films were characterized with XRD, XPS, Raman Spectroscopy and SEM techniques. A study of the deposition parameters is presented in the first section of this chapter. The deposition conditions examined include deposition mode, substrates, pH, temperature, and the influence of current density. 120 Sintering in air was conducted on the electrodeposited cerium oxide films. And corrosion test will be presented in the last part of the chapter, which will be used to illustrate one of the important applications of cerium oxide films on metal substrates. 3.2. Experimental 3.2.1. Electrochemical Deposition of CeO2 films The electrochemical deposition of CeO2 films was performed using an EG&G Princeton Applied Research (PAR) Model 273A potentiostat/galvanostat. The cell was kept at constant temperature with a Fisher Scientific Model 1016D circulator. The deposition was conducted in an undivided cell and regular three-electrode configuration (Figure 3.2). The working electrode was prepared by mounting stainless steel or platinum in epoxy, and then polished, rinsed and ultrasonicated before use. Platinum wire or mesh was used as the counter electrode and a saturated calomel reference electrode (SCE) as the reference. The electrolyte consisted of 0.1 M Ce(NO3)3·6H2O (99.5% pure, Alfa AESAR) and a ligand. All the solutions were dissolved in DI water. Ce(NO3)3 and ligand concentrations were 0.1 M throughout the study. Solution pH was adjusted with NaOH and monitored with a pH meter throughout the deposition experiments. Each deposition was completed in 24 - 48 hours. 121 Figure 2: Set up of CeO2 deposition Reference Electrode Counter Electrode Water Out pH Electrode Working Electrode Water In Figure 3.2 Setup of electrochemical deposition of CeO2 3.2.2. Characterization of CeO2 films 3.2.2.1. X-ray Diffraction (XRD) The structure and phase composition of electrochemical deposited films were identified by X-ray diffraction (XRD) analysis with a Siemens D500 diffractometer using Cu Kα radiation (λ = 0.15405 nm). The tube source was operated at 40 kV and 30 mA. The resultant patterns were analyzed with Jade software or Origin 6.0 so that the reflection positions, FWHM, intensity and other necessary information can be extracted. 3.2.2.2. Scanning Electron Microscopy (SEM) The surface morphologies of the films were characterized by a scanning electron microscope (SEM), JOEL JSM-T300 with an accelerating voltage of 5~25 kV. 3.2.2.3. X-ray Photoelectron Spectroscopy (XPS) The oxidation states of the cerium oxide films were determined with X-ray photoelectron spectroscopy (XPS). XPS data is obtained ex-situ with a VG Escalab MK II using Al Kα excitation at 1486.6 eV with a Hemispherical Analyzer, operated in the constant pass energy 122 mode at 20 eV. The analyzer was calibrated using Au, Cu or Ni. Binding energy scales in this work were calibrated according to ASTM standard E2108-00.26 For insulating samples, sample charging often occurs and is corrected with the C 1s peak set at 284.6 eV. The raw data was converted through Origin 6.0 software and plotted in Matlab, in which the Shirley background subtraction was applied on all the raw data. All peaks of the corrected plot were fitted with a Gaussian-Lorentzian shape function to deconvolute overlapping peaks. Peak positions were 27-29 assigned according to the literature assignment for the CeO2 XPS. Iterations were performed until a minimum for χ and χ2 factor were reached for each spectrum.30 3.2.2.4 Raman Spectroscopy Raman spectra were obtained by using Yvon Dilor XY800 Raman microprobe, with an excited wavelength of 514.54 nm laser radiation and 10 mW of power to prevent the local heating of the sample. The emitted backscattered light was observed with a nitrogen-cooled mulitchannel CCD detector. All the spectra were recorded between 200 ~ 4000 cm-1 with 3 cm-1 spectra resolution, 719 mm focal length, 600 lines/mm gratings and a x10 objective which focuses the laser beam in a spot of 10 µm. At least three spectra are randomly collected from different areas of the films and plotted. 3.2.3 Property and Application Studies 3.2.3.1 Sintering Test The deposited films together with substrates were sintered in air with a furnace. A temperature ramping procedure was used with a ramping rate of 1oC/min. The temperature was kept at a final value for a set period of time, after which the temperature was decreased at the ramping rate of 1oC. The beginning and ending temperature was 260oC for all samples. 123 Maximum temperatures varied with experiments. The structure and texture of CeO2 film before and after sintering are characterized and analyzed with XRD, XPS and SEM. 3.2.3.2 Corrosion Test Linear polarization and Tafel experiment were done with EG&G 273A potentiostat with EG&G 352 softcorr III. Figure 3.3 shows the employed corrosion cell configuration. The working electrode mounted in a Teflon gasket is attached at the end of a conductive rod, SCE reference is usually bridged with the test solution and the tip of the bridge tube is assigned as close as possible to the working electrode, two graphite rods are set at either side of working electrode at equal distance. A purge tube is prepared for the protection of the specimen from air. The whole system is designed tight sealed with various glass adapters. Counter Electrode Reference Electrode Purge Tube Bridge Tube Working Corrosion Electrode Flask Figure 3.3 Corrosion measurement cell 3.3. Results and Discussions 3.3.1. Deposition Condition Investigations 124 Cerium oxide films with fluorite structure can be identified by X-ray diffraction (XRD) analysis.3,31 The cathodic base generation deposition method is studied by XRD for the deposition of CeO2 film on stainless steel in which a 0.1M Ce(NO3)3 and 0.1M NH4NO3 solution in a divided cell at 0.60 mA/cm2 and 25oC was employed for the deposition.20 X-ray diffraction pattern of the cathodic deposited CeO2 film is shown in Figure 3.4A. The pattern exhibits a random orientation of CeO2 face-centered cubic (fcc) structure. The experiment reveals that the cathodic base generation produced CeO2 film is of powdery characteristic and weakly adherent to the substrate surface. As mentioned in the previous section, for the novel anodic electrochemical deposition of cerium oxide film method, a suitable ligand is used to complex Ce(III) in the solution at high pH, thus stabilizing free Ce(III) in solution, with the oxidation voltage or corresponding current applied to an electrode to produce CeO2 films on the substrate. Figure 3.4B and C demonstrate XRD patterns for CeO2 film that is anodically electrodeposited on stainless steel substrate with potentiostatic mode and galvanostatic mode, respectively. Compared with cathodic method, the anodically grown CeO2 films demonstrate superior properties of homogeneity and adherence to substrates. A systematic study of the experimental parameters has been done to optimize the conditions for anodic deposition of CeO2. The comparisons of the electrodeposited CeO2 XRD patterns with the JPDF are collected in Table 3.1-3.3. 125 500 ss(110) A 400 ) s p 300 (111) (c y it s 200 ss(211) n te (220) n I (200) (311) ss(200) 100 (422) (400) (331) 0 20 40 60 80 100 400 (111) B 350 300 ) 250 s p 200 (c (220) ss(110) ity s 150 n (200) (311) te ss(211) In 100 (400) ss(200) 50 (331) 0 20 40 60 80 100 400 C 300 (111) ss(110) ) s ss(211) p 200 (c ity s (220) n ss(200) e t 100 (200) (311) In (400) (422) 0 20 40 60 80 100 2 θ (degrees) Figure 3.4 X-ray diffraction patterns for electrochemically deposited CeO2 films by (A) cathodic 2 base generation method: 0.1M Ce(NO3)3, 0.5 M NH4NO3, current density is 0.60 mA/cm ; (B) anodic oxidation method: potentiostatic mode 0.1M Ce(NO3)3, 0.1M acetic acid, pH = 7.5 and applied potential is 1.1 V; (C) anodic oxidation method: galvanostatic mode, 0.1M Ce(NO3)3, 2 0.1M acetic acid, pH = 7.5 and current density is –0.06 mA/cm 126 Table 3.1 Comparison between experimental and PDF data (Experimental data correspond to the peaks in Figure 3.4A) Experimental Data PDF#81-0792 h k l 2θ I/I0 2θ I(f) 111 28.40 100 28.549 100 200 32.91 37 33.075 29 220 47.21 45 47.474 46 311 56.20 33 56.334 36 222 58.20 20.7 59.077 7 400 69.21 7.4 69.405 6 331 76.58 13.5 76.690 13 Table 3.2 Comparison between experimental and PDF data (Experimental data correspond to the peaks in Figure 3.4B). Experimental Data PDF#81-0792 h k l 2θ I/I0 2θ I(f) 111 28.40 100 28.549 100 200 33.20 28.4 33.075 29 220 47.40 50. 47.474 46 311 56.35 34.1 56.334 36 222 59.00 59.077 7 400 69.15 10.8 69.405 6 331 76.78 18.3 76.690 13 127 Table 3.3 Comparison between experimental and PDF data (Experimental data correspond to the peaks in Figure 3.4C). Experimental Data PDF#81-0792 h k l 2θ I/I0 2θ I(f) 111 28.54 100 28.549 100 200 33.03 29 33.075 29 220 47.35 47 47.474 46 311 56.34 33 56.334 36 222 59.16 7 59.077 7 400 69.21 6 69.405 6 331 76.78 13 76.690 13 Table 3.4 Cerium-ligand complexation constants Salicylic Lactic Citric Acid Ligands Acetate Oxalate EDTA Acid Acid 1.68 2.66 2.76 N/A 6.52 16.80 logK 2.69 6.18 A group of organic ligands was selected as the complexing agents for the Ce(III) in high pH environment (Table 3.4): acetic acid, lactic acid, citric acid, oxalic acid, salicylic acid and EDTA. The formation constants for the complexation of these ligands with Ce(III) all fall within a range of 5x102 to 3x106.32 The fairly high values of formation constants predicted that equilibrium favors complex formation. However, practically, acetic acid and lactic acid are the only two ligands that result in the deposition of the CeO2 on substrates. It is seen from Table 3.4 the formation constants for the complexation of acetic acid and lactic acid with Ce(III) are relatively lower in this group, which paves a way for the formed Ce(III)-L complex to dissociate; on the other hand, with EDTA as a ligand, it can complex Ce(III) tightly with little possibility of Ce(III) free species released easily. It is suggested the existence of Ce(III) free species is closely 128 related with the formation of CeO2 films on substrates. The CeO2 deposition mechanism will be discussed in chapter 5. The employed substrates in the anodic elctrodeposition of CeO2 involve stainless steel, nickel, platinum, glassy carbon, N-silicon and P-silicon. The deposition results are shown in Table 3.5 and Figure 3.5. The deposition succeeds in all the metal substrates and N-silicon, but failed with glassy carbon and P-silicon. 400 400 CeO (111) A 2 B 350 CeO (111) 300 2 300 250 ) ps c (cps) 200 ( y CeO (220) 200 y it 2 it s 150 n tens CeO (200) CeO (311) e n 2 2 I Int CeO (220) 2 100 CeO (331) 100 CeO (222) 2 2 50 CeO (400) 2 0 0 20 40 60 80 100 20 40 60 80 100 2θ (degrees) 2θ (degrees) 400 400 C D 300 300 ) ) s CeO (111) s p 2 p c ( 200 200 (c y y t it i s s n n e CeO (200) CeO (111) t 2 te 2 In In 100 100 0 0 10 20 30 40 50 60 70 80 90 100 110 20 40 60 80 100 2θ (degrees) 2θ (degrees) Figure 3.5 X-ray diffraction patterns of CeO2 film on different substrates (A): stainless steel; (B): nickel; (C) platinum; (D): n-silicon wafer. Deposition condition: 0.1M Ce(NO3)3, 0.1M acetic acid, pH=7~8, and applied potential is 1.10 V 129 Table 3.5 The employed substrates for deposition of CeO2 thin films Stainless N- P- Glassy Substrate Nickel Platinum Steel Silicon Silicon Carbon Plating Result Film Film No Film Film Film No Film Based on the experimental results for the ligands and substrates, further deposition parameter studies were carried out with acetic acid as complexing agent and stainless steel as the deposition substrate. Both potentiostatic mode and galvanostatic mode were utilized for the deposition of cerium oxide films. For galvanostatic deposition, a variety of current densities are studied. Figure o 3.6 shows the XRD of CeO2 films grown at 25 C for anodic current densities of (A) –0.06, (B) – 0.30, (C) –0.90 mA/cm2. The deposited films exhibit a random XRD pattern and with increasing current, the relative intensity of phase (111) decreases. This trend indicates that lower current (i.e. –0.06 mA/cm2) is conducive to the formation of preferred oriented crystals. This point has been confirmed in the following temperature effect investigation. In the potentiostatic mode deposition, the films demonstrate random orientation for the applied potential range from 0.7 V to 1.l0 V vs. SCE without distinct difference from each other. Deposition temperature and the effect on phase composition or microstructure is another important parameter for the electrodeposition of crystalline CeO2 films. It is well known that increasing temperature will increase the steady state current density, which can be indicated in the cyclic voltammetries for the cerium-acetic system without adjusting pH (Figure 3.7). The X- ray diffraction patterns are shown in Figure 3.8 for CeO2 films on stainless steel deposited by galvanostatic mode (Figure 3.8A) and potentiostaic mode (Figure 3.8B) at temperatures of (A) 25, (B) 50, (C) 70 and (D) 80oC. The anodically produced films deposited at 25oC exhibit 130 random XRD patterns for both potentiostatic and galvanostatic modes. However, with increasing temperature, for the galvanostatic deposition, the CeO2 (111) reflection increases in intensity producing films with a preferred (111) orientation as indicated in Figure 3.9A. The intensity ratios (I/Io) and the ratio of the CeO2 (111) reflection to the substrate (110) reflection of the patterns in Figure 3.8A are shown in Table 3.6. While the (111) intensity of the CeO2 increases, the substrate peak remains about the same intensity, indicating that the film thickness is approximately the same at all temperatures. The film thickness for all depositions tends to be thin since CeO2 is not very conducting. However, in the regard to the potentiostatic mode deposition, the deposition bath temperature increase did not induce the preferred orientation. Contrary to galvanostatic mode deposition, the ratio between intensities of CeO2 (111) and substrate main reflection gives random trend. It is concluded that the low current density at high temperature is the best condition for the formation of preferred oriented crystals of CeO2. This can be explained with the kinetic theory, low current density gives a low reaction rate while high current density keeps generating new crystals before the previously formed crystals have a chance to orient. And a certain critical temperature (e.g. 50oC) is required to form the crystals instead of agglomerate of amorphous phases. The potential for the galvanostatic deposition with current density of -0.06 mA/cm2 starts in the range of 0.7~1.0 V versus SCE and with progress of the deposition, the potential increases to 0.9~1.1 V. However, these corresponding potentials are applied in the potentiostatic deposition, which results in random structure not preferred oriented films. The discovery suggests that the formation of preferred oriented CeO2 film in two steps with galvanostatic deposition. At the beginning of the deposition, the random structure grows and later on at the potential range of 0.9~1.10V that is the preferred potential range for the generation of CeO2 film, 131 the oriented phase is formed with very slow reaction rate. From another aspect, when potentiostatic deposition is used and a low potential which is corresponding to the start potential (0.7~0.9 V) for galvanostatic deposition is applied, the film random growth dominates while a high potential (0.9~1.1 V) is applied, the starting reaction rate is too fast to allow the single- phase to grow larger. (ss) (ss) (111) y C it s n e B D t n A e rr u 10 20 30 40 50 60 70 80 90 100 110 C 2θ (Degrees) Figure 3.6 X-ray diffraction patterns of cerium oxide films deposited on stainless steel with galvanostatic deposition, from a solution of 0.1 M Ce(NO3)3 and 0.1 M acetic acid pH = 7.5 at applied currents densities of (A) –0.06, (B) –0.30, (C) –0.9 mA/cm2. Y-axis represents x-ray intensity in cps 132 1 A 0 -1 A) -3 0 0 -2 T = 70 C 1 0 ( T = 60 C t 0 n T = 50 C e -3 r 0 r T = 40 C 0 Cu T = 30 C -4 T = 250C -5 -6 1.5 1.0 0.5 0.0 -0.5 -1.0 Potential(V) 4 B 2 A) 0 -3 0 t (1 n -2 rre u o C 30 C -4 o 50 C 70oC -6 1.0 0.5 0.0 -0.5 Potential (V) Figure 3.7 Cyclic voltammetries for Ce(NO3)3-acetic acid system at different temperatures: (A) pH = 2~3; (B) pH = 7~8 133 a (111) D C re tu B ra e p A m e T 10 20 30 40 50 60 70 80 90 100 110 2θ (Degrees) ss b (111) (111) (111) C B re tu ra e A p m e T 0102030405060708090100110 2θ (degrees) Figure 3.8 X-ray diffraction patterns of cerium oxide films deposited on stainless steel with anodic galvanostat method from a solution of 0.1M Ce(NO3)3 and 0.1 M acetic acid pH = 7.5, Y-axis represents x-ray intensity in cps.(a) Galvanostatic mode: deposition current density of –0.06 mA/cm2 at (A) 25, (B) 50, (C) 70 and (D) 80oC. (b) Potentiostatic mode: deposition potential of 1.10V at (A) 25, (B) 50 and (C) 70oC 134 Table 3.6 The preferred orientation of CeO2 at temperatures of: 25, 40, 50, 60, 70 and 80oC (Data from Figure 3.7a). 2θ(o) hkl 25oC 40oC 50oC 60oC 70oC 80oC 28.54 111 100 100 100 100 100 100 33.03 200 29 ------47.51 220 47 12 14 3.0 ------56.34 311 33 2 2 ------59.36 222 7 4 3 3 3 3 CeO2(111)/ ss(110) 0.684 1.25 5.208 3.096 20.41 22.73 The Pourbaix diagram (Figure 3.1) for cerium indicates that CeO2 is stable at a pH above 9 and for potentials above 0.7 V vs. SHE. After a series of depositions, it was noticed that CeO2 films were produced starting above pH 7. Figure 3.9 illustrates the cyclic voltammograms for the cerium nitrate complexed with acetic acid at different pHs. Figure 3.10 show the influence of pH 2 o on the texture of the CeO2 films deposited on a stainless steel substrate at –0.06 mA/cm , 70 C and a pH of (A) 7.5, (B) 8.5, and (C) 10.5. As seen from the XRD patterns, for a solution pH between 7 and 9 (Figure 3.9A and B), the deposited CeO2 film exhibits a preferred (111) orientation, but at a solution pH of 9 to 11 (Figure 3.9C), CeO2 films form on the substrate with a random orientation. These films formed at pHs 9 to 11 probably exhibit a random orientation since at these conditions CeO2 powder is also generated and falls to the bottom of the cell during deposition. At solution pHs higher than 11, no CeO2 deposits on the electrode surface, although a CeO2 powder (confirmed by XRD) is generated and settles at the bottom of the reaction cell. 135 0.005 0.000 -0.005 -0.010 ) -0.015 A ( t -0.020 n rre -0.025 pH = 2.56 u C -0.030 pH = 7.77 pH = 8.73 -0.035 pH = 10.94 -0.040 pH = 12.30 -0.045 1.5 1.0 0.5 0.0 -0.5 -1.0 Potential (V) Figure 3.9 Cyclic voltammetries for Ce(NO3)3-acetic acid system at different pH values (111) A B C pH 10 20 30 40 50 60 70 80 90 2θ (Degrees) Figure 3.10 X-ray diffraction patterns of cerium oxide films deposited at a pH of (A) 7.5, (B) 8.5 and (C) 10.5. Deposition temperature is 70oC. j = -0.06 mA/cm2. Y-axis represents x-ray intensity in cps 136 3.3.2. Evaluation of Particle Sizes and Lattice Strains Particle size is a significant parameter for the properties and application of the cerium oxide films. It is revealed that smaller particle sizes and minimum film thickness of cerium oxide films lead to improvements of corrosion protection. Other researchers have demonstrated this for 33,34 CeO2 films deposited by sol-gel and reactive sputtering methods. Raman spectroscopy, as an effective tool to investigate the structure of cerium oxide, is 35-40 reported sensitive for the crystalline size of the CeO2. Spanier reported Raman first-order -1 -1 signals of bulk CeO2 with an intense peak at 464 cm , two weak peaks between 550 ~ 600 cm and second-order peaks at 261, 368, 670, 1184 and 1277cm-1. It is pointed out that major peak of 464 cm-1 shifts to lower frequency if the particle sizes are in nanoscale.40 In Wang’s report, when the crystallite is larger than 20 nm, there will be only one major shift at 463 cm-1 in Raman spectroscopy, however, with the crystal size decreasing below 10 nm, two more shifts at 270 (peak B), 315 cm-1 (peak C) appear and the intensities increase with the decreasing crystal 35 sizes. These two shifts are distinctive for nanosized CeO2 and cannot be distinguished in 35,37 single-crystal and polycrystalline CeO2. The Raman spectra of anodically deposited CeO2 are shown in Figure 3.11. The spectra (A) stand for the CeO2 film potentiostatically deposited on stainless steel substrate with applied potential of 1.10 V at 25oC. Spectra (B) are for 2 galvanostatically deposited CeO2 film with applied current density of –0.06 mA/cm . 137 10000 Peak A A 8000 ) s 6000 (cp ity s 4000 n e Int 2000 0 0 500 1000 1500 2000 Wavenumber (cm-1) B 1000 800 600 (cps) y t i s 400 ten In 200 0 0 500 1000 1500 2000 Wavenumber (cm-1) Figure 3.11 Raman spectra of CeO2 anodically electrodeposited on stainless steel by (A) potentiostatic mode: E = 1.10 V, T = 25 oC; (B) galvanostatic mode: j = -0.06 mA/cm2, T = 70 o C 138 The reproducible spectra in each Raman spectra figure demonstrate the homogeneities of -1 -1 the deposited CeO2 film samples. The peak A is observed around 450 cm , about ~14 cm lower -1 35 relative to the reported scattering signal (464 cm ) for microsized CeO2, indicating the nanocrystallites were produced in electrodeposition. The potentiostatic deposited CeO2 has higher arbitrary intensity than the galvanostatic deposited CeO2 because the film generated from potentiostatic deposition can result in thicker films than the galvanostatic deposition. There are two peaks appearing on the shoulder of peak A, in which the higher wavenumber shift can be attributed to a combination of peak B and C in Wang’s report. However, the shoulder peaks did not show up for the film deposited galvanostatically at temperature of 70oC which means that the CeO2 film formed at this deposition condition has higher crystallite size than the lower temperature production. The follow-up lattice calculation according to XRD patterns supports this observation from Raman spectra. Crystallite size and strain measurements can be calculated for these films from the XRD data. The crystallite sizes were estimated from the line broadening in the XRD patterns run at a scan rate of 1 o/min. Since several reflections can be applied in the calculation of crystallite size, Williamson and Hall method (Equation 2.36) is employed in calculation, a plot of βrcosθ versus sinθ yields a straight line with a slope of η and intercept of kλ/L.41 Figure 3.12 demonstrates the Williamson and Hall plots for the galvanostatic (A) and potentiostatic (B) deposited CeO2 film at temperature of 25 oC. The crystallite sizes calculated from the intercept of plot A and B are 6.0 and 7.5 nm, respectively. This measured crystallite size range for the electrodeposited films corresponds to sizes reported for cerium oxide particles prepared by a chemical solution technique.42 The slope of the lines are almost flat indicating minimal strain, η, in the films. 139 For XRD patterns which exhibit preferred orientation, Scherrer equation (Equation 2.37) 43 can be used to estimate the crystallite size. From (111) reflection of the CeO2 films, the crystallite sizes for preferred oriented films ranged from 6 to 20 nm (Figure 3.13). At higher deposition temperature, the crystallite sizes are in the µm range. 0.05 A 0.04 0.03 θ cos r 0.02 β 0.01 0.00 0.20.30.40.50.60.7 sinθ 0.06 B 0.05 0.04 0.03 θ 0.02 cos r B 0.01 0.00 -0.01 -0.02 0.4 0.5 0.6 0.7 sinθ Figure 3.12 Williamson-Hall plot of βcosθ versus sinθ for the XRD patterns for (A) 2 o galvanostatic deposited CeO2 film: j = -0.06mA/cm , T = 25 C and (B) potentiostatic o deposited CeO2 film: E = 1.10V, T = 25 C 140 18 16 ) 14 m n ( e 12 z n Si i 10 a Gr 8 6 20 30 40 50 60 70 80 Temperature (oC) Figure 3.13 Grain size versus deposition temperature for galvanostatic deposition of CeO2 on 2 stainless steel with j = -0.06mA/cm 3.3.3 Determination of Oxidation States with XPS Investigations of the oxidation state in CeO2 films fabricated with various synthesis techniques include electron energy loss spectroscopy (EELS), bremsstrahlung isochromat spectroscopy (BIS), UV photoelectron spectroscopy (UPS), X-ray absorption spectroscopy (XAS), and X-ray photoelectron spectroscopy (XPS).12,22,27-29,44-48 There have been a few studies of the electrodeposition of cerium oxide films. Balasubramanian and co-workers studied electrochemically deposited cerium hydroxide films and investigated the oxidation state of the cerium utilizing X-ray absorption fine structure 22 technique (XAFS). The anodic deposition resulted in films with stoichiometry close to CeO2, whereas cathodic base generation method produced a mixture of Ce(III)/Ce(IV) oxidation states in the films, with the Ce(III) phase oxidizing upon exposure to air. X-ray photoelectron spectroscopy (XPS) was employed to characterize thin films of cerium oxides and determine the chemical composition and oxidation state of the as-produced 141 cerium oxides, since the difference between Ce3+ and Ce4+ can be illustrated from the XPS signal shape for the Ce 3d signal and from the valence band spectra. The XPS signal for the Ce 4d also exhibits distinct characteristics for Ce4+ and Ce3+.49 Experimental investigations revealed XPS of both Ce 3d and Ce 4d are complex, and theoretical studies of XPS indicated different possible mechanisms for splitting in the Ce 3d and Ce 4d spectra.44,48,50-52 XPS Ce 3d spectra are illustrated in Figure 3.14 for a nanocrystalline cerium oxide film electrodeposited on stainless steel in which Figure 3.14A shows the rough spectrum and Figure 3.14B demonstrates the curving fitting of the resultant spectrum. There are 8 peaks assignments in the spectra, which are labeled according to the convention established by Burroughs,29 where peaks U, U″, U′′′ and V, V″, V′′′ refer to 3d3/2 and 3d5/2 respectively and are characteristic of Ce(IV) 3d final states; while U′, V′ refer to 3d3/2 and 3d5/2 respectively and are present for Ce(III) 3d final state.27,53,54 The high binding energy doublet V′′′/U′′′ at 916.9 eV and 898.5eV are assigned to the final state of Ce(IV) 3d94f0 O 2p6. The split-orbital separation of 18.4 eV is close to the established 18.3 eV. According to experimental studies, the high binding energy peak U′′′ is the best distinguished characteristic peak to differentiate Ce(IV) from Ce(III). The higher binding 2,55 V′″ and U″′ peaks are not observed for purely trivalent ionic cerium compounds (i.e. Ce2O3). Doublets V″/U″ at 907.5 eV and 888.9 eV were attributed to the hybridization state of Ce(IV) 3d94f1 O 2p5, and doublets V/U at 901.0 eV and 882.3 eV correspond to the state of Ce(IV) 3d94f2 O 2p4. The interaction between the Ce 3d and Ce 4f infers the decrease of the Ce 4f binding energy. The two lower binding energy doublets are called the “shake down” states in which the O2p electrons are transferred to Ce 4f orbital.27 142 Doublets V’/U’ at 904.2 eV and 885.9 eV are derived from the Ce(III) 3d94f2O2p5 final state. Compared to the other three doublet contributions for Ce (IV), these two peaks are relatively less intense. And another doublets Vo/Uo (corresponding to the Ce(III) 3d94f1 O 2p6 final state), typically located at 880.6 eV and 898.9 eV, do not appear for the experimental runs of the electrodeposited nanocrystalline films. The weak peaks of the V’/U’ and Vo/Uo absence in the XPS Ce 3d spectra indicate the electrodeposited CeO2 films consist mostly of the Ce(IV) 44,54,56 oxidation state; even the X-ray is prone to induce the reduction of the CeO2. The ratio between Ce(IV) and Ce(III) for cerium oxide has been shown to affect the catalytic activity of cerium oxide.57 Hence some researchers have establish a ratio between Ce(IV) and Ce(III) using XPS data. Strydom investigated combining modified Auger parameters with 4d5/2 and 4d3/2 peaks in XPS to indicate the ratio of Ce(IV) and Ce(III).57 Holgado evaluated 58 the proportion of Ce(III) and Ce(IV) in reduction of CeO2 with factor analysis method. El Fallah established a calculation of the ratio from contributions purely in the XPS spectra, which is convenient and will be adopted in this discussion.51 Since the features characteristic of the Ce(III) states are from the contributions of Uo, Vo, V′, and U′, the following equation can be used to estimate the cerium oxide stoichiometry: Ce(III) = (U +V +V ' +U ' ) /[(U +V ) + (U n' +V n' )] (Equation 3.1) 0 0 0 0 ∑n' where n′ is for all states. Using this method, the fitted peak areas in the XPS spectra (Figure 3.14B) can be used to estimate the contributions of Ce(IV) and Ce(III) for the electrodeposited nanocrystalline films, indicating the electrochemically deposited cerium oxide is non- stoichiomtric. It is this non-stoichiometry property that furnishes cerium oxide the ability to store or release oxygen as a catalyst in oxygen-related reactions59 and this defective characteristic is the primary reason for the electrical and ionic conductivity of cerium oxide.60 Although the 143 electrodeposited cerium oxide film exists as the Ce(IV) and Ce(III) mixed oxides, the 4+ Ce(IV)/Ce(III) ratio of 4.12 which is corresponded to CeO1.90 shows a preference for the Ce oxidation state using the anodic deposition method. 17000 U''' Ce 3d U V''' 16000 U' U'' V 15000 ) s V' p 14000 (c V'' ity 13000 s n te 12000 In 18.4 11000 10000 930 920 910 900 890 880 870 Binding Energy (eV) V B 4000 V''' 3000 U ) V' s U''' p (c 2000 ity U' s V'' n te U'' In 1000 0 930 920 910 900 890 880 870 Binding Energy (eV) Figure 3.14 X-ray photoelectron spectroscopy pattern of Ce 3d for a cerium oxide film deposited on stainless steel (A) raw data; (B) curve fitting result 144 Figure 3.15 shows the O 1s spectra (Figure 3.15A) and curve fitting (Figure 3.15B) for the electrodeposited CeO2 film. There is still controversy in the interpretation of O 1s XPS for cerium oxide.21,61,62 There are two peaks observed for the O 1s spectra. The first peak located at a binding energy of 529.6 eV is the low binding energy (LBE) component for O 1s XPS.54 This LBE peak has been assigned as corresponding to the lattice oxygen in Ce(IV) oxide.3,27,51,57 The second peak that appears at a binding energy of 531 - 532 eV is the high binding energy (HBE) component for the O 1s spectra. Some researchers state that the peak arises from hydroxyl groups on the surface, oxygen chemisorbed on the surface in other forms such as CO, CO2, or grain boundary impurities,27,51,53 and others have argued that this peak is related to the Ce(III) oxide on the surface.44,62,63 Recent grazing incident XPS studies indicate the presence of the peak 49 may be due to oxide ions in the defective CeOx (x < 2) film. The ratio of peak areas between the O 1s and Ce 3d can be used to estimate the cerium oxide stoichiometry, when corrected with atomic sensitivity factors.54 Atomic sensitivity factors (ASF) for the calculations were: 7.39 for Ce 3d and 0.73 for O 1s. For the electrodeposited film, the calculated stoichiometry is CeO1.86. This matches CeO1.90 calculated from the Ce 3d spectrum using the Ce(IV)/Ce(III) ratio. The area ratio between the low and high binding energy peaks has been employed to estimate Ce(IV)/Ce(III) ratio,44 it is not adopted owing to the present contention on the O 1s assignments. 145 7000 A 6500 6000 ) 5500 ps c ( y 5000 it ns e 4500 Int 4000 3500 3000 542 540 538 536 534 532 530 528 526 524 Binding Energy (eV) 3500 B 3000 2500 ) s 2000 p (c 1500 ity s n te 1000 In 500 0 -500 542 540 538 536 534 532 530 528 526 524 Binding Energy (eV) Figure 3.15 X-ray photoelectron spectroscopy pattern of O1s for cerium oxide film deposited on stainless steel. (A) raw data; (B) curve fitting result Compared with the Ce 3d spectra, the Ce 4d is relatively noisy (Figure 3.16A). There are six features present in the Ce 4d spectra for the electrodeposited cerium oxide film (Figure 3.16B). The appearance of the Ce 4d spectra for the nanocrystalline electrodeposited films is 29 similar to the Ce 4d spectra for CeO2 reported by Burroughs et.al., and Ce 4d spectra for CeO2 146 (100) samples reported by Mullins et.al..27,28 According to the notation and interpretation of Mullins, the doublets X″′ located at 122 eV and W′′′ located at 126 eV can be ascribed to the 9 6 0 hybridization state of Ce(IV) 4d O 2p Ce 4f , in which X′′′ and W′′′ correspond to Ce 4d5/2 and 27 Ce 4d3/2 components respectively. The intensities of X′″ and W′′′ are related to the proportion of Ce(IV) in the sample and are completely absent for spectra of Ce(III) oxides.28 The increase intensity of the high binding energy features (>120 eV) indicate an increase in Ce4+ states.64 There is no agreement for the assignment of the other peaks in the Ce 4d spectra. According to Burroughs, we can interpret the Ce 4d spectra similar to Ce 3d, but this interpretation confers some disagreement, and Mullins uses a different system of notation and different assignments for the lower binding energy features. Mullins assigns the peaks corresponding to A, B, and C as the Ce4+ with a Ce 4d9 O 2p5 Ce 4f1 + Ce 4d9 O 2p4 Ce 4f2 final state. Mullins et.al. also observed another peak (105.4eV) for CeO2 samples corresponding to the Ce(III) oxidation state, however this peak is not observed in our experiment. The emission between 117 and 120 eV present in the spectra was also seen by Mullins, but was not identified. 147 3600 B A 3400 X''' A 3200 W''' C 3000 Unk ps) (c 2800 nsity 2600 Inte 2400 2200 2000 140 130 120 110 100 Binding Energy (eV) 1200 B A B 1000 800 ) X''' s C p 600 (c W''' ity Unk s n 400 te In 200 0 140 130 120 110 100 Binding Energy (eV) Figure 3.16 X-ray photoelectron spectroscopy pattern of Ce 4d for cerium oxide film deposited on stainless steel. (A) raw data; (B)curve fitting result The valence band XPS of the samples were recorded as well. A representative valence band XPS is shown in Figure 3.17A for the electrodeposited sample. In the range of 0~10eV, two peaks can be distinguished (Figure 3.17B). The broad spectrum ranged between 3~8eV, attributed to the emission from the O 2p valence orbitals, is the characteristic of the fully 148 oxidized CeO2. It is suggested that this broad band can be deconvoluted into two distinct spectra features, the first is located at 4.5 eV which corresponds to the hybridization of O 2p with Ce 4f, and another at 6.5 eV which is usually ascribed to the hybridization of O 2p and Ce 5d bands.27 Emission from the localized Ce 4f orbital at a binding energy around 1.7 eV, though a little different from the reported value of Ce(III) characteristic peak (~2 eV), can be related to the evolution of Ce(III) in the film. However, the resolution was poor for the intensity of the x-ray source used making it difficult to observe fine features. Any conclusion from the valance band spectra is qualitative at best, since the spectra are very noisy making it difficult to deconvolute and curve fit the features.44,65 It is concluded that anodically electrodeposited cerium oxide is non-stoichiometric and Ce(IV) oxide dominates in composition. It is the non-stoichiometry property that promotes cerium oxide as an excellent catalyst for bulk oxygen-related reactions. 3.3.4 Morphology The morphology of the cerium oxide film is shown in Figure 3.18. The scanning electron microscopy (SEM) reveals that the films produced by anodic deposition has a smooth surface. From the SEM graph, extensive cracking can be seen of the film on the substrate surface. This is probably due to a large mismatch between the substrate and film20 or drying process.16 This type of cracking of CeO2 has also been observed for other wet chemical methods of deposition such as electrophoresis.15 The release of strain caused by this large mismatch is evident in the Williamson-Hall plot (Figure 3.12), where the slope is essentially horizontal, indicating strain in the system has been released. As the deposited film gets thicker, the morphology demonstrates the properties of agglomerated growth (Figure 3.18A). Usually this phenomenon occurs for potentiostatic deposition since the deposition rate gets lower as the deposition progresses. While 149 in the regard to galvanostatic growth, with low current density, it takes time to form thick films, and with high current density, the formation of the crystals is so fast that powder is generated instead of film growth at the substrate surface.66 54000 A 52000 50000 ) s p (c 48000 ity s n te n 46000 I 44000 42000 1086420 Binding Energy (eV) 7000 B 6000 5000 ) s 4000 p c ( 3000 ity s n te 2000 In 1000 0 1086420 Binding Energy (eV) Figure 3.17 X-ray photoelectron spectroscopy pattern of valence band for cerium oxide film deposited on stainless steel. (A) raw data; (B) curve fitting result 150 Figure 3.18 SEM morphologies for the CeO2 films on stainless steel (A) potentiostatic deposited: E = 1.10 V, T = 25oC, x10,000; (B) potentiostatic deposited: E = 1.10 V, T = 25oC, x10,000;(C) galvanostatic deposited: I = -0.06mA/cm2, T = 80oC, x7,500 151 3.3.5 Sintering Studies of Cerium Oxide Film on Stainless Steel Both random structured and (111) preferred oriented CeO2 films were subjected to sintering studies and then XRD, XPS and SEM techniques investigate the structure, composition and texture during the sintering process. A CeO2 film produced potentiostatically together with stainless steel substrate was sintered at 300, 500, 700, 900oC. The four graphs in Figure 3.19 illustrate the Ce 3d XPS differences between the original and sintered samples. As illustrated previously, all the spectra can be deconvoluted and attributed to eight separate peaks shown in the curve fittings. With the increase of sintering temperatures, the relative intensities of the doublet peaks V′′′ U′′′ and V″ U″ increase, these two features are among the six characteristic peaks (V, V’’,V’’’ and U, U’’,U’’’) for the Ce(IV) oxidation state. In contrast, the intensity of peaks V′ and U′, characteristic features for Ce(III), decrease with increasing of sintering temperature. This indicates a Ce(IV) oxide formation and an increase of Ce4+ concentration at the surface with increasing sintering temperature. Increase of the Ce(IV)/Ce(III) ratio with sintering indicates oxidation of the film. Table 3.7 illustrated this trend, in which the Ce(IV)/Ce(III) ratio changes from 4.12 for the o original sample to 24.25 for the sample sintered at 700 C and accordingly x in CeOx increases from 1.90 to 1.98. The oxidation trend is consistent the with the discovery by Preisler’s report which claimed that sintering in vacuum reduces as produced cerium oxide to Ce6O11 whereas in air as formed cerium oxide is further oxidized to a stoichometry close to 1: 2 for Ce:O.44 The reduction effect of the vacuum sintering on CeO2 has been discussed in detail by Roméo and his coworkers.67 152 V 1500 4000 A B V''' 3000 U 1000 ) U''' V' s p (c 2000 ity s U' n V'' 500 e t In 1000 U'' 0 0 930 920 910 900 890 880 870 930 920 910 900 890 880 870 2500 2000 C D 2000 1500 1500 ) s p (c 1000 ity 1000 s n e t In 500 500 0 0 930 920 910 900 890 880 870 930 920 910 900 890 880 870 Binding Energy (eV) Binding Energy (eV) Figure 3.19 The Ce 3d spectra for cerium oxide films before (A) and after sintering at temperatures of: 300oC (B), 500oC (C), and 700oC (D). Black lines represent the experimental spectra (after background subtraction) and the colored lines are results with peak fitting In the process of sintering, the O 1s peak gradually convolutes to an asymmetrical broad peak (Figure 3.20C, D) from the obvious two-peak layouts (Figure 3.20A, B). The intensity of the HBE O 1s peak decreases with increasing sintering temperature (Figure 3.21). Either scenario of decreasing Ce(III) oxide (or hydroxide) with increasing sintering temperature or a change in the stoichiometry of the Ce(IV) component is feasible for the electrodeposited films. With increasing temperature, the value for x in CeOx approaches 2, since Ce(III) species is converted to Ce(IV) at higher temperatures. These result agrees well with the trend for the Ce(IV)/Ce(III) ratio calculated in Table 3.7. For the Ce 4d XPS (Figure 3.21), the Ce(IV) 153 percentage increases with increasing sintering temperature as seen from the increasing W″′/X′″ intensities. Due to the poor sensitivity of valance band XPS, the valence band changes related with sintering can not be determined.44,65 2500 3500 B A 3000 2000 2500 ) 1500 s 2000 p c ( y 1500 it 1000 s n e t 1000 In 500 500 0 0 -500 540 535 530 525 540 535 530 525 4500 3000 C 4000 D 2500 3500 ) s 3000 p 2000 c ( e 2500 y tl i it T s 1500 s 2000 n i x e t A In 1500 1000 Y 1000 500 500 0 0 -500 540 535 530 525 540 535 530 525 Binding Energy (eV) Binding Energy (eV) Figure 3.20 The O 1s spectra for cerium oxide films before (A) and after sintering at temperatures of: 300 oC (B), 500 oC (C), and 700 oC (D). Black lines represent the experimental spectra (after background subtraction) and the colored lines are results with peak fitting 154 Table 3.7 Peak list of Ce 3d and calculation of x in CeOx* Peak Orignal Sample Sintering 300oC Sintering 500oC Sintering 700oC Cerium Assignm Contribution ent Peak Area Peak Area Peak Area Peak Area V IV 882.30 16175.00 882.40 4137.00 882.40 581.00 882.50 5443.00 V' III 885.90 6261.00 885.60 2154.00 884.80 1524.00 885.50 765.40 V'' IV 888.90 3992.00 888.90 2077.00 888.70 4850.00 889.10 5482.00 V''' IV 898.50 7766.00 898.40 3109.00 898.20 5088.00 898.30 5785.00 U IV 901.00 7429.00 901.00 2408.00 900.90 3416.00 901.10 3538.00 U' III 904.20 4669.00 904.10 1279.00 903.20 546.80 903.40 428.90 U'' IV 907.50 2773.00 907.60 892.20 907.50 3584.00 907.90 4410.00 U''' IV 916.90 6936.00 916.90 2479.00 916.80 4189.00 916.90 4306.00 Total 56001.00 18535.20 23778.80 30158.30 Ce(III) 10930.00 3433.00 2070.80 1194.30 Ce(IV)/Ce(III) 4.12 4.40 10.48 24.25 Ce(III)/Ce(IV) 0.2425 0.2273 0.0954 0.0412 Percentage of Ce(III) 19.5200 18.5000 8.7086 3.9601 x in CeOx 1.9024 1.9075 1.956457 1.980199 * Cerium oxide sample unsintered and sintered at 300, 500, 700oC. 155 1200 550 B A A 500 B 1000 450 400 800 X''' 350 ) s 300 p 600 C (c W''' 250 ity Unk s 200 n 400 te 150 In 200 100 50 0 0 -50 140 130 120 110 100 140 130 120 110 100 600 600 C D 500 500 400 400 ) s p 300 300 (c y t i s n 200 200 te In 100 100 0 0 140 130 120 110 100 140 130 120 110 100 Binding Energy (eV) Binding Energy (eV) Figure 3.21 The Ce 4d spectra for nanocrystalline cerium oxide films before (A) and after sintering at temperatures of: 300oC (B), 500oC (C), and 700oC (D). Black lines represent the experimental spectra (after background subtraction) and the colored lines are results with peak fitting Figure 3.22 shows the XRD patterns of the unsintered and sintered sample. Figure 3.23 illustrates the trend of particle size with sintering temperature. The particle sizes are calculated using CeO2 (111) reflection in Figure 3.23 with Scherrer equation (Equation 2.37). The o crystallite sizes of the electrodeposited CeO2 films remain less than 10 nm up to 500 C sintering temperature and increases as sintering temperature increases above 500oC. Crystallite sizes were not measured above sintering temperatures of 700oC, since the crystallite size was past the measurement range for X-ray diffraction calculations. It is noticed that sintering at 500oC and higher temperature induces stainless steel substrate components to migrate to the surface and further oxidize and corrode, combined by XRD patterns of CeO2 films (Figure 3.22D and E). 156 o Sintering at 900 C, the XRD reflections (marked in Figure E) for CeO2 fluorite structure are very sharp relative to those in 700oC-sintered sample pattern and the FWHMs are smaller than the instrumental broadening, proving that the particle size increases dramatically in high temperature sintering. Morphological changes of CeO2 film during sintering are recorded by SEM pictures (Figure 3.24). The texture of unsintered thick film is observed as agglomerates with pores or voids. As the sintering temperature increases, the grains gradually develop. Until at 700oC, the grains have uniform properties with round shape and ~500 nm size.20 Also the sequenced grains, some agglomerates even large crystal growth of CeO2 piece are revealed and simultaneous appearance as bright area features relative to the film surface indicates the sitting of the substrate components and the oxidized form (Figure 3.24G, H). Sintering study ends at temperature of 700oC because a large quantity of the substrate related impurities interferes with the surface of the film according XRD patterns (Figure 3.22D,E). 157 8 6 E 4 x cps) 2 2 0 0 I(1 4 D 2 x cps) 2 0 0 I(1 s) 20 C 10 x cp 2 0 I(10 6 s) 4 B x cp 2 2 0 I(10 3 A 2 x cps) 1 2 0 0 I(1 20 40 60 80 100 2θ (degrees) Figure 3.22 The XRD patterns for cerium oxide films electrodeposited in potentiostatic mode: E = 1.10V vs. SCE, T = 25oC. (A) unsintered and sintered at temperatures of: (B) 300 oC; (C) 500 oC , (D) 700 oC and (E) 900oC 158 25 20 ) m (n ze i 15 S le As deposited rtic a 10 P 5 0 200 400 600 800 Sintering Temperature (oC) Figure 3.23 Crystallite size versus sintering temperature calculated from x-ray diffraction data o of electrodeposited CeO2 films formed at T=25 C with applied potential of 1.10V vs. SCE 159 A B C D E F G H Figure 3.24 SEM morphologies of the CeO2 film electrodeposited with potentiostatic mode (A) unsintered: x10,000; (B) unsintered: x 5,000; (C) sintered at 300oC: x 10,000; (D) sintered at 300oC: x 5,000; (E) sintered at 500oC: x 10,000; (F) sintered at 500oC: x5,000; (G) sintered at 700oC: x 10,000; (H) sintered at 700oC: x 5,000 160 The (111) preferred oriented CeO2 film was sintered with the same parameters as well and the corresponding results are listed for comparison. Figure 3.25 and Figure 3.26 demonstrates the X-ray patterns and crystallite size calculation with sintering temperature increases. The CeO2 (111) in XRD patterns gets shaper and more intense as the sintering temperature increases until 900oC and accordingly the estimated crystallite size grows from under 10 nm unsintered to ~120 nm at 900oC. The composition change of the overlay is disguised in the main patterns due to the tremendous intensity of the preferred orientation of CeO2 (111). Insert in Figure 3.25D is the amplified pattern of the bottom portion of this XRD lines which include all other phases of CeO2 (marked) and some phases from substrate. And this ensures the contention discussed previously that the migration and oxidation of substrate occurs during thermal treatment.68 As the temperature is raised to 1100oC at a constant rate and the thin electrodeposited CeO2 films on substrate were further isothermally heated for 5h. XRD pattern of the film surface shows a dramatic change in both the composition and orientation. The CeO2 reverts to a random arrangement with 100-fold lower intensity. Accompanied is the evolution of strong substrate signals indicating the higher concentration of substrate components migrate to surface. Particle size change with sintering temperature of the preferred oriented CeO2 film (Figure 3.26) follows the same track as of the random structured CeO2 (Figure 3.23). Both films reach the particle size minimum at sintering temperature of 300oC. It is reasonable to contribute this to the drying and dehydration of green films.16 The microstructure features of the preferred oriented CeO2 film with sintering are displayed in Figure 3.27. Due to thin film and nanoscale crystal of the starting sample, the unsintered CeO2 film only exhibits as a very smooth surface with cracking from mismatch of the substrate and deposit (Figure 3.27A). It is stated that the sintering is an irreversible process, the defect initialized from the beginning would last during the 161 course, even probably develop.69 Therefore, the cracking of the film retains during this sintering and with the shrinkage of sample, the cracking extends its width (Figure 3.27B). As the sintering temperature reaches 900oC and even 1100oC, the grown crystal shape can be observed with SEM, in which the film is exposed with equaxial grains and well-developed grain boundaries. The homogeneous shape-and-sized grains are tightly arranged without voids and pores in between except the deposit-formed cracking (Figure 3.27C,D). Another apparent characteristic in o 1100 C-sintered sample is the corrosion progress on the CeO2 surface, which can be discerned from the cloudy bulk and light appearance contrast with the dark color of CeO2 film under scanning electron microscopy. 162 4 3 E 2 x cps) 2 1 0 0 I (1 2 (1 1 1 ) (400) (111) D (3 1 1 ) (422) x cps) (220) 4 1 (200) (3 3 1 ) 0 20 40 60 80 100 I (1 0 1.0 C x cps) 0.5 4 0 0.0 I (1 6 B 4 x cps) 3 0 2 0 I (1 8 A x cps) 3 4 0 0 I (1 20 40 60 80 100 2θ (degrees) Figure 3.25 The XRD patterns for cerium oxide films electrodeposited in galvanostatic mode: j = -0.06mA/cm2, T = 80oC. (A) unsintered and sintered at temperatures of: (B) 300 oC; (C) 500 oC , (D) 900 oC and (E) 1100oC 163 120 100 ) 80 m (n e z i 60 S le c 40 rti a P 20 0 0 200 400 600 800 1000 Sintering Temperaure (oC) Figure 3.26 Crystallite size versus sintering temperature calculated from x-ray diffraction data 2 of electrodeposited CeO2 films at temperature with applied current density of –0.06mA/cm 164 A B C D o Figure 3.27 SEM morphologies for CeO2 deposited with galvanostatic mode at 80 C, j = -0.06 mA/cm2 ; (A) unsintered: x 7,500; (B) sintered 500oC: x 7.500; (C) sintered at 900oC: x 7,500; (D) sintered at 1100oC: x 7,500 3.3 4. Corrosion Test Results Linear polarization and Tafel plots were applied to test the corrosion protection effect of the deposited cerium oxide films (Figure 3.28). Calculated from the Tafel plots, the corrosion current decreases from ~ 44 nA for the substrate only to 4 nA for the CeO2 film coated substrate in a 0.1 M NaCl solution. Correspondingly, the Rp decreases from 0.49 ΜΩ for the substrate to 5.42 MΩ for the film coated stainless steel. The corrosion of stainless steel in chloride solution can be categorized as passivation in which the corrosion current is exceptionally lower than that in aggressive acidic solution. A suggested mechanism for the corrosion of stainless steel in chloride solution is illustrated in 165 Figure 3.29. A passive metal hydrated oxide film forms on metal surface, which protects the further dissolution of metal. The metal ions not involved in the structure are also complexed by water molecules. The chloride ions of solution break down this intact protective coating through substituting water molecules in the complex structure and metal chloride is dissolved in solution. In addition, the corrosion of stainless steel in neutral chloride solution is characteristic of noble corrosion potential relative to the usual active-passive transition area in acidic solution, which is manifested in the positive potentials vs. SCE for both coated and uncoated stainless steel in 70 polarization plots (Figure 3.28). The CeO2 coated stainless steel exhibits a corrosion potential positive of the green stainless steel, indicating an anodic corrosion inhibition mechanism involved for CeO2 coating on stainless steel. A compact and intact CeO2 film on stainless steel surface impedes the ejection of metal ions into electrolyte, thereby inhibiting corrosion reactions. Generally the immersion experiment suggested cerium oxide/hydroxide as a cathodic inhibitor.10 However, Crossland and coworkers investigated the formation of anodized cerium oxide/hydroxide film out of Al/Ce alloys and substantiated that this cerium-rich layer provides anodic inhibition to aluminum corrosion,11 which agrees well with the conclusion of this study. 166 220 A 200 180 ) 160 V (m l 140 ----- Stainless steel ia t n ----- CeO coated stainless steel 120 2 te o P 100 80 60 -100 -80 -60 -40 -20 0 20 40 60 80 Current (nA) 500 ----- Stainless steel B 400 ----- CeO coated stainless steel 2 300 200 ) V 100 m E( 0 -100 -200 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 logI Figure 3.28 Corrosion test using 0.1M NaCl solution (A) Linear polarization plots (B) Tafel plots 167 Figure 3.29. Illustration of corrosion of stainless steel in chloride solution 3.4 Chapter Conclusions Cerium oxide thin films were successfully grown on stainless steel, nickel, platinum and N-silicon substrate with anodic electrodeposition technique. A solution of cerium (III) salt and a weak organic ligand (eg. acetate or lactic acid) with pH = 7.5 ~ 11 was employed for the deposition. X-ray diffraction reveals the deposited films of fluorite structure and the crystallite size of the as-produced films is in the scale of nanometer, which is further confirmed with Raman spectroscopy. X-ray photoelectron spectroscopy was applied to study the structure and oxidation, which indicates as-prepared cerium oxide is non-stoichiomtric but the percentage of CeO2 can be as high as 80%. Scanning electron microscopy pictures the surface texture and microstructure of deposits. Preferred orientation of CeO2 films was obtained by tailoring the deposition conditions, suggesting the optimized deposition parameters for oriented polycrystalline CeO2 include current density lower than –0.06 mA with galvanostatic deposition mode and deposition temperature of higher than 50oC. Sintering and corrosion experiments were 168 conducted on the formed films to further illustrate the properties and the possible applications of thin films. Detail investigations about optical, electrical and catalytic properties of cerium oxide films may acquaint us with this material and expand its application perspective. Anodic oxidation electrodeposition is relatively a novel synthesis technique for ceramic oxide and worthwhile to extend to other materials. Praseodymium and neodymium oxide deposition with this method is underway in our research group. 169 Chapter References (1) Özer, N.; Cronin, J. P.; Akyuiz, S. SPIE 1999, 3788, 103-110. (2) Wang, A.; Belot, J. A.; Tobin, J. M.; Markworth, P. R. Physica C 1999, 320, 154-160. (3) Trovareli, A. Catalysis Reviews: Science and Engineering. 1996, 38, 439-520. (4) Gerbinger, J.; Lohwasser, W.; Lampe, U.; Meixner, H. Sensors and Actuators B 1995, 26-27, 93-96. (5) Putna, E. S.; Stubenrauch, J.; Vohs, J. M.; Corte, R. J. Langmuir 1995, 11, 4832-4837. (6) Rudd, A. L.; Breslin, C. B.; Mansfeld, F. Corrosion Science 2000, 42, 275-288. (7) Wu, X. D.; Dye, R. C.; Muenchausen, R. E.; Foltyn, S. R.; Maley, M.; Rollett, A. D. Applied Physics Letters 1991, 58, 2165-2167. (8) Denhoff, M. W.; McCaffrey, P. Journal of Applied Physics 1991, 70, 3986-3988. (9) Inoue, T.; Yamamoto, Y.; Koyama, S.; Suzuki, S.; Ueda, Y. Applied Physics Letters 1990, 56, 1332-1333. (10) Arnott, D. R.; Hinton, B. R. W.; Ryan, N. E. Corrosion Science 1989, 45, 12-18. (11) Crossland, A. C.; Thompson, G. E.; Skeldon, P.; Smith, G. C.; et.al. Corrosion Science 1998, 40, 871-885. (12) Arenas, M. A.; Conde, A.; de Damborenea, J. J. Corrosion Science 2002, 44, 511-520. (13) Gorte, R.; Kim, H.; Vohs, J. M. Journal of Power Sources 2002, 106, 10-15. (14) Switzer, J. A. American Ceramic Society Bulletin 1987, 66, 1521-1524. (15) Zhitomirsky, I.; Petric, A. Materials Letters 1999, 40, 263-268. (16) Zhitomirsky, I. JOM 2001, 53, 48-50. (17) Li, F.; Newman, R. C.; Thompson, G. E. Electrochima Acta 1997, 42, 2455-2464. (18) Li, F.; Thompson, G. E. Journal of the Electrochemical Society 1999, 146, 1809-1815. 170 (19) Zhou, Y.; Phillips, R. J.; Switzer, J. A. Journal of American Ceramic Society 1995, 78, 981-985. (20) Zhou, Y.; Switzer, J. A. journal of Alloys and Compounds 1996, 237, 1-5. (21) Aldykiewicz, A. J. J.; Davenport, A. J.; Isaacs, H. S. Journal of the Electrochemical Society 1996, 143, 147-153. (22) Balasubramaian, M.; Melendres, C. A.; Mansour, A. N. Thin Solid Films 1999, 347, 178- 183. (23) Musiani, M. Chemical Communications. 1996, 21, 2403-2404. (24) Switzer, J. A.; Shumsky, M. G.; Bohannan, E. W. Science 1999, 284, 293-296. (25) Pourbaix, M.; NACE: Houston, TX, 1974, p 194. (26) E2108-00 Standard practice for calibration of the electron binding energy scale of an X- ray photoelectron spectroscopy. (27) Mullins, D. R.; Overbury, S. H.; Huntley, D. R. Surface Science 1998, 409, 307-319. (28) Mullins, D. R.; Radulovic, P. V.; Overbury, S. H. Surface Science 1999, 429. (29) Burroughs, P.; Hamnett, A.; Orchard, A. F.; et.al. Journal of the Chemical Society Dalton Transactions 1976, 17, 1686-1698. (30) Riviere, J. C.; Myhra, S. Handbook of Surface and Interface Analysis: Methods for Problem-Solving; Marcel: New York, 1998. (31) Wang, A. Q.; Golden, T. D. Journal of the Electrochemical Society 2003, in press. (32) Dean, J. A. Lance's Handbook of Chemistry; 13th ed.; McGraw_Hill Book Company: New York, 1985. (33) Papaiacovou, P.; Hussey, R. J.; Mitchell, D. F.; Graham, M. J. Corrosion Science 1990, 30, 451-455. 171 (34) Czerwinski, F.; Szpunar, J. A. Thin Solid Films 1996, 289, 213-219. (35) Wang, S.; Wang, W.; Zuo, J.; Qian, Y. Materials Chemistry and Physics 2001, 68, 246- 248. (36) Kanakaraju, S.; Mohan, S.; Sood, A. K. Thin Solid Films 1997, 305, 191-195. (37) Weber, W. H.; Hass, K. C.; McBride, J. R. Physical Reviews B 1993, 48, 178-185. (38) Francisco, M. S. P.; Mastelaro, V. R.; Nascente, P. A. P.; Florentino, A. O. Journal of Physical Chemistry B 2001, 105, 10515-10522. (39) Twu, J.; Chuang, C. J.; Chang, K. I.; Yang, C. H.; Chen, K. H. Applied Catalysis B: Environmental 1997, 12, 309-324. (40) Spanier, J. E.; Robinson, R. D.; Zhang, F.; Chan, S.-W.; Herman, I. P. Physical Review B 2001, 64, 245407-245415. (41) Cullity, B. D.; Stock, R. S. Elements of X- Ray Diffraction; 3rd ed.; Prentice Hall, 2001. (42) Zhang, F.; Chen, S.; Spanier, J. E.; Apak, E.; Jin, R. D.; Robinson, R. D.; Herman, I. P. Applied Physics Letters 2002, 80, 127-129. (43) Klug, H. P.; Alexander, L. E. X-ray Diffraction Procedures for Polycrystalline and Amorphous Materials; 2nd ed.; Wiley: New York, 1974. (44) Preisler, E. J.; Marsh, O. J.; Beach, R. A.; McGill, T. C. Journal of Vacuum Science and Technology B 2001, 19, 1611-1618. (45) Wuillound, E.; Delly, B.; Schneider, W.-D.; et.al. Physical Review Letters 1984, 53, 202- 205. (46) Pfau, A.; Schierbaum, K. D.; Göpel, W. Surface Science 1995, 331-333, 1479-1485. (47) Helms, C. R.; Spicer, W. E. Applied Physics Letters 1972, 21, 237-239. (48) Kotani, A.; Jo, T.; Parlebast, J. C. Advances in Physics 1988, 37, 37-85. 172 (49) Holgado, J. P.; Munuera, G.; Espinós, J. P.; al, e. Applied Surface Science 2000, 158, 164-171. (50) Kotani, A.; Haruhiko, O. Journal of Electron Spectroscopy and Related Phenomena 1992, 60, 257-299. (51) El Fallah, J.; Hilarie, L.; Roméo, M.; et.al. Journal of Electron Spectroscopy and Related Phenomena 1995, 73, 89-103. (52) Fujimori, A. Physical Review B 1983, 28, 2281-2283. (53) Schierbaum, K. Surface Science 1998, 399, 29-38. (54) Rana Rao, M. V.; Shripathi, T. Journal of Electron Spectroscopy and Related Phenomena 1997, 87, 121-126. (55) Normand, F. L.; Hilaire, L.; Killi, K.; al., e. Journal of Physical Chemistry 1988, 92, 2561-2568. (56) Paparazzo, E.; Ingo, G. M.; Zacchetti, N. Journal of Vacuum Science and Technology A 1991, 9, 1416-1421. (57) Strydom, C. Inorganica Chimica Acta 1989, 161, 7-9. (58) Holgado, J. P.; Alvarez, R.; Munuera, G. Applied Surface Science 2000, 161, 301-315. (59) Trovareli, A.; De Leitenbutrg, C.; Boaro, M.; Dolcetti, G. Catalysis Today 1999, 50, 353- 367. (60) Suzuki, T.; Kosacki, I.; Anderson, H. U. Journal of the American Ceramic Society 2001, 84, 2007-2014. (61) Elidrissi, B.; Addou, M.; Regragui, M.; et.al. Thin Solid Films 2000, 379, 23-27. (62) Praline, G.; Koel, B. E.; Hance, R. L.; al, e. Journal of Electron Spectroscopy and Related Phenomena 1980, 21, 17-30. 173 (63) Palmqvist, A. E. C.; Wirde, M.; Gelius, U.; et.al. NanoStructured Materials 1999, 11, 995-1007. (64) Overbury, S. H.; Huntley, D. R.; Mullins, D. R.; et.al. Journal of Vacuum Science and Technology A 1997, 15, 1647-1652. (65) Eck, S.; Castellarin-Cudia, C.; Surnev, S.; Ramsey, M. G.; Netzer, F. P. Surface Science 2002. (66) Paunovic, M.; Schlesinger, M. Fundamentals of Electrochemical Deposition; John Wiley & Sons, Inc.: New York, 1998. (67) Roméo, M.; Bak, K.; El Fallah, J.; et.al. Surface and Interface Analysis 1993, 20, 508- 512. (68) Ji, X.; Ponton, C. B.; Marquis, P. M. Journal of Materials Science: Materials in Medicine. 1992, 3, 283-287. (69) German, R. M. Sintering Theory and Practice; Johns Wiley & Sons, Inc., 1996. (70) Sedriks, A. J. Corrosion of Stainless Steels; 2nd ed.; John Wiley & Sons, Inc.: New York, 1996. 174 CHAPTER 4 ELECTROCHEMICAL SYNTHESIS OF NANOCRYSTALLINE CERIUM OXIDE POWDER 4.1 Introduction Cerium oxide is of interest in catalysis as structural and electronic promoters to improve the activity, selectivity and thermal stability of the catalysts as a transient oxygen storage material.1 Much effort has been dedicated to studying the role of ceria in well-established industrial processes such as three-way-catalysis (TWC) systems and fluid catalytic cracking (FCC) systems, where CeO2 is a key component in catalyst formulation. In the former systems, cerium oxide is used to remove automotive exhaust gas (NO, CO, CH, etc.) and for elimination 2,3 of SOx in the latter system. Ceria also demonstrates catalytic function in removal of soot from diesel engine exhaust, elimination of organics from wastewaters (catalytic wet oxidation), and cracking heavy oil in zeolite. On the other hand, CeO2 is a semiconducting and ionic-conducting oxide. It has the potential to substitute ZrO2 as the electrolyte material in solid oxide fuel cell (SOFC) systems and has been regarded as a model system for mixed ionic / electronic conductor investigation.4 Particle size plays a significant role in the unique properties and applications of cerium oxide. Generally speaking, the smaller the particle, the lower packing porosity and higher surface area. Therefore, small grain size can confer ceramics materials with improved properties compared with the coarse-grained counterpart, which has been demonstrated by different ceramic materials as enhanced strength in glass ceramics, increased fracture toughness in zirconia ceramics, raised dielectric constant in barium titanate, lower thermal expansion in lead zirconia, and reduced electrical fatigue in lanthanum zirconate titancate.5 Catalytic activity and electrical conductivity of crystalline cerium oxide strongly depend on particle size.6 Nano- 175 crystalline cerium oxide improves the catalytic properties significantly, which was demonstrated in the higher conversion of carbon monoxide to dioxide and sulfur dioxide to elemental sulfur at lower temperatures than its coarse-grained counterpart. The improved catalytic reactivity was also demonstrated in the oxidation of methane.7,8 As for conductivity, The electronic conductivity is predominated in nanocrystalline phase of cerium oxide whereas more ionic conductivity is in conventional microcrystalline structure.9 It is reported that nano-crystalline phase of cerium oxide is favorable to nonstoichiometric structure due to reduced enthalpy of defect formation and propensity for oxygen vacancy. From the view of catalytic activity, the fluorite structure of CeO2 favors oxygen vacancy formation in the lattices. It is the oxygen deficiency that is desirable as the oxygen buffer, which induces the catalytic reactivity of CeO2 in TWC, FCC and other gas phase oxidation and reduction reactions. And nanophase cerium oxide possesses increased level of non-stoichoimetry and enhances the storage / release oxygen ability, which results in the improvement of catalytic activity for redox reactions, giving lower reaction temperatures and elimination of catalytic activation hysteresis as compared with the stoichiometric counterpart. The reduced particle size increases the surface area and changes the morphology, providing a large number of reactive sites such as grain boundaries, interphase boundaries and dislocations. Nano-sized particles further increase the density of these defects, which creates many more catalytic sites and nano-scale interphase promotes reactive diffusion, consequently speeding up the reaction dynamics.10 In addition, the nanocrystalline cerium oxide catalyst exhibits higher resistance to CO2 poisoning during SO2 + CO reaction because non-stoichiometry renders a high percentage of reduced form of cerium oxide not only at the material surface but in the bulk 7,8,11,12 which cannot be reached by CO2 oxidation. 176 Electrical conductivity of cerium oxide is of small polaron transport style and usually determined by charge carrier density and mobility together with its electrical charge. Among these factors, the carrier density and mobility are determined by defect thermodynamics and microstructure of materials.10 The grain boundary is believed to be the site for electronic conductivity. Vacancy formation enthalpy at grain boundaries is comparably lower than bulk counterpart and the preferential reduction at interfacial sites leads to the formation of new electronic conductivity carriers – free electrons.4 Compared with microcrystalline phase, nanocrystalline CeO2 is of higher grain boundary density and even much more of this “donor” effect, therefore the activation energy for electronic conduction in nanocrystalline CeO2 is reduced to less than half of that of single crystalline CeO2, which substantially increases the levels of non-stoichoimetry and electronic carrier generation.13 Space charge and brick-layer models have been suggested to simulate and exemplify the grain size-dependent electrical 9,14 conductivity of cerium oxide. Both pure and doped nanophase CeO2 exhibited higher levels of 4 non-stoichiometriy and electronic conductivity than conventional microcrystalline CeO2. The sinterability of nanocrystalline cerium oxide increases with decreasing particle size. The reduction of sintering temperature for nanocrystalline cerium oxide overcomes processing temperature difficulty for cerium oxide and makes it a very promising candidate to substitute zirconia as electrolytes in solid oxide fuel cells.11,12 Since nanocrystalline cerium oxides exhibit improved properties in various applications, the production of nanocrystalline cerium oxide powders has attracted considerable interest in the past few years and many of these general methods to produce powder cerium oxide have been detailed in Chapter 1. The techniques including inert gas condensation,8,15 homogeneous precipitation,16,17 hydrothermal synthesis,18,19 thermal hydrolysis,20 microemulsion method21,22 177 and base generation electrochemical synthesis23 have been used to form nanocrystalline cerium oxide powders. The various properties of nanocrystalline cerium oxide powders are affected and can be manipulated by the synthesis. In this research work, anodic electrochemical synthesis was used to produce nanocrystalline cerium oxide powders. Microstructure and surface properties were studied with XRD, Raman Spectroscopy and BET techniques. Sintering behavior, particle size and lattice parameter changes with processing temperature were investigated as well in this research. 4.2 Experimental The anodic electrochemical synthesis of powder was performed using an EG&G Princeton Applied Research (PAR) Model 273A potentiostat/galvanostat. Constant temperature was maintained with a Fisher Scientific Model 1016D circulator. Three-electrode configuration was employed in the CeO2 synthesis in which platinum wire and platinum mesh were the working and counter electrode respectively, and the standard calomel reference electrode (SCE) as the reference. The electrolytes consisted of 0.1M Ce(NO3)3.6H2O (99.5% pure, purchased from Alfa AESAR), sodium acetate was the ligand. Solutions were made by dissolving chemicals in doubly distilled deionized water with a resistivity of 18.3 MΩ-cm and pH was adjusted with NaOH or KOH to pH > 10. Both galvanostatic and potentiostatic methods are employed to produce the CeO2 powders. The formed precipitate was filtered and rinsed with water and ethanol to remove possible impurities. Then the powder was dried in air at room temperature and grounded before it was subjected to any further characterization. 178 X-ray diffraction and Raman spectroscopy supplied the structure characteristics of the as- produced nanoscale powder. In addition, the surface area of the as-produced cerium oxide powder was calculated using Brunaauer-Emmett-Teller (BET) method from nitrogen sorption measurement on a Quantachrome Nova 2200 Absorption Analyzer with bath temperature of 77.40 K, assuming a molecular area of 0.163 nm2 for nitrogen gas and the relative pressure ranging from 0.05 to 0.35. And the pore size distributions were measured from desorption isotherms according to mathematic model developed by Barrett, Joyer and Halenda (BJH). The as-synthesized cerium oxide powders were pressed uniaxially under a pressure of ≈ 50 Mpa in a die (the diameter of the die = 1.0 cm) to prepare the sample for sintering. The sample was sintered either at a constant heating rate of 1oC/min or isothermally, the operation temperatures were in the range of 600 to 1100oC. The densities of the samples were measured from the used quantity and volume before and after sintering. 4.3 Results and Discussions 4.3.1 Electrochemical Synthesis of Cerium Oxide Powder With anodic electrodeposition, fabrication of cerium oxide film on a metal or semiconductor can be achieved in the pH range of 7.5 ~ 11.0. Simultaneous to film formation, cerium oxide powder can be precipitated at the bottom of cell. However, if the solution pH is higher than 11.0, cerium oxide powder is the only product formation in the electrochemical reaction. The acetate was dissolved to make aqueous solution and then a suitable quantity of cerium nitrate was added and a period of time was allowed for complete complexation. During the course of pH adjustment of the solution, the solution gradually turned opaque with a grey- brown color, which is the characteristic color of cerium (III) hydroxide. As potential or current is 179 applied, the color of colloid tended to be lighter and at the end of reaction, a light yellow signified the formation of cerium (IV) oxide. Both the powders made in the high pH solution and by-products of film deposition demonstrate similar XRD patterns (Figure 4.1A), which indicate a face-centered cubic (fcc) fluorite structure with a random orientation. The particle size of the synthesized CeO2 powder was in the range of 6 ~ 20 nm according to Williamson-Hall calculations (Equation 3.27), the result is shown in Figure 4.1B. The nanocrystalline size is demonstrated with a negative shift of -1 the CeO2 F2g frequency (~ 465 cm ) in Raman spectra as well. Figure 4.2 shows CeO2 F2g at frequency of ~ 451 cm-1, which is about 14cm-1 negative shift from normal ~465 cm-1 of microcrystalline counterpart. 180 200 (111) A 150 (220) ) p s 100 (311) (c ity (200) s n (222) (420) te 50 (331) In (400) (422) (511) 0 20 40 60 80 100 2θ (degees) 0.2 B 0.1 θ cos 0.0 r B -0.1 -0.2 0.20.30.40.50.60.70.8 sin θ Figure 4.1 (A) X-ray diffraction pattern of CeO2 powder produced from a solution of 0.1 M Ce(NO3)3 and 0.5 M acetate at pH ~ 11 at applied potential of 1.10V vs SCE. The synthesis occurs at room temperature. (B) Williamson-Hall plot βcosθ versus sinθ for the XRD pattern in (A) 181 12 451 cm-1 10 ) 3 8 X 10 (au 6 4 tensity In 2 0 0 500 1000 1500 2000 Wavenumber (cm-1) Figure 4.2 Raman spectra for electrosynthesized CeO2 powder produced from a solution of 0.1 M Ce(NO3)3 and 0.5 M acetate at pH ~ 11 at applied potential of 1.10 V vs SCE. The synthesis occurs at room temperature The specific surface area and pore size distribution were obtained using N2 sorption technique. Figure 4.3 shows the adsorption-desorption plot of cerium oxide powder. From the shape and characteristic feature, it belongs to the type IV isotherm in which there are basically three sections.24 In the lower-pressure region, the desorption overlaps with the adsorption (AB), and at a certain pressure (C), both the adsorption and desorption branches start to go up until a certain pressure (E), both branches converges together and level out. In between these two critical points (C & E), there is a range that the adsorbed amount at any given relative pressure is always greater along the desorption branch than that along the adsorption branch (CDE). And finally when the saturation vapor pressure is reached, both adsorption and desorption lines almost show zero slopes before a slight upward turn. (EFG). It is suggested from experimental 182 and theoretical studies that this type of isotherm usually is characteristic of inorganic oxide xerogels, which matches with the cerium oxide powder sample. 57.830 F 52.047 G 46.264 E 40.481 ) D g / 34.698 c (c 28.915 e C m 23.132 lu o B V 17.349 11.566 A 5.783 0.000 0.00.10.20.30.40.50.60.70.80.91.0 Relative Pressure, P/P o Figure 4.3 Adsorption-desorption isotherm plot of CeO2 powder produced from a solution of o 0.1 M Ce(NO3)3 and 0.5 M acetate at pH ~ 11, T = 25 C at applied potential of 1.10V vs SCE. X-axis represents relative pressure and Y- axis represents the adsorbed amount. In the plot, circle is for adsorption and square for desorption The explanation for the trend of the type IV isotherm is based on condensation theory proposed by Zsigmondy.16,25 In this model, it is assumed the pores are cylindrical shape, so the initial adsorption only covers the walls of the cylinder, until at the inflection point of hysteresis loop, the capillary condensations starts in the smallest pores, and with increasing pressure, the adsorption gradually fills larger pores until the saturation pressure, when the whole system is filled up with adsorptives. Based on this theory, some mathematical models were suggested and utilized to estimate the pore size distribution and surface area of the materials (Equation 2.42). Figure 4.4 shows the BJH desorption plot (A) and BET plot (B) derived from the isotherm. BJH plot indicates that the electrosyntheisized CeO2 powder has a narrow distribution of pore sizes. 183 From these two plots, the pore size and specific surface area of the electrosynthesized cerium oxide powder sample was calculated as 19.5 Å and 91.0 m2/g, respectively. 4.3.2 Nanosized Cerium Oxide Powder Sintering Since cerium oxide has a potential application as electrolytes in fuel cells, then sintering of the powder is a required process. Using ultrafine powders and doping solid solution additives are two adopted ways to reduce the high processing temperature for cerium oxide. It is well recognized that nanocrystalline particle favors higher sinterability. Two modes of sintering processes in air were employed in this research to determine the sintering ability of electrosynthesized nanocrystalline cerium oxide powders. Isothermal sintering, in which the required temperature was set before sintering started, was conducted at temperatures of 600, 700, 800, 900, 1000 and 1100oC. Each firing was kept for 5 hours. Ramping sintering, in which the sample was started sintering at temperature of 500oC, and heated up at rate of 1oC/min until the destination temperature was reached at which the sample was taken out of the sintering chamber. The employed temperatures are between 700 - 1100oC for comparison with the isothermal sintering result. The fired samples were subjected to XRD and BET measurement for grain size and specific surface area / pore size distribution analysis. 184 0.5980 0.5382 A 0.4784 g] 0.4186 c/ c [ 0.3588 og d) l ( 0.2990 v 0.2392 on D i 0.1794 sorpt De 0.1196 0.0598 0.0000 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 0.01 0.1 1 10 100 1000 Pore Diameter (A) 13.540 B 12.186 10.832 9.478 ] ) 8.124 1 - ) /P o 6.770 ((P 5.416 l/[W 4.062 2.708 1.354 0.000 0.00 0.07 0.14 0.21 0.28 0.35 Relative Pressure P/P o Figure 4.4. (a) BJH plot (b) BET plot of CeO2 powder produced from a solution of 0.1 M o Ce(NO3)3 and 0.5 M acetate at pH ~ 11, T = 25 C, at applied potential of 1.10 V vs SCE Isothermal sintering results are shown in Figure 4.5-4.7. Figure 4.5a illustrates the XRD patterns for the sample before sintering (A) and after (B-F). As expected, with increasing sintering temperature, the diffraction line intensity increases and the FWHM becomes narrower, 185 showing crystallization and grain growth with increasing sintering temperature. The Scherrer equation (Equation 2.37) was employed to estimate the grain size of the samples, which shows that the grain size increases from ~14.5 nm for the green powder to ~ 84.6 nm when the powder is subjected to 800oC isothermal sintering for 5h (Figure 4.5b insert). Ignoring the instrumental broadening effect, the crystallite size changes with sintering temperature is delineated in Figure 4.5b, indicating a > 180 nm crystal size after 5h sintering at 1100oC. The specific surface area decreases from ~ 91.0 m2/g for the green powder to ~1.0 m2/g for the 900oC-sintered powder (Figure 4.6A). The random variation of average pore as sintering temperature increases (Figure 4.6B) reflects the progress of pore size. It is known that during sintering, porosity is decreased but not always for the average pore size.26 Since the smaller pores nearby usually join each other to form large pores or are even removed by heating, it is natural for the small pores to shrink and the large pores to grow, resulting in the unexpected average pore variation with progress of sintering or at different sintering temperatures. 186 500 (111) 1000 ) A B s 400 p 800 a (c 300 (220) 600 ity (311) s 200 (200) n 400 te (331) (422) 100 (222) In (400) (420) (511) 200 0 0 1200 1600 ) C D s 1000 p 1200 800 (c ity 600 800 s n 400 te 400 In 200 0 0 E 1000 F ) 1200 s 800 p (c 800 600 ity s 400 n 400 te 200 In 0 0 20 40 60 80 100 20 40 60 80 100 2θ (degrees) 2θ (degrees) 200 180 90 80 b 160 70 ) 60 m n 140 ( 50 e z i S 40 n i m) 120 a n Gr 30 ( e 20 100 z 10 n Si 80 0 200 400 600 800 Sintering Temperature (oC) Grai 60 40 20 0 0 200 400 600 800 1000 1200 Sintering Temperature (oC) Figure 4.5. (a) X-ray patterns and (b) crystallite size variation for electrochemical synthesized o CeO2 powder (A) before sintering and after isothermal sintering at temperatures (B) 700 C; (C) 800oC; (D) 900oC; (E) 1000oC; (F) 1100oC for 5h. The grain sizes are estimated ignoring instrumental broadening, insert delineates the grain sizes estimated with standard correction 187 100 A 80 ) /g 2 (m 60 ea e Ar ac 40 f Sur c i f 20 i c Spe 0 0 200 400 600 800 1000 Sintering Temperature (oC) 400 B ) o A 300 er ( et am 200 100 Average Pore Di 0 0 200 400 600 800 1000 Sintering Temperature (oC) Figure 4.6 (A) Specific surface area vs. sintering temperature; (B) Average pore size vs. sintering temperature The rough sintering curve for the nano-sized CeO2 powder is shown in Figure 4.7, where the relative density (ratio between the observed density to the theoretical X-ray density for CeO2 of 7.184 g/cm3) is plotted against the sintering temperature. It indicates that the original sample 188 density is only around 40% of theoretical. When sintered at 1100oC for 5h, the relative density is close to 100%. Chen revealed that the hydrothermal synthesized nano-sized CeO2 can reach full o 16 density with 6 min of sintering at 1250 C. In Zhou’s report, the base generated CeO2 powder start sintering around 750oC and can reach 99.8% of theoretical density as sintering at 1300oC for 2h.23 Therefore the result of isothermal sintering is comparable to the literature conclusion that the nano-sized CeO2 powder has better sinterability as compared with other commercial 23 microsctructured CeO2 powders. 100 90 ) % 80 ( y t 70 60 ve Densi i t a 50 Rel 40 30 0 200 400 600 800 1000 1200 Sintering Temperature (oC) Figure 4.7 Sintering curve for the nano-sized CeO2 powder produced by anodic electrosynthesis from a solution of 0.1 M Ce(NO3)3 and 0.5 M acetate at pH ~ 11 at applied potential of 1.10 V vs SCE. The synthesis was performed at room temperature Ramping sintering was conducted parallel to isothermal counterpart. The XRD patterns (Figure 4.8a) for the sample before (A) and after sintering (B-H) demonstrate a similar trend for line broadening and crystallinity as the isothermal sintering case in Figure 4.5. And the grain growth verus sintering temperature in Figure 4.8b shows the crystallite size increases with 189 sintering temperature and reaches above 400 nm at 1100oC ramp sintering. The 1100oC sintered crystallite size in the series is estimated without the standard correction. Ramp sintering results in a lower densification rate but rapid grain growth at high temperature compared with the isothermal sintering. 1100oC ramp sintering can only achieve ~70% of theoretical density and the crystal grain size can be as large as ~ 400 nm. The results from these two sintering processes suggest that the prolonged sintering at 1100oC is required to obtain full sample density. Theoretically, heating rate is a critical parameter to control densification and grain growth in sintering for a heating-holding-cooling sintering cycle. Three stages are involved in a sintering process. At low temperature initial stage, surface diffusion dominates and neck bonds between grains form but a minor even no densification occurs. In intermediate stage and densification accentuates and grain grows. At high temperature final stage, coarsening complicates densification, resulting in high grain growth but declined densification rate.26 In low heat rate ramp sintering with no holding time process, sample will be heated to pass initial, intermediate stages and reach at the final stage at higher temperature, at which grain growth outsteps densification. Isothermal sintering without heating-up is the extreme case of the high heating rate. Ultra-short time to bring sample to the destination temperature results in many small grains without bonding, the grain growth is slower in comparison to the ramping procedure. However, long time holding and grain boundary diffusion provide energy to reshape the grain and remove pores, enhancing densification. Thinking of sintering mechanisms,26,27 sintering is based on various interfacial diffusions, including surface diffusion, volume diffusion and grain boundary diffusion and evaporation- condensation processes. Since the lower particle size has higher surface area and grain boundary per unit, and favors surface and grain boundary diffusion. The scaling law illustrates sensitivity 190 m t1 L1 of particle size to sintering time: = , t is for sintering time and L represents particle size t2 L2 and this equation indicates that the time ratio to reach the equivalent sintering level depends on the particle size ratio. Molecular dynamics simulations verified the higher surface instability with small particle size, attesting to particle size effect on surface diffusion. Therefore, nano-sized o CeO2 powders have lower temperature onset of sintering (~ 750 C) and lower temperature o o plateau (1100 C, 5h) compared with conventional micro-sized CeO2 powder (1500 C, at least 3h). In addition, XRD pattern demonstrates polycrystalline grain property for electrosynthesized cerium oxide powder and the powder has a narrow particle size range of 4 ~ 20 nm, valuable to high sinterability. Besides particle property, pore size is established as critical factor to control the densification rate by Mayo:28 1 dρ 1 1 − Q ∝ exp ρ(1− ρ) dt Ln r RT where ρ, L and r are sample density, particle size and pore radius respectively; and Q, R, T are activation energy, gas constant and temperature. This equation predicts the smaller pore size gives rise to higher densification rate. However, Figure 4.6B portrayed an irregular progress of pore size with different sintering temperature, which cautions non-ideal pore distribution in green powder. Controlling pore growth is an equivalent task as the particle size for successful sintering in the future synthesis of cerium. 191 200 400 (111) A B a ) 150 (220) 300 ps (c 100 (200) 200 y (311) it ns 50 (222) (331) (511) 100 te (400) (422) In 0 0 2500 2500 C ) D 2000 2000 ps (c 1500 1500 y it 1000 1000 ns te 500 500 In 0 0 3500 3000 4000 F ) E ps 2500 3000 (c 2000 ity 1500 2000 s 1000 ten 1000 In 500 0 0 20 40 60 80 100 20 40 60 80 100 2θ (degrees) 2θ (degrees) 500 b 400 300 m) n ( e z i 200 S n i a Gr 100 0 0 200 400 600 800 1000 1200 Sintering Temperature (oC) 192 Grain coarsening to micrometer range is still a problem for the nano cerium oxide powder even though the sintering temperature is reduced in the sintering. To retain grains in the nanometer range at full density is key for cerium oxide application in structural, electric, mechanical and catalytic areas. Doping has been explored to solve this problem.12,29 Sintering without final stage is advised as another choice, which has been theoretically analyzed and 30 experimentally applied in Y2O3 powder. To improve sintering efficacy of nanosized cerium oxide powder will be another future research path worth energy of scientists. 4.3.3 Lattice Relaxation of Nanocrystalline Cerium Oxide Another property for nanosized cerium oxide powders is that the lattice constant increases with decreasing particle size, which is known as lattice expansion or lattice relaxation.31,32 This property is in contrast to the lattice constant decrease trend for metal nano 39 particles but has been observed in some other rare metal oxides such as PbTiO3 and BaTiO3. X- ray diffraction data can be used to obtain particle size and lattice parameter so that it is an ideal technique to study lattice relaxation for nano cerium oxide.31-34 This lattice relaxation has also been observed in Raman as a negative shift and peak broadening in the spectra of CeO2 powders.35,36 4.3.3.1 X-ray Diffraction Results and Discussion Cerium oxide powders with sizes in the range of 5 ~ 20 nm were produced by anodic electrosynthesis and high temperature sintering was adopted to induce larger grain sizes of the powders for lattice relaxation investigation. The sintering practice is to heat green powder or compact pellets at 1oC/min from 25oC to the destination temperature (200, 300, till 1100oC) at which time the temperature is held for 5 hours and then cool to 25oC at a rate of 1oC/min. As- formed powders were x-rayed in the 2θ range of 60-150o.33,34 Particle sizes of the powders are 193 determined by X-ray peak broadening using Scherrer equation. And particle size verus sintering temperature is shown in Table 4.1. Precise lattice parameters are obtained with Bradley-Jay extrapolation, Nelson-Riley extrapolation and Cohen’s least square analysis that were detailed in Chapter 2. It is established that Cohen’s analysis has highest accuracy of 0.0004%, Bradley-Jay the lowest of 0.005% in this group and Nelson-Riley between of 0.002%. However, the application of Bradley-Jay extrapolation and Cohen’s method requires a θ range between 60-90o. Five sets of reflections (531), (600), (620), (533) and (622) are selected to determine lattice parameters with the Cohen’s method and Bradley-Jay extrapolation whereas all reflections of 2θ = 60-150o are used for Nelson-Riley extrapolation. The lattice parameter of green powder is determined as 0.54268, 0.54264 and 0.54253 nm by Cohen’s method, Bradley-Jay and Nelson- Riley, respectively. 33,34 Figure 4.9 illustrates the two extrapolation plots for “green” sample. Table 4.1 Sintering temperature and particle size of CeO2 powders Sample Sintering Temperature (oC) Grain size (nm) Green Powder 6.22 A 300 6.91 B 400 7.73 C 500 10.11 D 600 20.30 E 700 83.60 F 800 > 100 G 900 > 100 H 1,000 > 100 I 1,100 > 100 194 2 cos θ 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.58 Nelson-Riley Extrapolation 0.57 Bradley-Jay Extrapolation ) 0.56 0.55 ter (nm e 0.54 ram a 0.53 ice P 0.52 Latt 0.51 0.50 0.0 0.2 0.4 0.6 0.8 1.0 1.2 2 2 cos θ/sinθ+cos θ/θ cos 2 θ cos2 θ Figure 4.9 Plots of lattice parameter against + (Nelson-Riley sinθ θ 2 Extrapolation) and lattice parameter against cos θ (Bradley-Jay extrapolation) The obvious difference between N-R method and Cohen or B-J method can be imputed to the high-angle measurement limited by the instrument. The high-angle reflections cannot be resolved between Kα1 and Kα2, which leads to high-angle measurement error due to improper mixed patterns and insufficient peaks for Cohen’s method and Bradley-Jay extrapolation. To complement high-angle (θ > 60o) deficiency, reflections between θ = 30o and θ = 60o are used in the lattice parameter determination, i.e. Nelson-Riley extrapolation will be applied in the lattice parameter relaxation study. Cerium dioxide is well known as a f-electron metal oxide, which crystallizes into the fluorite-structure lattice. It is well recognized that the X-ray diffraction patterns of CeO2 are face-centered cubic symmetry, belonging to the space group Fm3m with the lattice parameter 195 5.41129 ± 0.0008Å for micro sized particles.37 X-ray diffraction patterns indicate that with decreasing particle size, the reflections have a shift to lower angle and a peak broadening (Figure 4.10). The peak broadening of cerium oxide is due to the decrease of cerium oxide powder grain size, as recognized and discussed in previous sections. Whereas the reflection negative shift is attributed to the lattice parameter expansion, which is a significant property for nanosized cerium oxide particle. X-ray diffraction lattice relaxation of nanosized CeO2 is revealed with Nelson- Riley extrapolation, indicating that with decreasing particle size of CeO2 from ~ 84 nm to ~ 6 nm, the lattice constant increases from 0.54101 nm to 0.54253 nm (Figure 4.11). At grain sizes greater than 100 nm, the lattice constant does not vary much and the variation is ± 0.0004 nm. Figure 4.11 indicates that the sharp expansion of lattice parameter usually occurs when the particle size is smaller than 20 nm, as seen in the previous literature report.32 196 A 6.2 nm 6.9 nm 20.3 nm 83.6 nm .) >100 nm .u (a ity ns te In 84 85 86 87 88 89 90 91 92 93 2θ (degrees) B 6.1 nm 6.9 nm 20.3 nm 83.6 nm ) > 100 nm . .u (a ity s n te In 67 68 69 70 71 72 2θ (degrees) Figure 4.10 X-ray spectra showing peak shift and broadening of CeO2 nanoparticles of (A) CeO2 (422) reflection and (B) CeO2 (400) reflection 197 0.5426 0.5424 0.5422 ) m 0.5420 (n r te 0.5418 e m a r 0.5416 a P e 0.5414 ttic a L 0.5412 0.5410 0.5408 0 102030405060708090 Grain Size (nm) Figure 4.11 Lattice parameter vs. grain size, blue line for theoretical lattice parameter of microcrystalline CeO2 Nonstoichiometry is stated as main cause for lattice relaxation in cerium oxide nanoparticles.32,35,38 Cerium oxide is well known for its nonstoichiometry and oxygen vacancy 1-3,7,32,35,38 formation. It is suggested that the C-type cerium sesquioxide (Ce2O3) co-exists with fluorite cerium dioxide in nanosized cerium oxide, so that the denotation of the nanosized cerium oxide should be CeO2-x instead of CeO2. This C-type cubic phase can be regarded as the combination of three orientations of defective fluorite-like unit cells, each of which functions as an octant with two oxygen atoms missed from the structure. The three oriented octants employed and arrangements in C-type structure are demonstrated in Figure 4.12A and 4.12B, respectively. Metal atoms occupy the same positions in C-type structure as in fluorite structure whereas the unit cell of C-type structure is two times the size of fluorite structure.36,38 The formation of C- type phase brings in the ordered oxygen vacancy structure. Tsundkawa proposed a method to estimate the nonstoichiometry of nanosized cerium oxide with hypothesis that C-type phase is 198 36 formed. The imaginary pure cerium C-type sesquioxide (CeO1.5) has a lattice constant of 1.123 nm. Correspondingly, microsized cerium (IV) dioxide (CeO2), assuming a C-type phase, has the lattice constant of 0.54101 x 2 = 1.08202 nm. A linear relationship results for x (x = 0 for CeO2 and x = 0.5 for CeO1.5) in CeO2-x and their lattice parameters according to Vegard’s rule (Figure 4.13A). Applying the C-type phase lattice parameters to the nanosized cerium oxide (CeO2-x), the stoichiometries of these nanosized CeO2-x can be located from this linear relationship. The result is listed in Table 4.2. Tsundkawa investigated the origin of the lattice relaxation in terms of both model simulations and ab initio calculations. The former emphasized on oxygen valence variation at the surface whereas the latter believed that the ionic cerium valence reduction is responsible for this abnormal phenomenon of some oxides.31,39 Z III III III III III II I Figure 4.12 C-type sesquioxide illustration. Left: octant type I. Clockwise rotation 90o around Z can result in octant type II and counterclockwise 90o is octant III. Two oxygen vacancies are formed with each Ce3+ replacing Ce4+ in octant structures. Circle represents cerium and solid ball represents oxygen. Right:arrangement of octants in C-type sesquioxide phase. Lattice of C-type phase is two times large as that of fluorite phase. Scale is not justified in this figure 199 Tasundkawa theory clarifies the lattice constant increase in nanosized ceria to a certain degree. However, it has difficulty in explaining the lattice constant change in trivant cation doped ceria, which has been found to form oxygen vacancies and is non-stoichiometric as well because of C-type phase existence.38 According to McBride’s research with a series of trivalent dopants, the ionic sizes of the dopants play a critical role in lattice variation compared to original dioxide, i.e. larger trivalent dopants (Compared with Ce4+) dilate the lattice constant whereas the smaller size cations contract the lattice constant. The phases apart from fluorite and C-type phase are evident with some dopants from XRD patterns. Hence, bearing in mind both the cation sizes and atomic fractions of dopants in the ceria is a relatively accepted approach to manifest the lattice variation of nanosized cerium oxide particles. Ce3+ can be considered as a trivalent dopant, which has an ionic radius of 0.1143 nm, larger than 0.097 nm of Ce4+. Using linear relationship between lattice expansion slope and ionic radius established by McBride and his colleagues, it is revealed the lattice of cerium oxide increases with Ce3+ doping at the rate of 0.02625 nm per atomic fraction. As lattice expansion values of nanosized cerium oxide are known, the compositions can be deferred from the linear relationship of lattice parameter vs. atomic doping of Ce3+ (Figure 4.13B), and stoichiometries of these cerium oxide particles are listed in Table 4.2 as well. Furthermore, oxygen vacancies are also estimated in accordance with the percentage of Ce3+ for both approaches (Table 4.2), since the replacement of Ce4+ with Ce3+ hypothesis is the structure basis for both approaches and it is well recognized that two substitutions of Ce3+ for Ce4+ results in an oxygen vacancy in the structure. It is found that the methods achieve similar results for Ce3+ percentage, stoichiometries and oxygen vacancies for nanosized cerium oxide powder. With the particle size decreasing and 200 the lattice constant increasing, the Ce3+ fraction increases and the O/Ce ratio in cerium oxide decreases. The slight difference between the sets of results comes from the divergence in hypotheses and methodologies of these two methods. Tsunekawa’s method completely constructs on the ideal condition of C-type sesquioxide whereas Mcbride’s approach attains the correlation between lattice parameter variation, atomic doping fraction and dopant cation size from experimental data. XPS measurement of nanosized cerium oxide film surface indirectly provided the evidence of nonstoichiometry and oxygen vacancies well elucidates the lattice relaxation in nanosized cerium oxide theoretically.40 To further validate above approaches, a quantity of systematic work are required, such as quantitative measurement of the Ce3+ concentration / oxygen vacancies, accurate determination of nonstoichiometry of nanosized cerium oxide particles, and even sufficient consideration of other attributions accounted for the lattice relaxation phenomenon in cerium oxide nanoparticle, etc. Nevertheless, aforementioned are out of the discussion scope of this research. 201 e d 1.0852 oxi 1.13 1.0850 squi e 1.0848 S 1.12 pe 1.0846 y t - C 1.0844 y 1.11 r a 1.0842 n agi 1.10 1.0840 m I 1.0838 n i r e 1.0836 t 1.09 e 0.020 0.024 0.028 0.032 0.036 ram a P 1.08 A ce i t 0.0 0.1 0.2 0.3 0.4 0.5 Lat X in CeO 2-x 0.5426 0.5424 ) m (n r 0.5422 te e m ra a 0.5420 P e ttic a L 0.5418 B 0.5416 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Ce3+ / (Ce3+ + Ce4+) Figure 4.13 (A) According to Tsunekawa’ hypothesis, applying Vegard’s rule: linear relationship between two compositions, CeO2 and CeO1.5 and their lattice parameters (solid square), insert indicates the experimental points for nanosized CeO2-x, from the known lattice to derive compositions (i.e. x); (B) According to McBride’s theory: linear relationship between Ce3+ doping fractions and lattice parameters. (0.02625 nm/Ce3+ atomic fraction was obtained from McBride’s paper) 202 Table 4.2 Estimation of Ce3+ percentage and oxygen vacancies in electrosynthesized nanosized CeOy Ce3+ / (Ce4+ +Ce3+) Oxygen Vacancies Grain Lattice y in CeO (%) (%) y Size Parameter (nm) (nm) M’s T’s M’s T’s M’s T’s 6.21752 0.54253 5.787 7.418 2.8935 3.7091 1.9710 1.9629 6.91599 0.54218 4.457 5.710 2.2285 2.8551 1.9777 1.9714 7.72579 0.54191 3.429 4.392 1.7145 2.1962 1.9829 1.9780 10.10909 0.54186 3.238 4.148 1.6190 2.0742 1.9838 1.9793 20.30101 0.54169 2.591 3.319 1.2955 1.6593 1.9870 1.9834 Note: M’s refers the results calculated using the method established by McBride and colleagues. T’s refers the results calculated using the method suggested by Tsunekawa and colleagues. Grain sizes and lattice parameters are measured by XRD. 4.3.3.2 Raman Result and Discussions From the spectroscopy view, the cerium dioxide f-electron structure has six phonon branches yielding three zone-center frequencies. These frequencies are doubly degenerate TO mode located around 270 cm-1, non-degenerate LO mode around 595 cm-1 and triply degenerate Raman active mode 465 cm-1.41 Among these three frequencies, 465 cm-1 gives a strong signal in 42-44 Raman spectroscopy and normally used to probe CeO2. The electrosynthesized cerium oxide powder and the sintered powders at 600, 800, 1000oC for 5h are subjected to Raman measurement. Figure 4.14 displays the Raman spectra for these samples. A strong Raman peak at 459.4 cm-1 and other peaks at 186.7, 263, 595.3, 724.7, 926.2, 1067.5, 1384.9 and 1439.3 cm-1 are observed. According to Long,42,45 263 and 595.3 cm-1 are assigned as the second order (2TA) peaks and other peaks are ascribed to the surface oxygen -1 2− species. It was reported that the peaks between 640-970 cm are due to peroxide species (O2 ) -1 − and the peaks between 1040-1172 cm are the absorption of superoxide species (O2 ). And the 203 -1 δ − peaks above 1300 cm could be due to other oxygen species (O2 ). All these surface oxygen peaks markedly decrease in relative intensity and negatively shift with increasing sintering temperature. Although there is no record about 186.7 cm-1, it is reasonable to believe that it can be assigned to oxygen species as well since it diminishes with the elevation of sintering temperature. Evolution of 870 and 1173 cm-1 in 1000oC sintered sample spectrum is another phenomenon worth noticing. These two peaks are also assigned to the surface oxygen species. These high temperature-oxygen species positions are different from spectra for these species before sintering, which may be due to the different origins. Before sintering, the adsorbed oxygen species come from the synthesis solution, which are easily desorbed during sintering. Whereas since the sintering is conducted in air, molecular O2 gas are adsorbed at high δ − − 2− temperature in the cerium oxide surface and partially converts to O2 ,O2 ,O2 . The double -1 peaks around 1546 and 1610 cm in the spectra are possible evidence for adsorbed molecular O2 species. From this discussion, we can draw the same conclusion as the reference that the adsorbed peroxide and superoxide species are the active components for cerium oxide as a catalyst at high temperature. For the Ce-O lattice, the second order peak around 595 cm-1 decreases in relative intensity in high temperature sintered-sample spectra, which agrees with that this absorption is the fingerprint signature for oxygen vacancies. The relative intensity of this signal enhances with the reduction of the particle size of cerium oxide. 204 B 1.0 A 1.0 0.8 0.8 ) d .) lize a 0.6 .u 0.6 a ( orm y N it ( s 0.4 0.4 n ity e t ns In te 0.2 0.2 In 0.0 0.0 0 500 1000 1500 2000 0 500 1000 1500 2000 -1 Wavelength (cm-1) Wavenumber (cm ) C D 1.0 1.0 0.8 0.8 ) d d) lize lize 0.6 a 0.6 m r o (N (Norma 0.4 0.4 y it ity s n e t 0.2 0.2 In Intens 0.0 0.0 0 500 1000 1500 2000 0 500 1000 1500 2000 -1 Wavenumber (cm-1) Wavenumber (cm ) Figure 4.14 Full range of Raman spectra for the anodic electrosynthesized cerium oxide (A) green powder and sintered at (B) 600oC; (C) 800oC and (D) 1000oC 205 The first-order peak at 465 cm-1 is the primary feature for cerium oxide, which presents 41,43 F2g symmetry. It has been noticed that this spectrum shifts to lower frequencies and shows a substantial broadening as the cerium oxide particle size reduces into the nano range.35,46 Figure 4.15 exhibits the Raman spectra for the electrosynthesized green cerium oxide powder (~ 6.2 nm), sintered at 600oC (~20.3 nm), 800oC (> 100 nm), and 1000oC (>100 nm). And the peak positions and normalized Raman linewidths are listed in Table 4.3. As the grain size of cerium oxide particle decreases from > 100 nm to a few nanometers, a decrease of first order Raman frequency from 466.8 cm-1 to 459.4 cm-1 and an increase of peak broadening from 7.9 cm-1 to 30.4 cm-1 are observed. o 1.0 --- 1000 C sintered --- 800oC sintered --- 600oC sintered 0.8 ) --- Green CeO d 2 e liz a 0.6 m r o (N 0.4 ity s n te 0.2 n I 0.0 250 300 350 400 450 500 550 600 650 Wavenumber (cm-1) Figure 4.15 Raman peak at ~ 465 cm-1 of cerium oxide green powder, sintered at 600, 800 and 1,000oC, indicating negative shift of Raman peak with grain size decreasing. Insert demonstrates the zoom-in picture of peak shift 206 Table 4.3 Comparison of peak breath of the sample at different sintering temperatures Sample No. Sintering Temperature (oC) Peak Position (cm-1) FWHM (cm-1) 1 1000 466.8067 7.9091 2 800 466.4627 8.6099 3 600 465.7767 11.4131 4 Green powder 459.4022 30.4350 Weber and Nakajma studied the trivalent cation doped cerium oxide Raman spectra and 41,46 calculated CeO2 phonon-dispersion curve respectively. Both of them put oxygen vacancy defects as the main contribution for Raman spectra change. Rigid-Ion Model calculation based on experimental oxygen vacancies fits the experimental Raman spectra results qualitatively. Phonon confinement, strain, defects and phonon relaxation have been hypothesized to cause to the Raman spectra change. To further evaluate the individual and combined contributions of different hypothesized factors for Raman negative shift and broadening with decreasing particle size, phonon confinement, strain and oxygen vacancies have been utilized to form simulated Raman spectra for different particle sizes. Phonon dispersion curves derived by Weber and Nakajima are applied to Combell phonon confinement model and Richter confinement phonon model respectively.35 The increasing lattice parameter with decreasing particle size in nansized cerium oxide explains the negative shifts in Raman lines. The frequency shift ∆ω of Raman line is related with lattice parameter variation ∆a as: ∆ω = −3γω 0 ∆a / a0 (Equation 4.1) where ω 0 and a0 are the Raman frequency and lattice parameter for CeO2, respectively, γ is the -1 mode Grüneisen constant. ω 0 = 466.8 cm , a0 = 0.54101 nm and γ = 1.24 are applied for CeO2 here. The first order Raman spectra shifts calculated from equation 1 are listed in Table 4.4. 207 Quantitatively, the calculated Raman shifts cannot tally very well with the observed results. The deviation may be due to the ignoring of other factors in the calculations, for instance surface stress, impurity defects in the synthesized cerium oxide particles, etc. Table 4.4 Comparison of calculated and experimental Raman shifts Calculated Grain Size Lattice Parameter Experimental Raman shift Raman shift (nm) (nm) (cm-1) (cm-1) ~100 0.54132 -0.344 -0.995 20.30 0.54169 -1.03 -2.18 6.22 0.54253 -7.40 -4.88 Particle size distribution and reduction of cerium oxide play significant roles in Raman line broadening. The dispersion of particle size induces inhomogeneous lattice strain. Oxygen vacancy concentration increases with particle size decreasing, which also adds to the density of strain. Combining strain and confinement phonon was used to simulate the Raman linewidth for different nanosized cerium oxides. The simulation detail can be found in reference.38 In summary, X-ray diffraction directly shows the lattice relaxation in nanosized cerium oxide particles whereas Raman spectrum also confirms the existence of it. The non- stoichiometry and oxygen vacancies are considered the main cause for this property of nanosized cerium oxide particles. From the application view, larger lattice parameter, i.e. high percentage of Ce3+ / oxygen vacancies in the lattice is beneficial to the storage and release of oxygen and caters to the catalytic function of cerium oxide. In addition, lattice expansion reduces oxygen vacancy migration enthalpy, which increases the ionic conductivity as mentioned. Therefore, compared with conventional cerium oxide, nanocrystalline powder will be more efficient and promising as catalyst or SOFC electrolyte. 208 4.4. Chapter Conclusions Nanocrystalline cerium oxide powders were produced with anodically electrochemical oxidation of cerium nitrate solution and acetate as a ligand. The coloring of the reaction system uncovers the involvement of chemical reaction prior to the electron transfer reaction in the electrosynthesis scheme. X-ray diffraction pattern reveals that the as-prepared cerium oxide powder has the fluorite structure with crystallite size of in the range of nano-regime (5 ~ 20 nm). Negative shift of Raman spectra line verifies the nano-scale of the particle size. And pore size of 2 ~20 Å and specific surface size of ~90 m /g were obtained through N2 sorption. Isothermal and ramping sinterings of as-produced nanocryatallite cerium oxide powder were studied. It is found as-produced cerium oxide possesses a higher sintering ability compared with conventional chemical methods. It is found sintering of as-formed sample at 1100oC for 5h can result in the near-full density compared with theoretical. However, this one-step sintering is limited by the sharp increase of the particle size and quick reach at micro-scale and out-of- control pore size progress. Careful scrutinizing of synthesis process and improving sintering techniques are suggested to overcome the problems. The lattice expansion of nanosized cerium oxide particles is studied with X-ray diffraction and Raman spectroscopy. The CeO2-x stoichiometry and oxygen vacancies are theoretically estimated with two different approaches and applied to explain origin of the incidence. 209 Chapter References (1) Trovareli, A. Catalysis Reviews, Science and Engineer 1996, 38, 439-520. (2) Trovareli, A.; De Leitenbutrg, C.; Boaro, M.; Dolcetti, D. Catalysis Today 1999, 50, 353- 367. (3) Trovareli, A.; Boaro, M.; Rocchini, E.; DeLeitenburg, C.; Dolcetti, G. Journal of Alloys and Compounds 2001, 323-324, 584-591. (4) Chiang, Y.-M.; Lavik, E. B.; Blom, D. A. Nanostructured Materials 1997, 9, 633-642. (5) Subbarao Colloids and Surface A: Physicochemical and Engineering Aspects 1998, 133, 3-11. (6) Tschope, A.; Sommer, E.; Birringer, R. Solid State Ionics 2001, 139, 255-265. (7) Tschope, A.; Liu, W.; Flytzani-Stephanopoulos, M.; Ying, J. Y. Journal of Catalysis 1995, 157, 42-50. (8) Tschope, A.; Schaadt, D.; Birringer, R.; Ying, J. Y. Nanostructured Materials 1997, 9, 423-432. (9) Tschope, A. Solid State Ionics 2001, 139, 267-280. (10) Zhang, Y.; Anderson, Y.; Muhammed, M. Applied Catalysis B: Environment 1995, 6, 325-337. (11) Zhou, Y.; Rahaman, M. N. Acta Materialia 1997, 45, 3635-3639. (12) Kleinlogel, C.; Gaukler, L. J. Solid State Ionics 2000, 135, 567-573. (13) Tuller, H. L. Solid State Ionics 2000, 131, 143-157. (14) Tschöpe, A.; Ying, J. Y.; Tuller, H. L. Sensors and Actuators B 1996, 31, 111-114. (15) Ying, J. Y.; Tschöpe, A. The Chemical Engineering Journal 1996, 64, 225-237. (16) Chen, P. L.; Chen, L. W. Journal of the American Ceramic Society 1993, 76, 1577-1583. 210 (17) Djuricic, B.; Pickering, S. Journal of European Ceramic Society 1999, 19, 1925-1934. (18) Wang, C.; Qian, Y.; Xie, Y.; Wang, C.; et.al. Materials Science and Engineering B 1996, 39, 160-162. (19) Yang, X.; Rahaman, N. Journal of European Society 1997, 17, 525-535. (20) Hirano, M.; Fukuda, Y.; Iwata, H.; Horta, Y.; Inagaki, M. Journal of the American Ceramic Society 2000, 83, 1287-1289. (21) Zhang, J.; Ju, X.; Wu, Z. Y.; Liu, T.; Du, T. D.; Xie, Y. N. Chemistry of Materials 2001, 13, 4192-4297. (22) Masui, T.; Fujiwara, K.; Machida, K.; Adachi, G. Chemistry of Materials 1997, 9, 2197- 2204. (23) Zhou, Y.; Philips, R. J.; Switzer, J. A. Journal of the American Ceramic Society 1995, 78, 981-985. (24) Gregg, S. J.; Sing, K. S. W. Adsorption, Surface area and Porosity; 2nd ed.; Academic Press: New York, 1982. (25) Zsigmondy, A. Zeitschrift Fuer Anorganische und Allgemeine Chemie 1911, 71, 356- 377. (26) German, R. M. Sintering Theory and Practice; John Wiley & Sons, Inc.: New York, 1996. (27) Groza, J. G. Nanostructured Materials 1999, 12, 987-992. (28) Mayo, M. International Materials Review 1996, 41, 85-115. (29) Zhang, T.; Hing, P.; Huang, H.; Kilner, J. Journal of Materials Processing Technology 2001, 113, 463-468. (30) Chen, I.-W.; Wang, X.-H. Nature 2000, 404, 168-171. 211 (31) Tsunekawa, S.; Sahara, R.; Kawazoe, Y.; Ishikawa, K. Applied Surface Science 1999, 152, 53-56. (32) Zhang, F.; Chan, S.-W.; Spanier, J. E.; Apak, E.; Jin, Q. Applied Physics letters 2002, 80, 127-129. (33) Suryanarayana, C.; Norton, M. G. X-ray Diffraction A Practical Approach; Plenum Press: New York, 1998. (34) Klug, H. P. A.; Alexander, L. E. X-ray Diffraction Procedures for Polycrystalline and Amorphous Materials; 2nd ed.; Wiley: New York, 1974. (35) Spanier, J. E.; Robinson, R. D.; Zhang, F.; Chan, S.-W.; Herman, I. P. Physical Review B 2001, 64, 245407-245415. (36) Tsunekawa, S.; Sivamohan, R.; Ito, S.; Kasuya, A.; Fukuda, T. Nanostructured Materials 1999, 11, 141-147. (37) Standard Reference Materials; National Institute of Standards and Technology: Gaithersburg, MD, 1983. (38) McBride, J. R.; Hass, K. C.; Poindexter, B. D.; Weber, W. H. Journal of Applied Physics 1994, 76, 2435-2441. (39) Tsunekawa, S.; Ishikawa, K.; Li, Z.-Q.; Kawazoe, Y.; Kasuya, A. Physical Review B 2000, 85, 3440-3443. (40) Wang, A. Q.; Punchaipetch, P.; Wallace, R. M.; Golden, T. D. Journal of Vacuum Science and Technology B 2003, 21, 1169-1175. (41) Weber, W. H.; Hass, K. C.; McBride, J. R. Physical Review B 1993, 48, 178-185. (42) Long, R. Q.; Wan, H. L. Journal of the Chemical Society: Dalton Transactions 1997, 93, 355-358. 212 (43) Franciso, M. S. P.; Mastelaro, R.; Nascente, P. A. P.; Florentino, A. O. Journal of Physical Chemistry B 2001, 105, 10515-10522. (44) Twu, J.; Chuang, C. J.; Chang, K. I.; Yang, C. H.; Chen, K. H. Applied Catalysis B: Environment 1997, 12, 309-324. (45) Long, R. Q.; Huang, Y. P.; Wan, H. L. Journal of Raman Spectroscopy 1997, 28, 29-32. (46) Nakajima, A.; Yoshihara, A.; Ishigame, M. Physical Review B 1994, 50, 13297-13307. 213 CHAPTER 5 MECHANISM STUDIES ON ELECTRODEPOSITION OF CERIUM OXIDE FILMS AND ELECTRSYNTHESIS OF CERIUM OXIDE POWDER IN SOLUTION OF CE (III) COMPLEXE 5.1 Introduction Base generation at cathode by electrochemical process was suggested as a crucial step for cathodic electrosynthesis of metal oxides by Switzer.1,2 The in-depth study of the complete mechanism for CeO2 deposition using the base generation method has been conducted by Aldykiewicz3 and further supported by Li and coworkers.4-6 Aldykiewicz et. al. employed deposition of cerium-rich films on copper, gold and iron under cathodic polarization as a model system to study the corrosion inhibition of copper-containing aluminum alloys.3 And it is suggested that the chemical mechanism of deposition is through the reduction of oxygen in the electrolyte. Ce(III) is oxidized by the hydrogen peroxide intermediate to Ce(IV) hydroxide. Concurrently, the pH increases at the surface of the electrode due to the generation of OH- ions. As a result of the base generation reaction at the working electrode, Ce(IV) hydroxide ions in the - solution close to the electrode surface reacts with OH and CeO2 precipitates on the surface. The mechanism was supported by rotation disk electrode experiments and XANES studies.3,7 And Li’s AFM assisted study further pointed out that the nucleation and growth of the CeO2 crystals 4-6 are the rate-determining step for the CeO2 film formation. Anodic oxidation is a novel technique for the fabrication of CeO2 film and film deposition studies in Chapter 3 reveal that both cerium ion complexation and pH are critical factors for successful formation of the film onto the substrate. According to Pourbaix diagram for cerium (Figure 3.1), at pH’s above 7, Ce(OH)3 precipitate (gray-green) forms toward 214 reducing voltages and CeO2 precipitate (yellow) forms at oxidizing voltages. In aqueous solutions below pH of 7, Ce3+ ions are stable between the reduction and oxidation limits of the 3+ ++ electrolyte. There is a small region for oxidization of Ce to Ce(OH)2 ions, which can help in the formation of CeO2, though direct anodic deposition of CeO2 as a film is difficult in aqueous solution. If the solution pH is kept below 7, in a 0.1 M cerium nitrate solution, at an oxidation 3+ ++ voltage above ~ 1.0 V, Ce ions form a hydroxide species, Ce(OH)2 turning the solution light orange.8 3+ 2+ + − Ce + 2H 2O → Ce(OH ) 2 + 2H + e (Equation 5.1) ++ This Ce(OH)2 species can undergo a chemical reaction to form CeO2 in solution, which then drops to the bottom of the cell as a powder (yellow precipitate). 2+ + Ce(OH ) 2 → CeO2 + 2H (Equation 5.2) Likewise, if the solution pH is above 7, then the Ce3+ ions in solution form a precipitate, 8 Ce(OH)3 (gray-green precipitate), by a chemical reaction pathway, 3+ + 2Ce + 3H 2O → Ce2O3 + 6H (Equation 5.3) which can be oxidized to form a CeO2 precipitate. + − Ce2O3 + H 2O → 2CeO2 + 2H + 2e (Equation 5.4) The Ce3+ ion is stabilized with a complexing agent in solution to prevent the formation of the Ce(OH)3 precipitate in high pHs, then higher oxidation voltages can be applied for the 3+ conversion of Ce to CeO2. To validate the chemical or electrochemical processes involved in the formation of CeO2 films and powders, the mechanism of the anodic oxidation electrosynthesis is studied. Among electrochemical analytical techniques, voltammetry is a good way to probe or monitor the electron transfer.9 Mass transfer of chemical agents in the region close to the 215 electrode is amendable with voltammetry by rotation disk electrode.9 Therefore, cyclic voltammetry combined with deposition was employed to study the mechanism of anodic deposition of CeO2 at various pHs and ligand conditions to further understand the formation of the cerium oxide films. Rotating disk electrode experiments were employed to illustrate the thermodynamic properties implicated in this electrochemical process. The peculiar powder formation related scheme at high pH value is presented as well in this chapter. 5.2 Experimental 5.2.1. Electrochemical Techniques Cyclic and linear voltammeries were utilized in experiments to investigate the mechanism of the cerium oxide film deposition. The rotating disk experiments were done using a Pine Instrument analytical rotator and stainless steel electrode as the substrate (Figure 5.1). Cyclic voltammetry of the cerium-ligand solution systems were performed with an EG&G Princeton Applied Research (PAR) Model 273A potentiostat/galvanostat and Model 270/250 Research Electrochemistry Software. A three-electrode setup was utilized in the experiment, which included a SCE reference electrode, platinum counter electrode and various working electrodes. The temperature was controlled with a Fisher Scientific Model 1016D circulator. 5.2.2 Fourier Transfer Infrared Spectroscopy (FTIR) Transmission FTIR was used to study the structure of cerium-acetate complexation in aqueous solution. A trace amount of cerium-acetate aqueous solution was dropped onto a BaF2 disc and covered with another BaF2 disc for IR testing. Infrared spectra were obtained using Perkin-Elmer 1760X FTIR spectrophotometer. For each sample, 40 scans were collected in the range of 4000 ~ 400 cm-1 with a resolution of 4 cm-1. 216 Counter Electrode Reference Electrode Figure 2: Set up of CeO2 deposition Water Out Working Electrode Water In Figure 5.1 Setup of rotating disk electrode experiment 5.3 Results and Discussions 5.3.1. Ligand-pH Effect on Film or Powder Formation Figure 5.2 A shows a typical CV obtained in 0.1M Ce(NO3)3•6H2O (pH ≅ 4.7), which demonstrates the redox process for Ce(III) to Ce(IV). It is notable that the principal peak position at a potential around 0.9 V (vs. SCE) is ascribed to oxidation of Ce(III) to Ce(IV). On the reverse scan of CV, no reduction peak exists for the cerium species but a relatively smaller reduction peak located at potential -0.05 V due to substrate conditioning effect. Since the substrate is stainless steel, this can be covered with an oxide thin layer.10 When the potential reverses, the formed thin layer will be stripped. The cyclic voltammgram changes with different ligands at various pHs and temperatures. Figure 5.2B illustrates the difference between the pure cerium species system and the cerium- ligand system. As expected, CV shape for cerium-ligand system does not change much from the original, only the potential for the oxidation peak shifts to a higher position relative to the system 217 without the ligand. Figure 5.2C demonstrates the CV scans at different pH from very acidic to very basic solution. Since CeO2 formation occurs at pH’s above 7, the solution pH for this study is kept between 7 to 12. And to eliminate the interference from the substrate, some CVs are done with platinum substrate. Figure 5.3 shows the cyclic voltammetry for the cerium-EDTA complex (which represents the stronger complexes) in solution at pH 8. The reversible wave at ~1.05 V corresponds to the redox change of the metal center in the complex. The metal-ligand complex can be oxidized and reduced, however the metal is held tightly in the complex and free metal ion is not available to form cerium oxide precipitate or films from solution. By increasing the pH, the Ce(III)-L to Ce(IV)-L oxidation is pushed past the O2 evolution and no anodic peak is observed. In fact for pH’s 7 to 12, there is no CeO2 film deposited onto the electrode and no powder formed in the solution during electrolysis for the cerium-EDTA complex at applied anodic currents. Similar results were obtained for other complexes previously listed in Table 3.6: oxalic and citric acid produced no films of CeO2 at anodic currents. 218 0.002 A 0.000 -0.002 -0.004 -0.006 Current (A) -0.008 -0.010 1.5 1.0 0.5 0.0 -0.5 -1.0 0.002 B 0.000 -0.002 -0.004 -0.006 Current (A) -0.008 No Ligand Acetate-Com plexed -0.010 C 0.00 -0.01 -0.02 ---- 2.58 (A) t ---- 4.53 ---- 5.58 -0.03 ---- 7.77 -----8.73 Curren ---- 9.64 -0.04 ---- 10.94 ---- 12.30 -0.05 1.51.00.50.0-0.5-1.0 Potential (V) Figure 5.2 Cyclic voltammgrams for (A) CV of 0.1M Ce(III)(NO3)3 at pH ~ 4.7 and (B) comparison between CV of pure 0.1M Ce(III)(NO3)3 and CV of 0.1M acetate complexed 0.1M Ce(III)(NO3)3, pH ~ 4.7; (C) CV of 0.1M acetate complexed 0.1M Ce(III)(NO3)3, pHs are in range of 2 ~ 13 219 0.02 0.01 0.00 ) -0.01 rrent (A -0.02 u C -0.03 -0.04 -0.05 1.5 1.0 0.5 0.0 -0.5 -1.0 Potential (V) Figure 5.3 Cyclic voltammetry of cerium-EDTA complex in 0.1 M Ce(NO3)3 and 0.1 M o EDTA, pH = 8.0, T = 25 C, and scan rate = 50 mV/sec In contrast, ligands with weaker formation constants, i.e. acetic and lactic acid are able to produce CeO2 powder and films at certain experimental conditions. Figure 5.4 shows the cyclic voltammogram around pH 12 for the cerium-acetic acid complex, which displays a peculiar curve: the reverse cathodic peak positive of anodic peak is seen. And the CV on the stainless steel substrate is more symmetrical than on platinum in regard to peak shape and height. It is found that anodic deposition from this solution does not form a CeO2 film onto the stainless steel electrode, however a variety of precipitates can be formed in the solution. If the solution is stirred with no voltage applied, a gray-green precipitate forms. It is Ce(OH)3. However if the solution undergoes anodic electrolysis at ~ 1.1 V, then a yellow precipitate forms in the cell. X- ray diffraction of this collected powder corresponds to CeO2. Using these data a mechanism is proposed for the CV curve in Figure 5.4: The oxidation peak, at ~ 0.9 V is preceded by a chemical reaction as follows: 2Ce(III) − L ⇔ 2Ce3+ + 2L (Equation 5.5) 220 3+ + 2Ce + 3H 2O → Ce2O3 + 6H (Equation 5.6) Since the formation constant for the cerium-ligand complex in equation 5.5 is relatively small, then free metal ions in solution are available at the higher pH to form Ce2O3, this species can be oxidized to CeO2 at voltages above ~ 0.8 V vs. SHE (Figure 3.1). + − Ce2O3 + H 2O → 2CeO2 + 2H + 2e (Equation 5.7) The thermodynamic voltage calculated for the reaction in equation 5.7 is:8 E = 1.559 − 0.059 pH (Equation 5.8) in which pH ~ 12, resulting to E = 0.85 V vs. SHE or 0.61 V vs. SCE 0.02 0.00 A ) -0.02 A ( t -0.04 en r -0.06 Cur -0.08 0.00 B -0.01 ) A ( -0.02 -0.03 rrent u C -0.04 -0.05 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 Potential(V) Figure 5.4 Cyclic voltammetry of cerium-acetic acid complex in 0.1 M Ce(NO3)3 and 0.1 M acetic acid, pH = 12.0, T = 25 oC, and scan rate = 50 mV/sec, (A) Stainless steel as substrate; (B) Pt as substrate From equation 5.7, the hydrogen ions are formed simultaneously as the oxidation of Ce2O3 to form CeO2, neutralizing the hydroxide at electrode surface so as to lower the local pH 221 ++ at the electrode surface. A chemical reaction can then form Ce(OH)2 ions at the immediate vicinity of the electrode (Equation 5.9), which can then be reduced to Ce2O3 (Equation 5.10). + 2+ CeO2 + 2H → Ce(OH) 2 (Equation 5.9) 2+ + 2Ce(OH) 2 + 2e− → Ce2O3 + H 2O + 2H (Equation 5.10) The reduction peak positive of the oxidation peak on the reverse scan of the CV is the representation of the second step (Equation 5.10). The thermodynamic potential for equation 5.10 can be calculated as:8 2+ E = 0.422 + 0.059 pH + 0.059log[Ce(OH) 2 ] (Equation 5.11) 2+ in which pH ~ 12 and Ce(OH)2 can be estimated from 2+ log[Ce(OH) 2 ] = 19.22 − 2 pH (Equation 5.12) 2+ As bulk solution pH is around 12, log[Ce(OH) 2 ] contribution will be a very small number, which can be omitted in Equation 5.11, leading to 1.13 V vs. SHE or 0.89 V vs. SCE for reaction of Equation 5.10. This calculated potential for the reverse CV scan is 0.28 V positive of the oxidation potential for the anodic CV scan. The ∆V between the two peaks in Figure 5.4 is ~ 0.25 V (1.53-1.28 V for Figure 5.4 (A) and 1.20-0.95 V for Figure 5.4(B)), which matches the calculated thermodynamic potential difference for the equations above. Experimentally, two powder products, Ce2O3 and CeO2, form in this solution. Only by adjusting the pH between 7 to 10 in the cerium-acetate solution, does a CeO2 film deposits onto the electrode. Figure 5.5 shows the CV of the Ce(III)-acetate complex at a pH of ~ 8.0. At a potential of ~ 0.95 V there is a small peak for oxidation. This peak is diminished at pH’s between 7.5 to 10 compared to the CV’s of the complex at all other pH’s. It is proposed that the small peak in Figure 5.5 is attributed to conversion of Ce(III) to Ce(IV) species (Equation 5.15), and this only free ion of Ce3+ involved electrochemical process is limited by a 222 slow release of free Ce3+ in solution from the complex available for oxidation reaction (Equation 5.14). 0.000 -0.003 ) (A t -0.006 rren u C -0.009 -0.012 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.2 Potential (V) Figure 5.5 Cyclic voltammetry of cerium-acetic acid complex in 0.1 M Ce(NO3)3 and 0.1 M acetic acid, pH = 8.0, T = 25 oC, and scan rate = 50 mV/sec Accordingly, the deposition process can be illustrated with following three steps: (1) Complexation and slow release of Ce(III) ions: Ce(III) + xL → Ce(III)Lx (Equation 5.13) Slow 3+ − Ce(III)Lx + H 2O → Ce + HLx + OH (Equation 5.14) (2) Oxidation at the electrode interface; 3+ − 2+ Ce + 2OH → Ce(OH)2 + e − (Equation 5.15) (3) Deposition of CeO2 films; 2+ − Ce(OH) 2 + 2OH → CeO2 + 2H 2O (Equation 5.16) In first step Ce3+ is complexed with a ligand (Equation 5.13), which helps stabilize 3+ Ce(III) in solution and prevent Ce2O3 precipitation. The dissociation of the Ce(III)Lx to Ce is a 223 slow step (Equation 5.14) followed by electron transfer at the electrode. It is the slow released Ce3+ ions that are responsible for the film formation instead of the powder precipitation because the slow supply of reactants tarry the electrochemical oxidation, growing the crystals in order as a film. On the other hand, a large quantity of reactants can result in quick reaction with powder formation. It is important to note that there is a simultaneous formation of CeO2 powder during the deposition of the CeO2 film and other competing pathways such as in Equation 5.6 and 5.7 may occur at the same time. 5.3.2 Mass Transfer Study using Rotation Disk Electrode Linear sweep voltammetry for deposition of cerium oxide onto a CeO2 prelayer gives steady state currents at a potential range from 0.7 to 1.0 V. Figure 5.6 exemplifies the linear sweep voltammograms for such deposition reaction with rotating disk electrode. It shows that the steady-state currents for the deposition of CeO2 are in potential range of 0.7-1.0 V. The same o experiments were conducted at temperature of 25, 40, 50, 70, 80 C on the prelayers of CeO2 in solution of the cerium species and ligand. Levich plots (current density versus the square root of the rotation speed) for a series of temperatures (25 - 80 oC) and rotation rates (0 - 3000 rpm) are shown in Figure 5.7. The series of Levich plots at different temperatures are relatively flat, demonstrating a non-dependence of current on the rotation rate. The deposition of cerium oxide is not under diffusion control, but there is an exponential increase in current with increasing temperature. 224 4 2 0 -2 A) -4 -4 rpm0 rpm50 -6 rpm100 rpm200 rrent (10 -8 rpm500 u rpm1000 C -10 rpm1500 rpm2000 -12 rpm2500 rpm3000 -14 1.0 0.5 0.0 -0.5 Potential (V) Figure 5.6 Linear sweep voltammetry of CeO2 prelayer in solution of 0.1M Ce(NO3)3, 0.1M o acetic acid, pH ~ 7.5 adjusted with NaOH, T = 60 C, rotation rate = 0 ~ 3000 rpm 7.0 6.5 6.0 5.5 5.0 4.5 A) 4.0 -4 3.5 3.0 2.5 25oC rrent (10 o u 2.0 40 C C 1.5 50oC 1.0 60oC 0.5 70oC 0.0 80oC -0.5 0 102030405060 1/2 ω Figure 5.7 Levich plots for Ce(III)/Ce(IV) oxidation at temperatures of 25, 40, 50, 60, 70 and o 80 C at rotation rates from 0 to 3000 rpm Schotty-like rectifying barrier which usually forms at the interface of bulk semiconductor and solution can be borrowed to illustrate this phenomenon in rotation disk experiment.11 The semiconductor/solution electrochemistry uncovers that rectifying barrier of p-type semiconductor 225 can be formed towards cathodic reactions whereas n-type semiconductor towards anodic reactions. The temperature-dependent limiting current could be regarded as the dark saturation current in a solid-state diode. A barrier height for the CeO2/solution interface can be determined from the following equation:11 2 ** ln(I / T ) = ln(Ae A ) − qφb / kT (Equation 17) 2 where I is the current density (A/cm ), T is the temperature (K), Ae is the electrically active area, A** is the effective Richardson’s constant (A/cm2K2), q is the elementary charge on the electron -19 -23 (1.60218 x 10 C), φB is the barrier height (eV), and k is Boltzman’s constant (1.38066 x 10 J/K). Plotting ln(I/T2) versus 1/T at a set rotation rate, the barrier height can be determined from the slope, qφB/kT. From the slope of the plot of Figure 5.8, the barrier height for the cerium oxide/solution interface is 0.377 eV. -18.5 -19.0 -19.5 -20.0 ) 2 (I/T -20.5 ln -21.0 -21.5 -22.0 0.0028 0.0029 0.0030 0.0031 0.0032 0.0033 0.0034 1/T Figure 5.8 Activation plot from linear sweep voltammetry data at a rotation rate of 1000 rpm This temperature-dependent limiting current may be related to a CE mechanism, where C represents a homogeneous chemical reaction and E represents an electron transfer reaction. A slow chemical step (i.e. ligand dissociation) would generate an electrochemical species prior to 226 electron transfer at the electrode. The rate constant of the chemical step would be temperature dependent and has an activation energy of ~0.377 eV. This conclusion is consistent with the mechanism proposed according to CV studies in the previous section. Combining rotation disk voltammetry and cyclic voltammetry, we can summarize the formation of CeO2 films and powders in cerium series and ligand system as follows: At pH < 7: anodic peaks in the CV scan indicate the oxidation of Ce(III) to Ce(IV). However, neither CeO2 film or powder is formed at this condition due to the requirement for the formation of CeO2 is that pH > 7.5 Around 7.5 < pH < 11: a small oxidation peak is detected in CV scan but CeO2 film can be produced as well as CeO2 powder and this is the only range to fabricate CeO2 film on substrate. It is the low complexation of ligand (acetic acid, lactic acid) to cerium species that favors the slow release of Ce3+ from Ce-L complex dissociation (Equation 5.14). These produced Ce3+ free ions are the primary reactant supply for the electrochemical reaction on electrode (Equation 5.15). The slow production of Ce3+ process retards the followed electrochemical oxidation of Ce3+ to form Ce(IV) compound, which makes film formation preferable over powder. This chemical reaction rate-determining CE mechanism is hypothesized according to CVs and LVs with weak ligand acetic acid used and supported by the reversibility of CV spectrum with strong ligand EDTA used. At pH > 11: a reverse reduction peak locates positive of anodic oxidation and the CeO2 powder can be produced but not film. The distinctive CV scan manifests that both oxidation and 2+ reduction occur but with different pathways: Ce2O3 to CeO2 for oxidation and Ce(OH)2 to Ce2O3 for reduction. 227 Other researchers have shown that the oxidation state of cerium ions in solution is sensitive to pH and aeration of the solution.7 In deaerated solutions, the cerium remains in a +3 oxidation state during precipitation on surfaces regardless of pH for cathodic deposition. Also, oxygen reduction in solution plays an important role for the deposition of cerium oxide using cathodic deposition.3 For this reason, we checked anodic deposition of cerium oxide films from deaerated solutions. For deaerated solutions, polycrystalline CeO2 films still form on the working electrode at the same rate. There is a possibility of a small amount of oxidant, i.e. H2O2, formed from the oxidation of water at this pH and voltage. However this would have the formation of the 2+ Ce(OH)2 species by a chemical reaction and not support the CE mechanism. 5.3.3 Coordination Investigation through Varying Ligand Concentration The cerium (III) –ligand complex is so significant for the anodic deposition of CeO2 film. As aforementioned, weak ligands such as acetate (or acetic acid) and lactic acid result in successful growth of film. Further coordination study will furnish us with deeper understanding of the complex system. Ce3+ ion, has the electronic structure of 4f15d06s0, and as a trivalent member of lanthanide series, the most stable ligands are oxygen-ligand with chelating property such as citrate and EDTA that have been explored in our research. Citrate, as a ligand, both carboxylate and hydroxo groups are involved in the complexation and EDTA processes 2 ~ 6 complexing positions corresponding to different forms (basic to acidic). As cerium (III) coordination number can exceed 6, the aforementioned mulitdentate can function as a chelating agent with possible 1:1 molar ratio (e.g. basic form of EDTA).12 Monodentate oxygen ligands are less stable than the chelating ligands and such complexes are prone to dissociation. According to the electronic structure layout, acetate ion can 228 be as a one-electron donor or three-electron donor (Figure 5.9).12 As a complexing agent, it may bind with a metal as unidentate, bidentate or bridging type in which the later two types can be classified as symmetrical and unsymmetrical (Figure 5.10).13-15 The complexes of cerium and acetate were studied in the early 50-60’s.16-18 Since trivalent rare earth metal ions tend to hydrolyze in solution, determination of complex stability constants and coordination numbers are prevented or biased.19 Figure 5.9 Coordination scheme for acetate as ligand: M = metal; (A) one-electron donator; (B) three-electron donator 229 Figure 5.10 Coordination modes of acetate with metal, M = metal; (A) unidentate; (B) bidentate, symmetrical; (B’) bidentate, unsymmetrical; (C) bridging, symmetrical; (C’) bridging, unsymmetrical Figure 5.11A shows CV responses for acetic acid-only and cerium (III) ion-only solutions. Figure 5.11B are the CVs for the complexes in acetic acid and cerium (III) in molar ratios from 0 to 10 for acetic acid: cerium (III) ions. To be comparable with deposition conditions, the solutions for these CV plots are all adjusted to pH 7~8. It is notable (Figure 5.11A) that at the similar pH ~ 7.5, there is no oxidation peaks for the ligand acetate blank or Ce(III) species blank. This is consistent with our predictions: since at pH around 7~8, most Ce3+ - cannot exist in solution by itself, it will combine with OH in solution to form Ce(OH)3. For the acetate blank, no oxidation peak occurs for either methyl group or carbonyl group, because neither group can be oxidized in this range of potentials at this pH. However, when ligand is added in the Ce(III) solution system, in the range of coordinator number below 6, there is always a flat oxidation peak. For acetate/Ce(III) ratio greater than 5, the oxidation peak vanishes (Figure 5.11B). It is expected that at higher coordination numbers, the Ce(III) species are complexed so tightly that there is no free Ce3+ ions available for oxidation. Therefore, it is argued that Ce(III) 230 free ions are the most important participants for the oxidation of Ce(III) to Ce(IV) on substrate to form cerium oxide films in this system. The cerium oxide films can only be electrodeposited on the stainless steel substrate for the ligand/Ce(III) ratios between 1 to 5. FTIR is another technique used to probe the coordination between acetate and cerium (III) species.12,14,20-22 Symmetrical and antisymmetric C-O stretch modes at ~ 1415 ± 20 and ~1570 ± 20 cm-1 respectively are two characteristic IR bands that are subjected to alter in position and shape with different coordination modes. If acetate is used as a unidentate ligand, one of oxygens in C-O is used in bonding with metal, leaving the enhancement of double-bond property in another C-O (C=O) with the emergence of a high frequency emission at 1590 ~1650 cm-1 for the antisymmetric stretch mode. The large increase in antisymmetric band and the corresponding decrease in symmetrical band enlarge the frequency difference between these two vibrations. While acetate acted as a bidnetate, the two C-O bonds retain equivalent in stretching, therefore the two bands are shown without preferred enhancement. Coupling between different acetate groups leads to multiple bands in range of 1400 ~ 1550 cm-1,14 if more than one acetates are bonded to a metal ions.12 Bridging can be regarded as an intermediate mode between uni- and bi-dentate coordinations.12 231 0.005 A 0.000 -0.005 -0.010 -0.015 ) A ( t -0.020 n rre -0.025 u C -0.030 -0.035 Ce Species only -0.040 Ligand (Acetate) only -0.045 1.5 1.0 0.5 0.0 -0.5 -1.0 Potential(V) 10 B 0 -10 A) -4 C :C -20 Ce(NO ) acetate 3 3 -30 1:1 1:2 rrent (10 u 1:3 C -40 1:5 1:10 -50 -60 1.51.00.50.0-0.5-1.0 Potential (V) Figure 5.11 CV diagrams for Ce-acetate complex with different molar ratio (A) Blank; (B) Molar ratio 1:1 to 1: 10 for cerium nitrate to acetate Solutions of cerium (III) nitrate-acetate with molar ratio of cerium (III) to acetate in the range of 1:1 to 1:10 were tested using FTIR. The results are summarized in Figure 5.12 and 5.13. Since the acetate or low concentration of acetic acid in solution exists as ions and the charge on carboxylic ions is delocalized to two C-O functional groups so that the two C-O stretch bands exhibit equivalently indicating weak bidentate-like property by itself,12 which can be seen in 232 Figure 5.12A. Figure 5.12B shows the IR spectrum alterations with the molar ratio between cerium (III) and acetate increasing from 1:1 to 1:10. It is known that at lower concentration of acetate, the lower frequency (~ 1415 cm-1) bands are enhanced and splitted into two bands, for instance, two emissions are displayed at 1404 and 1352 cm-1, respectively for low frequency whereas high frequency peak appears around 1560 and 1577 cm-1 (Figure 5.13A). The pair of 1404 and 1560 cm-1, which results in the difference of ~ 156 cm-1 (Table 5.1), is identical to the emission couple for free acetate ions in solution and is ascribed to the bridging complexing.14 The higher frequency of 1577 cm-1 combined with the low frequency of 1352 cm-1 give a spacing of 228 cm-1(Table 5.1), which can be assigned to unidentate type of coordination;23 though the higher frequency value is comparably lower than that of typical unidentate complex. High intensity of lower frequencies may be attributed to the large quantity of free ions possessing symmetrical properties. However, identical intensities of splitted 1352 and 1404 cm-1 emissions also support the conclusion from above discussion that unidentate and bridging coexist in this system. With the increase of amount of acetate, the lowest band (~1350 cm-1) gradually gets less intense but the splitting is still sustained (Figure 5.13B), indicating the coordination is still two modes mixed but with the bridging gradually dominates. It is demonstrated from Figure 5.12B that from molar ratio of 1:2 to 1:7, similar IR spectra were obtained suggesting the saturating the coordination with increasing the amount of acetate. When the molar ratio is raised to as high as 8 ~ 10, the higher frequency bands of ~1570 cm-1 and ~ 1550 cm-1 merge and present higher intensity than the low frequency bands (Figure 5.13C), indicating more of bridging coordination involvement or free acetate ion presence. In addition, the very low frequency band (~1350 cm-1) reduces the intensity, also indicating the decrement of unidentate mode percentage. Since 233 complexation is saturated when coordination number is as high as 8 ~ 10, extra free acetate ions give rise to new characteristic IR spectra. In conclusion, the coordination number of Ce(III)- acetate is estimated between 6 ~ 8 and the mixed unidentate and bridging is the primary coordination mode for this complex in solution. Therefore, as molar ratio between acetate and cerium (III) is lower than 6, there will be free Ce(III) ions in solution so as that cerium oxide films can form on the substrate, which is consistent with the deposition scheme (Equation 5.13 ~ 5.16). More accurate determination of complex coordination number can be acquired with combining IR or NMR with crystal structure technique as X-ray diffraction on solid-state compound. Table 5.1 IR Assignment for Ce(III)-acetate complex and calculation of frequency difference ( Peak position in cm-1) according to Figure 5.12B Emission A B C D Diff1 Diff2 Emission Type Sym Sym An-sym An-sym Ce : Acetate 0:1 1417.6 1552 134.4 1:1 1352.8 1404 1560.0 1581.2 156 228.4 1:2 1350.9 1411.2 1556.4 1577.6 145.2 226.7 1:3 1350.9 1414.7 1554.6 1577.6 139.9 226.7 1:4 1353.9 1414.7 1556.1 1576.5 141.4 222.6 1:5 1350.9 1416.5 1554.6 1577.6 138.1 226.7 1:6 1347.8 1416.5 1554.2 1576.5 137.7 228.7 1:7 1351.6 1416.5 1556.1 1576.5 139.6 224.9 1:8 1349.7 1414.7 1556.1 1576.5 141.4 226.8 1:9 1347.8 1414.7 1556.1 1576.5 141.4 228.7 1:10 1350.7 1415.1 1556 1576.1 140.9 225.4 234 110 A 108 106 104 -1 102 1416 cm %T 1550 cm-1 100 98 96 94 1600 1500 1400 1300 1200 1100 1000 Wavelength 140 120 100 C :C Ce(NO )3 acetate 3 80 1:1 1:2 %T 1:3 60 1:4 1:5 1:6 1:7 40 1:8 1:9 1:10 20 Acetate Blank 1600 1400 1200 1000 Wavenumber (cm-1) Figure 5.12 FTIR spectra for (A) acetate blank; (B) cerium(III)-acetate complex at different molar ratio: Ce:acetate = 1:1~1:10 together with blank 235 105 A 100 95 90 %T 85 80 75 140 B 130 120 110 %T 100 90 80 120 C 110 100 90 %T 80 70 60 1600 1500 1400 1300 1200 1100 1000 -1 W avenumber (cm ) Figure 5.13 FTIR spectra for cerium (III)-acetate complex at the range of 1000-1650 cm-1 and molar ratio of cerium and acetate (A) 1:1; (B) 1:6; (C) 1:10 5.4 Chapter Conclusions Cyclic voltammetry and rotation disk experiment were conducted to investigate the electrodeposition of CeO2, which suggests a CE mechanism and the possible pathway of 24 formation of CeO2 film through the following three steps: Step 1: Weak association and slow dissociation of Ce(III)-Ligand: 236 Ce(III) + xL → Ce(III)Lx Slow 3+ − Ce(III)Lx + H 2O → Ce + HLx + OH Step 2: Electron transfer at electrode to oxidize Ce(III) to Ce(IV): 3+ − 2+ Ce + 2OH → Ce(OH) 2 + e − Step 3: Final formation of CeO2 films on electrode: 2+ − Ce(OH) 2 + 2OH → CeO2 + 2H 2O The conspicuous CV scan for high pH region, where only powders are formed, recommends that a pair of distinct redox reactions occur in this range.24 The chemical reaction to 3+ convert Ce to Ce2O3 is followed by the electrochemical reaction on the forward oxidation: + − Ce2O3 + H 2O → 2CeO2 + 2H + 2e 2+ Whereas chemically transform of CeO2 to Ce(OH)2 is ahead of electrochemical reduction corresponding to the peak on the reverse scan: 2+ + 2Ce(OH) 2 + 2e− → Ce2O3 + H 2O + 2H Preliminary investigation of the coordination between Ce(III) and ligand acetate insulates a mixed complex mode for the system in deposition solution. Simulations can be exerted to the experimental CV scan to confirm the aforementioned hypothesis of CeO2 film growth. In-situ monitor of the deposition process and film growth is another possibility to deeply appreciate the formation of CeO2 films. 237 Chapter References (1) Therese, G. H. A.; Kamath, P. V. Chemistry of Materials 2000, 12, 1195-1204. (2) Switzer, J. A. American Ceramic Society Bulletin 1987, 66, 1521-1524. (3) Aldykiewicz, A. J. J.; Davenport, A. J.; Isaacs, H. S. Journal of the Electrochemical Society 1996, 143, 147-153. (4) Li, F.; Newman, R. C.; Thompson, G. E. Electrochima Acta 1997, 42, 2455-2464. (5) Li, F.; Thompson, G. E.; Newman, R. C. Applied Surface Science 1998, 126, 21-33. (6) Li, F.; Thompson, G. E. Journal of the Electrochemical Society 1999, 146, 1809-1815. (7) Davenport, A. J.; Isaacs, H. S.; Kendig, M. W. D. In The Electrochemical Society Proceeding Series; Baer, R., Clayton, C. E., Davis, G. D., Eds.: Pennington, NJ, 1991; Vol. 433, pp 91-97. (8) Pourbaix, M.; NACE: Houston, TX, 1974, p 194. (9) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; 1st ed.; John Wiley & Sons, Inc: New York, 1980. (10) Crossland, A. C.; Thompson, G. E.; Skeldon, P.; Smith, G. C.; et.al. Corrosion Science 1998, 40, 871-885. (11) Sze, S. M. Physics of Semiconductor Devices; 1st ed.; Wiley-Interscience: New York, 1969. (12) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry: a Comprehensive Text; 4th ed.; John Wiley & Sons: New York, 1980. (13) Alcock, N. W.; Tracy, V. M. Journal of the Chemical Society Dalton Transactions: Inorganic Chemistry 1976, 2238-2243. 238 (14) Alcock, N. W.; Tracy, V. M. Journal of the Chemical Society Dalton Transactions: Inorganic Chemistry 1976, 2243-2246. (15) Alcock, N. W.; Tracy, V. M. Journal of the Chemical Society Dalton Transactions: Inorganic Chemistry 1976, 2246-2249. (16) Sonesson, A. Acta Chemica Scandinavica 1958, 12, 165-181. (17) Sonesson, A. Acta Chemica Scandinavica 1960, 14, 1495-1498. (18) Sonesson, A. Acta Chemica Scandinavica 1958, 12, 1937-1954. (19) Kolat, R. S.; Powell, J. E. Inorganic Chemistry 1962, 1, 293-296. (20) Mathey, Y.; Greig, D. R.; Shriver, D. F. Inorganic Chemistry 1982, 21, 3409-3413. (21) Stubenranch, J.; Vohs, J. M. Journal of Catalysis 1996, 159, 50-57. (22) Vratny, F.; Rao, C. N. R.; Dilliy, M. Analytical Chemistry 1961, 33, 1455-1455. (23) Liu, S.; Liu, F.; Guo, H.; Zhang, Z.; Wang, Z. Physics Letters A 2000, 211, 128-133. (24) Golden, T. D.; Wang, A. Q. Journal of The Electrochemical Society 2003, in press. 239 CHAPTER 6 ELECTROCHEMICAL SYNTHESIS AND CHARACTERIZATION OF CeO2 / MONTMORILLONITE NANOCOMPOSITES 6.1 Introduction Recently, layered silicates have been popular in the formation of polymer/layered silicate nanocomposites.1-6 Layered silicates naturally exist as platelets. The polymer/layered silicate nanocomposites exhibit improved mechanical properties in hardness and elastic modulus, improved heat resistance and lower gas permeability. All the above properties are due to the intercalation and delamination of layered silicates into small platelets within the polymer matrix.1,3 Layered silicates have been combined with ceramic materials to form composites. However, most ceramics /layered silicate nanocomposites that have been studied belong to ceramics intercalated clay.7-10 In this structure, the layered silicate is the structure matrix and another ceramic material acts as the additive. As described, the ceramics particles intercalate into the interlayer of layered silicates so that the interlayer spacing increases. The employed ceramics 11 12 13 14 15 16 17 7 18 19 includes CdS, Fe3O4, , SiO2, TiO2, ZrO2, Al2O3, Fe2O3, Cr2O3, CeO2, V2O5 and some mixed oxide clusters.8,20 There is rarely a report on the composites with the ceramic material as the main structure matrix and clay platelet as the additive. The reinforcement of the polymer with silicate platelet encourages the first attempt to synthesize cerium oxide / layered silicate nanocomposites. Anodic electrochemical synthesis was employed in the research to form cerium oxide/ layered silicate nanocomposites since this method achieved success in obtaining nanocrystalline cerium oxide powders and thin films. 240 The objective of this chapter is to illustrate the fundamental studies of three types of nanoclays used in the composite formation and presents the preliminary results in regard to cerium oxide/layered silicate nanocomposite powders. Powder X-ray diffraction, FTIR and differential scanning calorimetry (DSC) were used to demonstrate the structure characteristics and properties of the employed layered silicates and the composites. Chapter 7 will focus on the composite thin film fabrication and property evaluation. 6.2 Experimental 6.2.1 Materials and Electrosynthesis of Composites Three types of layered silicates have been applied to the study of formation and properties of the composites: sodium cloisite supplied by Southern Clay Products, treated and untreated montmorillonite provided by Nanocor. A certain concentration of layered silicate suspension in D.I. water was stirred or ultrasonicated overnight for sufficient exfoliation of clay, which was followed by slow decanting of the suspension into a mixture of cerium salt and ligand. This draft reports the composite results from nitrate salt and acetate ligand. The applied concentration of layered silicate is between 0.5 ~ 50% (w/w) of cerium nitrate concentration. A three-electrode configuration was employed for the synthesis with stainless steel disk, chromel wire and the saturated calomel electrode (SCE) as the working, counter, and reference electrodes, respectively. A galvanostatic method (-10 mA, electrode area is 1.77 cm2) was used to produce the nanocomposite powders. 6.2.2 Composite Property Analysis 6.2.2.1 X-ray diffraction A Siemens D500 diffractometer with Cu Kα radiation (λ = 0.15405 nm) was used to investigate X-ray diffraction (XRD) behavior of pure layered silicate and ceramics / silicate 241 composites. All tests were carried out at a 2θ ranging from 2o to 100o, in which the primary reflections of layered silicate usually appear under 40o. To diagnostically identify clay, both air dried-Na-cloisite and ethylene glycol solvated-Na-cloisite were analyzed using a step size of 0.01o 2θ and a dwell time of 3 second per step. As obtained layered silicate was exposed to ethylene glycol atmosphere at 60oC at least overnight to accomplish ethylene glycol solvation of layered silicate,21 which orders the natural clay. For the composites, the employed step size and dwell time are 0.01o 2θ and 1 second, respectively. 6.2.2.2 Fourier Transfer Infrared Spectroscopy (FTIR) FTIR was used to study the structure of layered silicate and further probe the existence of layered silicate in composites. Both transmission and reflection techniques were employed to characterize layered silicate / ceramics composites. In transmission mode, 1~2 mg of powder sample and ~ 100 mg of KBr were mixed and then pressed to disc of d = 1.0 cm under deaerated condition. Such discs were heated at 150oC overnight to eliminate the adsorbed moisture in the sample. Infrared spectra were obtained using Perkin-Elmer 1760X FTIR spectrophotometer. For each sample, 40 scans were collected in the range of 4000 ~ 400 cm-1 with a resolution of 4 cm-1. Attenuated total reflectance (ATR) spectra of unsintered and sintered pellets were collected on a Nicolet Nexus 470 FTIR spectrometer that is equipped with IR source, silicon splitter. To minimize the inference of silicon, 32 scans were recorded in the range of 4000 ~ 500 cm-1 with a resolution of 4 cm-1. Raman spectra of CeO2 / Na-cloisite composites were collected using the same method as employed for CeO2 powder. 6.2.2.3 Differentiated Scanning Calorimetry (DSC) 242 Differentiated scanning calorimetry (DSC) was to discern the thermal properties of the nanocomposites with the variation of the clay concentration in composites. The tests were conducted on a Perkin-Elmer DSC 6 instrument in the temperature range of 20 ~ 400 oC with a constant heating rate of 10 oC/min and cooling rate of 25oC/min and 5 ~ 20 mg of powder sample was loaded for each experimental run. The system was calibrated with elemental indium. 6.3 Results and Discussions 6.3.1 CeO2 / Na-cloisite Composites It is noted that “green” solid powders were formed by anodic electrosynthesis of cerium oxide powder and concurrent exfoliation and insertion of Na-cloisite into cerium oxide. The formed powder demonstrated the appearance of pure cerium oxide powder with a light yellow color. The further analyses based on the X-ray diffraction and FTIR will illustrate the effect of Na-cloisite on cerium oxide powders. 6.3.1.1 XRD Results The crystallographic structures of Na- montmorllonites were detected with XRD. Figure 6.1 illustrates the XRD patterns of Na-cloisite in comparison to nanocrystalline cerium oxide powder. The CeO2 powder is a random pattern with typical fcc fluorite structure and the reflection assignment detail is discussed in Chapter 4. The basal diffractions (00l) are significant for interpreting montmorillonite. The glycol-solvated preparation of Na-cloisite gives a strong 001 reflection at 5.18o 2θ, and using Bragg’s law, the basal spacing can be calculated as 17.1 Å. Accordingly, the series of (00l) reflection position and corresponding interspacing are established and tabulated in Table 6.1. The calculated (00l) are equidistantly distributed in XRD pattern, which is a critical diagnostic property for clay sample diffraction patterns21 and this characteristics is demonstrated as blue lines in Figure 6.1. Experimentally, (002) and (003) 243 reflections emerge at 10.06 and 15.75o 2θ, respectively in the glycol-treated pattern, which matches the calculated patterns (Table 6.1). However, in air–dried samples, (001) and (002) shift to 7.23 and 14.73o 2θ respectively, and most higher-order reflections are not detectable, which reflects the inherent disorder of natural clay samples and interstratified property. The calculated basal spacing for (001) from the position 7.23o is 12.2 Å, close to the Southern Clay Products published d-spacing parameter for this clay (11.7Å). The (060) reflection is another important peak in XRD for detecting clays and can provide supplementary information for distinguishing dioctahedral and trioctahedral types according to different d-spacing and 2θ position.21 Experiment shows that the (060) reflection of Na-cloisite is at 61.87o 2θ with a d-spacing of 1.498Å, which corresponds to the typical dioctahedral clay –montmorillonite (2θ (060) = 62.22 ~ 61.67o and d (060) = 1.492 ~1.504 Å). In addition, there are three two-dimensional bands (02; 11) at 19.75o 2θ, (20;13) at 35.23o 2θ and (04;22) at 53.92o 2θ present in cloisite XRD pattern (Figure 6.1, Table 6.1) due to turbostratic disordered stacking of the layers. The shape of bands is observed to be asymmetric with a tail and structure factor plays a crucial role in the shape variation of the bands. Band (20;13) has been recognized as the best to evaluate turbostratic disorder since it is more asymmetric or stronger than the other two common bands. Since the reflections for (00l) in a pure typical clay would possess identical width, three narrow reflections in XRD patterns indicate presence of impurity, which is identified as quartz silica in Na- cloisite.21,22 244 0 20406080100 001 002 003 004 005 006 007 008 009 glycol solvated Na-cloisite air dried-Na-cloisite CeO Powder 2 y it s n 02;11 quartz te 20;13 060 In quartz quartz 04;22 (111) (220) (311) (200) (331) (511) (222) (400) (420) (422) 0 20406080100 2θ (degrees) Figure 6.1. XRD patterns for air-dried Na-cloisite and ethylene glycol-solvated Na-cloisite in comparison to CeO2 powder. Blue lines represent the calculated (00l) ordered reflection position for clay. The corresponding peak positions and d-spacings are listed in Table 6.1. 245 Table 6.1 Experimental and calculated XRD pattern position and d-spacing for Na-cloisite sample in Figure 6.1, c & m refers calculated and measured, respectively. Glycol treated Na-cloisite Air dried Na-cloisite Reflection 2θ (c) d-spacing 2θ (m) d-spacing 2θ (m) d-spacing (degrees) (nm) (degrees) (nm) (degrees) (nm) 001 5.18 17.06 5.18 17.06 7.23 12.23 002 10.37 8.53 10.06 8.79 14.73 6.22 003 15.58 5.69 15.75 5.63 N/A N/A 004 20.83 4.27 N/A N/A N/A N/A 005 26.12 3.41 N/A N/A 28.32 3.15 006 31.46 2.84 N/A N/A N/A N/A 007 36.88 2.44 N/A N/A N/A N/A 008 42.39 2.13 42.58 2.12 N/A N/A 009 47.99 1.90 47.91 1.90 N/A N/A 60 61.87 1.450 61.86 1.450 02;11 19.75 19.72 20;13 35.23 34.98 04;22 53.92 53.95 Quartz 21.92 21.9 Quartz 31.62 31.57 Quartz 45.4 45.33 The electrosynthesized CeO2 / Na-cloisite composites retain the fcc fluorite structure characteristic of CeO2 for different contents of Na-cloisite in composites (1, 2, 3, 5, 10, 20, 30%). XRD patterns of composites in Figure 6.2 indicate that as-produced cerium oxide powder is randomly ordered, i.e. participation of Na-cloisite in reaction has no influence in orientation of structure. On the contrary, with formation of the composite, Na-cloisite experienced tremendous changes in its structure. It is apparent that strong (001) reflection vanishes from lower angle regions of XRD. It can be attributed to the exfoliation of the nanoclay layer structure when forming composite. The same phenomenon has been observed in the production of metal / Na- cloisite composite.23 However, for different percentages of Na-cloisite content in composites, the 246 (002) reflection is detected in diffraction. The (002) reflection at ~ 12.00o 2θ is around 2o positive to the glycol-solvated Na-cloisite. A small reflection emerges at ~ 64.00o 2θ, which is not present in pristine clay. This peak corresponds to α-quartz and could result from the breakdown of silicates because of drastic agitation or ultrasonication when the clay suspension was prepared and electrochemical synthesis was carried out. The basal swelling and shrinking should not interfere the diffraction property of x/y direction, but the agitation or ultrasonication not only delaminate the layered structure in z-stacking direction but also break down the order in x/y-extending direction of the clay to some degree and form small silica platelets. The clay platelets disperse within the cerium oxide structure with every orientation, not limited to horizontal direction. This would be reasonably explains the upward dislocations of stacking and the structural breakdown of extending dimension. In short, the shifted (002) reflection and silica peak appearance in Na-cloisite associated with intact CeO2 pattern indicate the insertion of the clay platelets into the matrix of CeO2 instead of otherwise intercalation of cerium oxide into clay gallery. The particle sizes of the composites were estimated from the FWHM of CeO2 (111) reflection by using Scherrer equation (Equation 2.37) and listed in Table 6.2. The composites are in nanoscale, which qualifies as-produced composites being called nanocomposites. Compared with pure cerium oxide powder made using the same approach and on the same condition, cerium oxide particle in composites exhibits larger size. But with increasing content of clay, the particle size becomes smaller. This can be explained from the view of the exfoliation and size of inserted platelets. When the clay addition is in low content, ultrasonication and agitation can delaminate the nanoclay homogeneously and exfoliate it to small platelets,24,25 which are readily dispersed into the structure matrix when cerium oxide crystals form. The incorporation of the 247 exfoliated clay platelet with cerium oxide forms tactoids with larger size than pristine cerium oxide particles. As nanoclay concentration increases, the extent of homogeneous and complete exfoliation of nanoclay is impeded somehow, which results in some degree of relatively larger size of platelets that cannot be inserted into cerium oxide structure. Therefore, at higher percentage of clays, composites exist as cerium oxide particles and clay platelets, respectively. The average particle size of the cerium oxide in the higher clay-composite is larger than that at lower clay-composite and closer to pure cerium oxide particle size as the concentration of clay in nanocomposite increases. Table 6.2 carries out this hypothesis in phenomenon: when the nanoclay content is as high as 20 ~ 30%, the particle size is as small as 4.8 nm, which is close to that of pure nanocrystalline cerium oxide (4.6 nm). Table 6.2 Particle size vs. content of Na-cloisite in composites Sample CeO2 1% 2% 3% 5% 10% 20% 30% Reflection 28.45 28.80 28.70 28.75 28.69 28.75 28.70 28.70 Particle Size 4.63 5.45 5.45 5.11 5.06 5.2 4.79 4.79 248 100 100 A B ) ps (c (cps) y y it NTC (002) ns nsit NTC (002) te e t In n I α -quartz α -quartz 0 0 0 20406080100 0 20406080100 Two Theta (degrees) Two Theta (degrees) 100 100 C D ) ) s p ps c c ( ( y t NTC (002) ity i s n ns NTC (002) te In Inte α -quartz α -quartz 0 0 0 20406080100 0 20406080100 Two Theta (degrees) Two Theta (degrees) 100 100 E F ) ) s s p p (c (c y y it it s s NTC (002) NTC (002) n n te te In In α -quartz α -quartz 0 0 0 20406080100 0 20406080100 Two Theta (degrees) Two Theta (degrees) 100 G ) s p (c y it s n NTC (002) te In α -quartz 0 020406080100 Two Theta (degrees) Figure 6.2 XRD patterns for CeO2 / Na-cloisite (NTC) composites at different clay concentration (w/w): (A) 1%; (B) 2%; (C) 3%; (D) 5%; (E) 10%; (F) 20%, (G) 30% 249 6.3.1.2 FTIR Results For layered silicates, infrared spectra can give information on the occupancy of octahedral and tetrahedral sites as well as stacking and arrangement of the sheets.26 Infrared spectra of Na-cloisite were collected in the region of 4000-400 cm-1 using KBr methods, since the absorbance bands due to structural OH and Si-O groups have significant roles in identification and analysis of unique clay using infrared method. The KBr spectra of Na-cloisite are shown (Figure 6.2) as well as for CeO2 / Na-cloisite nanocomposites (Figure 6.3) in the regions for OH stretching (A) and Si-O stretching (B). To eliminate the adsorbed water, the detected samples were heated at 150oC overnight.26,27 The spectra for Na-cloisite before and after heating treatment are illustrated in Figure 6.3. To simplify the discussion, the spectra of composites before heating are not shown in this report. The infrared spectra of pure CeO2 are plotted together with Na-cloisite in Figure 6.4 for comparison. Na-cloisite is a di-octahedral smectite, in which the position and shape of the OH stretching band is determined by the octahedral atoms coordinated with the hydroxyl groups.26,28,29 Other related factors include the cations radius, cation type valence charge and hydration energies of the exchangeable cations.12,30 The band at 3631 cm-1 can be ascribed to the inner OH out-of-plane stretching vibration of cloisite clay. And the shoulder peak around ~ 3622 cm-1 in the heated sample is also assigned to the structural OH stretching vibration, which indicates relative high percentage of Al in the octahedral structure. This type of inner OH lies between the tetrahedral and octahedral sheets and the distribution of OH stretching bands varies slightly with the interlayer water content, which can be seen from the 3626 cm-1 absorption for air-dried clay shifting to 3622 cm-1 when heated. Another broad absorption around 3447 cm-1 is due to the water moisture in clay sample. This band is observed to level off when a KBr- 250 containing-pellet was heated overnight at 150oC, as shown of the black line in Figure 6.2A. The OH bending vibrations can be found in low frequency region of infrared spectra and the positions of such inflections reflect the occupancy of octahedral sheet in the layered silicate structure. Dioctahedral clays have been observed to have this vibration in the range of 950 ~ 800 cm-1, different from the lower wavenumber vibrations at 700 ~ 600 cm-1 for trioctahedral ones.26,27 In Na-cloisite layered clay, the 917 cm-1, 883 cm-1 and 848 cm-1 absorption bands represent inner hydroxyl bending in Al-Al-OH, Al-Fe-OH and in Al-Mg-OH structure respectively, reflecting partial magnesium and iron substitution for aluminum in octahedral sheets.26,27 Furthermore, the intensity of Al-Al-OH band is observed higher than that of Al-Mg- OH and Al-Fe-OH, indicating that only a small portion of octahedral aluminum is replaced with magnesium or iron, agreeing well with high aluminum content inference from OH stretching frequency previously. In addition, the hydroxyl deformation band in regard to water locates at ~ 1635 cm-1 (not shown), intensity of which reduces substantially as the sample was subjected to 150oC-heating. -1 -1 The OH stretching band (3631 cm ) and bending band (917 cm ) appear in CeO2 / Na- cloisite nanocomposites, indicating the presence of the Na-cloisite in the composites. As expected, when concentration of Na-cloisite in composite increases, band signals become more obvious, though intensities cannot be compared here because it was not quantitative. The position of the hydroxyl stretching shifts negatively with decreasing clay and increasing CeO2 content in the composites whereas the hydroxyl bending band moves upward along frequency (Table 6.3). It is known that isomorphous substitution of Al3+ by Mg2+ or Fe2+ in octahedral structure brings negative charge on the apical oxygens of the tetrahedra of the original sodium cloisite clay. This oxygen interacts with hydroxyl group by OH….Oap hydrogen bond formation. 251 This hydrogen bond strength plays a significant role in the OH vibration frequency, which can be seen in the cation intercalation experiments. Cation (Li+, Cu2+, Cd2+) migration and intercalation into layer structure modify the environment and orientation of hydroxyl group and compensate 31 for the positive charge deficit. In this case, the absence of the negative Oap weakens hydrogen bond interaction related structural OH, OH stretching shifts to a higher wavenumber with increasing the concentration of inserted cations. Therefore, the downward shift in the position of the OH stretching band associated with an increase in OH bending frequency usually evidences relative hydrogen bonding forming or enhancing for hydroxyl containing structure. In the formation of CeO2 / Na-cloisite composites, the Na-cloisite was suspended by ultrasonication or drastic agitation before it was subjected to mixing with cerium salt solution, clay loses the long range structural order and delaminates into small platelets. Dilution is another way to enhance delamination of clay,25 therefore, the lower concentration of Na-cloisite, the higher percentage of the delaminated forms. As clay was exfoliated, more of apical oxygen atoms will be exposed in the surface of octahedral or tetrahedral sheets, creating more chances of hydrogen bonding between these tetrahedral apical oxygen atoms and the inner hydroxyl group in octahedral sheets. In short, the delaminated form in composites explains downward and upward shift in OH stretching and bending absorption, respectively. Since acetate was employed in the formation of composite, oxygens in acetate are possible to participate hydrogen bonding. To eliminate acetate effect, the complexation of cerium (III) ions with acetate was arranged prior to the addition of clay suspension. The presence of Si-O stretching and bending in Na-cloisite and CeO2 / Na-cloisite composites is demonstrated as well as those of O-H. The deviation of Si-O network in clay tetrahedral structure from the ideal hexagonal symmetry renders the complex infrared spectra in 252 clay samples.28 The absorption position and shape for Si-O stretching and bending vibration also fingerprint different arrangements and stackings of layered silicates. Theoretically, montmorillonite shows three well-defined in-plane vibrations and one perpendicular vibration for stretching mode with respect to the structure sheets. The perpendicular mode lies around 1080 cm-1, two degenerated vibration modes locate at lower frequencies ~ 1034 and 1003 cm-1 respectively, and the third in-plane vibration lies at higher frequency ~ 1100 cm-1, which is a longitudinal mode.28 A broad band between 1130 and 1000 cm-1 in Figure 6.2B can be assigned to the complex Si-O stretching vibration, which is usually the characteristic band for montmorillonite and this band is seldom deconvoluted.26,27 The diffuseness and broadness of the band can be ascribed to the structure irregularity of the natural Na-cloisite. And the non- symmetry of the band is due to quartz impurity, which can be distinguished by Si-O stretching vibration in amorphous silica at ~1100 cm-1 27 and its presence has been verified by X-ray diffraction.32 Si-O bending vibrations appear at 522 cm-1 and 468 cm-1 in Na-cloisite. These two bands correspond to Si-O-Al (octahedral Al) and Si-O-Si deformations respectively. The band at ~ 624 cm-1 can be assigned to Si-O and Al-O coupled out-of-plane vibrations according to literature.27 The presence of quartz impurity in this natural montmorillonite can be identified with an amorphous silica band (~ 800 cm-1) in infrared spectra as well. It is recognized that ~ 550 cm-1 band is a characteristic peak in phonon modes spectral range for cubic crystalline forms of rare earth oxides. Generally, the stoichiometric cerium dioxide can be optically transparent in the wavenumber range of 4000 ~ 600 cm-1.33-35 The 1300 ~ 600 cm –1 region spectrum of cerium oxide powder (Figure 6.3B) show: two bands at 1052 and 1021 cm-1; one small band at 931 cm-1; two sharp bands at 856 and 837 cm-1; one small peak at 726 cm-1; and two bands at ~ 654 and 621 cm-1. The 726 cm-1 can be attributed to Ce-O 253 vibration, the 1021 cm-1 is usually assigned as the first overtone mode of the fundamental vibration at ~550 cm-1. A band at ~ 2103 cm-1 is found in IR spectrum (Figure 6.3A), which can be regarded as a possible overtone for 1021 cm-1. The peaks at 856 and 837 cm-1 can be ascribed − to adsorbed superoxide species, (O2ads ) which is the characteristic frequency of O-O vibration with bond order of 1.05.36,37 Some other researchers also assign these two peaks to the 35,38 carbonates due to the adsorbtion of CO2 in air. Since in this electrochemical synthesis, acetate was employed to be as a ligand, 1021 and 1052 cm-1 can also be ascribed to the residual acetate. Positions of the Si-O-Al and Si-O-Si bendings at 522 and 468 cm-1 do not change much with the formation of composites. The infrared signals diminish as clay content in composites decreases from 50% to 10%, and as the content of the Na-cloisite reaches below 5%, there is no obvious inflection for Si-O bending vibrations shown. Similar trend is available for the Si-O and Al-O coupled vibration bands. In contrast, infrared absorption in the range of 1000 ~ 1100 cm-1 can be used to distinguish CeO2 / Na-cloisite composites from pure Na-cloisite. Upon exfoliated, long-range order of the layered silicate is destroyed in the composites. Small tetrahedral sheet platelets possess less tension and exhibit more regularity in structure compared with large ordered layered silicate, resulting in a relative sharper absorption band centered around 1049 ~ 1035 cm-1 in Si-O infrared spectra. There is no obvious shift for the band position as composites form and clay content varies, even with heating at 150oC heating overnight. This observation contrasts the previously reported Si-O absorption upward migration phenomenon associated with cations insertion into octahedral and tetrahedral sites31and further consolidates the view that the clay platelets fill into matrix of cerium oxide crystals instead of the intercalation of oxide particles into the interlayer of clay. In addition, the absorption between 1000 and 1100 cm-1 is 254 observed differently when varying the clay content in composites. For high content of clay, the Si-O band in clay predominates over vibration from cerium compound, this band is observed as a band basically centered at ~ 1045 cm-1 (e.g. 50%, 30%, 20% and 10% in this report). As the clay concentration in composition becomes lower (e.g. 5%, 3%, 2%, 1%, 0.5%), the band splits into two peaks, which locate at ~1046 and ~1023 cm-1, respectively (Figure 6.5). Impurity acetate absorption overlapping with Si-O from clay could contribute to the emerging of these two peaks. A B 1038 468 522 432 3631 3626 3447 917 624 883 3622 ce ce 848 ban 798 sorban Absor b A 3800 3700 3600 3500 3400 1200 1000 800 600 400 Wavenumber (cm-1) Wavenumber (cm-1) Figure 6.3. IR spectra of Na-cloisite in (A) OH stretching region (3850-3350 cm-1) and (B) 1300- 400 cm-1 region. Red line: before heating; Black line: after heating at 150oC overnight; Blue line: CeO2 255 + + Table 6.3 OH vibration band shifts versus concentration of Na -cloisite in CeO2/Na - cloisite nanocomposites Na-cloisite Clay, % 100 50 30 20 10 5 3 Peak, cm-1 3631.00 3630.00 3628.00 3627.00 3626.98 3626.11 3621.01 Stretching Shift, cm-1 N/A -1.00 -3.00 -4.00 -4.02 -4.89 -9.99 Peak, cm 917.07 920.84 920.96 921.07 922.03 N/A 934.08 Bending -1 Shift, cm-1 N/A 3.77 3.89 4.00 4.98 N/A 17.01 A B n n o o i i t p 0.5% r o sorpt s 1% Ab Ab 2% 3% 5% 10% 20% 30% 50% 3800 3700 3600 3500 3400 1200 1000 800 600 400 Wavenumber (cm-1) Wavenumber (cm-1) Figure 6.4 IR spectra of CeO2/Na-cloisite nanocomposites in (A) OH stretching (3850-3350 -1 -1 cm ) and 1300-400 cm region 256 1023.1 cm-1 1046.7 cm-1 Absorbance 1060 1050 1040 1030 1020 1010 Wavenumber (cm-1) -1 Figure 6.5 Zoom-in of infrared spectra in the range of 1060 ~ 1010 cm for CeO2 / Na-cloisite composite at 0.5% Na-cloisite illustrates the overlapping of Si-O stretching vibration in clay and vibration in CeO2 Photo-luminance interferes with the Raman signals for dispersive Raman spectroscopy for the clay samples. Raman spectra of the CeO2/ Na-cloisite nanocomposites exhibit characteristics of pure CeO2 powder samples that have been discussed in Chapter 4. With increasing the concentration of Na-cloisite in the nanocomposites, the spectrum patterns present more luminance characteristics. Figure 6.6 illustrates the Raman spectra for pure CeO2 powder (labeled as 0%), pure Na-cloisite (labeled as 100%) and CeO2 / Na-cloisite nanocomposites at varied concentrations of clays. 257 y t i s 0% 1% Inten 2% 3% 5% 10% 20% 30% 100% 500 1000 1500 2000 Wavenumber (cm-1) Figure 6.6 Raman spectroscopic measurements for the CeO2 / Na-cloisite nanocomposites 6.3.1.4 DSC results Figure 6.7 demonstrates the peak height-normalized heating thermographs of Na-cloisite, cerium oxide and nanocomposite samples at different contents of clay. Cerium oxide powder made from electrosynthesis shows endo- and exo- thermal signals during heating and cooling process respectively at temperature of ~ 273oC, which are ascribed to the reduction and re- oxidation of non-stoichiomtric CeO2-x. Usually for microcrystalline CeO2, the reduction occurs at temperature higher than 500oC. However, nanocrystalline cerium oxide has non-stoichiometric structure and large surface area, which confer such powder with high capability to desorb and then absorb surface species such as superoxides (Oδ-) in this oxygen-deficient compound.39 Therefore, nanocrystalline cerium oxide power has heat flow signals at lower temperature than that required for microstructured counterpart, even as low as < 150oC, which is due to reduction reaction. It is concluded from Chapter 4 that the smaller particle size usually leads to higher non- stoichiometry. So the smaller size of nanocrystalline cerium oxide (5.56 nm) has the lower 258 reduction reaction temperature (143.4oC) compared with relatively larger nano cerium oxide particle (6.20 nm, 273.3oC). In addition, reduction-oxidation reversibility totally depends on the availability of the oxygen deficiency. The broad peak between 350oC and 400oC for 6.20 nm cerium oxide during heating process can be regarded as the rearrangement of reduced CeO2-y to a stable state, which usually releases an amount of heat. Pure Na-cloisite has a single volatile emission at ~ 211.5oC, which is assigned to the loss of interlayer water in layered structure together with adsorbed CO2 species according to previous investigation of thermal property of montmorillonite.40,41 For composites, a single endo- signal shows around temperatures between 150oC and 200oC. It is indicative that Na-cloisite and cerium oxide have merged into one phase. During the composite formation, nanoclay experienced exfoliation and gallery spacing domain is destroyed, which has been verified by the XRD result that (001) reflection of clay vanished from composite patterns. Accordingly, the interlayer water, as the main evolution product for Na-cloisite in the temperature range of 200 ~ 500oC, is lost accompanying the vanishing of interlayer prior to that the composite was subjected to DSC test. Therefore, the corresponding endo-peak for pristine Na-cloisite does not show up in DSC spectrum of the composite and the single endo-signal is basically attributed to the reduction of cerium oxide component. The presence of structure rearrangement spectrum in composites validates above postulation about the main spectrum assignment. With increasing clay content in composites, compared with the 6.2 nm of pure cerium oxide powder, the particle size of the nanocrystalline cerium oxide in composites decreases from 5.4 nm to 4.8 nm (Table 6.2), which generally induces higher percentage of non- stoichiometry. As mentioned, the temperature for reduction to occur is relevant to non- stoichiometricity and size of the particles. The higher non-stoichiometry, the easier for surface 259 superoxide species on cerium oxide to desorb, which corresponds to lower reduction temperature. The experimental results correlate to this general trend. However, the change of nanoclay content does not affect much on the variation of reduction temperature. The temperature drop for nanocomposites (180 -> 168oC as particle size 5.45 -> 4.79 nm) is not as large as that for pure cerium oxide (273->143 oC as particle size 6.20->5.56 nm). Nanoclay platelets may play another role in the system, possibly the adsorption of species, which remains to be verified in future work. 211.48 Na-cloisite, 100 % 168.82 30 %, 4.79 nm 162.86 20 %, 5.20 nm ve i t 165.39 10 %, 5.06 nm Posi 3 %, 5.11 nm 174.53 , Endo - ow 2 %, 5.45 nm 181.32 Fl at e 180.18 H 1 %, 5.45 nm CeO , 0 %, 6.20 nm 273.28 2 143.40 CeO , 0 %, 5.56 nm 2 50 100 150 200 250 300 350 400 Sample Temperature (o C) Figure 6.7 Differential scanning calorimetry curves of pure Na-cloisite (labeled as 100%), cerium oxide (labeled as 0%) and their nanocomposites 6.3.1.5 Sintering results To illustrate the high temperature heating effect on the nanocomposites, the nanocomposites were sintering using ramp method from 500oC to destination temperatures (600, 260 700, 800, 900, 1000, 1100oC), which is parallel to ramp sintering of pure cerium oxide powders. The unsintered and sintered nanocomposite pellets were subjected to XRD and FTIR measurement. XRD patterns for Na-cloisite and nanocomposites are displayed in Figure 6.8 and Figure 6.9, respectively. Before sintering 700oC ) s p (c o 800 C ity s n te In 900oC 1000o C 1100o C 20 40 60 80 100 2 θ (degrees) Figure 6.8 XRD patterns for Na-cloisite subjected to different heating temperatures 261 300 100 300 100 1%, Before Sintering 20%, before sintering 200 200 0 (cps) (cps) 0 20406080100 y t i 0 20 40 60 80 100 s Two Theta (degrees) ten 100 100 n Intensity I 0 0 20 40 60 80 100 20 40 60 80 100 Two Theta (degrees) Two Theta (degrees) 500 40 o 30 o 1%, 700 C 35 20%, 700 C 30 400 25 400 25 20 20 15 10 300 ) 15 300 5 ) ps c 0 ps ( 40 60 c 10 ( ity 200 ity ns 200 5 e 40 60 ns e Int Int 100 100 0 0 0 20406080100 020406080100 2θ (degrees) 2θ (degrees) 1000 40 50 o o 45 1%, 800 C 35 20%, 800 C 40 30 400 35 800 25 30 20 25 15 20 300 15 ) 10 ) s 600 s 10 p 5 p 5 c c ( 0 ( 0 y 40 60 y 40 60 t t i i 200 s s n 400 n e e t t In In 200 100 0 0 0 20406080100 0 20406080100 2θ (degrees) 2θ (degrees) Figure 6.9 XRD patterns of CeO2 / Na-cloisite composites (Left: 1%; Right: 20%) before sintering after sintering at 700, 800 oC 262 1400 30 50 o 1%, 900oC 800 20%, 900 C 25 40 1200 20 700 30 15 1000 600 20 10 10 5 500 ps) 800 ps) 0 (c 0 (c 40 60 y y 40 60 400 600 2θ (degrees) nsit nsit 300 Inte 400 Inte 200 200 100 0 0 0 20406080100 0 20406080100 2θ (degrees) 2θ (degrees) 40 50 o 20%, 1000oC 1%, 1000 C 35 1000 40 2000 30 25 30 20 800 20 1500 15 10 ) 10 5 ps ps) 0 600 0 (c 30 60 (c 40 60 y y it 1000 it s ns e ten 400 n I Int 500 200 0 0 0306090 0 20406080100 2θ (degrees) 2θ (degrees) 1200 40 30 o o 35 20%, 1100 C 1%, 1100 C 1000 30 20 1000 25 20 800 10 800 15 ) 10 ) s ps 5 0 40 60 p 600 c (c 0 600 40 60 ( y it s nsity n 400 e 400 t Inte In 200 200 0 0 0 20406080100 0 20406080100 2θ (degrees) 2θ (degrees) Figure 6.9 Cont. XRD pattern of CeO2 / Na-cloisite composites (Left: 1%; Right: 20%) after sintering at 900, 1000, 1100 oC 263 As can be seen from Figure 6.8, Na-cloisite can retain the layered structure until heating to 800oC, but when heated at temperatures higher than 900oC, the layered structure of the clay o 32 collapses, leaving a continuous band around ~ 20 2θ, which characterizes the amorphous SiO2 and indicates the non-reversible loss of long range order of the nanoclay. This collapsed structure will retain throughout the following heating process and these clay samples sintered at temperature higher than 900oC exhibit XRD patterns characteristics for amorphous silica.32 Compared with unsintered composites, the sintered composites (Figure 6.9) demonstrate more obvious clay inserted into the CeO2 structure. In addition to previous discussed reflection at ~ 64o, the (007) around ~ 37o and (008) around 43o also reflect the presence of clay platelets in CeO2 structure. Contrast to pure standalone clay, these three reflections in composites can be observed even at sintering temperature of 1100oC, which is indicative that the clay platelets are well dispersed in matrix of the CeO2 and the stable structure of CeO2 protects clay platelets from being decomposing. It is noted that the peak broadening of 1%-nanocomposite is obviously faster than that of 20%-nanocomposite, indicating the faster crystal growth of corresponding composites relative to the other (Figure 6.9). Particle sizes of pure cerium oxide and nanocompoistes at different sintering temperatures are calculated by using Scherrer Equation and listed in Table 6.4, in which the particle sizes after 1000oC and 1100oC sintering are not available because they are too large and out of detection limit. It is seen that with increasing sintering temperature, sintering of 1%- montmorillonite-nanocomposite results in a much faster crystallization than pure cerium oxide nanoparticle; whereas for nanocomposite with 20% montmorillonite, the crystal growth is barely slower than that of pure cerium oxide. This result coincides with the XRD pattern FWHM prediction. The 20% and 30% nanocomposites exhibit lower crystallization rate than pure CeO2 264 powder when sintering at temperature 700oC and 800oC, but the 2~10% samples have comparable particle sizes after 700oC sintering and smaller crystals after 800oC sintering. For 900oC sintering, the results are reversed, the 20% and 30% samples comes out with similar particle sizes as pure CeO2 powder while higher sizes are obtained for 2~10% nanocomposites. Table 6.4. Crystallization of pristine CeO2, CeO2/Na-cloisite nanocomposites with different percentages of Na-cloisite at different sintering temperatures Before Sintering Temperature, oC 700 800 900 1000 1100 Sintering CeO2, nm 4.17 15.92 53.72 53.98 N/A N/A CeO2 / 1% Na-Cloisite, nm 5.44 26.57 80.87 N/A N/A N/A CeO2 / 2% Na-Cloisite, nm 5.11 17.14 24.26 80.75 N/A N/A CeO2 / 3% Na-Cloisite, nm 5.06 14.64 30.21 88.89 N/A N/A CeO2 / 5% Na-Cloisite, nm 5.20 12.69 22.61 80.48 N/A N/A CeO2 / 10% Na-Cloisite, nm 4.80 8.70 16.71 66.05 N/A N/A CeO2 / 20% Na-Cloisite, nm 4.79 9.31 15.85 45.55 N/A N/A CeO2 / 30% Na-Cloisite, nm 4.80 6.37 14.37 54.26 N/A N/A At low concentrations of clay, clay is well exfoliated, and the exfoliated platelets can be inserted into the structure of cerium oxide. When this nanocompsite is subjected to heating, the swelling of platelets inside cerium oxide aggregates accordingly increases the size of cerium oxide. Higher concentration of clay cannot be delaminated as well as at low concentration, leaving some relative larger size of platelets that are not possible to incorporate into the matrix of cerium oxide. These platelets will be dispersed in pores between cerium oxide crystals. As heating, the swelling of these platelets outside cerium oxide aggregates inhibits cerium oxide crystals from further growth. This phenomenon again confirms the incorporation of clay with 265 cerium oxide at low concentration is the insertion of platelets into cerium oxide matrix. The critical concentration to divide these two totally different effects of incorporated clay in cerium oxide crystallization needs further study. The sintered composites were also examined with FTIR techniques, using pressed pellets. Transmission mode was used in the test for comparison. Different forms of water, including free adsorbed water, the hydration water coordinated with exchangeable cation (e.g. Na+ in Na- cloisite) in clay interlayer are gone in temperature range of 200 ~ 500oC. And the hydroxyl group covalently bonded in clay crystal is dehydrated in the temperature range of 500 ~ 800oC.40 Therefore no FTIR absorption of OH can be detected for all the sintered Na-cloisite and composites. Si-O stretching and bending vibration spectra for different temperature-sintered samples will be focused in the following discussion. Figure 6.10 compares pellet ATR, pellet-grounded powder transmission, powder transmission FTIR spectra of CeO2 / Na-cloisite composites (e.g. 3%) with transmission FTIR spectrum for pure Na-cloisite. All the composites demonstrates Si-O stretching active around 950 ~ 1100oC. Relative to Na-cloisite and composite powder, the band center from pellets downshifts to ~990 cm-1 and most of bending signals are missing probably because pressure twists the tetrahedral structure of clay crystals. As observed, both ATR and transmission gave Si-O absorption for high temperature sintered nanocomposites (Figure 6.11). 266 --- a --- b --- c --- d rbance o Abs 1000 800 600 Wavenumber (cm-1) Figure 6. 10 Comparison of the Si-O stretching at different conditions: (a) CeO2 / Na- cloisite (w/w 3%) composite pellet ATR spectra; (b) CeO2 / Na-cloisite (w/w 3%) composite pellet transmission spectra; (c) CeO2 / Na-cloisite (w/w 3%) composite powder transmission spectra; (d) Na-cloisite transmission spectra 6.3.2 Nanocomposites from CeO2 and Other Nanoclays Untreated and treated montmorillonites from Nanocor were also used to form CeO2 / montmorillonite nanocomposites. Untreated montmorillonite has Na+ as the exchangeable cation as Na-cloisite does whereas the treated clay is hydrophobic type with cation replaced by + octadecylammonium group (CH 3 (CH 2 )12 (NH 4 ) . 267 A B a b c a d e anc e b b Absorbance Absor c d e 1200 1000 800 600 400 1100 1000 900 800 700 600 500 Wavenumber (cm-1) Wavenumber (cm-1) o o Figure 6.11. IR spectra of CeO2 / Na-cloisite (3% in w/w) composite pellet before (a) and after sintering at 700 C (b), 800 C (c), 1000oC (d) and 1100oC (e) in Si-O stretching region with (A) transmission method, (1300 ~ 400 cm-1); (B) ATR method (1150 ~ 500 cm-1) 268 As observed, untreated montmorillonite/CeO2 composites exhibit the characteristics of both CeO2 matrix and inserted nanoclay platelets when they were subjected to XRD and FTIR measurement. In addition, the crystallization of CeO2/untreated clays also shows the similar property as that of Na-cloisites. However, for treated montmorillonite-incorporated- nanocomposites, FTIR is a more sensitive technique to detect the presence of clay platelets in the composites compared with X-ray diffraction. The details of these two types of nanocomposites are addressed in a separate draft.42 6.3.3 Structural Model for CeO2 / Montmorillonite Nanocomposites As discussed previously, montmorillonite exhibits a 2:1 layered structure in which two silica tetrahedral sheets are linked by alumina octahedral sheet through apical oxygen atoms.21,43 When dramatic agitation or ultrasoncation is applied, the layered structure even the sheets will collapse, resulting in small platelets which retain the basic tetrahedral and octahedral unit structures. On the other hand, cerium oxides crystallize as fluorite structure and usually exhibit non-stoichiometry when the particle size in nanoscale. The exfoliated platelets are so small as to be inserted in the matrix of the cerium oxide to form the nanocomposite particles. The proposed composite formation model is shown in Figure 6.12. 269 Layered silicates Platelets Exfoliation ++++ + Electrochemical Synthesis Ce3+ solution Figure 6.12 Proposed structural model for the formation of the CeO2 / Na-cloisite nanocomposite, demonstrating the exfoliation of nanoclay and the incorporation of clay platelet into cerium oxide structure matrix. Empty circles represent cations, empty ellipses represent cerium oxide, gray ellipses represent Ce3+ ions and solid circles represent water between clay interlayers 6.4. Chapter Conclusions The CeO2 / nanoclay nanocomposites have been successfully produced by using anodic electrochemical synthesis. FTIR measurement confirmed the presence of nanoclay in the composites whereas X-ray diffraction and thermal analysis (DSC & sintering) suggest the inclusion of exfoliated clay platelets in cerium oxide particles. Finally, the nanocomposites formation scheme and possible structural model are proposed. 270 Chapter References (1) Alexandre, M.; Dubois, P. Materials Science and Engineering 2001, 28, 1-63. (2) Fukushima, Y.; Inagaki, S. Journal of Inclusion Phenomena 1987, 5, 473-482. (3) Kim, T. H.; Jang, L. W.; Lee, D. C.; Choi, H. J.; John, M. S. Macromolecular Rapid Communications 2002, 23, 191-195. (4) LeBaron, P. C.; Wang, Z.; Pinnavaria, T. J. Applied Clay Science 1999, 15, 11-29. (5) Ranade, A.; D'Souza, N. A.; Gnade, B. Polymer 2002, 43, 3759-3766. (6) Lan, T.; Kavirarna, P. D.; Pinnavaia, T. J. Journal of Physics and Chemistry of Solids 1996, 57, 1003-1010. (7) Pinnavaia, T. Science 1983, 220, 365-371. (8) Kiš, E. E.; Ranogajec, J. C.; Marinkovic-Neducin, R. P.; Vulic, T. J.; Mesaroš-Borbelj, A. A.; Oprijan, D. A. APEFF 2001, 32, 71-78. (9) Kristof, J.; Mink, J.; Hováth, E.; Gábor, M. Vibational Spectroscopy 1993, 5, 61-67. (10) Hernando, M. J.; Pesquera, C.; Blanco, C.; González, F. Chemistry of Materials 2001, 13, 2154-2159. (11) Dekány, I.; Turi, L.; Király, Z. Applied Clay Science 1999, 15, 221-239. (12) Mamedov, A.; Ostrander, J.; Aliev, F.; Kotov, N. A. Langmuir 2000, 16, 3941-3949. (13) Han, Y.-S.; Matsunoto, H.; Yamanaka, S. Chemistry of Materials 1997, 9, 2013-2018. (14) Ooka, C.; Akita, S.; Ohashi, Y.; Horiuchi, T.; et.al. Journal of Materials Chemistry 1999, 9, 2943-2952. (15) Yamanaka, S.; Brindley, G. W. Clays and Clays Minerals 1979, 27, 119-124. (16) Brindley, G. W.; Sempels, R. E. Clay Minerals 1977, 12, 229-237. (17) Rightor, E. G.; Tzou, M. S.; Pinnavaia, T. J. Journal of Catalysis 1991, 130, 29-40. 271 (18) Garcia-Rodriguez, A.; del Rey-Bueno, F.; del Rey-Perez-Caballero, F. J.; Ureña-Amate, M. D.; Mata-Arjiona, A. Materials Chemistry and Physics 1995, 39, 269-277. (19) Anaissi, F. J.; Demets, G. J.-F.; Toma, H. E.; Dovidauskas, S.; Coelha, A. C. V. Materials Research Bulletin 2001, 36, 289-306. (20) Mishra, T.; Parida, K. Applied Catalysis A; General, 1998, 174, 91-98. (21) Moore, D. M.; Reynolds, R. C. J. X-ray Diffraction and the Identification and Analysis of Clay Minerals; Oxford university press: New York, 1997. (22) Chipera, S. J.; Bish, D. L. Clays and Clay Mineral 2001, 49, 398-409. (23) Horch, R. A.; Golden, T. D.; D'Souza, N. A.; Riester, L. Chemistry of Materials 2002, 14, 3531-3538. (24) Katdare, S. P.; Ramaswamy, V.; Ramaswamy, A. V. Catalysis Today 1999, 49, 313-320. (25) Sivakumar, A.; Damodaran, A. D.; Warrier, K. G. K. Ceramics International 1995, 21, 85-88. (26) Madejová, J. Vibrational Spectroscopy 2003, 31, 1-10. (27) Madejová, J.; Komadel, P. Clays and Clay Mineral 2001, 49, 410-432. (28) Farmer, V. C.; Russell, J. D. Spectrochima Acta 1964, 20, 1149-1178. (29) Farmer, V. C. Spectrochima Acta A 2000, 56, 927-930. (30) Yan, L.; Roth, C. B.; Low, P. Journal of Colloid and Interface Science 1996, 184, 663- 670. (31) Madejová, J.; Arvaiová, B.; Komadel, P. Spectrochima Acta A 1999, 55, 2467-2476. (32) Jenkins, R.; Snyder, R. L. Introduction to X-ray Powder Diffractometry; John Wilery & Sons, Inc.: New York, 1996; Vol. 138. 272 (33) Al-Robaee, M. S.; Rao, K. N.; Mohan, S. Journal of Applied Physics 1992, 71, 2380- 2386. (34) Elidrissi, B.; Addou, M.; Regragui, M.; Monty, C.; Bougrine, A.; Kachouane, A. Thin Solid Films 2000, 379, 23-27. (35) Binet, C.; Daturi, M.; Lavalley, J.-C. Catalysis Today 1999, 50, 207-225. (36) Li, C.; Domen, K.; Maruya, K.; Onishi, T. Journal of the American Chemical Society 1989, 111, 7683-7687. (37) Li, C.; Domen, K.; Maruya, K.; Onishi, T. Journal of Catalysis 1993, 131, 540-547. (38) Bozon-Verduraz, F.; Bensalem, A. Journal of Chemical Society: Faraday Transactions 1994, 90, 653-657. (39) Ying, Y.; Tschöpe, A. The Chemical Engineering Journal 1996, 64, 225-237. (40) Xie, W.; Gao, Z.; Liu, K.; Pan, W.; Vaia, R.; Hunter, D.; Singh, A. Thermochimica Acta 2001, 367-368, 339-350. (41) Temuujin, J.; Okada, K.; MacKenzie, K. J. D.; Jadambaa, T. Journal of European Ceramic Society 1998, 19, 105-112. (42) Wang, A. Q.; D'Souza, N. A.; Golden, T. D. Composite B 2003, Submitted. (43) Tan, K. H. In Principles of Soil Chemistry; Marcel Dekker, Inc., 1998, pp 177-259. 273 CHAPTER 7 ELECTRCHEMICAL FORMATION OF CeO2 / LAYERED SILICATE NANOCOMPOSITE FILMS 7.1 Introduction Cerium oxide (CeO2) is a very important ceramic material, which has been discussed in previous chapters (Chapter 1, 3, 4, 5). Studies indicate that the cerium oxide film has shown excellent corrosion resistant properties for metals and alloys.1-7 And cerium oxide films act as an anodic electrode in solid fuel oxide cells8,9 and counter electrode in electrochromic devices.10,11 In addition, epitaxially deposited cerium oxide films on silicon or even insulating materials renders it a competitive candidates as a buffer layer material for silicon-on-insulator structure12-15 16-18 and superconductor-on-silicon structure. Thermodynamic stability and reactivity of CeO2 also gives this material a promising potential in the application of dielectric and catalytic industry.19-21 However, the applications of cerium oxide films are limited by its strength and toughness. Improving the strength of the film will lead to a longer lifetime for the protective coatings and reinforcing the films will broaden the application of CeO2 as a gate dielectric material, catalyst, electrodes and buffers layers.22-24 The known technique to improve the mechanical properties of ceramics is mainly by dispersing nanometer-sized material particles into the ceramic-matrix grains or grain boundaries. The reinforcing agents can be metals, whiskers, short fibers and platelet structured materials.25-28 The processing procedures and reinforcing effects of these second phases have been reviewed in the introduction section. It is revealed that electrochemical synthesis is one of ideal techniques to suspend a second phase into the ceramic matrix and assure the improvement of mechanical 274 properties.29 In addition, higher toughness can be achieved as the aspect ratios of the materials increase.30 Montmorillonite layered silicates, when exfoliated, have dimensions that range from 100- 1000 nm in length and width while the thickness is only 1-2 nm. This high aspect ratio (~ 1000) property gives this type of materials the feasibility to disperse within the structure matrix and leads to significant improvements in mechanical properties, which has been extremely successful in polymer / montmorillonite composites. And the electrochemical insertion of the nanoclay platelets within CeO2 matrix has been demonstrated in the formation of powdery CeO2 / Na- cloisite nanocomposite and CeO2 / untreated montmorillonite nanocomposite as well. This chapter presents electrochemical deposition of CeO2 / nanoclay nanocomposite. The crystalline structure, morphology and spectroscopic property of the fabricated films are addressed. In addition, the mechanical property testing results of these films are included. 7.2 Experimental 7.2.1 Electrodeposition of Nanocomposite Films Na-cloisite provided by Southern Clay Products, untreated Na-montmorillonite and + octadecylammonium group (CH 3 (CH 2 )12 (NH 4 ) - treated montmorillonite supplied by Nanocor were employed in the experiments. Nanoclay is mixed with Ce(NO3)3 and ligand acetic acid to form the deposition solution, pH was adjusted with NaOH in the range of 7.5~8.0. The employed concentrations of nanoclay are 1%, 2%, 3% 5%, 10%, 20%, 30%, 50% (w/w, nanoclay / Ce(NO3)3.H2O). Anodic electrodeposition is used to form the films. 7.2.2 Characterization of Nanocomposite Films 7.2.2.1 Nanoindentation 275 Nanoindentation was conducted on a NANO III nanoindentator (nano-diamond tip) and the attached Kikon optical microscope can be used to obtain optical micrographs. 7.2.2.2 SEM / EDX JEOL JSM-T300 scanning electron microscopy (SEM) was applied to obtain topological morphologies of composite film surfaces and cross-sections. The attached TN 5500 X-ray analyzer provided the elemental information of detected areas. X-ray diffraction, ATR-FTIR and Raman spectra were used to characterize deposited nanocomposite films. The experimental parameters were set the same as those in previous sections. 7.3 Results and Discussions 7.3.1 Treated and Untreated Montmorillonites as Second Phases 7.3.1.1 Delamination of Montmorillonites Prior to electrochemical deposition of nanocomposites, the octadecylammonium + + (CH 3 (CH 2 )12 (NH 4 ) ) intercalated and Na intercalated montmorillonites are subjected to X-ray diffraction. Figure 7.1 shows the resulting patterns from the treated (TC) (A) and untreated monmoriillonite (UTC) (B). The individual reflection position and d-spacing assignments are tabulated in Table 7.1 and Table 7.2 for TC and UTC, respectively. Since the intercalating ion in treated montmorillonite is the 13-carbon ammonium ion, which is of larger volume compared with Na+ ion in untreated montmorillonite, the (001) reflection locates at ~ 1.95o 2θ (calculated), corresponding to 32.1 Å in d-spacing, while the (001) reflection of untreated montmorillonite appears at ~ 6.85o 2θ (d = 12.9 Å). The (00l) reflections for air-dried clay samples shift to higher degrees while the (060) and bands (02;11) (20;13) (04;22) retain the same positions. Quartz is the main impurity in the clay samples. 276 A Glycol Soaked TC B Glycol-soaked UTC Air Dried TC Air-dried UTC CeO Powder 2 CeO powder 2 ity s ity s ten n In Inte 0 20406080100 0 20406080100 2θ (degrees) 2θ (degrees) Figure 7.1 XRD patterns for (A) Ethylene glycol-soaked and air-dried octadecylammonium montmorillonite (TC) and (B) Ethylene glycol-untreated montmorillonite in comparison with that of pure CeO2 powder 277 Table 7.1 Experimental and calculated XRD pattern positions and d-spacings for treated montmorillonite sample in Figure 7.1 (m & c refers to measured & calculated) Glycol soaked TC Air dried TC Reflection 2θ (c) d-spacing 2θ (m) d-spacing 2θ (m) d-spacing (degrees) (nm) (degrees) (nm) (degrees) (nm) 001 1.45 32.12 N/A N/A 002 2.90 15.42 2.90 15.42 3.61 24.48 003 4.35 10.21 4.37 10.16 N/A 004 5.80 7.64 N/A N/A 005 7.25 6.10 7.20 6.15 8.05 10.98 006 8.70 5.08 N/A N/A 007 10.15 4.36 N/A N/A 008 11.60 3.81 N/A N/A 009 13.05 3.39 N/A N/A 0010 14.50 3.05 15.01 5.90 16.20 5.47 28.58 060 61.67 1.5040 61.76 1.5021 02;11 19.58 19.72 20;13 34.80 34.82 04;22 53.82 53.80 Quartz 26.48 26.56 Quartz Quartz 45.34 278 Table 7.2 Experimental and calculated XRD pattern positions and d-spacings for untreated montmorillonite (UTC) sample in Figure 7.2 (m & c refers to measured & calculated, respectively) Glycol treated UTC Air dried UTC Reflection 2θ (c) d-spacing 2θ (m) d-spacing 2θ (m) d-spacing (degrees) (nm) (degrees) (nm) (degrees) (nm) 001 6.85 12.9 6.85 12.9 6.96 12.7 002 13.7 6.46 14.05 6.303 14.09 6.29 003 20.66 4.30 20.30 4.38 N/A 3.12 004 27.66 3.22 28.33 3.15 28.60 005 34.77 2.58 N/A N/A 006 41.99 2.15 40.16 2.25 40.25 2.24 007 49.41 1.84 N/A N/A 060 61.87 1.450 61.89 1.499 02;11 19.79 19.74 20;13 34.93 34.84 04;22 53.93 54.04 Quartz 26.63 21.90 To study the delamination properties of the employed clays, these two types of clays were mechanically mixed with CeO2 by 50:50 volume ratio and X-rayed. As shown, the 001 reflection totally disappears in treated montmorillonite-CeO2 mixture and becomes less intense in untreated montmorillonite-CeO2 mixture. In addition, 002 and the higher order of (00l) of both montmorllonites vanish. As known, the pure clays exhibit intense reflections in the 2θ range of 2 to 9o, corresponding to d-spacing from 17 to 40 Å, which represent the interlayer spacing.31 The absence or considerable intensity decrease of these peaks in CeO2-treated montmorillonite and CeO2 – untreated montmorillonite mixtures (Figure 7.2) indicates the separation of the layers. The exfoliation from this mechanical mixing cannot destroy the x/y dimension structure, which 279 can be revealed from the retaining of reflection (060) and characteristic bands for clay in XRD patterns of the mixtures. CeO powder 2 CeO / UTC powder mixture (50:50) 2 CeO / TC powder mixture (50:50) 2 x x x ) d e liz x a m r x o (N x x ity s n te In 0 20406080100 2θ (degrees) Figure 7.2 XRD patterns for CeO2 / UTC powder mixture and CeO2 / TC powder mixture in comparison with pure CeO2 powder 7.3.1.2 Electrodeposited CeO2 / Montmorillonite Nnanocomposite Films 7.3.1.2.1 Crystal Structure and Surface Morphologies of Films Figure 7.3 displays the XRD patterns of CeO2 / treated montmorillonite nanocomposite films and CeO2 / untreated montmorillonite nanocomposits films with the montmorillonite content range from 0.5 ~ 60% of cerium nitrate (w/w). The films were fabricated using potentiostatic anodic electrodeposition at room temperature. The montmorillonite addition in the films does not change the fcc structure of CeO2, indicating an intact matrix of CeO2 in the composites. These two types of nanocomposite films share the same property in XRD: at lower concentrations (0.5 ~ 5 %) of clay in the system, XRD patterns exhibit only characteristics of fcc CeO2. As the concentration of montmorillonite in composites is higher than 10%, the properties 280 of clay are demonstrated in the emergence of the reflection at ~ 11.50 ~ 12.00o 2θ. This reflection corresponds (002) of untreated montmorillonite and (008) of treated montmorillonite according to calculation. The as-produced composite films have particle sizes in the range of 5 ~ 10 nm calculated from FWHM of CeO2 reflections using the Scherrer equation (Equation 2.37). All the calculated particle sizes corresponding to XRD patterns in Figure 7.3 are listed in Table 7.3. The morphologies of as-formed composites were observed using scanning electron microscopy. Figure 7.4 illustrates the selected SEM pictures of the CeO2 / TC composite films and CeO2 / UTC composite films. Since untreated montmorillonite has sodium ions in the interlayer in comparison to octadecylammonium ions as the intercalated cations in treated montmorillonite, untreated montmorillonite is more hydrophilic whereas treated montmorillonite is hydrophobic. Therefore UTC is more readily to interact with the cerium oxide system in an aqueous reaction environment, which can be verified in the morphology of the films. CeO2 / UTC composites, at a certain content of clay, usually give a homogenous surface with the clay incorporated with the cerium oxide surface. However, for CeO2 / TC composites, clay sparsely resides on the top of cerium oxide film crack surface even at higher concentrations of clays in the reaction system. The condition for preferred oriented CeO2 film on stainless steel was also tested in electrodeposition of composite films. XRD patterns in Figure 7.5 and optical micrographs in Figure 7.6 demonstrate the difference between the deposition results from the commonly used potentiostatic mode and orientation inducing galvanostatic mode. TC and UTC based composites present the same results in this regard. Both reaction modes delaminated interlayer structure of clays. And as predicted, the galvanostatic deposition with applied current –0.1 mA at 50oC leads 281 to more CeO2 (111) preferred structure compared with potentiostatic deposition with applied potential 1.10 V vs. SCE at 25oC. In addition, the former approach gives rise to the particle sizes of the composites in the range of 10 ~ 15 nm, a few nanometer larger than those from the potentiostatic deposition at room temperatures (Table 7.3). Figure 7.6 illustrates the CeO2 / UTC composite film (5%) appearance under optical microscopy. The potentiostatic deposited composites film appears mottled on top of the CeO2 film due to a layer of clay coating whereas the galvanostatic fabricated film surface shows globular glassy character in the appearance from the accumulation of clay on top of CeO2 surface. EDX on cross-section and surface of composite films reveals presence of elemental Si and Al in the composite films. 7.3.1.2.2 Mechanical Properties Nanoindentation was employed to test the mechanical properties of as-produced nanocomposite films. The loading procedure can be divided into six sections: holding at surface, loading to ~ 100 nm, holding at this displacement, loading to ~ 200 nm, holding at this displacement, and unloading. The hardness was calculated from two holding sections and Young’s modulus was obtained from the unloading section. Figure 7.7 and 7.8 illustrate the hardness and Young’s modulus as the function of displacement for holding section at ~ 100 nm (A) and ~ 200 nm (B). It is shown that the 1% untreated clay composite films exhibit the similar hardness and elastic modulus as those of pure CeO2 film. As the percentage of clay increases, the hardness and elastic modulus show a slight decrease. Treated clay composite gives the comparable result as the untreated clay ones. The reduction of the mechanical properties for nanocomposites compared with pure ceramic oxide film can be blamed on the clay coatings on the top of films. 282 ss ss X A B ss X ss ss * ss * * * 60% X 50% * * * * 60% 40% 50% 30% 40% 30% 20% 20% 10% 10% 5% 5% 3% 3% 2% 2% 1% 1% 0.5% 0.5% 0 20406080100 0 20406080100 2θ (degrees) 2θ (degrees) Figure 7.3 XRD patterns of CeO2 / treated montmorillonite nanocomposite films (A) and CeO2 / untreated montmorillonite nanocomposite films (B) at different concentrations of montmorillonites in the composites. Red X refers to reflections from montmorillonite and black ∗ represents CeO2 reflections in Figures 283 A C B D Figure 7.4 SEM of CeO2 / TC composite films (A) 20%, (B) 10% and SEM of CeO2 / UTC composite films (C) 20%, (D) 5% A B Figure 7.6 Optical pictures of nanocomposites of (A) CeO2 / 5% UTC by potentiostatic o deposition (E = 1.10V), x 800; (B) CeO2 / 5% UTC by galvanostatic deposition (T = 50 C, I = - 0.1mA), x 800 284 o A T = 50oC; Gal: I= -0.1mA B T = 50 C; Gal: I= -0.1mA o T = 25oC; Pot: E = 1.10V T = 25 C; Pot: E = 1.10V (100) ss ity s ity n s te n In Inte ss (220) ss (200) (311) ss (100) (200) (220) 0 20 40 60 80 100 0 20406080100 2θ (degrees) 2θ (degrees) Figure 7.5 XRD patterns for comparison between potentiostatic and galvanostatic deposition for composite film formation. (A) CeO2 / TC composite film (5%); (B) CeO2 / UTC composite film (5%) 285 Table 7.3 Particle sizes of CeO2 / montmorillonite nanocomposite films calculated from XRD patterns in Figure 7.3 and Figure 7.5 using Scheerr equation on CeO2 (111) 5 % 0.5 1 2 3 10 20 30 40 50 60 Clay Type P, 25oC, G, 50oC Untreated 6.51 5.11 5.50 6.00 5.33 13.62 5.40 4.53 5.06 4.54 5.99 6.00 Montmorillonite Treated 6.88 6.56 8.79 7.44 6.40 12.49 6.72 4.40 8.53 8.08 5.53 5.77 Montmorillonite At 5% of clay in composites, the particle sizes from two electrodeposition modes are tabulated in this table. P, 25oC refers to that the composite was fabricated with potentiostatic deposition at 25oC. G, 50oC refers to that the composite was fabricated with galvanostatic deposition at 50oC. 286 3.0 A CeO 2 2.5 CeO / 1% U TC 2 CeO / 3% U TC 2 2.0 CeO / 5% U TC 2 CeO / 5% TC 1.5 2 Pa) G 1.0 0.5 0.0 Hardness ( -0.5 -1.0 -1.5 100 104 108 112 116 120 124 128 Displacement (nm) 4 B CeO2 CeO / 1% U TC 2 3 CeO / 3% U TC 2 CeO / 5% U TC 2 CeO / 5% TC 2 2 Pa) G 1 Hardness ( 0 -1 200 204 208 212 216 220 224 228 232 Displacem ent (nm) Figure 7.7 Hardness vs. Loading Displacement for CeO2 / nanoclay nanocomposite films at different loading depth (A) 100 nm; (B) 200 nm 287 200 A CeO 2 CeO / 1% U TC 2 150 CeO / 3% U TC 2 Pa) CeO / 5% U TC 2 G CeO / 5% TC 100 2 us ( 50 s Modul 0 Young' -50 -100 100 104 108 112 116 120 124 128 Displacement (nm) 250 B CeO2 CeO / 1% U TC 2 200 CeO / 3% U TC 2 CeO / 5% U TC 2 Pa) CeO / 5% TC 150 2 G us ( 100 odul 50 s M 0 Young' -50 -100 200 204 208 212 216 220 224 228 232 Displacement (nm) Figure 7.8 Young’s Modulus vs. Loading Displacement for CeO2 / nanoclay nanocomposite films at different loading depth (A) 100 nm; (B) 200 nm 288 7.3.2 Electrodeposited CeO2 / Na-Cloisite Nanocomposite Films The preliminary studies on untreated and treated montmorillonite based nanocomposites indicate that hydrophilic untreated montmorillonite incorporated into ceramic oxide film can result in more homogeneous films. And the usual potentiostatic electrodeposition leads to good adhesion and does not change the basic property of clays, compared to galvanostic deposition. Therefore, the aqueous type clay (Na-cloisite) from Southern Clay Products was applied in the composite film deposition. The basic properties of this clay have been discussed in Chapter 6 and will not be repeated here. 7.3.2.1 Crystal structures and Morphologies of Composite Films XRD patterns for CeO2 / Na-cloisite composite films indicate that this type of composites demonstrates the same property and trend as other two types discussed in previous section. Table 7.4 lists the particle sizes of composites, which indicates that potentiostatic mode deposition at 25oC results in composites of particle 5 ~ 8 nm and the galvanostatic deposition at I = -0.1 mA and 50oC leads to a relatively more (111) oriented film with particle size of ~ 12 nm. To validate the incorporation of Na-cloisite within ceramic oxide films, both film surface and cross-section were subjected to SEM / EDX detection. The surface (Figure 7.10(A) (B)) exhibits a smooth and tight construction. Magnification of 75 image of cross-section (Figure 7.10 (C) manifests the boundary between stainless steel region (lower portion) and film region (upper portion). Under magnification of 2000 (Figure 7.10 (D)), the layered structure of clay can be seen. Dispersive X-ray analysis on the cross-section reveals elemental Ce, Si, Al, Fe and Ni is present in the film. Among those, Fe and Ni are attributed to stainless steel substrate while Ce, Si and Al are assigned to film components. 289 ss A T = 50oC; Gal: I= -0.1mA X (111) T = 25oC; Pot: E = 1.10V ss B ss * (220) * * * 60% 50% (311) (200) 40% ity ss (222) ns 30% ss ss(400) 20% Inte 10% ss 5% ss 3% (111) ss 2% 1% (200) (220) 0.5% (311) (400) 0 20406080100 0 20406080100 2θ (degrees) 2θ (degrees) Figure 7.9 XRD patterns for CeO2 / Na-cloisite composite films Table 7.4 Particle sizes of CeO2 / Na-cloisite composite films 5 % 0.5 1 2 3 P, 25oC G, 50oC 10 20 30 40 50 60 Clay Type Na-Cloisite 6.86 6.58 7.06 6.88 6.88 12.32 6.26 5.53 5.30 5.33 5.02 5.65 P, 25oC: potentiostatic deposition (E = 1.10V) at 25oC; G, 50oC: galvanostatic deposition (I = -0.1mA) at 50oC. 290 A B C D Figure 7.10. SEM images of 10% NTC, (A) film surface x 5,000; (B) film surface, x 10,000; (C) film cross-section, x 75; (B) film cross-section, x 2,000 7.3.2.2 Ultrasonication Studies Compared with conventional method, ultrasound offers advantage to delamination of layered silicate and pillaring of oxides into the interspacing of these clays with faster time and high concentration of clays.32-35 To investigate the effect of ultrasonification on the composite film, Na-cloisite suspension was prepared beforehand and was ultrasonicated for 24, 216, 432, 672 hours before it is added in the cerium salt reaction system. The clay amount is 0.2% and 0.5% corresponding to the volume of the solution, converted as 4.6% and 11.5% in w/w relative to cerium nitrate. Reflections in XRD patterns for films (Figure 7.11A) indicate characteristic fcc fluorite structure of cerium oxide.36 The reflections corresponding to cloisite clay are missing in all of 291 these patterns, since the layered silicate was delaminated and also no signs of silicon oxide in these XRD patterns of composite films, which is contrary to the phenomena in powdery composites, possibly because the films are thin and the signals are not strong enough for detection. Raman absorption (Figure 7.11B) shows the main component F2g of cerium oxide at ~ 459 cm-1, whereas the clays have no activity in Raman frequency region.37 ATR-IR spectra of the films are shown in Figure 7.11C. ATR configuration in infrared spectroscopy is a sensitive method that can be used to investigate surface adsorbed species and thin films.38 It allows to study various spectral range of oxides39 and also has been used to investigate clay minerals in any form like films, dispersions, gels or pastes.40 There are three absorptions in range of 1100 ~ -1 -1 1000 cm . The peak ~ 1080 cm is due to Si-O stretching in the amorphous SiO2, which was obtained from the breakdown of silicate bonding in clay during ultrasonication prior to deposition and agitation during the process. This Si-O vibration belongs to TO of As1 mode, which locates at the lowest frequency compared to LO As2, TO As2, LO As1 modes at progressively increasing frequencies.38,39,41 The ATR-IR study on silicon oxide films indicates that the Si-O gradually shifts upwards, prone to the vibration modes in high frequencies with the 41 growth of SiO2 film. The longer ultrasonication results in a larger amount of SiO2 and more homogeneous films, hence the Si-O peak shifts upwards from 1078 cm-1 for 24 h ultrasonication to 1083 cm-1 for 672 h ultrasonication, which matches the discussed upward relocation of Si-O vibration frequency. The peaks ~ 1025 cm-1 and 1045 cm-1 can be ascribed to the overlapping of Si-O-Si stretching of cloisite and residual acetate impurity. Possibly Si-O vibration in silica from clay breakdown contributes as well. Bands centered at ~ 935 cm-1 and 840 cm-1 are mainly due to cerium acetate in this regard.42 Whether the aluminum octahedral network is broken-down cannot be determined, even though the previous ATR results of composites manifest the collapse 292 of silicon tetrahedral sheet in cloisite. However, the Al-Al-OH (916 cm-1) and Al-Mg-OH (846 cm-1) bending vibrations are possibly “buried” in these two broad bands. The gradual decrease of relative intensity of band 840 cm-1 and vanishing of band 935 cm-1 with longer ultrasonication time seems from the reduction of Al-Mg-OH and Al-Al-OH in composites, which can be related to weakening of bonding in octahedral sheet. The corresponding surface optical micrographs of these films shown in Figure 7.12 indicate that the deposited nanocomposites have thick and homogeneous surface with different concentrations of Na-cloisite. The variation of the ultrasonication time has not much influence in the morphology of the surface of the deposits. The cloudy morphologies of the films are attributed to the accumulation of extra clay at surface of the deposit and this can be used to explain that hardness and Young’s modulus of these nanocomposite films are lower than pure cerium oxide film. More research is needed to resolve this problem. Compared with nanocomposite made without ultrasonification (~ 6.9 nm for 5%, and ~ 6.3 nm for 10%), ultrasonification resulted in smaller particle size (Table 7.5). It illustrates that the particle size of the nanocomposites has decreasing tendency with increasing ultrasonication time from 24 h to 632 h with the same concentration of clay contained in the nanocomposites, but the variation is slight, which shows that the ultrasonification time beyond 24 h is not necessary. Higher concentration of cloisite also leads to lower particle size, which is consistent with the conclusion from untreated and treated montmorillonite nanocomposites. 7.3 Chapter Conclusions This chapter briefly described the electrochemical deposition of CeO2 / layered silicate nanocomposites on metal substrates. XRD and Raman spectroscopy indicate that CeO2 is still the structural matrix of the composite films and ATR-FTIR reveals the silicate platelet insertion into 293 the matrix of composites. The influence of ultrasonication on nanocomposite was investigated with Na-cloisite as the secondary phase, which results that 24 h sonication will lead to homogeneous and thick deposition, and with elongation of the sonification time, the deposit has lower particle sizes. It is concluded from nanoindentation study that the hardness and modulus of the composite films are on par with the cerium oxide film itself due to accumulation of extra clay on the surface. 294 24 h A 216 h 432 h 672 h y t i s n e t In 0 20406080100 2 θ (degrees) B 24 h 216 h 432 h y t i 672 h s n e t n I ed z i al m r No 500 1000 1500 2000 W avenum ber (cm -1 ) C 24 h 216 h 432 h ce 672 h n a b r so b A ed z i al m r o N 1200 1100 1000 900 800 700 600 W avenum ber (cm -1 ) Figure 7.11 CeO2 / Na-cloisite composite films at different clay ultrasonication time (A) XRD pattern; (B) Raman spectra; (C) FTIR spectra 295 A D BB E C F Figure 7.12 Optical micrographs of CeO2 / NTC nanocomposite filmes x 800 (A) 0.2%, 24 h; (B) 0.2%, 216 h (C) 0.2%, 432 h (D) 0.5%, 24h; (E) 0.5%, 216h (F) 0.5%, 432 h 296 Table 7.5 Particle sizes of composites at different unltrasonication time Ultrasonication Time (h) 24 216 432 672 0.2% NTC 4.87 4.66 4.92 4.41 0.5% NTC 4.47 4.30 3.89 3.97 A B C D Figure 7.13 SEM image of CeO2/ Na-cloisite (NTC) nanocomposite films after clay was ultraconicated for 216 h. (A) 0.2% x 1,000; (B) 0.5%, x 1,000; (C) 0.2%, x 5,000; (D) 0.5%, x 5,000 297 Chapter References (1) Aramaki, K. Corrosion Science 2001, 43, 2201-2215. (2) Aramaki, K. Corrosion Science 2002, 44, 2621-2634. (3) Arenas, M. A.; Conde, A.; de Damborenea, J. J. Corrosion Science 2002, 44, 511-520. (4) Arnott, D. R.; Hinton, B. R. W.; Ryan, N. E. Corrosion Science 1989, 45, 12-18. (5) Breslin, C. B.; Chen, C.; Mansfeld, F. Corrosion Science 1997, 39, 1061-1073. (6) Lu, Y. C.; Ives, M. B. Corrosion Science 1993, 34, 1773-1785. (7) Lu, Y. C.; Ives, M. B. Corrosion Science 1995, 37, 145-155. (8) Marina, O. A.; Bagger, C.; Primdahl, S.; Mogensen, M.; Sammes, S. M.; Tompsett, G. A. Solid State Ionics 1999, 123, 199-208. (9) Putna, E. S.; Stubenrauch, J.; Vohs, J. M.; Corte, R. J. Langmuir 1995, 11, 4832-4837. (10) Krishna, M. G.; Hartridge, A.; Bhattacharya, A. K. Materials Science Engineering B 1998, 55, 14-20. (11) Özer, N. Solar Energy Materials & Solar Cells 2001, 68, 391-400. (12) Hirschauer, B.; Göthelid, M.; Janin, E.; Lu, H.; Karlsson, U. O. Applied Surface Science 1999, 148, 164-170. (13) Inoue, T.; Yamamoto, Y.; Koyama, S.; Suzuki, S.; Ueda, Y. Applied Physics Letters 1990, 56, 1332-1333. (14) Inoue, T.; Yamamoto, Y.; Satoh, M. Thin Solid Films 1999, 343/344, 594-597. (15) Li, M.; Wang, Z.; Fan, S.; Zhao, Q.; Xiong, G. Nuclear Instruments and Methods in Physics Research B 1998, 135, 535-539. (16) Luo, L.; Wu, X. D.; Dye, R. C.; Muenchausen, R. E.; Foltyn, S. R.; et.al. Applied Physics Letters 1991, 59, 2043-2045. 298 (17) Wu, X. D.; Dye, R. C.; Muenshasusen, R. E.; Foltyn, S. R.; et.al. Applied Physics Letters 1991, 58, 2165-2167. (18) Denhoff, M. W.; McCaffrey, P. Journal of Applied Physics 1991, 70, 3986-3988. (19) Tschöpe, A.; Ying, J. Y.; Tuller, H. L. Sensors and Actuators B 1996, 31, 111-114. (20) Trovareli, A.; De Leitenbutrg, C.; Boaro, M.; Dolcetti, G. Catalysis Today 1999, 50, 353- 367. (21) Trovareli, A.; Boaro, M.; Rocchini, E.; DeLeitenburg, C.; Dolcetti, G. Journal of Alloys and Compounds 2001, 323-324, 584-591. (22) Foreman, J.; NexTech Materials, 2003. (23) Zhou, Y.; Phillips, R. J.; Switzer, J. A. Journal of American Ceramic Society 1995, 78, 981-985. (24) Shemilt, J. E.; Williams, H. W.; Edirisinghe, M. J.; Evans, J. R. G.; Ralph, B. Scipta Materialia 1997, 36, 929-934. (25) Poorteman, M.; Descamps, P.; Cambier, F.; Poulet, A.; Descamps, J. C. Journal of European Ceramic Society 1999, 19, 2375-2379. (26) Kaya, C.; Gu, X.; Al-Dawery, I.; Butler, E. G. Science and Technology of Advanced Materials 2002, 3, 35-44. (27) Grigge, J.; University of Florida: Gainsville, 1998. (28) Lee, T.-C.; Deng, J. Ceramics International 2001, 27, 755-760. (29) Horch, R. A.; Golden, T. D.; D'Sousa, N. A.; Riester, L. Chemistry of Materials 2002, 14, 3531-3538. (30) Butzloff, P. R. In Materials Science; University of North Texas: Denton, 2000, p 97. 299 (31) Moore, D. M.; Reynolds, R. C. J. X-ray Diffraction and Identification and Analysis of Clay Minerals; 2nd ed.; Oxford University Press, Inc.: New York, 1997. (32) Katdare, S. P.; Ramaswamy, V.; Ramaswamy, A. V. Catalysis Today 1999, 49, 313-320. (33) Sivakumar, A.; Damodaran, A. D.; Warrier, K. G. K. Ceramics International 1995, 21, 85-88. (34) Franco, F.; Pérez-Maqueda, L. A.; Pérez-Rodríguez, J. L. Thermochimica Acta 2003, 4, in press. (35) Katdare, S. P.; Ramaswamy, V.; Ramaswamy, A. V. Journal of Materials Chemistry 1997, 7, 2197-2199. (36) Wang, A. Q.; Golden, T. D. Journal of the Electrochemical Society 2003, 9, in press. (37) Wang, S.; Wang, W.; Zuo, J.; Qian, Y. Materials Chemistry and Physics 2001, 68, 246- 248. (38) Nagai, N.; Hashimoto, H. Applied Surface Science 2001, 172, 307-311. (39) Rochat, N.; Chabli, A.; Bertin, F.; Vergnaud, C.; Mur, P.; Petitdidier, S. Materials Science and Engineering B 2003, in press. (40) Madejová, J.; Komadel, P. Clays and Clay Minerals 2001, 49, 410-432. (41) He, L.; Hasegawa, S. Thin Solid Films 2001, 384, 195-199. (42) Socrates, G. Infrared Characteristic Group Frequencies; 2nd ed.; John Wiley & Sons: Chichester, 1998. 300 BIBLIOGRAPHY Aguiar, R.; Sánchez, F.; Ferrter, C.; Varela, M. Thin Solid Films 1997, 306, 74-77. Ahn, B. K.; Curtin , W. A.; Parthasarathy, T. A.; Dutton, R. E. Composites Science and Technology 1998, 58, 1775-1784. Aita, C. R. In Intermetallic and Ceramic Coatings; Dahotre, N. B., Sudarshan, T. S., Eds.; Marcel Dekker. Inc.: New York, 1999, pp 181-231. Alcock, N. W.; Tracy, V. M. Journal of the Chemical Society Dalton Transactions:Inorganic Chemistry 1976, 2238-2243. Alcock, N. W.; Tracy, V. M. Journal of the Chemical Society Dalton Transactions: Inorganic Chemistry 1976, 2243-2246. Alcock, N. W.; Tracy, V. M. Journal of the Chemical Society Dalton Transactions: Inorganic Chemistry 1976, 2246-2249. Al-Dhhan, Z. T.; Hashen, T.; Hogarth, C. A. Spectrochimica Acta. 1989, 44B, 205-208. Aldykiewicz, A. J. J.; Davenport, A. J.; Isaacs, H. S. Journal of the Electrochemical Society 1996, 143, 147-153. Alexandre, M.; Dubois, P. Materials Science and Engineering 2001, 28, 1-63. Al-Robaee, M. S.; Rao, K. N.; Mohan, S. Journal of Applied Physics 1992, 71, 2380-2386. Amirhaghi, S.; Li, Y. H.; Kilner, J. A.; Boyd, I. A. Materials Science and Engineering B 1995, 34, 192-198. Anaissi, F. J.; Demets, G. J.-F.; Toma, H. E.; Dovidauskas, S.; Coelha, A. C. V. Materials Research Bulletin 2001, 36, 289-306. Aramaki, K. Corrosion Science 2001, 43, 2201-2215. Aramaki, K. Corrosion Science 2002, 44, 2621-2634. 301 Arenas, M. A.; Conde, A.; de Damborenea, J. J. Corrosion Science 2002, 44, 511-520. Arkinson, A.; Ramos, T. M. G. M. Solid State Ionics 2000, 129, 259-269. Arnott, D. R.; Hinton, B. R. W.; Ryan, N. E. Corrosion Science 1989, 45, 12-18. Audebrand, N.; Auffrédic, J.-P.; Louër, D. Chemistry of Materials 2000, 12, 1791-1799. Awaji, H.; Choi, S.-M.; Yagi, E. Mechanics of Materials 2002, 34, 411-422. Azen, A.; Kullman, L.; Ragan, D. D.; Granqvist, C. G. Solar Energy Materials & Solar Cells 1998, 54, 85-91. Balasubramanian, M.; Melendres, C.; Mansour, A. N. Thin Solid Films 1999, 347, 178-183. Bamwenda, G. R.; Arakawa, H. Journal of Molecular Catalysis A: Chemical 2000, 161, 105- 113. Bansal, N. P.; Setlock, J. A. Composites A 2001, 32, 1021-1029. Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; 1st ed.; John Wiley & Sons, Inc: New York, 1980. Bassini, L.; Marchionna, M.; Aragno, A. The Journal of Physical Chemistry B 1992, 96, 9431- 9441. Baur, J. F.; Spaine, T. W. Journal of Electroanalytical Chemistry 1998, 443, 208-216. Bellon, O.; Sammes, N. M.; Staniforth, J. Journal of Power Sources 1998, 75, 116-121. Bene, R.; Perczel, I. V.; Réti, F.; Meyer, F. A.; Fleiseher, M.; Meixner, H. Sensors and Actuators B 2000, 71, 36-41. Bensadon, E. O.; Nascente, P. A. P.; Olivi, P.; Bulhoes, L. O. S.; Pereira, E. C. Chemistry of Materials 1999, 11, 277-280. Berner, U.; Schierbaum, K. Thin Solid Films 2001, 400, 46-49. Bhaduri, S.; Bhaduri, S. B. JOM 1998, 50, 44-51. 302 Binet, C.; Daturi, M.; Lavalley, J.-C. Catalysis Today 1999, 50, 207-225. Birss, V. I.; C., B.; Eizanowska, H. Canadian Journal of Chemistry. 1997, 75, 1687-1693. Boivin, J. C.; Mairesse, G. Chemistry of Materials 1998, 10, 2870-2888. Bozon-Verduraz, F.; Bensalem, A. Journal of the Chemical Societ Faraday Transactions 1994, 90, 653-657. Brefogle, B. E.; Hung, C. J.; Shumsky, M. G.; Switzer, J. A. Journal of the Electrochemical Society 1996, 143, 2741-2746. Breslin, C. B.; Chen, C.; Mansfeld, F. Corrosion Science 1997, 39, 1061-1073. Brindley, G. W.; Sempels, R. E. Clay Minerals 1977, 12, 229-237. Bunluesin, T.; Putna, E. S.; Stubenrauch, J.; Vohs, J. M.; Gorte, R. J. Catalysis Letters 1996, 41, 1-5. Burroughs, P.; Hamnett, A.; Orchard, A. F.; et.al. Journal of the Chemical Society Dalton Transactions 1976, 17, 1686-1698. Butzloff, P. R. In Materials Science; University of North Texas: Denton, 2000, p 97. Camino, D.; Deroo, D.; Salardenne, J.; Treuil, N. Solar Energy Materials & Solar Cells 1995, 39, 349-366. Carrado, K. A.; Xu, L.; Gregory, D. M.; Song, K.; Seifert, S.; Botto, R. E. Chemistry of Materials 2000, 12, 3052-3059. Carter, W. B.; Book, G. W.; Pollery, T. A.; Stollberg, D. W.; Hampikian, J. M. Thin Solid Films 1999, 347, 25-30. Casella, I. G.; Gatta, M. Journal of Electroanalytical Chemistry 2002, 534, 31-38. Casellas, D.; Ràfols, I.; Llanes, L.; Anglada, M. International Journal of Refractory Metals & Hard Materials 1999, 17, 11-20. 303 Chen, I.-W.; Wang, X.-H. Nature 2000, 404, 168-171. Chen, P. L.; Chen, L. W. Journal of the American Ceramic Society 1993, 76, 1577-1583. Chen, S.; Ye, F.; Zhou, Y. Ceramics International 2000, 28, 51-58. Chiang, Y.-C. Composites Science and Technology 2001, 61, 1743-1756. Chiang, Y.-M.; Lavik, E. B.; Blom, D. A. Nanostructured Materials 1997, 9, 633-642. Chikyow, T.; Bedair, S. M.; Tye, L.; El-Masry, N. A. Applied Physics Letters 1994, 65, 1030- 1032. Chin, C. C.; Lin, R. J.; Yu, Y. C.; Wang, C. W. Physica C 1996, 260, 86-92. Chin, C. C.; Lin, R. J.; Yu, Y. C.; Wang, C. W. IEEE Transactions on Applied Superconductivity 1997, 7, 1403-1406. Chipera, S. J.; Bish, D. L. Clays and Clay Minerals 2001, 49, 398-409. Choa, Y.-H.; Nakahira, A.; Niihar, K. Journal of Materials Science 2000, 35, 3143-3149. Chou, Y.-S.; Green , D. Journal of the American Ceramic Society 1993, 76, 1452-1458. Chou, Y.-S.; Green, D. J. Journal of the American Ceramic Society 1993, 76, 1985-1992. Cotell, C. M.; Hirvonen, J. K. Surface and Coatings Technology 1996, 81, 118-124. Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry: A Comprehensive Text; 4th ed.; John Wiley & Sons: New York, 1980. Crnjak, Z.; Huchins, M. G.; McMeeking, G. Solar Energy Materials & Solar Cells 1995, 30, 327-350. Crossland, A. C.; Thompson, G. E.; Skeldon, P.; Smith, G. C. Corrosion Science 1998, 40, 871- 885. Cullity, B. D.; Stock, R. S. Elements of X- Ray Diffraction; 3rd ed.; Prentice Hall, 2001. Cunningham, J.; Cullinane, D.; Sanz, J.; Rojo, J. M.; Soria, X. A.; Fierro, J. L. G. Journal of the 304 Chemical Society Faraday Transactions 1992, 88, 3233-3240. Czerwinski, F.; Szpunar, J. A. Thin Solid Films 1996, 289, 213-219. Davenport, A. J.; Isaacs, H. S.; Kendig, M. W. Corrosion Science 1991, 32, 653-663. Davenport, A. J.; Isaacs, H. S.; Kendig, M. W. D. In The Electrochemical Society Proceedings Series; Baer, R., Clayton, C. E., Davis, G. D., Eds.: Pennington, NJ, 1991; Vol. 433, pp 91-97. Dean, J. A. Lance's Handbook of Chemistry; 13st ed.; McGraw_Hill Book Company: New York, 1985. Dekány, I.; Turi, L.; Király, Z. Applied Clay Science 1999, 15, 221-239. Denhoff, M. W.; McCaffrey, P. Journal of Applied Physics 1991, 70, 3986-3988. Deroo, D.; Petit, J.-P.; Kéoman, D. Solar Energy Materials & Solar Cells 1995, 36, 397-408. Develos, K. D.; Yamasaki, H.; Nakagawa, Y.; Sawa, A. Physica C 2002, 372-376, 642-648. Develos, K. D.; Yamasaki, H.; Sawa, A.; Nakagawa, Y. Physica C 2001, 361, 121-129. Djuricic, B.; Pickering, S. Journal of European Ceramic Society 1999, 19, 1925-1934. Doerner, M. F.; Nix, W. D. Journal of Materials Research 1986, 1, 601-609. Dokiya, M. Solid State Ionics 2002, 152-153, 383-392. Drozdov, V. A.; Doronin, V. P. S.; Porokina, T. P.; Gulyaeva, T. I.; Duplyakin, V. K. Kinetics and Catalysis 2001, 42, 117-125. Eck, S.; Castellarin-Cudia, C.; Surnev, S.; Ramsey, M. G.; Netzer, F. P. Surface Science 2002, 520, 173-185. El Fallah, J.; Hilarie, L.; Roméo, M.; et.al. Journal of Electron Spectroscopy and Related Phenomena 1995, 73, 89-103. Elidrissi, B.; Addou, M.; Regragui, M.; Monty, C.; Bougrine, A.; Kachouane, A. Thin Solid 305 Films 2000, 379, 23-27. Etsel, T. H.; Flengas, N. Chemical Reviews 1970, 70, 339-376. Evans, A. G.; Heure, A. H. Journal of the American Ceramic Society 1980, 63, 241-248. Faber, K. T.; Evans, A. G. Acta Metallurgica 1983, 31, 565-576. Fahrenholtz, W. G.; O'Keefe, M. J.; Zhou, H.; Grant, J. T. Surface and Coatings Technology 2002, 155, 208-213. Farmer, V. C. Spectrochima Acta 2000, 56, 927-930. Farmer, V. C.; Russell, J. D. Spectrochima Acta 1964, 20, 1149-1178. Favier, V.; Chanzy, H.; Cavaillé, J. Y. Macromolecules 1995, 28, 6365-6367. Foreman, J.; NexTech Materials, 2003. Forsyth, M.; Forsyth, C. M.; Wilsom, K.; Behrsing, T.; Deacon, G. B. Corrosion Science 2002, 44, 2651-2656. Francids, L. F. In Intermetallic and Ceramic Coatings; Dahotre, N. B., Sudarshan, T. S., Eds.; Marcel Dekker, Inc: New York, 1999, pp 31-82. Francisco, M. S. P.; Mastelaro, V. P.; Nascente, P. A. P.; Florebtino, A. O. The Journal of Physical Chemistry B 2001, 105, 10515-10522. Franco, F.; Pérez-Maqueda, L. A.; Pérez-Rodríguez, J. L. Thermochimica Acta 2003, 4, in press. Frost, R. L.; Rintoul, L. Applied Clay Science 1996, 11, 171-183. Fujimori, A. Physical Review B 1983, 28, 2281-2283. Fukushima, Y.; Inagaki, S. Journal of Inclusion Phenomena 1987, 5, 473-482. Gal-Or, L.; Sibermar, I.; Chaim, R. Journal of the Electrochemical Society 1991, 138, 1939- 1942, 1942-1946. Gao, L.; Yang, H.; Yuan, R.; Huang, P. Journal of Materials Processing Technology 2001, 115, 306 298-301. Garcia-Rodriguez, A.; del Rey-Bueno, F.; del Rey-Perez-Caballero, F. J.; Ureña-Amate, M. D.; Mata-Arjiona, A. Materials Chemistry and Physics 1995, 39, 269-277. Gates, S. M. Chemical Reviews 1996, 96, 1519-1532. Gerbinger, J.; Lohwasser, W.; Lampe, U.; Meixner, H. Sensors and Actuators B 1995, 26-27, 93- 96. German, R. M. Sintering Theory and Practice; John Wiley & Sons, Inc.: New York, 1996. Ghodsi, F. E.; Tepehan, F. Z.; Tepehan, G. G. Solar Energy Materials & Solar Cells 2001, 68, 355-364. Giannelis, E. Chemistry of Materials 1998, 2, 627-629. Glasner, A.; Steinberg, M. Journal of Inorganic and Nuclear Chemistry 1961, 16, 279-287. Glasner, A.; Steinberg, M. Journal of Inorganic and Nuclear Chemistry 1961, 22, 39-48. Gnanarajan, S.; Savvides, N. Thin Solid Films 1999, 350, 124-129. Golden, T. D.; Wang, A. Q Journal of the Electrochemical Society 2003, 9, in press Goldenstein, J. I.; Newbury, D. E.; Echlin, P.; et.al. Scannig Electron Microscopy and X-ray Microanalysis, A Text for Biologists, Materials Scientists, and Geologists; 2nd ed.; Plenum publishing Corportation: New York, 1992. Gopalan, S.; Singhal, S. C. Scripta Materialia 2000, 42, 993-996. Gorte, R.; Kim, H.; Vohs, J. M. Journal of Power Sources 2002, 106, 10-15. Gosser, D. K. Cyclic Voltammetry: Simulation and Analysis of Reaction Mechanisms; VCH: New York, 1993. Green, D. J.; Hannink, R. H. J.; Swain, M. V. Transformation Toughening of Ceramics; CRC press, Inc.: Florida, USA, 1989. 307 Gregg, S. J.; Sing, K. S. W. Adsorption, Surface Area and Porosity; 2nd ed.; Academic Press: London & New York, 1982. Grigge, J.; University of Florida: Gainsville, 1998. Groza, J. G. Nanostructured Materials 1999, 12, 987-992. Gum, J. C.; Ye, F.; Zhou, Y.; Lei, T. C. Ceramics International 2000, 26, 855-863. Günay, V.; Hampshire, S. Journal of Materials Processing Technology 1995, 54, 348-354. Gunnarsson, O.; Schönhammer, K. Physical Review B. 1983, 28, 4315-4341. Haensel, R.; Rabe, P.; Sonntag, B. Solid State Communications 1970, 8, 1845-1848. Hamagami, J.; Inda, Y.; Umegaki, T.; Yamashita, K. Solid State Ionics 1998, 113-115, 235-239. Han, Y.-S.; Matsunoto, H.; Yamanaka, S. Chemistry of Materials 1997, 9, 2013-2018. Harnoic, C.; Hervieu, M.; Monot-Laffez, I.; Desgardin, G. Journal of European Ceramic Society 2001, 21, 1909-1912. He, L.; Hasegawa, S. Thin Solid Films 2001, 384, 195-199. Heineman, W. R.; Kissinger, P. T. Current Separations 1989, 9, 15-18. Heineman, W. R.; Kissinger, P. T. In Laboratory Techniques in Electroanalytical Chemistry; 2nd ed.; Kissinger, P. T., Heineman, W. R., Eds.; Marcel Dekker, Inc.: New York, 1996, pp 51-125. Helms, C. R.; Spicer, W. E. Applied Physics Letters 1972, 21, 237-239. Hernando, M. J.; Pesquera, C.; Blanco, C.; González, F. Chemistry of Materials 2001, 13, 2154- 2159. Hirano, M.; Fukuda, Y.; Iwata, H.; Horta, Y.; Inagaki, M. Journal of the American Ceramic Society 2000, 83, 1287-1289. Hirschauer, B.; Göthelid, M.; Janin, E.; Lu, H.; Karlsson, U. O. Applied Surface Science 1999, 308 148, 164-170. Ho evar, S.; Kra ovec, U. O.; Antonino, B. O. Applied Catalysis B: Environmental 2000, 28, 113-125. Hoflund, G. B. In Handbook of Surface and Interface Analysis; Riviere, J. C. a. M., S., Ed.; Marcel Dekker Inc.: New York, 1998, pp 57-86. Holgado, J. P.; Alvarez, R.; Munuera, G. Applied Surface Science 2000, 161, 301-315. Holgado, J. P.; Munuera, G.; Espinós, J. P.; et.al. Applied Surface Science 2000, 158, 164-171. Horch, R. A.; Golden, T. D.; D'Sousa, N. A.; Riester, L. Chemistry of Materials 2002, 14, 3531- 3538. Hu, C.-C.; Tsou, T.-W. Electrochemistry Communications 2002, 2, 105-109. Huang, D.; Qin, F.; Yao, Z.; Ren, Z. Applied Physics Letters 1995, 67, 3725-3726. Huang, W.; Shuk, P.; Creenblatt, M. Solid State Ionics 1997, 100, 23-27. Hughes, A. E.; Gorman, J. D.; Patterson, P. J. K.; Carter, R. Surface and Interface Analysis 1996, 24, 634-640. Huijsmans, J. P. P. Current Opinion in Solid State and Materials Science, 2001, 5, 317-323. Hussein, G. A. M. Journal of Analytical and Applied Pyrolysis 1996, 37, 111-149. Hutchioson, J. W. Acta Metallurgica 1987, 35, 1605-1619. Hwang, B.-J.; Shyang, S.-K.; Wei, C.-Y. Journal of The Chinese Institute of Chemical Engineers 1995, 26, 37-45. Ihaba, H.; Tagawa, H. Solid State Ionics 1996, 83, 1-16. Inoue, T.; Yamamoto, Y.; Koyama, S.; Suzuki, S.; Ueda, Y. Applied Physics Letters 1990, 56, 1332-1333. Inoue, T.; Yamamoto, Y.; Satoh, M. Thin Solid Films 1999, 343/344, 594-597. 309 Izu, N.; Shin, W.; Murayama, N.; Kanzaki, S. Sensors and Actuators B 2002, 87, 95-98. Jacobsen, S. N. In Lingköping Studies in Science and Technology; Lingköping University,: Lingköping, Sweden, 1998. Janssen, R.; Heussner, K.-H. Advanced Ceramics 1995, 23, 241-245. Jenkins, R.; Snyder, R. L. Introduction to X-ray Powder Diffractometry; John Wilery & Sons, Inc.: New York, 1996; Vol. 138. Jeong, Y.-K.; Choa, Y.-H.; Niihara, K. Journal of the American Ceramic Society 1998, 81, 1453- 1260. Ji, X.; Ponton, C. B.; Marquis, P. M. Journal of Materials Science: Materials in Medicine. 1992, 3, 283-287. Jurado, J. R. Journal of Materials Science 2001, 36, 1133-1139. Kanakaraju, S.; Mohan, S.; Sood, A. K. Thin Solid Films 1997, 305, 191-195. Kang, J.; Liu, X.; Lian, G.; Zhang, Z. Microelectronic Engineering 2001, 56, 191-194. Kapat, J. S.; Kumar, A. In Intermetallic and Ceramic Coatings; Dahotre, N. B., Sudarshan, T. S., Eds.; Marcel Dekker. Inc.: New York, 1999, pp 441-484. Katdare, S. P.; Ramaswamy, V.; Ramaswamy, A. V. Journal of Materials Chemistry 1997, 7, 2197-2199. Katdare, S. P.; Ramaswamy, V.; Ramaswamy, A. V. Catalysis Today 1999, 49, 313-320. Kavan, L.; O'Regan, B.; Kay, A.; Gratzel, M. Journal of Electroanalytical Chemistry 1993, 346, 291-307. Kaya, C.; Gu, X.; Al-Dawery, I.; Butler, E. G. Science and Technology of Advanced Materials 2002, 3, 35-44. Kaya, C.; Kaya, F.; Marsoglu, M. Journal of Materials Science Letters 1997, 892-896. 310 Kaya, C.; Kaya, F.; Trusty, P. A.; Boccaccini, A. R.; Marsoglu, M. Ceramics International 1999, 25, 359-366. Kepinski, L.; Wolcyrz, M. Catalysis Letters 1992, 15, 329-337. Kharton, V. V.; Figueiredo, F. M.; Navarro, L. Journal of Materials Science 2001, 36, 1105- 1117. Khodadadi, A.; Mohajerzadeh, S. S.; Mortazavi, Y.; Miri, A. M. Sensors and Actuators B 2001, 80, 267-271. Kiekeng, P.; Verplaetse, H.; Temmerman, E. Journal of Electroanalytical Chemistry 1980, 113, 233-238. Kim, C. G. Journal of Vacuum Science & Technology B 2000, 18, 2650-2652. Kim, T. H.; Jang, L. W.; Lee, D. C.; Choi, H. J.; John, M. S. Macromolecular Rapid Communications 2002, 23, 191-195. Kiš, E. E.; Ranogajec, J. G.; Marinkovic-Neducin, R. P.; Vulic, T. J. APTEFF 2001, 32, 71-78. Kleinlogel, C.; Gaukler, L. J. Solid State Ionics 2000, 135, 567-573. Klug, H. P. A.; Alexander, L. E. X-ray Diffraction Procedures for Polycrystalline and Amorphous Materials; 2nd ed.; Wiley: New York, 1974. Koelling, D. D.; Boring, A. M.; Wood, J. H. Solid State Communications 1983, 47, 227-232. Kolat, R. S.; Powell, J. E. Inorganic Chemistry 1962, 1, 293-296. Komarnenl, S. Journal of Materials Chemsitry 1992, 2, 1219-1230. Konishi, Y.; Murai, T.; Asai, S. Industrial & Engineering Chemsitry Research 1997, 36, 2641- 2645. Kosynkin, V. D.; Arzgatkina, A. A.; Ivanov, E. N.; Chtoutsa, M. G.; Grabko, A. I. Journal of Alloys and Compounds 2000, 303-304, 421-425. 311 Kotani, A.; Haruhiko, O. Journal of Electron Spectroscopy and Related Phenomena 1992, 60, 257-299. Kotani, A.; Jo, T.; Parlebast, J. C. Advances in Physics 1988, 37, 37-85. Kotoul, M.; Profant, T. Mechanics of Materials 2000, 32, 203-234. Krishna, M. G.; Hartridge, A.; Bhattacharya, A. K. Materials Science Engineering B 1998, 55, 14-20. Kristof, J.; Mink, J.; Hováth, E.; Gábor, M. Vibrational Spectroscopy 1993, 5, 61-67. Laachir, A.; Perrichon, V.; Badri, A.; Lamotte, J.; Catherine, E.; Lavalley, J. C.; El Fallah, J. H.; Hilaire, L. Journal of the Chemical Society Faraday Transactions 1991, 87, 1601-1609. Lan, T.; Kavirarna, P. D.; Pinnavaia, T. J. Journal of Physics and Chemistry of Solids 1996, 57, 1003-1010. Laubversheimer, J.; Ritzhaupt-Kleissl, H.-J.; Haubelt, J.; Emit, G. Journal of European Ceramic Society 1998, 18, 255-260. Lavik, E. B.; Chiang, Y.-M. Material Research Society Symposium Proceedings 1997, 457, 63- 68. Law, W. L.; Hill, R. H. Materials Research Bulletin 1998, 33, 69-80. LeBaron, P. C.; Wang, Z.; Pinnavaria, T. J. Applied Clay Science 1999, 15, 11-29. Lee, G. R. Journal of Materials Chemistry 1996, 6, 187-192. Lee, T.-C.; Deng, J. Ceramics International 2001, 27, 755-760. Levin, D.; Soled, S. L.; Ying, J. Y. ACS Symposium Series 1996, 237-249. Lewis, I. R.; Edwarsds, H. G. M. Handbook of Raman Spectroscopy, from the Research Laboratory to the Process Line; Marcel Dekker, Inc.: New York, 2001; Vol. 28. Lewis, R.; Kuroda, H. Solid State Ionics 1989, 32/33, 373-377. 312 Li, C.; Domen, K.; Maruya, K.; Onishi, T. Journal of the American Chemical Society 1989, 111, 7683-7687. Li, C.; Domen, K.; Maruya, K.; Onishi, T. Journal of Catalysis 1993, 131, 540-547. Li, F.; Newman, R. C.; Thompson, G. E. Electrochima Acta 1997, 42, 2455-2464. Li, F.; Thompson, G. E. Journal of the Electrochemical Society 1999, 146, 1809-1815. Li, F.; Thompson, G. E.; Newman, R. C. Applied Surface Science 1998, 126, 21-33. Li, J.-F.; Watanabe, R. Journal of the American Ceramic Society 1995, 78, 1079-1082. Li, M.; Wang, Z.; Fan, S.; Zhao, Q.; Xiong, G. Nuclear Instruments and Methods in Physics Research B 1998, 135, 535-539. Li, S.-B.; Zhang, B.-S.; Wen, G.-W.; Xie, J.-X. Materials Letters 2002, 57, 1445-1452. Liu, H.; Yang, J.; Liang, H.-H.; Zhuang, J.-H.; Zhou, W.-F. Journal of Power Sources 2001, 93, 230-233. Liu, R.; Li, D. Y. Journal of Materials Science 2000, 35, 633-641. Liu, S.; Liu, F.; Guo, H.; Zhang, Z.; Wang, Z. Physics Letters A 2000, 211, 128-133. Livage, J.; Ganguli, D. Solar Energy Materials & Solar Cells 2001, 68, 365-381. Livermore, S. J. A.; Cotton, J. W.; Ormerod, R. M. Journal of Power Sources 2000, 86, 411-416. Long, R. Q.; Huang, Y. P.; Wan, H. L. Journal of Raman Spectroscopy 1997, 28, 29-32. Long, R. Q.; Wan, H. L. Journal of the Chemical Society Faraday Transactions 1997, 93, 355- 358. Lovric, M.; Osteryoung, J. Journal of Electroanalytical Chemistry 1986, 197, 63-75. Lu, Y. C.; Ives, M. B. Corrosion Science 1993, 34, 1773-1785. Lu, Y. C.; Ives, M. B. Corrosion Science 1995, 37, 145-155. Lu, Z.; Hiskes, R.; DiCarolis, S. A.; Nel, A. Journal of Crystal Growth 1995, 156, 227-234. 313 Luo, L.; Wu, X. D.; Dye, R. C.; Muenchausen, R. E.; Foltyn, S. R. Applied Physics Letters 1991, 59, 2043-2045. Luo, M.; Lu, G.; Zheng, X. Journal of Materials Science Letters 1998, 17, 1553-1557. Madejová, J. Vibrational Spectroscopy 2003, 31, 1-10. Madejová, J.; Arvaiová, B.; Komadel, P. Spectrochima Acta A 1999, 55, 2467-2476. Madejová, J.; Janek, M.; Komadel, P.; Herbert, H.-J.; Moog, H. C. Applied Clay Science 2002, 20, 255-271. Madejová, J.; Komadel, P. Clays and Clay Minerals 2001, 49, 410-432. Majumder, S. B.; Mohapatra, Y. N.; Agrawal, D. C. Journal of Materials Science 1997, 32, 2141-2150. Mamedov, A.; Ostrander, J.; Aliev, F.; Kotov, N. A. Langmuir 2000, 16, 3941-3949. Mamontov, E.; Egami, T. Journal of Physics and Chemistry of Solids 2000, 61, 1345-1356. Mariani, M. P.; Bartolozzi, M. Journal of Electroanalytical Chemistry 1996, 209, 275-282. Marina, O. A.; Bagger, C.; Primdahl, S.; Mogensen, M.; Sammes, S. M.; Tompsett, G. A. Solid State Ionics 1999, 123, 199-208. Marzouk, S. A. M.; Ufer, S.; Buck, R. P. Analytical Chemistry 1998, 70, 5054-5061. Masetti, E.; Varsano, F.; Decker, F.; Krasilnikova, A. Electrochimica Acta 2001, 46, 2085-2090. Masui, T.; Fujiwara, K.; Machida, K.; Adachi, G. Chemistry of Materials 1997, 9, 2197-2204. Mathey, Y.; Greig, D. R.; Shriver, D. F. Inorganic Chemistry 1982, 21, 3409-3413. Maul, M.; Schulte, B.; Häussler, P.; Frank, G.; Seinborn, T.; Fues, H.; Adrian, H. Journal of Applied Physics 1993, 74, 2942-2944. Mayo, M. International Materials Review 1996, 41, 85-115. McAleese, J.; Plakatouras, J. C.; Steele, B. C. H. Thin Solid Films 1996, 280, 152-159. 314 McBride, J. R.; Hass, K. C.; Poindexter, B. D.; Weber, W. H. Journal of Applied Physics 1994, 76, 2435-2441. Meguid, S. A. Journal of Materials Processing Technology 1996, 56, 978-989. Mishra, T.; Parida, K. Applied Catalysis: General, 1998, 174, 91-98. Mogensen, M.; Sammes, S. M.; Tompsett, G. A. Solid State Ionics 2000, 129, 63-94. Mölsä, H.; Niionstö, L. Materials Research Society Symposium Proceedings 1994, 335, 341-350. Molycorp Cerium: A Guide to its Role in Chemical Technology; Molycorp, Inc.: Mountain Pass, CA U.S.A, 1992. Monk, P. M. S.; Mortimer, R. J.; Rosseinsky, D. R. Electrochromism: Fundamentals and Applications; VCH: Weinheim, 1995. Moore, D. M.; Reynolds, R. C. J. X-ray Diffraction and Identification and Analysis of Clay Minerals; 2nd ed.; Oxford University Press, Inc.: New York, 1997. Moosath, S. S.; Abraham, J.; Swaminathan, T. V. Zeitschrift Fuer Anorganische und Allgemeine Chemie 1963, 324, 103-105. Morshed, A. H.; Moussa, M. E.; Leonard, R.; Liu, S. X.; El-Masry, N. Applied Physics Letters 1997, 70, 1647-1649. Mullins, D. R.; Overbury, S. H.; Huntley, D. R. Surface Science 1998, 409, 307-319. Mullins, D. R.; Radulovic, P. V.; Overbury, S. H. Surface Science 1999, 429, 186-198. Musiani, M. Chemical Communications. 1996, 21, 2403-2404. Nachimuthu, P.; Shih, W.-C.; Liu, R.-S.; Jang, L.-Y.; Chen, J.-M. Journal of Solid State Chemistry 2000, 149, 408-413. Nagai, N.; Hashimoto, H. Applied Surface Science 2001, 172, 307-311. Nakajima, A.; Yoshihara, A.; Ishigame, M. Physical Review B 1994, 50, 13297-13307. 315 Narayanan, R.; Seshadri, S. K. Metal Finishing 2001, 2, 84-89. Nigro, R. L.; Malandrino, G.; Fragalà, I. L. Chemistry of Materials 2001, 13, 4402-4404. Nihara, K. Journal of the Ceramic Society of Japan 1991, 99, 974-981. Nörenberg, H.; Briggs, G. A. D. Surface Science 1999, 433-435, 127-130. Normand, F. L.; Hilaire, L.; Killi, K.; et.al.. Journal of Physical Chemistry 1988, 92, 2561-2568. Occelli, M. L.; Auroux, A.; Ray, G. J. Microporous and Mesoporous Materials 2000, 39, 43-56. Ohji, T. Ceramic Engineering and Science Proceedings. 1997, 18, 353-360. Ohtsuka, K.; Koga, J.; Tsunoda, M.; Suda, M.; Ono, M. Journal of the American Ceramic Society 1990, 73, 1719-1725. Oliver, W. C.; Pharr, G. M. Journal of Materials Research 1992, 7, 1564-1583. Ooka, C.; Akita, S.; Ohashi, Y.; Horiuchi, T. Journal of Materials Chemsitry 1999, 9, 2943- 2952. Orel, Z. C. The Internet Journal of Vibrational Spectroscopy 1999, 3. http://www.ijvs.com/volume3/edition4/section3.htm Orel, Z. C.; Hutchins, M. G.; McMeeking, G. Solar Energy Materials & Solar Cells 1993, 30, 327-337. Ovel, Z. C.; Ovel, B. J. Journal of Materials Science 1995, 30, 2284-2290. Overbury, S. H.; Huntley, D. R.; Mullins, D. R.; et.al. Journal of Vacuum Science and Technology A 1997, 15, 1647-1652. Özer, N. Solar Energy Materials & Solar Cells 2001, 68, 391-400. Özer, N.; Cronin, J. P.; Akyuiz, S. SPIE 1999, 3788, 103-110. Paama, L.; Pitkãnen, I.; Perãmãki, P. Talanta 2000, 51, 349-357. Paillet, M.; Dufresne, A. Macromolecules 2001, 34, 6527-6530. 316 Palmqvist, A. E. C.; Wirde, M.; Gelius, U.; et.al. Nanostructured Materials 1999, 11, 995-1007. Pan, M.; Meng, G. Y.; Xin, H. W.; Chen, C. S. Thin Solid Films 1998, 324, 89-93. Papaiacovou, P.; Hussey, R. J.; Mitchell, D. F.; Graham, M. J. Corrosion Science 1990, 30, 451- 455. Paparazzo, E.; Ingo, G. M.; Zacchetti, N. Journal of Vacuum Science and Technology A 1991, 9, 1416-1421. Park, A.; Gorte, R. J.; Vohs, J. M. Applied Catalysis A: General 2000, 200, 55-61. Paunovic, M.; Schlesinger, M. Fundamentals of Electrochemical Deposition; John Wiley & Sons, Inc.: New York, 1998. Petit, M. A.; Plichon, V. Journal of Electroanalytical Chemistry 1998, 444, 247-252. Pfau, A.; Schierbaum, K. D.; Göpel, W. Surface Science 1995, 331-333, 1479-1485. Pinnavaia, T. J. Science 1983, 220, 365-371. Pinnavaia, T. J.; Tzou, M. S.; Landau, S. D. Journal of the American Chemical Society 1985, 107, 4783-4785. Pollard, K. D.; Jenkins, H. A.; Puddephatt, R. J. Chemistry of Materials 2000, 12, 701-710. Poorteman, M.; Descamps, P.; Cambier, F.; Poulet, A.; Descamps, J. C. Journal of European Ceramic Society 1999, 19, 2375-2379. Pospíšil, M.; Capkivá, P.; Weiss, Z.; Malác, Z.; Šimoník, J. Journal of Colloid and Interface Science 2002, 25, 126-132. Pourbaix, M. Atlas of Electrochemical Equilibria in Aqueous Solutions; NACE: Houston, TX, 1974. Praline, G.; Koel, B. E.; Hance, R. L.; et.al. Journal of Electron Spectroscopy and Related Phenomena 1980, 21, 17-30. 317 Preisler, E. Journal of Applied Electrochemistry 1989, 19, 559-565. Preisler, E. J.; Marsh, O. J.; Beach, R. A.; McGill, T. C. Journal of Vacuum Science and Technology B 2001, 19, 1611-1618. Putna, E. S.; Stubenrauch, J.; Vohs, J. M.; Gorte, R. J. Langmuir 1995, 11, 4832-4837. Qiao, J.; Yang, C. Y. Materials Science and Engineering R 1995, 14, 157-202. Ralph, J. M.; Scheler, A. C.; Krumpelt, M. Journal of Materials Science 2001, 36, 1161-1172. Ramírez-Duverger, A.; Ruiz-Salvador, A.; Hernández-Sáchez, M. P.; García-Sánchez, M. F.; Rodríguez-Gattorno, G. Solid State Ionics 1997, 96, 89-93. Rana Rao, M. V.; Shripathi, T. Journal of Electron Spectroscopy and Related Phenomena 1997, 87, 121-126. Ranade, A.; D'Sousa, N. A.; Gnade, B. Polymer 2002, 43, 3759-3766. Rao, G. R.; Sahu, H. R. Proceedings - Indian Academy of Sciences, Chemical Sciences 2001, 113, 651-658. Reimanis, I. E. Materials Science and Engineering A 1997, 237, 159-167. Rightor, E. G.; Tzou, M. S.; Pinnavaia, T. J. Journal of Catalysis 1991, 130, 29-40. Riviere, J. C.; Myhra, S. Handbook of Surface and Interface Analysis: Methods for Problem- Solving; Marcel: New York, 1998. Rochat, N.; Chabli, A.; Bertin, F.; Vergnaud, C.; Mur, P.; Petitdidier, S. Materials Science and Engineering B 2003, in press. Roméo, M.; Bak, K.; El Fallah, J.; et.al. Surface and Interface Analysis 1993, 20, 508-512. Roy, R. Science 1987, 238, 1664-1669. Rudd, A. L.; Breslin, C. B.; Mansfeld, F. Corrosion Science 2000, 42, 275-288. Rühle, M.; Evans, A. G.; McMeeking, R. M.; Charlambides, P. G. Acta Metallurgica 1987, 35, 318 2701-2710. Sauma, V. E.; Natekar, D.; Sbaizero, O. Materials Science and Engineering 2002, 323, 129-137. Savvides, N.; Thorley, A.; Gnanarajan, S.; Katssros, A. Thin Solid Films 2001, 338, 177-182. Schemilt, J. E.; Williams, H. M.; Edirisinghe, M. J.; Evans, J. R. G.; Ralph, B. Scripta Materialia 1997, 36, 929-934. Schierbaum, K. Surface Science 1998, 399, 29-38. Schrader, B. Infrared and Raman Spectroscopy: Methods and Applications; VCH publishers, Inc.: New York, 1995. Sedriks, A. J. Corrosion of Stainless Steels; 2nd ed.; John Wiley & Sons, Inc.: New York, 1996. Shemilt, J. E.; Williams, H. W.; Edirisinghe, M. J.; Evans, J. R. G.; Ralph, B. Scripta Materialia 1997, 36, 929-934. Shen, W.; Matsumura, Y. Journal of Molecular Catalysis A: Chemical 2000, 153, 165-168. Shi, D. Q.; Ionescu, M.; McKinnon, J.; Dou, S. X. Physica C 2001, 356, 304-310. Shuk, P.; Ramanujachary, K. V.; Greenblatt, M. Solid State Ionics 1996, 85, 257-263. Shuk, P.; Ramaswamy, A. V.; Greenblatt, M. Solid State Ionics 1996, 86-88, 1115-1120. Sing, K. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2001, 187-188, 3- 9. Sivakumar, A.; Damodaran, A. D.; Warrier, K. G. K. Ceramics International 1995, 21, 85-88. Skoog, D. A.; Leary, J. L. Principles of Instrumental Analysis; 4th ed.; Saunders College Publishing,: Forth Worth, USA, 1992. Smith, G. C. In Handbook of Surface and Interface Analysis, Methods for Problem-Solving; Riviere, J. C., Myhra, S., Eds.; Marcel Deker, Inc.: New York, 1998, pp 159-208. Socrates, G. Infrared Characteristic Group Frequencies; 2nd ed.; John Wiley & Sons: 319 Chichester, 1998. Sonesson, A. Acta Chemica Scandinavica 1958, 12, 165-181. Sonesson, A. Acta Chemica Scandinavica 1958, 12, 1937-1954. Sonesson, A. Acta Chemica Scandinavica 1960, 14, 1495-1498. Song, G. M.; Zhou, Y.; Sun, Y. Ceramics International 1999, 25, 257-260. Spanier, J. E.; Robinson, R. D.; Zhang, F.; Chan, S.-W.; Herman, I. P. Physical Review B 2001, 64, 245407-245415. Stambouli, A. B.; Traversa, E. Renewable and Sustainable Energy Reviews 2002, 6, 433-455. Steele, B. C. H. Solid State Ionics 2000, 129, 95-110. Stefanik, T. S.; Tuller, H. L. Journal of European Ceramic Society 2001, 21, 1967-1970. Strydom, C. Inorganica Chimica Acta 1989, 161, 7-9. Stubenranch, J.; Vohs, J. M. Journal of Catalysis 1996, 159, 50-57. Subbarao, E. C. Colloids and Surface A: Physicochemical and Engineering Aspects 1998, 133, 3-11. Suemasu, H.; Kondo, A.; Itatani, K.; Nozue, A. Composites Science and Technology 2001, 61, 281-288. Suryanarayana, C.; Norton, M. G. X-ray Diffraction: A Practical Approach; Plenum Press: New York, 1998. Suzuki, T.; Kosacki, I.; Anderson, H. U. Journal of the American Ceramic Society 2001, 84, 2007-2014. Switzer, J. A. Journal of the Electrochemical Society 1986, 133, 722-728. Switzer, J. A. American Ceramic Society Bulletin 1987, 66, 1521-1524. Switzer, J. A.; Phillips, R. J. Material Research Society Symposium Proceedings 1988, 121, 111- 320 114. Switzer, J. A.; Shumsky, M. G.; Bohannan, E. W. Science 1999, 284, 293-296. Sze, S. M. Physics of Semiconductor Devices; 1st ed.; Wiley-Interscience: New York, 1969. Szostak, R.; Ingram, C. Studies in Surface Science and Catalysis 1995, 94, 13-38. Szücs, A.; Berger, F.; Dekány, I. Colloids and Surface A: Physicochemical and Engineering Aspects 2000, 174, 387-402. Tait, W. S. An Introduction to Electrochemical Corrosion Testing for Practicing Engineers and Scientists; Pair O Docs Pubns: Racine, Wisonsin, 1994. Tan, H.; Yang, W. Mechanics of Materials 1998, 30, 111-123. Tan, K. H. In Principles of Soil Chemistry; Marcel Dekker, Inc.: New York, 1998, pp 177-259. Tan, Y.-J.; Lim, K. Y. Surface and Coatings Technology 2003, 167, 255-262. Temuujin, J.; Okada, K.; MacKenzie, K. J. D.; Jadambaa, T. Journal of European Ceramic Society 1998, 19, 105-112. Teterin, Y. A.; Teterin, A. Y.; Lebedev, A. M.; Utkin, I. O. Journal of Electron Spectroscopy and Relative Phenomena 1998, 88-91, 275-279. Therese, G. H. A.; Kamath, P. V. Chemistry of Materials 2000, 12, 1195-1204. Thomsen, N. B.; Karihaloo, B. L. Journal of the Mechanics and Physics of Solids 1996, 44, 207- 231. Tian, Y. J.; Linzen, S.; Schmidl, F.; Matthes, A. Thin Solid Films 1999, 338, 224-230. Trovareli, A. Catalysis Reviews: Science and Engineering. 1996, 38, 439-520. Trovareli, A.; Boaro, M.; Rocchini, E.; DeLeitenburg, C.; Dolcetti, G. Journal of Alloys and Compounds 2001, 323-324, 584-591. Trovareli, A.; De Leitenbutrg, C.; Boaro, M.; Dolcetti, D. Catalysis Today 1999, 50, 353-367. 321 Tschöpe, A. Solid State Ionics 2001, 139, 267-280. Tschöpe, A.; Liu, W.; Flytzani-Stephanopoulos, M.; Ying, J. Y. Journal of Catalysis 1995, 157, 42-50. Tschöpe, A.; Schaadt, D.; Birringer, R.; Ying, J. Y. Nanostructured Materials 1997, 9, 423-432. Tschöpe, A.; Sommer, E.; Birringer, R. Solid State Ionics 2001, 139, 255-265. Tschöpe, A.; Ying, J. Y.; Tuller, H. L. Sensors and Actuators B 1996, 31, 111-114. Tsunekawa, S.; Fukuda, T.; Kasuya, A. Surface Science 2000, 457, L437-440. Tsunekawa, S.; Ishikawa, K.; Li, Z.-Q.; Kawazoe, Y.; Kasuya, A. Physical Review B 2000, 85, 3440-3443. Tsunekawa, S.; Sahara, R.; Kawazoe, Y.; Ishikawa, K. Applied Surface Science 1999, 152, 53- 56. Tsunekawa, S.; Sivamohan, R.; Ito, S.; Kasuya, A.; Fukuda, T. Nanostructured Materials 1999, 11, 141-147. Tuller, H. L. Solid State Ionics 2000, 131, 143-157. Tulun, T.; Akinci, H. Ceramics International 1991, 23, 141-148. Turrel, G.; Corset, J. Raman Microscopy, Developments and Applications; Academic Press, Harcourt Brace & Company: San Diego, 1996. Twu, J.; Chuang, C. J.; Chang, K. I.; Yang, C. H.; Chen, K. H. Applied Catalysis B: Environment 1997, 12, 309-324. Šulcova, P.; Trojan, M.; Šolc, Z. Dyes and Pigments 1998, 37, 65-70. Varsano, F.; Decker, F.; Masetti, E.; Cardellini, F.; Licciulli, A. Electrochimica Acta 1999, 44, 3149-3156. Vratny, F.; Rao, C. N. R.; Dilliy, M. Analytical Chemistry 1961, 33, 1455-1455. 322 Walker, J. K.; Kinkoph, J.; Saha, C. K. Journal of Applied Electrochemistry 1989, 19, 225-230. Wang, A.; Belot, J. A.; Tobin, J. M.; Markworth, P. R. Physica C 1999, 320, 154-160. Wang, A. Q.; D'Souza, N. A.; Golden, T. D. Composite B 2003, Submitted. Wang, A. Q.; Golden, T. D. Journal of the Electrochemical Society 2003, 9, in press. Wang, A. Q.; Punchaipetch, P; Wallace, R. M.; Golden, T. G. Journal of Vacuum Science and Technology B 2002, 21, 1169-1175. Wang, C.; Qian, Y.; Xie, Y.; Wang, C. Materials Science and Engineering B 1996, 39, 160-162. Wang, K. L.; Zhang, Q. B.; Sun, M. L.; Zhu, Y. M. Surface and Coatings Technology 1997, 96, 267-271. Wang, L. P.; Tang, B. Y.; Huang, N.; Tian, X. B.; Chu, P. K. Materials Science and Engineering A 2001, 308, 176-179. Wang, S.; Wang, W.; Liu, Q.; Zhang, M.; Qian, Y. Solid State Ionics 2000, 133, 211-215. Wang, S.; Wang, W.; Zuo, J.; Qian, Y. Materials Chemistry and Physics 2001, 68, 246-248. Weber, W. H.; Hass, K. C.; McBride, J. R. Physical Reviews B 1993, 48, 178-185. Wendlandt, W. W. Analytical Chemistry 1958, 30, 58-61. Wildan, M.; Edrees, H. J.; Hendry, A. Materials Chemistry and Physics 2002, 75, 276-283. Williamson, G. K.; Hall, W. H. Acta Metallurgica 1952, 1, 22-31. Woirgard, J.; Tromas, C.; Giard, J. C.; Audurier, V. Journal of the European Society 1998, 18, 2297-2305. Wu, X. D.; Dye, R. C.; Muenshasusen, R. E.; Foltyn, S. R. Applied Physics Letters 1991, 58, 2165-2167. Wu, Z.; Huang, D.; Yang, X. Vacuum 1998, 51, 397-401. Wu, Z.; Huang, D.; Yang, X.; Wang, J.; Qin, F.; Zhang, J.; Yang, Z. Vacuum 1998, 49, 133-137. 323 Wuillound, E.; Delly, B.; Schneider, W.-D.; et.al. Physical Review Letters 1984, 53, 202-205. Xavier, K. O.; Sreekala, R.; Rashid, K. K. A.; Yusuff, K. K. M.; Sen, B. Catalysis Today 1999, 49, 17-22. Xie, W.; Gao, Z.; Liu, K.; Pan, W.; Vaia, R.; Hunter, D.; Singh, A. Thermochimica Acta 2001, 367-368, 339-350. Xu, C.; Ai, X. International Journal of Refractory Metals & Hard Materials 2000, 19, 85-88. Xu, H. H. K.; Quinn, J. B.; Smith, D. T.; Antonucci, J. M. Biomaterials 2002, 23, 735-742. Yamamoto, O. Electrochima Acta 2000, 45, 2423-2435. Yamanaka, S.; Brindley, G. W. Clays and Clays Minerals 1979, 27, 119-124. Yan, L.; Roth, C. B.; Low, P. Journal of Colloid and Interface Science 1996, 184, 663-670. Yang, X.; Rahaman, M. N. Journal of European Ceramic Society 1997, 17, 525-535. Yao, C.; Weng, H. Industrial & Engineering Chemsitry Research 1998, 37, 2647-2653. Ye, F.; Lei, T. C.; Zhao, Y. C.; Zhou, Y.; Yang, J. M.; Zhang, L. T. Ceramics International 1998, 24, 181-188. Ying, J. Y.; Tschöpe, A. The Chemical Engineering Journal 1996, 64, 225-237. Yu, Z. B.; Thompson, D. P.; Bhatti, A. R. Composites A, 2002, 33, 621-629. Yuan, Z.; Vleugels, J.; Biest, O. V. D. Materials Letters 2000, 46, 249-254. Zhang, F.; Chan, S.-W.; Spanier, J. E.; Apak, E.; Jin, Q. Applied Physics Letters 2002, 80, 127- 129. Zhang, J.; Ju, X.; Wu, Z. Y.; Liu, T.; Du, T. D.; Xie, Y. N. Chemistry of Materials 2001, 13, 4192-4297. Zhang, T.; Hing, P.; Huang, H.; Kilner, J. Journal of Materials Processing Technology 2001, 113, 463-468. 324 Zhang, Y.; Anderson, Y.; Muhammed, M. Applied Catalysis B: Environment 1995, 6, 325-337. Zhao, X.; et.al. Instrumental Analysis; Beijing University Publishing: Beijing, 1997. Zhitomirsky, I. Materials Letters 1998, 33, 305-310. Zhitomirsky, I. Electrochemical Society Proceedings 2000, 4, 381-388. Zhitomirsky, I. Electrochemical Society Proceedings 2000, 4, 338-345. Zhitomirsky, I. The American Ceramic Society Bulletin 2000, 9, 57-60. Zhitomirsky, I. JOM 2001, 53, 48-50. Zhitomirsky, I.; Gal-Or Journal of the Ceramic Society 1996, 16, 819-824. Zhitomirsky, I.; Gal-Or In Intermetallic and Ceramic Coatings; Dahortre, N. B., Sudarshan, T. S., Eds.; Marcel Dekker Inc.: New York, 1999, pp 83-145. Zhitomirsky, I.; Petric, A. Materials Letters 1999, 40, 263-268. Zhitomirsky, I.; Petric, A. Journal of the European Ceramic Society 2000, 20, 2055-2061. Zhitomirsky, I.; Petric, A. Ceramics International 2001, 27, 149-155. Zhou, J.-J.; Zhang, H.-T.; Chen, L.-X.; Qi, P. Journal of Materials Science Letters 1994, 13, 846-848. Zhou, Y.; Philips, R. J.; Switzer, J. A. Journal of the American Ceramic Society 1995, 78, 981- 985. Zhou, Y.; Rahaman, M. N. Acta Materialia 1997, 45, 3635-3639. Zhou, Y.; Switzer, J. A. Journal of Alloys and Compounds 1996, 237, 1-5. Ziehfreunt, A.; Maier, W. F. Chemistry of Materials 1996, 8, 2721-2729. Zsigmondy, A. Zeitschrift Fuer Anorganische und Allgemeine Chemie 1911, 71, 356-377. ASTM, Philadelpha, Pennsyvivania. E2108-00, Standard Practice for Calibration of the Electron Binding Energy Scale of an X-ray 325 Photoelectron Spectroscopy Standard Reference Materials; National Institute of Standards and Technology:Gaithersburg, MD, 1983. http://www.nanocor.com/nano_struct.asp http://ap.polyu.edu.hk/pld/PLD.htm 326