Koro 5 Chap One

CHEMICAL SENSORS
SIMULATION AND MODELING

VOLUME 5: ELECTROCHEMICAL
SENSORS
CHEMICAL SENSORS
SIMULATION AND MODELING
VOLUME 5: ELECTROCHEMICAL
SENSORS
EDITED BY
GHENADII KOROTCENKOV
GWANGJU INSTITUTE OF SCIENCE AND TECHNOLOGY
GWANGJU, REPUBLIC OF KOREA
MOMENTUM PRESS, LLC, NEW YORK
Chemical Sensors: Simulation and Modeling Volume 5: Electrochemical Sensors
Copyright Momentum Press
, LLC, 2013
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted in any form or by any meanselectronic,
mechanical, photocopy, recording or any otherexcept for brief quotations, not to
exceed 400 words, without the prior permission of the publisher.
First published by Momentum Press
, LLC
222 East 46th Street, New York, NY 10017
www.momentumpress.net
ISBN-13: 978-1-60650-596-0 (hard back, case bound)
ISBN-10: 1-60650-596-3 (hard back, case bound)
ISBN-13: 978-1-60650-598-4 (e-book)
ISBN-10: 1-60650-598-X (e-book)
DOI: 10.5643/9781606505984
Cover design by Jonathan Pennell
Interior design by Derryeld Publishing, LLC
10 9 8 7 6 5 4 3 2 1
Printed in the United States of America
v
CONTENTS
PREFACE xiii
ABOUT THE EDITOR xvii
CONTRIBUTORS xix
PART 1: SOLID-STATE ELECTROCHEMICAL SENSORS
1 SURFACE AND INTERFACE DEFECTS IN IONIC CRYSTALS 3
N. F. Uvarov
1 Introduction 3
1.1 Solid Electrolytes and Electrodes for Electrochemical Sensors:
A Brief Overview 3
1.2 Surface and Interface Properties of Ionic Solids 6
2 Calculation of the Surface Potential and Surface Defects Using the
Stern Model 8
2.1 Description of the Model 8
2.2 Pure Crystals of the NaCl Type 10
2.3 Surface Potential in NaCl Crystals Containing Divalent Cations 13
2.4 Comparison with Experimental Data 15
2.5 Surface Potential and Concentration of Point Defects on Grain
Boundaries of Superionic Oxide Ceramics 15
2.6 Surface Disorder in Terms of Energy Diagrams 23
2.7 Defects on Interfaces 25
3 Size Effects in Nanocomposite Solid Electrolytes 29
4 Applications in Sensors 30
5 Conclusions 34
References 34
vi CONTENTS
2 SOLID-STATE ELECTROCHEMICAL GAS SENSORS 41
C. O. Park
I. Lee
D. R. Lee
J. W. Fergus
N. Miura
H. J. Yoo
1 Introduction 41
2 Electrode Potentials 42
3 Types of Electrochemical Sensors 46
3.1 Equilibrium Potentiometric Sensors 46
3.2 Mixed Potentiometric Sensors 49
3.3 Amperometric Sensors 53
4 Applications 57
4.1 Oxygen Sensors 57
4.2 Carbon Dioxide Sensors 64
4.3 NO
x
Sensors 66
4.4 SO
x
Sensors 76
4.5 Hydrogen Sensors 77
Acknowledgments 86
References 86
PART 2: ELECTROCHEMICAL SENSORS FOR LIQUID ENVIRONMENTS
3 MODELING AND SIMULATION OF IONIC TRANSPORT PROCESSES THROUGH
IDEAL ION-EXCHANGE MEMBRANE SYSTEMS 95
A. A. Moya
1 Introduction 95
2 Theoretical Description 98
2.1 Ionic Transport in Ideal Ion-Exchange Membrane Systems 98
2.2 Electric Current Perturbations 101
2.3 Analytical Solutions 102
3 The Network Model 106
4 Network Simulation 108
4.1 Transient Response 109
4.2 Electrochemical Impedance 113
5 Conclusion 121
Nomenclature 122
CONTENTS vii
Appendix 123
Acknowledgments 124
References 124
4 MECHANISM OF POTENTIAL DEVELOPMENT FOR POTENTIOMETRIC SENSORS,
BASED ON MODELING OF INTERACTION BETWEEN ELECTROCHEMICALLY
ACTIVE COMPOUNDS FROM THE MEMBRANE AND ANALYTE 131
R.-I. Stefan-van Staden
1 Introduction 131
2 The MembraneSolution Interface 132
3 Membrane Conguration 132
4 New Theoretical Model for Potential Development Based on
Membrane Equilibria 133
5 Mechanism of the Potential Development 134
6 ModelingA Theoretical Approach to Predict the Response and
Mechanism of Potential Development 137
7 Selectivity of Potentiometric Sensors: Explanation through
Membrane Equilibria 149
7.1 Inuence of the Composition of the Membrane on the
Selectivity of Potentiometric Sensors 150
8 Conclusions 151
References 152
5 COMPUTER MODELING OF THE POTENTIOMETRIC RESPONSE OF
ION-SELECTIVE ELECTRODES WITH IONOPHORE-BASED MEMBRANES 155
K. N. Mikhelson
1 Introduction 155
2 Physical Models of Ionophore-Based Membranes 158
2.1 Levels of ISE Membrane Modeling 158
2.2 One-Dimensional Approach to ISE Membrane Modeling 160
2.3 Segmented Model of the ISE Membrane 161
2.4 Integral Model of the ISE Membrane 164
3 Computer Modeling for the Phase Boundary Theory 166
3.1 Description of the ISE Response in Mixed Solutions
Containing Differently Charged Ions 166
3.2 Description of Apparently Non-Nernstian Response Slopes
of Ion-Selective Electrodes 168
viii CONTENTS
4 Modeling Using the Multispecies Approximation 170
4.1 The Essence of the Multispecies Approximation 170
4.2 System of Equations for Implementation of the Multispecies
Model 171
4.3 Selected Results of Modeling Using the Multispecies
Approximation 174
5 Diffusion Layer Model: Example of Local Equilibrium Modeling 179
6 Advanced Nonequilibrium Modeling in Real Time and Space 181
7 Conclusions 194
Acknowledgments 194
References 195
6 MODELS OF RESPONSE IN MIXED-ION SOLUTIONS FOR ION-SENSITIVE
FIELD-EFFECT TRANSISTORS 201
Sergio Bermejo
1 Introduction 201
2 ISFET Basics 202
2.1 Principles of Electrochemical Operation 202
2.2 Structures and Materials 206
3 Electrochemical Models 211
3.1 The MetalSolution Junction 211
3.2 The OxideSolution Junction 219
3.3 Membrane-Based ISFETs 225
3.4 A General Approach for ISFET Modeling in Mixed-Ion Solutions 232
4 Conclusions 242
Appendix: SPICE Models 242
References 243
PART 3: ELECTROCHEMICAL BIOSENSORS
7 NANOMATERIAL-BASED ELECTROCHEMICAL BIOSENSORS 251
N. Jaffrezic-Renault
1 Introduction 251
2 Nanomaterials: Fabrication, Chemical and Physical Properties 252
2.1 Conducting Nanomaterials 252
2.2 Nonconducting Nanomaterials: Magnetic Nanoparticles 254
CONTENTS ix
3 Conception and Modeling of Amplication Effect in
Nanomaterial-Based Enzyme Sensors 255
3.1 AuNPs-Based Amperometric Sensors 255
3.2 CNT-Based Amperometric Sensors 258
3.3 MNP-Based Amperometric Biosensors 262
3.4 Potentiometric Sensors 265
3.5 Conductometric and Impedimetric Biosensors 265
Nanomaterial-Based Immunosensors 267
4.1 AuNP-Based Amperometric Immunosensors 267
4.2 AuNP-Based Potentiometric Sensors 272
4.3 Impedimetric Sensors 273
4.4 Conductometric Sensors 276
Nanomaterial-Based DNA Biosensors 277
5.1 Amperometric Sensors 277
5.2 Impedimetric Sensors 283
6 Conclusion 284
References 285
8 ION-SENSITIVE FIELD-EFFECT TRANSISTORS WITH NANOSTRUCTURED
CHANNELS AND NANOPARTICLE-MODIFIED GATE SURFACES: THEORY,
MODELING, AND ANALYSIS 295
V. K. Khanna
1 Introduction 295
2 Structural Congurations of the Nanoscale ISFET 297
2.1 The Nanoporous Silicon ISFET 297
2.2 The CNT ISFET 298
2.3 The Si-NW ISFET 299
3 Physics of the Si-NW Biosensor 299
3.1 Basic Principle 299
3.2 Analogy with the Nanocantilever 300
3.3 Preliminary Analysis of Micro-ISFET Downscaling to
Nano-ISFET 301
3.4 Single-Gate and Dual-Gate Nanowire Sensors 304
3.5 Energy-Band Model of the NW Sensor 305
4 Nair-Alam Model of Si-NW Biosensors 307
4.1 The Three Regions in the Biosensor 307
4.2 Computational Approach 308
x CONTENTS
4.3 Effect of Nanowire Diameter (d) on Sensitivity at Different
Doping Densities, with Air as the Surrounding Medium 310
4.4 Effect of Nanowire Length (L) on Sensitivity at Different
Doping Densities, with Air as the Surrounding Medium 310
4.5 Effect of the Fluidic Environment 310
4.6 Overall Model Implications 315
5 pH Response of Silicon Nanowires in Terms of the Site-Binding
and Gouy-Chapman-Stern Models 316
6 Subthreshold Regime as the Optimal Sensitivity Regime of
Nanowire Biosensors 321
7 Effective Capacitance Model for Apparent Surpassing of the
Nernst Limit by Sensitivity of the Dual-Gate NW Sensor 324
8 Tunnel Field-Effect Transistor Concept 326
9 Role of Nanoparticles in ISFET Gate Functionalization 328
9.1 Supportive Role of Nanoparticles 328
9.2 Direct Reactant Role of Nanoparticles 330
10 Neuron-CNT (Carbon Nanotube) ISFET Junction Modeling 332
11 Conclusions and Perspectives 334
Dedication 335
Acknowledgments 335
References 335
9 BIOSENSORS: MODELING AND SIMULATION OF DIFFUSION-LIMITED
PROCESSES 339
L. Rajendran
1 Introduction 339
1.1 Enzyme Kinetics 339
1.2 Basic Scheme of Biosensors 340
1.3 The Nonlinear Reaction-Diffusion Equation and Biosensors 340
1.4 Types of Biosensors 342
1.5 Michaelis-Menten Kinetics 343
1.6 NonMichaelis-Menten Kinetics 343
1.7 Importance of Modeling and Simulation of Biosensors 344
2 Modeling of Biosensors 345
2.1 Michaelis-Menten Kinetics and Potentiometric Biosensors 345
2.2 Michaelis-Menten Kinetics and Amperometric Biosensors 346
2.3 Michaelis-Menten Kinetics and Amperometric Biosensors
for Immobilizing Enzymes 348
CONTENTS xi
2.4 Michaelis-Menten Kinetics and the Two-Substrate Model 349
2.5 NonMichaelis-Menten Kinetics 353
2.6 Other Enzyme Reaction Mechanisms 356
2.7 Kinetics of Enzyme Action 361
2.8 Trienzyme Biosensor 362
3 Microdisk Biosensors 363
3.1 Introduction 363
3.2 Mathematical Formulation of the Problem 364
3.3 First-Order Catalytic Kinetics 366
3.4 Zero-Order Catalytic Kinetics 370
3.5 For All Values of K
M
372
3.6 Conclusions 373
4 Microcylinder Biosensors 373
4.1 Introduction 373
4.2 Mathematical Formulation of the Problem 374
4.3 Analytical Solutions of the Concentrations and Current 376
4.4 Comparison with Limiting Case of Rijiravanichs Work 378
4.5 Discussion 379
4.6 Conclusions 381
4.7 PPO-Modied Microcylinder Biosensors 382
5 Spherical Biosensors 383
5.1 Simple Michaelis-Menten and Product Competitive
Inhibition Kinetics 383
5.2 Immobilized Enzyme for Spherical Biosensors 385
5.3 Conclusion 386
Appendix: Various Analytical Schemes for Solving Nonlinear
Reaction Diffusion Equations 386
A. Basic Concept of the Variational Iteration Method 386
B. Basic Concept of the Homotopy Perturbation Method 387
C. Basic Concept of the Homotopy Analysis Method 388
D. Basic Concept of the Adomian Decomposition Method 391
References 392
INDEX 399
xiii
PREFACE
This series, Chemical Sensors: Simulation and Modeling, is the perfect comple-
ment to Momentum Presss six-volume reference series, Chemical Sensors:
Fundamentals of Sensing Materials and Chemical Sensors: Comprehensive Sensor
Technologies, which present detailed information about materials, technologies,
fabrication, and applications of various devices for chemical sensing. Chemical
sensors are integral to the automation of myriad industrial processes and every-
day monitoring of such activities as public safety, engine performance, medical
therapeutics, and many more.
Despite the large number of chemical sensors already on the market, selec-
tion and design of a suitable sensor for a new application is a difcult task for
the design engineer. Careful selection of the sensing material, sensor platform,
technology of synthesis or deposition of sensitive materials, appropriate coatings
and membranes, and the sampling system is very important, because those deci-
sions can determine the specicity, sensitivity, response time, and stability of the
nal device. Selective functionalization of the sensor is also critical to achieving
the required operating parameters. Therefore, in designing a chemical sensor, de-
velopers have to answer the enormous questions related to properties of sensing
materials and their functioning in various environments. This ve-volume com-
prehensive reference work analyzes approaches used for computer simulation and
modeling in various elds of chemical sensing and discusses various phenomena
important for chemical sensing, such as surface diffusion, adsorption, surface
reactions, sintering, conductivity, mass transport, interphase inter actions, etc.
In these volumes it is shown that theoretical modeling and simulation of the pro-
cesses, being a basic for chemical sensor operation, can provide considerable
assistance in choosing both optimal materials and optimal congurations of
sensing elements for use in chemical sensors. The theoretical simulation and
model ing of sensing material behavior during interactions with gases and liquid
surroundings can promote understanding of the nature of effects responsible for
high effectiveness of chemical sensors operation as well. Nevertheless, we have to
understand that only very a few aspects of chemistry can be computed exactly.
xiv PREFACE
However, just as not all spectra are perfectly resolved, often a qualitative or ap-
proximate computation can give useful insight into the chemistry of studied phe-
nomena. For example, the modeling of surface-molecule interactions, which can
lead to changes in the basic properties of sensing materials, can show how these
steps are linked with the macroscopic parameters describing the sensor response.
Using quantum mechanics calculations, it is possible to determine parameters
of the energetic (electronic) levels of the surface, both inherent ones and those
introduced by adsorbed species, adsorption complexes, the precursor state, etc.
Statistical thermodynamics and kinetics can allow one to link those calculated
surface parameters with surface coverage of adsorbed species corresponding to
real experimental conditions (dependent on temperature, pressure, etc.). Finally,
phenomenological modeling can tie together theoretically calculated characteris-
tics with real sensor parameters. This modeling may include modeling of hot plat-
forms, modern approaches to the study of sensing effects, modeling of processes
responsible for chemical sensing, phenomenological modeling of operating char-
acteristics of chemical sensors, etc.. In addition, it is necessary to recognize that
in many cases researchers are in urgent need of theory, since many experimental
observations, particularly in such elds as optical and electron spectroscopy, can
hardly be interpreted correctly without applying detailed theoretical calculations.
Each modeling and simulation volume in the present series reviews model-
ing principles and approaches particular to specic groups of materials and de-
vices applied for chemical sensing. Volume 1: Microstructural Characterization and
Modeling of Metal Oxides covers microstructural characterization using scanning
electron microscopy (SEM), transmission electron spectroscopy (TEM), Raman
spectroscopy, in-situ high-temperature SEM, and multiscale atomistic simulation
and modeling of metal oxides, including surface state, stability, and metal oxide
interactions with gas molecules, water, and metals. Volume 2: Conductometric-
Type Sensors covers phenomenological modeling and computational design of
conductometric chemical sensors based on nanostructured materials such as
metal oxides, carbon nanotubes, and graphenes. This volume includes an over-
view of the approaches used to quantitatively evaluate characteristics of sensitive
structures in which electric charge transport depends on the interaction between
the surfaces of the structures and chemical compounds in the surroundings.
Volume 3: Solid-State Devices covers phenomenological and molecular model-
ing of processes which control sensing characteristics and parameters of various
solid-state chemical sensors, including surface acoustic wave, metal-insulator-
semiconductor (MIS), microcantilever, thermoelectric-based devices, and sensor
arrays intended for electronic nose design. Modeling of nanomaterials and nano-
systems that show promise for solid-state chemical sensor design is analyzed as
well. Volume 4: Optical Sensors covers approaches used for modeling and simu-
lation of various types of optical sensors such as ber optic, surface plasmon
resonance, Fabry-Prot interferometers, transmittance in the mid-infrared region,
PREFACE xv
luminescence-based devices, etc. Approaches used for design and optimization
of optical systems aimed for both remote gas sensing and gas analysis cham-
bers for the nondispersive infrared (NDIR) spectral range are discussed as well.
A description of multiscale atomistic simulation of hierarchical nanostructured
materials for optical chemical sensing is also included in this volume. Volume 5:
Electrochemical Sensors covers modeling and simulation of electrochemical pro-
cesses in both solid and liquid electrolytes, including charge separation and
transport (gas diffusion, ion diffusion) in membranes, protonelectron transfers,
electrode reactions, etc. Various models used to describe electrochemical sensors
such as potentiometric, amperometric, conductometric, impedimetric, and ion-
sensitive FET sensors are discussed as well.
I believe that this series will be of interest of all who work or plan to work in
the eld of chemical sensor design. The chapters in this series have been prepared
by well-known persons with high qualication in their elds and therefore should
be a signicant and insightful source of valuable information for engineers and
researchers who are either entering these elds for the rst time, or who are al-
ready conducting research in these areas but wish to extend their knowledge in
the eld of chemical sensors and computational chemistry. This series will also be
interesting for university students, post-docs, and professors in material science,
analytical chemistry, computational chemistry, physics of semiconductor devices,
chemical engineering, etc. I believe that all of them will nd useful information in
these volumes.
G. Korotcenkov
xvii
ABOUT THE EDITOR
Ghenadii Korotcenkov received his Ph.D. in Physics and Technology of
Semiconductor Materials and Devices in 1976, and his Habilitate Degree (Dr.
Sci.) in Physics and Mathematics of Semiconductors and Dielectrics in 1990. For
a long time he was a leader of the scientic Gas Sensor Group and manager of
various national and international scientic and engineering projects carried out
in the Laboratory of Micro- and Optoelectronics, Technical University of Moldova.
Currently, Dr. Korotcenkov is a research professor at the Gwangju Institute of
Science and Technology, Republic of Korea.
Specialists from the former Soviet Union know Dr. Korotcenkovs research
results in the eld of study of Schottky barriers, MOS structures, native oxides, and
photoreceivers based on Group IIIV compounds
very well. His current research interests include
materials science and surface science, focused on
nanostructured metal oxides and solid-state gas
sensor design. Dr. Korotcenkov is the author or
editor of 11 books and special issues, 11 invited
review papers, 17 book chapters, and more than
190 peer-reviewed articles. He holds 18 patents,
and he has presented more than 200 reports at
national and international conferences.
Dr. Korotcenkovs research activities have
been honored by an Award of the Supreme
Council of Science and Advanced Technology
of the Republic of Moldova (2004), The Prize of
the Presidents of the Ukrainian, Belarus, and
Moldovan Academies of Sciences (2003), Senior
Research Excellence Awards from the Technical
University of Moldova (2001, 2003, 2005), a
fellowship from the International Research Exchange Board (1998), and the
National Youth Prize of the Republic of Moldova (1980), among others.
xix
CONTRIBUTORS
Nikolai F. Uvarov (Chapter 1)
Institute of Solid State Chemistry and Mechanochemistry
Siberian Branch of the Russian Academy of Sciences
Novosibirsk 630128, Russia
Chongook Park (Chapter 2)
Department of Materials Science & Engineering
KAIST
Dae-jeon 305-701, South Korea
Inkun Lee (Chapter 2)
KAIST
Dearo Lee (Chapter 2)
KAIST
Jeffrey Fergus (Chapter 2)
Materials Research and Education Center
Auburn University
Auburn, Alabama 36849-5341, USA
Norio Miura (Chapter 2)
Art, Science and Technology Center for Cooperative Research
Kyushu University
Fukuoka 816-8580, Japan
xx CONTRIBUTORS
Hyungjun Yoo (Chapter 2)
Department of Electrical Engineering
KAIST
Antonio ngel Moya Molina (Chapter 3)
Departamento de Fsica
Universidad de Jan, Campus de las Lagunillas
Jan 23071, Spain
Raluca-Ioana Stefan-van Staden (Chapter 4)
Laboratory of Electrochemistry and PATLAB Bucharest
National Institute of Research for Electrochemistry and Condensed Matter
Bucharest 060021, Romania
Konstantin N. Mikhelson (Chapter 5)
Ionometry Laboratory, Chemical Faculty
St. Petersburg State University
St. Petersburg, Russia
Sergio Bermejo (Chapter 6)
Department of Electronic Engineering
Universitat Politcnica de Catalunya (UPC)
Barcelona 08034, Spain
Nicole Jaffrezic-Renault (Chapter 7)
Institute of Analytical Chemistry, UMR CNRS 5280
Claude Bernard University Lyon 1
Villeurbanne 69100, France
Vinod Kumar Khanna (Chapter 8)
MEMS & Microsensors
CSIRCentral Electronics Engineering Research Institute
Pilani 333031 (Rajasthan), India
L. Rajendran (Chapter 9)
Department of Mathematics
The Madura College (Autonomous)
Madurai 625011, Tamil Nadu, South India
PART 1
SOLID-STATE ELECTROCHEMICAL SENSORS
3 DOI: 10.5643/9781606505984/ch1
CHAPTER 1
SURFACE AND INTERFACE DEFECTS
IN IONIC CRYSTALS
N. F. Uvarov
1. INTRODUCTION
1.1. SOLID ELECTROLYTES AND ELECTRODES FOR
ELECTROCHEMICAL SENSORS: A BRIEF OVERWIEW
Among all known gas sensors, electrochemical sensors are the most compact,
economical, and suitable for developing gas-sensing devices. They have wide
appli cation for control of air quality and impurity concentration in biotechnology,
medical, brewing technologies, and in various industrial, chemical, and biologi-
cal processes. The action of electrochemical sensors is based on clearly known
fundamental principles, and they are reliable and easily calibrated. Solid-state
electrochemical gas sensors exhibit some outstanding properties which makes
them of special ineterst for applications. A prime example is yttria-stabilized zir-
conia potentiometric sensors used to control the air-to-fuel ratio in automobile
engines. These sensors contain all-solid-state electrochemical components; they
are extremely robust and operate for a long time even after exposure to tempera-
tures as high as 1000C. There are two main types of electrochemical gas sensors:
potentiometric and amperometric ones. The rst type operate as concentration
or chemical cells; with the second type one measures the electrical current value
4 - CHEMICAL SENSORS SIMULATION AND MODELING: VOLUME 5
through the cell as a function of the concentration of the gas to be analyzed. In all
cases the electrochemical cell includes two metallic or semiconducting sensitive
electrodes and a solid electrolyte in between.
Solid electrolytes are necessary components of electrochemical sensors. Solid
electrolytes with conductivity higher than 10
3
S/cm are classied as superionic
conductors. By conductivity level, superionic conductors are close to molten
salts. However, in contrast to the latter, they have unipolar conductivity. The
high ionic conductivity of superionic conductors is due to a high concentration
of point defects, vacancies or interstitial ions, in one of the sublattices of the
ionic compound. The defects may be structurally ordered or distributed randomly,
resulting in sublattice melting. At present there are known many superionic
conductors which exhibit conductivity via various cations (H
+
, Li
+
, Na
+
, Ag
+
) or
anions (F
, O
2
, S
2
) (Hangenmuller and Van Gool 1978; Chebotin and Perliev
1978; Salamon 1979; Vashishta et al. 1979; Chandra 1981; Takahashi 1989;
Hull 2004; Thangadurai and Weppner 2006; Ishihara 2009). A general strategy to
create new superionic compounds and to control their transport properties is to
search for suitable host phases and then dope them with aliovalent ions in order
to increase the concentration of extrinsic defects. This approach traditional works
fairly for some host crystal lattices which are capable of carrying solute at tens
of mole percent defects without global structural reconstruction. As a rule, such
compounds have structurally disordered high-temperature phases, and doping
allows one to stabilize them at low temperatures. Examples of such structures
are o-AgI, -Li
3
PO
4
, o-Li
4
SiO
4
, perovskites, uorites, tysonites, pyrochlores, and
structures related to them (Hull 2004). The most widely used solid oxygen-ion
conductor, yttria-stabilized zirconia (YSZ), has a uorite structure containing
510 mol% anionic vacancies in the oxygen sublattice. The situation with proton
solid conductors is more complicated. High-temperature proton conductors (with
the range of high conductivity of 7001000 K) may be obtained by dissolution
of water or hydrogen in some acceptor-doped perovskite-related oxides contain-
ing oxygen vacancies, e.g., BaCeO
3
, BaZrO
3
, etc. (Kreuer 2003; Norby 2009). In
low-temperature proton conductors (stable below 400 K), conductivity is due to
the presence of hydrate water in the crystal structure. There is a special class
of intermediate-temperature proton conductors including acid salts such as
CsHSO
4
, CsH
2
PO
4
, etc., which have high proton conductivity in high-temperature
phases (at 400500 K) with dynamical disorder of hydrogen bonds in the anionic
sublattice (Haile 2001; Ponomareva 2011).
The choice of solid electrolyte for use in the sensor is determined by a particu-
lar mechanism of electrode processes taking place on the electrodes. The main
types of electrode processes which may proceed in gas sensors are reported else-
where (Fabry and Siebert 1997; Park et al. 2003; Bhoga and Singh 2007). For sta-
ble operation of the sensor, it is preferable if gas species to be detected participate
in an electrode reaction with an ionic species presenting in a solid electrolyte. As
SURFACE AND INTERFACE DEFECTS IN IONIC CRYSTALS - 5
a result, in potentiometric sensors, equilibrium is established between gas species
and ionic species. In amperometric sensors, the electrode reactions proceed in
stationary mode, providing steady current from one electrode to another through
the solid electrolyte. For example, oxygen-containing molecules can be easily oxi-
dized (hydrocarbons, CO, N
2
O, NO, SO
2
, etc.) or reduced (O
2
, H
2
O, CO
2
, N
2
O, NO,
etc.) electrochemically on electrodes in the cell with solid oxide electrolytes. These
reactions are complex and include several preliminary stages such as adsorption,
dissociation, diffusion, or formation of intermediates (Chebotin and Perliev 1978;
Verkerk et al. 1983; Adler 2004; Sunarso et al. 2008). This is why the kinetics and
specic mechanism of summary electrode reactions depend mostly on the chemi-
cal nature and morphology of the electrode material.
To control gas-sensing properties of electrochemical sensors, semiconduct-
ing oxide electrodes may be used instead of traditional metallic electrodes such
as Pt or Pd. Using standard approaches of defects chemistry, one can control the
nonstoichiometry of oxides, dominating related physical and chemical properties,
including electrical conductivity and catalytic activity (Krger 1964; Schoonman
1997; Tuller 2003). It is important that the oxide electrode have high ionic con-
ductivity together with electronic conductivity. The use of such mixed electronic-
ionic conductors (MIECs) as electrodes leads to a strong extension of the triple
electrode boundary (gaselectrodeelectrolyte) up to full surface of solid electro-
lyte. This results in a strong decrease in the specic surface resistance of the
electrode and, in the case of a solid oxide fuel cell, enhances the specic current
and power values of the cell. In the case of gas sensors, the use of MIECs offers a
possibility to vary the catalytic properties of the electrode material (Adler 2004),
control the rate of a particular chemical stage of the process, and achieve higher
selectivity of the sensors toward detection of a given gas.
Solid electrolytes and electrode materials are usually prepared in polycrystal-
line form and contain a large number of grain boundaries. These planar defects
can be electrically active, resulting in the blocking of charge carriers. This ef-
fect is attributed to depletion of charge carriers resulting from the compensation
of electrically charged grain boundary cores by adjacent space charge regions
of opposite charge. Depending on the charge of the grain boundary core, defect
concentrations of one type can be dramatically depressed, whereas others are
dramatically enhanced. Solid electrolytes should be sintered at high tempera-
tures to diminish the contribution of grain boundaries to overall resistivity of
the electrolyte. In nanocrystalline solids, the grain size is less than the space
charge length. This can lead to size effects, when properties of the solid differ dra-
matically from those exhibited by materials with identical composition but with
larger grains or wider lm spacings. Similarly, thin lms are susceptible to the
chemi sorption of gaseous species and the consequent depletion or accumulation
of charge carriers in the lm. Such phenomena are important for the design and
operation of semiconducting oxide gas sensors. Thus, the inuence of surface-
and interface- related defects on the transport properties of ionic conductors and
oxide-conducting materials is an area of growing interest.
1.2. SURFACE AND INTERFACE PROPERTIES OF IONIC SOLIDS
Surface or interfacial phenomena play an important role in solid-state ionics.
Space charge regions formed by point defects and impurity ions localized on the
crystal surfaces or grain boundaries strongly inuence the transport properties
of real ceramic materials and contribute to the electrode polarization. The sur-
face potential value depends on adsorption of the gaseous species and therefore
strongly inuences the sensitivity and selectivity of gas sensors.
According to the model of Frenkel and Kliever (FK model) (Frenkel 1946;
Grimley 1950; Kliever and Koehler 1965a, 1965b), the surface potential, a key
parameter dening space charge properties in pure ionic crystals, is determined
mainly by the difference in formation energies of the individual defects in the bulk
of the crystal. Defects having higher energy accumulate on the surface, whereas
the oppositely charged ones form the space charge inside the crystal near the sur-
face. Impurities of bivalent cations inuence the surface potential of ionic crystals,
and in some cases an isoelectric point exists (Kliever and Koehler 1965a, 1965b),
i.e., a temperature at which the surface potential changes sign. The existence of
isoelectric points in doped alkali halide crystals has been experimentally proved
by studies of dislocation charge (Urusovskaya 1969; Whitworth 1975; Tallon et
al. 1985). Although the FK model is relatively simple and representative, it has an
essential disadvantage: This model includes only bulk parameters and does not
take into account explicitly the specic characteristics of the surface. Attempts to
solve this problem have been made by Poeppel and Blakely (the PB model) (Poeppel
and Blakely 1969; Blakely and Daniluk 1973), who have studied the inuence of
a limited number of surface sites on the surface potential. Macdonald et al. (1980)
substantially improved the model by taking into consideration statistics of surface
defects and calculating the surface potential and double-layer capacitance in AgCl.
It was shown that the thermodynamic equilibrium of surface defects in terms of
the lattice gas model leads to Langmuir adsorption of the defects on the surface.
On the other hand, an alternative approach has been proposed by Lifschitz
and Geguzin, who postulated that the surface potential is determined only by
parameters of species located on the surface of the crystal (Lifschits et al. 1967).
This approach was extended by Chebotin et al. (1984) by consideration of the
elementary mechanism of surface disordering. In fact, both these models are vari-
ants of the Stern model of the electrochemical double layer on electrodes in con-
tact with liquid electrolytes (Stern 1924).
In papers of Maier (1987, 1995) the space charge model was represented in
terms of level diagrams for standard chemical and electrochemical potentials of
point defects in both the bulk and surface, including the case of composite solid
electrolytes. More sophisticated approaches (i.e., a combination of the FK and
Stern models) have been considered by Jamnik et al. (1995) and Khaneft et al.
(Khaneft et al. 1990a, 1990b; Khaneft 1992); their models take into account the
defect chemistry in the bulk and the surface core. It was again shown that the
space charge potential is determined by the difference between the core and bulk
standard chemical potentials of individual defects.
Space charge regions localized on the crystal surfaces or grain boundaries
strongly inuence the transport properties of real ceramic materials. In superionic
oxides of the M
IV
1-c
Me
III
c
O
2-c/2
type the grain boundaries are known to have strong
blocking effect on conductivity. The effect is caused by the depletion of the grain-
boundary space charge layer (SCL) in anionic vacancies due to the positive sign of
the surface potential on the grain boundary (Heyne 1983; Maier 1986; Burgraaf
and Winnbust 1988; Tschppe 2001a; Guo et al. 2002, 2003; Kim and Maier 2003;
Guo and Wasser 2006). The value of the surface potential at intergrain boundar-
ies as estimated using the Mott-Schottky model from the conductivity data varies
from 0.20.3 V for Zr
1-x
Y
x
O
2-x/2
(Guo et al. 2002; Guo and Wasser 2006) to 0.30.9
V for heavily doped CeO
2
(Tschppe 2001; Tschppe et al. 2001, 2004; Kim and
Maier 2003; Guo et al. 2003). The barrier height is determined by the value of
the electrical potential on the grain boundary, which is an analog of the surface
potential. In contrast to classical ionic crystals, in superionic oxides the defect
formation energies are close to zero (Goodenough 2003), while the concentrations
of impurities and oxygen vacancies are very high. In this case, the conventional
FK and PB models are not effective, and only the Stern model is applicable.
Interface defects play an important role in transport properties of composite
solid electrolytes, a promising new class of ionic conductors with high ionic con-
ductivity. The combination of high conductivity with the enhanced mechanical
strength together with the wide prospects for the purposeful modication of the
electrolyte properties by varying the type and concentration of the dopant makes
these composites promising materials for real electrochemical systems. Since the
rst report on the effect of heterogeneous doping on ionic conductivity (Liang
1973), many reviews have been published devoted to the description and the anal-
ysis of the ion transport in polycrystalline and composite solid electrolytes (Shahi
and Wagner 1981; Wagner 1985, 1989; Maier 1985, 1987, 1989a, 1995, 2002,
2003, 2004, 2005; Khandkar and Wagner 1986; Chen 1986; Shukla et al. 1986;
Dudney 1989; Uvarov et al. 1992; Uvarov 1996, 2007a, 2008a, 2011; Agrawal
and Gupta 1999; Yarostalvtsev 2000; Uvarov and Vanek 2000; Jamnik and Maier
2003; Heitjans and Indris 2003; Schoonman 2005). The increase in the ionic con-
ductivity upon heterogeneous doping can be explained within the framework of
the space charge model proposed by Wagner and Maier (Jow and Wagner 1979;
Maier 1985). This model allows the interpretation of many phenomena observed
in composites and is the best suited for the explanation of experimental data for
composites containing oxides with relatively coarse grains. However, the space
charge model in its classical version is correct only for ideal crystals in contact
with vacuum or a structure-free medium and obviously ignores the real features
of the interphase contact, namely, changes in the structures of ionic crystals
(e.g., for epitaxial contacts), the effect of elastic strains, the formation of disloca-
tions, etc. Moreover, if the surface concentration of defects is sufciently high,
it is impossible to ignore the interaction between the defects, which results in
their ordering and the formation of superstructures and even metastable surface
phases. It is known that the conductivity of composites increases as the size of
dopant particles increases. Hence, composites with nanosized grains (about 10
nm) are of particular interest for practice. Obviously, uniform mixing of such an
oxide with an ionic component should produce a nanocomposite the properties of
which depend strongly on the energy of surface interaction and the peculiarities
of the interface between the phases. For composites with coarse-grained addi-
tives, the presence of surfaces or interphase contacts has virtually no effect on
the bulk properties of the ionic salt; hence, the increase in the conductivity is of
purely surface nature. However, in many cases, it still remains unclear whether
the enhanced conductivity is caused primarily by the specic interactions at the
interface or by the trivial increase in the surface conductivity as such. To answer
this question, information on the conductivity of polycrystals is necessary. In
nanocomposites, virtually all volume of the ionic salt is located at the interface,
leading to formation of disordered interface-stabilized phases not inherent to the
pure salt.
The aim of this chapter is to review surface and interface properties of ionic
crystals in terms of the surface/interface defects and quasi-chemical mechanisms
of their formation. For this purpose a classic variant of the Stern model is used
for the case of strong adsorption of point defects on the surface for ordinary ionic
crystals of the NaCl type (Uvarov 2007b, 2008a) and superionic oxides (Uvarov
2007b, 2008b, 2008c). Surface- and interface-related properties of different ionic
systems are also analyzed, with an emphasis on size effects in nanocomposites.
2. CALCULATION OF THE SURFACE POTENTIAL AND SURFACE
DEFECTS USING THE STERN MODEL
2.1. DESCRIPTION OF THE MODEL
Due to the lattice distortion associated with the assymetric eld at the surface,
all characteristics of point defects and impurity ions located on the surface dif-
fer from those in the bulk. The difference is the reason for the specic adsorption
(positive or negative) of defects at the crystal surface. If the adsorption energies
Ag
i
of oppositely charged defects are different, then an excess number of defects
that have the most negative value of Ag
i
appear on the surface. Defects of the op-
posite charge form a diffuse layer, or space charge layer (SCL), under the surface.
According to the Stern model, the surface charge Q
S
is determined by the sum of
contributions made by all defects adsorbed on the surface. In the simplest case of
the Langmuir adsorption isotherm, the surface charge is given by the expression
(Stern 1924)

1
,
,
1 exp
i i S
S i S i
i
i
N g q
Q q N
n kT
, ]
+ j \
+
, ] , (
( ,
, ]
]
(1.1)
where N
S
is the concentration of available surface sites (adsorption centers), q
is the effective charge of the defect, n
is the concentration of the defects in the

bulk of the crystal, N is the total number of sites of the crystal lattice, and the
summation is done over all possible charged defects of i type. As follows from
Eq. (1.1), the surface charge is determined by both the bulk properties (n
,i
) and
surface- related parameters [the adsorption energies Ag
i
and the surface potential
S
, which is approximately equivalent to the potential at the inner Helmholtz layer
(IHL) in electrochemistry]. The surface charge is balanced by the charge of the
SCL, Q
d
, formed under the crystal surface. If the probability of nding an ion at a
particular point depends on the local potential through a Boltzmann distribution,
n
i
(x) = n
i,
exp[q(x)/kT], then the charge density distribution and the potential
gradient must satisfy the Poisson-Boltzman equation. The solution of this equa-
tion is the Gouy-Chapman formula for the charge of the SCL,

1/2
1
,
exp 1
i S
d i
i
q
Q A n
kT

, ] j \

, ( , ]
( ,
]

(1.2)
where A = (2cc
0
kT )
1/2
;
S1
is the potential of the outer Helmholtz layer (OHL).
In superionic conductors, due to the high concentration of charged defects,
the thickness of the SCL is very small, of the order of 13 nm. This strongly re-
stricts the applicability of the Gouy-Chapman model. In a special case when the
concentration of the impurity ions is constant throughout the crystal up to the
OHL, the Mott-Schottky approximation may be used and the charge of the SCL is
given by (Gurevich and Pleskov 1983; Wasser 1995)

( ) ( )
1/2 1/2
0 1
2
d c S
Q q n

(1.3)
The characteristic length of the SCL estimated by the Mott-Schottky model is
several times larger than for the Gouy-Chapman case.
Thus, both charges, i.e., Q
S
and Q
d
, are dened by the surface potential. One
can determine its value from the condition of overall neutrality, Q
S
+ Q
d
= 0, i.e.,
by solving the transcendental equation derived by summing the right-hand parts
of (1.1) and (1.2) or (1.3).
2.2. PURE CRYSTALS OF THE NaCl TYPE
First, let us consider the case of an ideally pure crystal of the NaCl type. As follows
from Eqs. (1.2) and (1.3), the surface potential depends on temperature, concen-
tration of the surface sites N
S
, partial energies of adsorption Ag
i
, and the concen-
tration of the i defects in the bulk of the crystal, x
i
. The equilibrium concentration,
n
0
, of Schottky defects in pure MX is given by Arrhenius dependence,

0
0 0
exp
2
g
n n n N
kT
+
j \

, (
( ,
(1.4)
where n
+
and n
are the concentrations of the anion and cation vacancies, respec-

tively; N
0
is the total volume concentration of MX molecules; g
0
= h
0
TS
0
is the
Gibbs energy of the formation of the Schottky defect determined by the defect
formation enthalpy h
0
and the entropy S
0
. From Eqs. (1.2), (1.3), and (1.4) one can
express the Stern equation for a NaCl-type crystal as follows:

1 1
0 0
1 exp 1 exp
2 2
S S
S
g g e g g e
e N
kT kT kT kT

+
, ] , ] j \ j \ +
+ + + +
, ] , ]
, ( , (
( , ( ,
, ] , ]
] ]

0
0
exp sh
4 2
S
g e
A N
kT kT
j \ j \

, ( , (
( , ( ,
(1.5)
Here Ag
+
and Ag
are values of the adsorption energy for anion vacancy and

cation vacancy, respectively, and sh is the hyperbolic sine. General analysis of
this equation shows that the character of the
S
(T) curves essentially depends
on the values of Ag
i
, g
0
, and N
S
. If adsorption energies are positive (Ag
> 0, Ag
+

> 0), then the surface is depleted by defects and practically uncharged, hence
the surface potential is close to zero. In the case of negative adsorption energies
(Ag
< 0, Ag
+
< 0), both cationic and anionic vacancies are accumulated on the
surface, resulting in formation of negative (at Ag
> Ag
+
) or positive (at Ag

< Ag
+
) surface charge. For simplicity of data presentation, all entropy terms in
g
0
, Ag
+
, and Ag
are taken equal to zero. Theoretical temperature dependencies of

the surface potential are plotted in Figures 1.1a1.1d. The curves were obtained
at different values of Ag
+
< 0 and N
S
; g
0
and Ag
were taken to be equal to 1.0 and

0.1 eV, respectively.
Functions
S
(T) for various N
S
calculated at a low value of the adsorption en-
ergy, Ag
+
= 0.2 eV, are presented in Figure 1.1a. Analysis of the data shows that
in this case the surface potential tends to zero at T 0 and increases monotoni-
cally with temperature. The temperature dependence
S
(T) of this type cannot be
interpreted in terms of the FK and PB models (Kliever and Koehler 1965a; Poeppel
and Blakely 1969), but it is well approximated by the following expression ob-
tained from Eq. (1.5) for small potentials e
S
/kT << 1:

0
0
0
2
exp exp exp
4
S S
g kT g g
N N
AN kT kT kT
+
, ] j \ j \ j \

, ] , (
, ( , (
( ,
( , ( ,
]
(1.6)
A similar equation was reported earlier by Lifshitz and Geguzin (Lifschits et al.
1967). It is seen that the surface potential is determined not only by the difference
Figure 1.1. Dependence of the surface potential on temperature as a function of g
+
and N
S
at
constant g
0
= 1.0 eV and g
= 0.1 eV. Graphs (a), (b), (c), and (d) correspond to values of g
+

equal to 0.2, 0.3, 0.4, and 0.6 eV, respectively. Theoretical curves (1), (2), (3), (4), (5), (6), and
(7) are calculated at N
S
values of 1 10
12
, 3 10
12
, 1 10
13
, 3 10
13
, 1 10
14
, 3 10
14
, and 1
10
15
cm
2
, respectively. (Reprinted with permission from Uvarov 2007b. Copyright 2007 Pleiades
Publishing, Ltd.)
between Ag
+
and Ag
but also depends on the defect formation energy g

0
. The
value of
S
increases linearly with N
S
and proportionally to the square root of the
concentration of Schottky defects in the bulk of the crystal. Thus at low
S
the
process of the formation of the surface charge is limited not only by the concentra-
tion of available surface sites, but also by the defect density in the diffuse layer.
At sufciently high values of the adsorption energy, Ag
+
> g
0
/4 (Ag
+
< 0),
the shape of the
S
(T) curves changes substantially. As seen from Figures 1.1b,
1.1c, and 1.1d, in this case, at low temperatures
S
tends to a limiting value
S
*
which does not depend on N
S
. At high values of Ag
+
and Ag
+
> Ag
, anion
vacancies prevail in the surface adsorption layer, and Eq. (1.5) may be rewritten
in the form

0 0
0
exp exp exp
2 4 2
S S
S
g g e g e
e N A N
kT kT kT kT
+
j \ + j \ j \

, ( , (
, (
( , ( ,
( ,
(1.7)
From this equation one nds that at T 0,
S

S
*, where the value of
S
* is
given by

0
2
*
3 4
S
g
e g
+
j \

, (
( ,
(1.8)
For example, in the case of Ag
+
= 0.4 eV, Ag
= 0.1 eV, and g

0
= 1.0 eV, the value
of
S
* is equal to 0.10 V (Figure 1.1c). The surface potential may increase or de-
crease with the temperature as a function of Ag
+
and N
S
values. At very high values
of Ag
+
, temperature dependencies of the surface potential (this situation is il-
lustrated in Figure 1.1d for Ag
+
= 0.6 eV) resemble ones obtained by PB (Poeppel
and Blakely 1969), who have shown that at low N
S
the number of surface sites
becomes a limiting factor and
S
falls to zero with temperature, whereas at high
N
S
the function
S
(T) tends to a high-temperature limit given by Eq. (1). A similar
limiting value obtained in frames of the Stern model is equal to

S
= (Ag
Ag
+
)/2e (1.9)
This expression becomes identical to the equation proposed by Frenkel and Kliever
(Kliever and Koehler 1965a). Thus, at sufciently high values of Ag
+
and N
S
, the
FK, PB, and Stern models lead to qualitatively the same result.
In both cases of low and high adsorption energies [Eqs. (1.6) and (1.9), respec-
tively], the surface potential is determined by the difference between the partial
adsorption energies of the oppositely charged defects. The dependencies
S
= f(Ag

Ag
+
) obtained by the variation of Ag
+
at xed values of Ag
are shown in Figures

1.2a and 1.2b for values of g
0
= 1.0 and 2.0 eV, respectively. The general character
of the
S
curves does not depend on the absolute value of g
0
but is determined
by the Ag
i
/g
0
ratio and temperature. From this plot one can see that the surface
potential values approximate to ones described by Eq. (1.9) at extreme values of
N
S
10
15
cm
2
and at sufciently high values of Ag
Ag
+
.
Similar analysis may be done for the case when the surface is enriched in
cationic vacancies (Ag
+
, Ag
< 0; Ag
> Ag
+
). In this case the value of the sur-
face potential is negative and the term Ag
+
in Eq. (1.8) is be substituted for Ag
.
2.3. SURFACE POTENTIAL IN NaCl CRYSTALS CONTAINING
DIVALENT CATIONS
Concentration of the defects in an MX crystal containing divalent cations may be
calculated on the basis of a classical approach (Krger 1964) from known values
of the molar fraction of the impurity, c, the formation energy of Schottky defects,
g
0
, and the Gibbs energy, g
ass
, of the formation of dipole complexes [V
M
M
Na
]
x
. In
order to obtain realistic values of the surface potentials and compare our results
Figure 1.2. Dependence of the surface potential
S
on the value of (g
g
+
)/2e for g
0
= 1.0 eV (a)
and g
0
= 2.0 eV (b) at T = 500 K. Theoretical curves (1), (2), (3), (4), and (5) are calculated at N
S
values
of 1 10
11
, 1 10
12
, 1 10
13
, 1 10
14
, and 1 10
15
cm
2
, respectively. Solid lines are calculated with
Eq. (9). (Reprinted with permission from Uvarov 2007b. Copyright 2007 Pleiades Publishing, Ltd.)
with those obtained earlier, g
0
and g
ass
are taken in the form g
0
= 2.12 6.4keV,
g
ass
= 0.4 eV (Kliever and Koehler 1965a). The surface of pure NaCl is known to
be charged positively (Urusovskaya 1969; Tallon et al. 1985), however, no reliable
experimental data on
S
in NaCl are available in the literature. In this work we
have used in calculations the values Ag
+
= 0.2 eV and Ag
= 0.51 eV; the energy

of the impurity segregation was taken to be zero. Temperature dependencies of
S

calculated with the above parameters as a function of N
S
and c are presented in
Figures 1.3a and 1.3b. As one can see, the presence of the impurity atoms has a
dramatic effect on the surface potential. On the temperature dependencies
S
(T)
there are isoelectric points, T
e
, where
S
changes sign. The value of T
e
does not
depend on the concentration of the surface sites N
S
(Figure 1.3a) and is dened by
the molar fraction of the impurity (Figure 1.3b). Similar effects have been reported
earlier (Kliever and Koehler 1965; Kliever 1965; Tallon et al. 1985) on the basis of
the FK model. It should be noted that in NaCl the effect of the impurity becomes
appreciable at extremely small (less then 0.01 ppm) concentration of impurity.
Figure 1.3. Temperature dependence of
S
in NaCl doped with bivalent impurities obtained from
Stern model at g
+
= 0.2 eV; g
= 0.51 eV: (a) at constant concentration of dopant c = 1 ppm and

different concentrations of surface sites at N
S
: 10
11
, 10
12
, 10
13
, 10
14
, and 10
15
cm
2
, curves 1, 2, 3, 4,
and 5, respectively; (b) at constant concentration of surface sites N
S
= 10
15
cm
2
and different concen-
trations of the dopant: 0.01, 0.1, 1, 10, and 100 ppm, curves 1, 2, 3, 4, and 5, respectively. (Reprinted
with permission from Uvarov 2007b. Copyright 2007 Pleiades Publishing, Ltd.)
2.4. COMPARISON WITH EXPERIMENTAL DATA
At present, reliable data on the surface potential are available only for the silver ha-
lides AgCl and AgBr (Wonnel and Slifkin 1995). The FK model has been employed
for their interpretation, and the following values of defect formation energies for
interstitial cations (g
0
+
) and cationic vacancy (g
0
) have been obtained (in eV):

AgCl: g
0
+
= 1.34 22.0kT g
0
= 0.13 + 12.2kT (1.10a)
AgBr: g
0
+
= 0.86 14.5kT g
0
= 0.30 + 7.2kT (1.10b)
(additional small kTln 2 terms are omitted). The reported values of the forma-
tion entropy for interstitial cations seem to be exceptionally high, whereas the
corresponding values for cationic vacancies are negative, which is not typical for
vacancies in solids. The Stern model was then applied for the interpretation of the
experimental data mentioned above. The calculated curves were obtained using
Eqs. (1.1) and (1.2). The concentration of Frenkel defects was estimated using data
reported in the literature (Poeppel and Blakely 1969; Blakely and Daniluk 1973;
Wonnel and Slifkin 1995): g
0
= 1.44 9.7keV, g
ass
= 0.2 eV for AgCl and g
0
= 1.10
7.6keV, g
ass
= 0.4 eV for AgBr; the concentration of bivalent metal impurities in
the both cases was taken as 100 ppm and N
S
was ranged within 10
13
10
15
cm
2
.
In all our calculations we have neglected the inuence of the adsorption energy of
the impurity, i.e., the energy of segregation.
The results of the calculations are illustrated in Figure 1.4; theoretical curves
were obtained for different N
S
values using the following parameters:
AgCl: Ag
= 0.18 6kT Ag
+
0 + 2kT (1.11a)
AgBr: Ag
= 0.05 10kT Ag
+
0 + 0kT (1.11b)
The analysis of the data shows that the concentration of the surface sites is nearly
10
14
cm
2
. The surface potential in silver halides is determined mainly by the ad-
sorption energies of cationic vacancies, Ag
. These values may be evaluated with

more or less appropriate accuracy, whereas values of Ag
+
may be estimated only
roughly. Nevertheless, obtained entropy values of 610 kT seem to be more reli-
able than those estimated by the FK model.
2.5. SURFACE POTENTIAL AND CONCENTRATION OF POINT
DEFECTS ON GRAIN BOUNDARIES OF SUPERIONIC
OXIDE CERAMICS
In contrast to ordinary ionic crystals, the concentration of defects in superionic
conductors is high, comparable to the total number of ions. In this case the value
of defect formation energy is assumed to be zero and the oxide M
IV
1-c
Me
III
c
O
2-c/2

may be regarded as extrinsic ion conductor. Oxygen vacancies V
O
form as a re-
sult of the dissolution of the Me
III
2
O
3
in the M
IV
O
2
matrix to compensate for the
effective negative charge of Me
III
cations in the crystal lattice:
Me
III
2
O
3
2Me
M
+ V
O
+ 3O
O
x
(1.12)
With decreasing temperature, defects may undergo association into complexes.
The association effects have been reported to be responsible for the change in
the slope of the Arrhenius plot for conductivity in superionic oxygen conductors
(Goodenough 2003; Wang et al. 1981; Arachi et al. 1999). It is of interest to in-
vestigate an inuence of the complexes formed on the surface potential value. For
simplicity it is assumed that only electrically neutral complexes are formed.
2Me
M
+ V
O
[2Me
M
+ V
O
]
x
(1.13)
Then the bulk concentrations of defects may be calculated at any temperature
from known values of the dopant concentration c and the free energy of the
Figure 1.4. Experimental values of
S
in AgCl (a) and AgBr (b) in comparison with theoretical values
obtained using the Stern model for N
S
= 10
15
, 10
14
, 10
13
, and 10
12
cm
2
, curves (1), (2), (3), and (4), re-
spectively, and concentration of Me
2+
impurities of 100 ppm. Values of the adsorption energies used
in calculations are pointed out in expressions (1.11a and 1.11b). (Experimental data from Wonnel and
Slifkin 1995; theoretical curves have been reported in Uvarov 2007b.)
complex dissociation g
dis
using a standard procedure (Krger 1964). For the asso-
ciation constant of dipole complexes K
d
we used the expression K
d
= exp(-g
ass
/kT),
where g
ass
was varied from 0 to 0.4 eV.
The surface of solid solution M
IV
1-c
Me
III
c
O
2-c/2
may adsorb defects, just as is the
case with common ionic crystals. For this situation the Stern equation may be
represented in the form:
1
1
O
M
2 1 1
2 1 exp 1 exp
[V ]
[Me ]
S S
S
g e g e
e N
kT kT
+

, ]
, ] j \ j \ +
, ] + +
, ]
, ( , (
( , ( , , ] , ]
]
]

( )
1/2
1 1
1 0 M O
2
sign [Me ] exp 1 [V ] exp 1
S S
S
e e
A N
kT kT

, ] , ] j \ j \
+
, ( , ( , ] , ]
( , ( ,
] ]
(1.14)
for Gouy-Chapman SCLs, and

1
1
O
M
2 1 1
2 1 exp 1 exp
[V ]
[Me ]
S S
S
g e g e
e N
kT kT
+

, ]
, ] j \ j \ +
, ] + +
, ]
, ( , (
( , ( , , ] , ]
]
]

0 0
sign( ) 4
S S
e cN (1.15)
for Mott-Schottky SCLs. Here N
S
is the concentration of the surface sites available
for the adsorption of oxygen vacancies, which is twice as large as the correspond-
ing value for impurity cations; Ag
+
and Ag
are the adsorption energies of the oxy-

gen vacancies and the impurity cations, respectively; N
0
= 3 10
22
cm
3
is the total
volume concentration of M ions; c = 10c
0
. Defect concentrations in the bulk of the
crystal, [V
O
] and [Me
M
], are interrelated by a ratio [Me
M
] = 2[V
O
]; the sign of the

right-hand members is chosen to be opposite to the sign of the left-hand member.
For simplicity it is assumed that
S
=
S1
, i.e., the change in electric potential
takes place in the diffuse layer only. In spite of the fact that such approximation
is rough, it sufces for qualitative evaluations. The results of the calculation of the
surface potential with Eqs. (1.14) and (1.15) are presented and discussed below.
Gouy-Chapman SCL. Figure 1.1 shows the temperature dependencies of the
surface potential in M
IV
1-c
Me
III
c
O
2-c/2
calculated from Eq. (1.14) for c = 0.10 and
different values of Ag
+
and Ag
. For simplicity it was assumed that the entropy

terms in the free Gibbs energies (Ag
= Ah
T AS
) were also put to be zero, and

practically all surface sites are supposed to be available for the adsorption of
the defects: N
S
= 10
15
cm
2
. The surface potential is very small if both Ag

> 0,
and becomes measurable only when at least one of the Ag
terms is negative. As
seen from Figure 1.5, the surface potential is determined by values of Ag
and
Ag
+
similarly to ordinary ionic crystals of the NaCl type. Analysis of the
S
curves
shows that at low temperatures the surface potential tends to zero. At high tem-
peratures the absolute values of the surface potential are limited by a parameter
S
* obtained by extrapolation of the high-temperature parts of the
S
(T) curves to
the ordinate. Values of
S
* may be estimated from Eq. (1.6) for the case of high
potentials when one term in the sum of two exponentials can be neglected. In this
case Eq. (1.6) may be rewritten in the following forms:

O 0 M
2
2 [V ] exp [Me ] exp
2
S S
S
g e e
e N A N
kT kT
+
, ] j \ + j \

, ] , (
, (
( ,
( ,
, ]
]
(1.16)
for
S
>0 (in this case Ag
+
< 0, Ag
+
> Ag
), and

M 0 O
[Me ] exp [V ] exp
S S
S
g e e
e N A N
kT kT
j \ j \

, (
, (
( ,
( ,
(1.17)
Figure 1.5. Temperature dependence of the surface potential in M
IV
1-c
Me
III
c
O
2-c/2
calculated from
Eq. (1.14) at c = 0.10; N
S
= 10
15
cm
2
; G
ass
= 0; g
= 1 eV, and different values of g

+
. Dash lines
were obtained by extrapolation of the high-temperature parts up to
S
* values in the limit T 0. (Data
from Uvarov 2008b.)
for
S
< 0 (when Ag
< 0, Ag
> Ag
+
). Under the assumption that all nonexpo-
nential terms are weakly dependent on temperature, from Eqs. (1.16) and (1.17)
one can nd:

S
* 2/5Ag
+
(
S
> 0; Ag
+
< 0) (1.18a)

S
* 1/2Ag
(
S
< 0; Ag
< 0) (1.18b)
Values of
S
* may be used only for rough estimation of upper and lower limits of
the surface potential at given Ag
values. Approximate relation (1.18a) gives more

reliable values than Eq. (1.18b).
In superionic oxides M
IV
1-c
Me
III
c
O
2-c/2
, the value of
S
is positive, hence the
rst case (1.18a) is realized. At Ag
+
= 0.6 eV, Ag
+
> Ag
, the absolute val-

ues of the surface potential, ~0.20.3 V, agrees with the experimental values for
Zr
1-x
Y
x
O
2-x/2
obtained from conductivity measurements (Guo et al. 2002; Guo and
Wasser 2006), therefore this value of Ag
+
was used in all further calculations. For
simplicity it was proposed that Ag
= 0 eV. Figure 1.6 presents the

S
dependen-
cies obtained at various concentrations of the dopant. On the same gure, data
are plotted for the cases of the complete absence of defects association (for g
ass
= 0)
and the formation of associates with g
ass
= 0.2 eV. The inuence of the defects
association on
S
(T) dependencies is visible only at low temperatures; at T > 500 K
this effect can be neglected. It is to be noted that the straight lines, obtained by
the extrapolation of the high-temperature linear segments of the
S
(T) curves to
the ordinate, converge at the point
S

S
*, regardless of the concentration of
the dopant.
Mott-Schottky SCL. For comparison, the Stern equation with the Mott-
Schottky SCL, Eq. (1.15), was solved with the same parameters Ag
+
and Ag
. The
dependencies of the surface potential on temperature are plotted in Figure 1.7.
In contrast to the case of Gouy-Chapman SCL,
S
values do not decrease with
temperature at low temperatures. At high temperatures both Gouy-Chapman and
Mott-Schottky models give approximately the same result: at T > 500700 K, the
values of the surface potential calculated using the models differ by less than
5 mV (see Figure 1.7). Therefore, in the practically important case of high tem-
peratures, the surface potential values do not depend essentially on the particular
form of the (x) function in the diffuse layer and are determined mainly by the
adsorption energies of the defects.
Cation segregation at the surface. One would expect that as the surface
of the oxide M
IV
1-c
Me
III
c
O
2-c/2
is charged positively, it should be enriched in the an-
ionic vacancies V
O
and depleted in negatively charged dopant cations Me

M
. This,
however, contradicts many experimental observations. Precise chemical micro-
probe analyses with a spatial resolution of less than 1 nm using methods of Auger
electron spectroscopy, electron energy-loss spectroscopy, energy-dispersive x-ray
spectroscopy, and x-ray photoelectron spectroscopy show that the surface core
is enriched in dopant cations (Ikuhara et al. 1997; Lei et al. 2002; Wilkes et al.
2003; Zhu et al. 2005; Norrman et al. 2006). The Stern model provides a clear
explanation of these phenomena. Figure 1.8 shows the dependence of the concen-
tration of the defects on the surface (in the inner Helmholz layer at
S

S-I
) on
the total concentration of the dopant, c, at 1000 K for the case Ag
+
= 0.6 eV of
positive adsorption at the surface of predominantly oxygen vacancies; the segre-
gation energy of the Me cation, Ag
, is taken to be zero. It is seen that the surface

potential increases monotonically with c, and the surface concentration of the
oxygen vacancies [V
O
]
S
is higher than that in the bulk ([V
O
] = c/2). In parallel,
the concentration of the dopant cations at the surface [Me
M
]
S
increases with c
and becomes comparable to the [V
O
]
S
value at high c. The cation segregation at
S
> 0 is a peculiarity of the defect adsorption isotherm and Eq. (1.1), which may
be repre sented for the case of the adsorption of cations in the form
Figure 1.6. Temperature dependence of the surface potential obtained at various concentrations of
dopant in the absence of defects association (for g
ass
= 0, solid lines) and formation of associates
with g
ass
= 0.2 eV (dash-dot lines). Curves 1, 2, 3, 4, and 5 correspond to c = 0.001, 0.003, 0.01,
0.03, and 0.1, respectively; g
+
= 0.6 eV; g
= 0 eV; N
S
= 10
15
cm
2
. Dash lines obtained by rough
extrapolation of the high-temperature parts up to
S
* values in the limit T 0. (Reprinted with permis-
sion from Uvarov 2008b. Copyright 2008 Elsevier.)

M
M
M
[Me ]
[Me ]exp
1 [Me ]
S S
S
e g
kT kT
j \
+
, (
( ,

(1.19)
From this equation one nds that at positive values of the surface potential, [Me
M
]
S
>
[Me
M
] even at Ag
= 0 (in this case the surface potential is given by a negative value

of Ag
+
). At c = 0.10, concentrations of both [Me
M
]
S
and [V
O
]
S
are very high, 0.32
and 0.44, respectively. Nevertheless, the total charge of the surface is determined
by an excess charge of the oxygen vacancies. This situation agrees completely
with experimental data reported in the literature (see Figure 1.8).
It is interesting to compare Eq. (1.19) with an equation proposed by McLean
(1957) for qualitative description of impurity segregation in metals:
[ ]
[ ] ( )
[ ] ( )
seg
seg
M exp
M
1 M exp
S
g kT
g kT
+
(1.20)
Figure 1.7. Comparison of
S
values calculated using Gouy-Chapman (solid lines) and Mott-
Schottky (dot lines) models of SCL. Curves (1, 1), (2, 2), (3, 3), (4, 4) and (5, 5) correspond to c =
0.001, 0.0032, 0.01, 0.032, and 0.1, respectively; g
+
= 0.6 eV; g
= 0 eV; g
ass
= 0; N
S
= 10
15
cm
2
.
(Reprinted with permission from Uvarov 2008b. Copyright 2008 Elsevier.)
where [M] and [M]
S
are concentrations of M atoms in the bulk and on the surface
of the crystal; Ag
seg
is the segregation energy, i.e., the difference of the energy of
the crystal when the M atom is located deep inside the crystal and on the surface.
The McLean equation may be applied to ionic crystals. In particular, Eqs. (1.19)
and (1.20) become equivalent under the condition

seg
S
e g
g
kT kT

+ (1.21)
In metals the surface potential is close to zero, whereas in ionic crystals it
gives an additional (positive or negative) contribution to the segregation energy.
Nevertheless, similar to metals, the segregation of impurity atoms (ions, defects)
in ionic crystals is dened mainly by the corresponding adsorption energies Ag
+

and Ag
.
A possible reason for the difference in the defect adsorption energies is the
difference in excess elastic energy of the crystals accumulated at formation of the
defects. Kingery (1984) proposed the following relation between the segregation
energy and the size of the impurity atom dissolved in the crystal:
Figure 1.8. Defect concentrations in the bulk ([ Me
M
], [ V
O
]), on the surface ([ V

O
]
S
, [ Me
M
]
S
), and
the surface potential (
S
) at 1000 K as a function of the composition of solid solution M
IV
1-c
Me
III
c
O
2-c/2
.
Values calculated from Eq. (1.6) at g
+
= 0.6 eV; g
= 0 eV; g
ass
= 0.4 eV; N
S
= 10
15
cm
2
. Symbols
correspond to experimental data on the surface concentration [ Me
M
]
S
reported in the literature. (Data
from Uvarov 2008b.)
( )
2
1 2
seg 1 2
1 2
24
4 3
KGr r
g r r
Gr Kr

+
(1.22)
where K is the bulk modulus of the solute; G is the shear modulus of the solvent,
and r
1
and r
2
are the ionic radii of the solvent and solute, respectively. From this
equation it follows that, in oxides containing impurity cations, the segregation
energy of the dopant cations increases with the relative difference in cationic radii
of the host oxide and the dopant, Ar
cat
2
. The segregation energy of oxygen vacan-
cies is dened by the difference in the radius of the oxygen anion and the effective
radius of the oxygen vacancy, Ar
O
2
. In superionic oxides of M
IV
1-c
Me
III
c
O
2-c/2
type,
the surface potential is positive, which can be explained by a high volume change
at the formation of extra oxygen vacancies when Ar
O
2
> Ar
cat
2
.
2.6. SURFACE DISORDER IN TERMS OF ENERGY DIAGRAMS
The results of calculations of the surface potential may be summarized as follows:
1. The surface potential is nonzero only if at least one of Ag
i
< 0 and is de-
termined by the difference in the defect adsorption energies Ag
and Ag
+
of
positively and negatively charged dominant defects. If Ag
= Ag
+
, then
S
= 0.
2. The surface potential increases monotonically with N
S
value.
3. In pure ionic crystals the temperature behavior of
S
depends on the value
of Ag
o
, where Ag
o
is the most negative of the Ag
+
and Ag
energies, and differs

for two cases:
(a) Ag
o
< g
0
/4; the surface potential is small,
S
increases monotonically
with the temperature and rises as a function of the defect concentration
in the bulk and N
S
.
(b) Ag
o
> g
0
/4; the case of high surface potentials. The values of
S
may
increase or decrease with temperature as a function of N
S
values. At
sufciently high values of adsorption energy Ag
o
and N
S
, the surface
potential decreases with temperature and does not depend on the defect
concentration.
4. At high N
S
and Ag
o
the surface potential tends to the value of
S
(Ag

Ag
+
)/2e, which can be regarded as the upper limit. It is to be noted that this
expression is formally similar to the equation
S
(g
g
+
)/2e obtained in
frames of the FK approach, where g
and g
+
are the defect formation energies
in the bulk of the crystal.
5. In MX crystals doped with MeX
2
, the situation is more complicated:
S
val-
ues depend on the type, the concentration of the impurity ions (given by the
energy of dissociation of complexes [impurity ion cation vacancy]), and
their adsorption energy (or segregation energy) Ag
Me
+
. At Ag
+
> Ag
>
Ag
Me
+
or Ag
Me
+
> Ag
+
> Ag
(Ag
+
, Ag
, Ag
Me
< 0), an isoelectric point
(i.e., temperature where
S
change sign) exists on the
S
(T) dependence.
6. In superionic oxides M
IV
1-c
Me
III
c
O
2-c/2
, values of
S
are generally bounded by
limiting parameters given by values of Ag
+
and Ag
Me
(the adsorption energy

of oxygen vacancies and the segregation energy of Me cations, respectively).
The values of
S
obtained using the Gouy-Chapman and Mott-Schottky mod-
els of the SCL differ strongly in the low-temperature limit and tend to level
off with a temperature increase. At high temperatures the surface potential
does not in practice depend on the particular form of the (x) function in the
SCL and is determined mainly by the adsorption energies of the defects.
7. According to the literature (Heyne 1983; Guo and Maier 2001; Guo and
Wasser 2006), in oxides of the M
IV
1-c
Me
III
c
O
2-c/2
type the surface potential is
positive, therefore Ag
+
< 0 (Ag
+
> Ag
Me
) and the surface should be enriched

in anionic vacancies and depleted in cations. In the Stern model the adsorp-
tion of oxygen vacancies and the segregation of extrinsic cations are inter-
related, and even at Ag
Me
= 0 a strong segregation of cations takes place on

the surface. Nevertheless, the surface as a whole remains positively charged
and the diffusion layer is depleted of anionic vacancies.
Figure 1.9. Energy diagrams of surface and bulk defects, charge distribution near the surface, and
electrical potential proles for pure (a, b) and doped (c) ionic crystal MX with Schottky defects.
Diagrams obtained in terms of the Stern model. Cases (a) and (b) correspond to the pure crystal with
different g
+
and g
values. Cases (b) and (c) relate to the MX crystal doped with bivalent impurity
MeX
2
(dopant concentration equal to c) in intrinsic (b) and extrinsic (c) conductivity regions. (Data
from Uvarov 2011.)
Representing the chemical potential of the ith defect,
i
, in the standard form,
i
= g
i
+ kTln[ ]
i
(here [ ]
i
is the fraction of the defects; the superscript corresponds
to the sign of the defect) and taking into account the electrical neutrality condi-
tion, one can obtain values of g
i
for all the defects. Physically, each g

i
value cor-
responds to the energy necessary for generation of a single defect. One can plot
energy diagrams (Figure 1.9) illustrating the difference between the defect ener-
gies in the bulk of the crystal and at the surface for MeX
2
-doped MX crystals in
different temperature regions. This diagram differs from the diagrams reported
earlier (Maier 1995; 2003; Jamnik et al. 1995) because in the bulk of the crystal in
the intrinsic region the defect formation energies for both defects are taken to be
equal to g
0
/2. As seen from the diagram, at negative values of adsorption energies
the surface is enriched in defects even in the case of zero surface charge (which is
possible at Ag
+
= Ag
). In general, the surface can be considered as an independent

subsystem characterized by intrinsic surface disordering with an effective defect
formation energy equal to

0
2
S
g g
g g
+
+
+ (1.23)
The surface is more or less disordered than the bulk, depending on the sign of the
second term of this equation.
2.7. DEFECTS ON INTERFACES
Interfaces comprise phase boundaries with more complicated structure dened
by several interrelated chemical and morphological factors. Chemical factors in-
clude the bonding type, the crystal structure of contacting phases, and inter-
actions between them. Morphological factors include the type of lattice surfaces,
their orientation, and a mist between the crystal lattice parameters of adjacent
phases. In systems consisting of small particles (or grains), size effects should also
be taken into account.
From general conditions of mass and charge conservation it follows that con-
centrations of ions are interrelated to fractions of corresponding point defects. As
conductivity of ordinary ionic salt MX is carried out by point defects, it is con-
venient to express the parameters of the chemical adsorption of ions in terms of
the adsorption isotherms of the corresponding defects (including impurity ions)
of ith type with the adsorption energies of Ag
i
. Chemical adsorption of charged
species seems to be a general phenomenon typical for any polar media (includ-
ing ionic crystals) and may take place for free surfaces of MX, grain boundary
MXMX, and MXA interfaces. Therefore the same phenomenological approach
based on the Stern model can be applied to surface- or interface-related effects in
polycrystalline samples and composites.
In heterogeneous systems consisting of two ionic salts (of the MXMX type),
there are diffuse layers in both phases. Therefore, for calculation of the surface
potential using the Stern model one has to take into account the adsorption
of defects belonging to each phase to a common interface. Such a situation is
demon strated in the energy diagram depicted in Figure 1.10 for the contact of two
intrinsic conductors MX and MX with Schottky defects. In this case the surface
potential is determined by six independent parameters: (1) the defect formation
energies in the bulk of the phases MX and MX, g
01
and g
02
, respectively; (2) the
energies Ag
1
+
and Ag
1
of the defect adsorption for one of the phases (the corre-

sponding adsorption energies for another phase are given by the relations Ag
2
+
=
(g
01
+ g
02
)/2 Ag
1
+
and Ag
2
= (g
01
+ g
02
)/2 Ag
1
; (3) the number of active surface

sites, N
S
; (4) the potential difference between MX and MX, or Galvani potential,
A
c
, which cannot be measured directly. Work is in progress on analysis of the
Stern equation for this case. Preliminary estimates show that the following quali-
tatively different situations, demonstrated in Figure 1.10, may be realized:
(a) If Ag
i
> 0, then the interface charge is close to zero, and two oppositely
charged diffuse layers may be formed due to the existence of the potential
drop between the phases. The potential prole shown in Figure 1.10a is
Figure 1.10. Energy diagrams of surface and bulk defects, charge distribution, and electrical poten-
tial proles near MXMX (a, b) and MXA (c) interfaces. Diagrams obtained in terms of the Stern
model. Cases (a) and (b) obtained at different values of g
+
and g
. (Data from Uvarov 2011.)

typical for classical boundaries between semiconductor and liquid electro-
lyte (Gurevich and Pleskov 1983; Sato 1998) and was earlier applied to ex-
planation of the interface-related properties of MXMX systems by Maier
(1985, 1995).
(b) If Ag
i
< 0, the surface potential and the interface charge are high; the
potential has an extremum at the interface; and two double layers form
near the interface in both phases, with the space charge opposite in sign
to the interface charge. Such a prole is presented in Figure 1.10b; it is
qualitatively similar to one at the intergrain boundaries of ionic crystals,
but the charge distribution is not symmetricalhigher space charge is ac-
cumulated in the phase with the lower value of defect formation energy.
These variants correspond to some limiting cases. The real situation is inter-
mediate and the interface potential is dened by all the independent parameters
mentioned above. In the case of doped ionic crystals, one can also take into ac-
count the concentration of dopants in the contacting phases and the segregation
energy of dopants to the interface. As no data on the interface potential of MXMX
systems are available in the literature, it is hard to verify the model.
The defect equilibrium on the interface between an ionic salt MX and an oxide
A may be regarded as a chemical adsorption of ions of the MX phase onto the sur-
face of the oxide. The concentration of defects inside the bulk of the oxide is neg-
ligible, and only the rst oxide layer takes part in the interface interaction, which
includes pair interactions between ions of MX and A. This results in a change of
the defect adsorption energies, the surface potential, and, consequently, the con-
centration of point defects in the diffuse layer. The increase in the concentration
of the adsorbed ions on the interface (as well as the point defects concentration
in the diffuse layer of the ionic crystal) leads to enhancement of the conductivity
of the composites. Such a mechanism has been proposed by Maier (1985, 1995).
Most of the experimental data reported suggest that the interface interaction con-
sists of a selective chemical adsorption of M
+
cations, i.e., their shift from the MX
bulk to the MXA interface. Physically, this is equivalent to the change in the ad-
sorption energy of positively charged defects (anionic vacancies V
X

or interstitial
cations M
i
) and formation of high positive charge at the interface. As a result, a

diffuse layer built of cationic vacancies forms near the interface. The energy dia-
gram for the MXA interface is presented in Figure 1.10c. Selective adsorption
of cations can be regarded as interaction of M
+
cations as Lewis acid particles
with the Lewis base centers O
2
or OH
on the surface of the oxide. Therefore, the

adsorption energy should depend on the type of cation (the acidity of the cation
increases with a decrease in the ionic radius) as well as on the presence and
strength of basic groups on the oxide surface. It has been shown (Uvarov 2007a,
2008a, 2011) that lithium salts in systems with alumina exhibit a stronger in-
crease in conductivity as compared to rubidium and cesium salts, while iodides
more readily form nanocomposites as compared to chlorides and uorides. This
suggests that the polarizing ability and the polarizability of ions in the ionic com-
ponent play an important role in the surface interaction mechanism. The physi-
cal reason for the surface interaction in a composite of the ionic saltoxide type
lies in the trend of both substances to decrease their surface energy due to the
interaction of surface ions with the ions of the neighboring phase. Because of the
difference in interionic energies and peculiarities of the crystal structures in the
interface layers, the ideal structure inherent in individual phases will be distorted
in such a way as to provide a gain in the surface energy due to the mutual ap-
proach or removal of surface atoms. The relative displacement of ions from their
ideal positions is determined by the balance of the interaction energies. In alu-
mina and in the majority of the salts MX under discussion, anions exceed cations
in size, so it can be expected that for close packing, the interface cations will have
the larger free volumes and will be displaced by longer distances than the anions.
As a result, in the space between the surface layers, an intermediate, positively
charged layer enriched with cations is formed, the charge of which is compen-
sated by the cationic vacancies that constitute the diffuse layer. This process may
be regarded as a chemical adsorption (Maier 1985, 1995) and may be presented
as the following quasi-chemical reaction:
0 V
M
+ (MA)
S
(1.24)
which describes the stage of the surface disordering of MX at the MXA interface.
If an anion is adsorbed on the surface, another reaction may proceed:
0 M
i
+ (XA)
S
(1.25)
0 V
X
+ (XA)
S
(1.26)
The isoelectric point of an oxide pE was proposed by Shukla et al. (1988) as a
measure of its surface activity. Indeed, equations similar to Eqs. (1.25)(1.26) may
be written for the surface interaction of an oxide with water:
H
2
O OH + (HA)
S
(1.27)
2H
2
O H
3
O
+ (OHA)
S
(1.28)
The rst reaction predominates for oxides with pE > 7, for instance, MgO, Al
2
O
3
,
CeO
2
; the second prevails for oxides with pE < 7 (ZrO
2
, SiO
2
). By analogy with
aqueous solutions, one can expect that the surface reaction (1.25) will occur in
composites containing basic oxides (pE > 7), whereas for acidic oxides (pE < 7)
the interface interaction will follow the mechanism (1.26). The isoelectric point
of any oxide changes according to its doping with different dopants or by direct
modication of its surface by acidic or basic agents. This enables one to increase
the conductivity of the composite by variation of just its surface properties. A
more general approach is to change the Lewis acidity/basity of the oxide surface.
Both approaches were successful for the improvement of transport properties of
different composites (Saito and Maier 1995).
Recently, we have carried out a comparative study of electrical properties and
7
Li NMR data of the composites LiClO
4
A (A = o-Al
2
O
3
, -Al
2
O
3
, o-LiAlO
2
, -LiAlO
2
)
(Ulihin et al. 2006, 2008). It was shown that the conductivity depends not only
on the specic surface but also on the chemical nature and the structure of the
additive: At the same value of specic surface area, composites with -Al
2
O
3
and
-LiAlO
2
have lower activation energy for conductivity than composites containing
o-Al
2
O
3
and o-LiAlO
2
additives. A possible reason for such behavior could be the
presence of tetrahedrally coordinated cationic positions in the crystal structure
of the -phases. These sites, being located on the surface, seem to be strong basic
centers and favor chemical adsorption of lithium cations, leading to enhancement
of the concentration of defects in the vicinity of the LiClO
4
A interface. It has
been recently demonstrated (Ulihin 2009) that the conductivity of composite solid
electrolytes LiClO
4
A at close values of the specic surface area and the volume
fraction of oxide increases in a series SiO
2
Al
2
O
3
MgO in parallel to an increase
in the basicity. That is in agreement with the chemical adsorption approach men-
tioned above. Several papers (Nakamura and Saito 1992; Saito et al. 1988; Singh
et al. 1995) have reported on conductivity enhancement in composites containing
the ferroelectric oxides BaTiO
3
, LiNbO
3
, and KTaO
3
. The effect was proposed to be
strengthened due to high dielectric permittivity of the oxide. The concentration of
charge carriers in the surface or interface region of MX can be also varied using
modication of the surface by electronic donor molecules (Lauer and Maier 1990;
Saito and Maier 1995).
3. SIZE EFFECTS IN NANOCOMPOSITE SOLID ELECTROLYTES
When the particle size of a substance becomes smaller than 10100 nm, its physi-
cal properties change appreciably due to size effects. In the last two decades huge
progress has been made in research on nanosystems of different types, such as
nanostructured pure and composite materials, metal nanoparticles, carbon nano-
tubes, mesoporous systems, etc. Size effects may be very strong in nanocom-
posite solid electrolytes. In particular, new phases that are not typical for pure
components may be stabilized in the nanocomposites. Such effects have been
observed in such varied systems as AgIAl
2
O
3
(Uvarov et al. 1990, 1993, 1996b,
2000a, 2000b, 2000c), Li
2
SO
4
Al
2
O
3
(Uvarov et al. 1994), MNO
3
Al
2
O
3
(M = Li, Na,
K) (Uvarov et al. 1996c), MNO
3
Al
2
O
3
(M = Rb, Cs) (Uvarov et al. 1996d, 1996e),
RbNO
3
SiO
2
(Lavrova et al. 2000), CsHSO
4
SiO
2
(Ponomareva et al. 1996, 1998,
2000), LiClO
4
A (A = Al
2
O
3
, LiAlO
2
, SiO
2
) (Vinod and Bahnemann 2002; Ulihin et
al. 2006, 2008) and may be explained by the occurrence of amorphous phases of
the ionic salt in the composites. The existence of the amorphous state is evidenced
by the following effects observed in the composites:
1. Strong decrease in the integral intensity of x-ray diffraction peaks attributed
to all crystalline phases of MX and appearance of a wide halo on electron
diffraction patterns.
2. Diminishing of molar enthalpies of all phase transitions of MX, including the
melting enthalpy. Instead, a diffuse peak appears at a temperature much
lower than the melting point.
3. Disappearance of abrupt conductivity changes due to phase transitions of MX
in the MXA composites. Arrhenius dependencies of conductivity are nonlin-
ear for some composites. Estimates show that the charge carrier concentra-
tion in such nanocomposites is comparable to the overall number of cations
that is typical for superionic conductors or ion-conducting glasses. This is
also conrmed by the absence of the conductivity change upon melting.
4. Effects 1, 2, and 3 increase systematically with the total number of MXA
interfaces in the composite. This suggests that the amorphous phase is
an interface-induced, nonautonomous phase, which occurs only at MXA
interfaces.
The formation of the nanocomposites proceeds spontaneously and the self-
dispersion proceeds at noticeable rates even at temperatures substantially below
the melting point of MX, i.e., where the ionic salt is in the crystalline state. Due
to the solid-phase spreading, the amorphous phase is formed, i.e., the crystal-
line phase spontaneously transforms into the amorphous state. It is known that
composites MXA exhibit enhanced conductivity at temperatures below the melt-
ing point or superionic phase transition of MX. The properties of nanocomposite
systems have been analyzed in detail in several review papers (Heitjans and Indris
2003; Maier 2004, 2005; Schoonman 2005; Uvarov 2007a, 2008a, 2011).
The formation of the amorphous interface-stabilized phase cannot be explained
by standard models which involve surface point defect concepts. Nevertheless,
one can expect that interface interaction includes the adsorption of ionic species
from the ionic salt to the oxide surface. Therefore, at least on a qualitative level,
a chemical adsorption concept may be used for a rough estimate of the surface
potential and the interface effects in MXA composites.
4. APPLICATIONS IN SENSORS
There are two main types of resistivity sensors: (1) sensors based on a change
of the equilibrium concentration of ionic or electronic defects in the bulk of the
conductor and (2) adsorption sensors the resisitivity of which is determined by
surface (or interface) properties of the sensing material. The specic features of
these sensors can be compared using the examples of binary ionic salts MX and
metal oxides.
Sensors of the rst type are based on the dependence of the conductivity on
the activity of the X (for ionic salt MX) or oxygen (in oxides). This dependence
is caused by the variation of equilibrium concentrations of bulk ionic and elec-
tronic defects as a function of the components activity. In the concentration range
where conductivity depends on the oxygen activity, the conductivity is caused by
electronic defects which have much higher mobilities than ionic defects. Defect
concentrations are commonly represented by Krger-Vink diagrams (Krger
1964), i.e., plots of log defect concentration versus log component activity (Figure
1.11), which quantitatively express the variation of defect content within a given
phase. The character of the diagram depends on the ratio between the formation
enthalpy g
0
for ionic defects (Schottky or Frenkel ones) and the band gap E
g
. When
E
g
> g
0
(Figure 1.11a), ionic defects prevail, and their equilibrium denes the con-
centration of electronic defects. In the opposite case, E
g
< g
0
(Figure 1.11b), defect
proerties are dened by electronhole equilibrium. The concentration of electronic
and ionic defects in the bulk of the material may be comparable, and the equili-
bration time is limited by slow diffusion of ionic defects. To shorten the response
Figure 1.11. Kreger-Vink diagrams for pure ionic salt MX with dominant ionic (a) and electronic
(b) defects. Black lines correspond to bulk defects, dot lines correspond to the concentration of
surface defects.
time, the sensor should be made as thin lms. Similar to oxides, suldes may be
used for determination of sulfur-containing substances.
Sensors of the second type, operating due to the adsorption phenomena, are
more widespread (Lannto 1992; Gopel et al. 1995; Shimizu 1999; Korotchenkov
2007, 2008; Barsan et al. 2007; Wang 2010). Sensitive elements comprise semi-
conducting oxides, for example, doped SnO
2
, ZnO, and TiO
2
, which show changes
in electrical resistivity in the presence of small concentrations of ammable gases
such as propane, and toxic gases such as carbon monoxide. The sensors detect
gases because of a change in electrical resistance which accompanies the reaction
between adsorbed species such as O
2
, O
and O
2
and the gases to be detected.
The chemical adsorption leads to a change in the surface potential which, in turn,
results in a resisitivity increase or decrease (as a function of the double-layer
structure). Therefore, the surface potential is a key parameter which denes the
performance of the adsorption-type resisitive sensors, as this parameter denes
the concentration of the defects (including electronic ones) near the surfaces or
grain boundaries. Maier (1989b, 1995) has presented the Kreger-Vink diagram for
surface and bulk defects in MX crystals with Frenkel defects. Figure 1.11a shows
schematically the Kreger-Vink diagram for a MX crystal with bulk and surface
Schottky defects when E
g
> g
0
. One sees that concentrations of electrons and holes
deviate signicantly from the bulk values. In this case the surface potential is de-
termined by the defect adsorption energies as discussed above (see Section 2.2).
In particular, the process of the adsorption of intrinsic anionic vacancy may be
represented in the form
V
X
V
XS
(1.29)
hereafter, an additional subscript S denotes a species located at the surface. The
chemical adsorption of some gaseous species Y at the anionic sites of the surface
may be represented by a reaction
Y(g) + V
XS
Y
XS
(1.30)
with formation of a surface acceptor centers Y
XS
. Reaction (1.30) competes with

(1.29), leading to lling the surface by vacancies and results in a change in the
surface potential. A qualitatively similar situation takes place in semiconductors
MX with E
g
< g
0
(Figure 1.11b). In this case the surface potential is dened by
the chemical adsorption (accumulation) of electron donor (M
MS
) or acceptor (X
XS
)
states at the surface or grain boundaries.
M
MS
x
+ e M
MS
(1.31a)
X
XS
x
+ e
X
XS
(1.31b)
X
XS
x
+ e
X(g) + V
XS
(1.31c)
The formation of surface impurity defects Y
XS
due to reaction (1.30) shifts

equilibrium of the processes (1.31) and affects the surface potental. In contrast to
the bulk defects, the concentration of the surface defects may be easily varied, as
the surface potential depends not only on specic properties of the surface, but
also on the type and concentration of adsorbed species. In oxides the chemosorp-
tion of oxygen proceeds through the reactions
O
2
(g) + V
OS
+ e O
OS
(1.32a)
O
2
(g) + V
OS
O
OS
+ e
(1.32b)
for conductors of n-type and p-type, Eqs. (1.32a) and (1.32b), respectively, with
formation of chemisorbed oxygen ions O
(ad) or surface anionic defect O

OS
(in
Kreger notation). Concentration of chemisorbed oxygen may be easily affected by
adsorption of other reducing or oxidative agents, resulting in a change of the sur-
face potential and surface conductivity (or resisitivity).
Similar processes may proceed not only in pure single- and polycrystalline
oxides but also in composites where the surface potential forms on the interfaces.
Fast response time may be achieved by reducing the crystal size, porosity, or lm
thickness (allowing an increase in the adsorption surface and the contribution of
the surface conductivity). Selectivity is dened by the relative value of the adsorp-
tion energy of the gas to be detected. It depends on the atomic structure of the
adsorption centers and adsorption complex. By variation of the microstructure
(orientation and type of grains, porousity, etc.), type and concentration of dop-
ants, temperature, and optimal geometry of the sensor, one can nely tune ad-
sorption properties of the gas-analyte and to gain the best selectivity.
In recent years there has been increased interest in thin-lm sensors. Thin
lms, typically 100 nm thick, of doped tin oxide can be deposited by sputtering,
evaporation, sol-gel techniques, etc., with subsequent annealing leading to the
development of a nanocrystalline porous structure. Although thin-lm technol-
ogy, in comparison with thick-lm technology, would appear to offer the advan-
tage of relatively lower production costs because of the avoidance of powder and
screen-printing paste processing steps, there are drawbacks. Thin lms offer a
much smaller reactive area than sintered thick-lm ceramics, and so the sensitiv-
ity of sensors fabricated from them is correspondingly lower; for the same reason,
they are more susceptible to surface contamination.
Despite many applications of ceramic gas sensors and great progress manu-
facturing techniques, no quantitative models have been proposed for theoretical
description of their characteristics. This lack hinders the prediction of optimal
materials, dopants, concentrations, and temperatures. The best performance of
sensors is achieved mainly by a trial-and-error strategy. Calculation of the surface
potential (or potential prole on the interface) would open ways to qualitatively
describe the surface properties of sensor materials and nd new possibilities to
improve sensors characteristics.
5. CONCLUSIONS
It has been demonstrated that the Stern model provides a common basis for the
quantitative description of the surface and interface effects, including the surface
potential formation and the point defects equilibrium at free surfaces and inter-
faces. In particular, in classical ionic crystals the model explains describes quali-
tatively the absolute values of the surface potential and the presence of isoelectric
points. The model correctly describes the defect equilibrium at the interfaces and
is applicable to the interfaces in composite solid electrolytes in which the double
layer is formed by the point defects in the interface region of the ionic salt. In
nanocomposites, structural reconstruction or the formation of interface phases
takes place. Analysis of the experimental data shows that in most cases, amor-
phous interface phases exist in nanocomposites. Due to their high ionic con-
ductivity and other characteristics, the composite solid electrolytes, especially
nanocomposites, will nd many applications in future technologies, in particular
in various gas-sensing devices.
REFERENCES
Adler S.B. (2004) Factors governing oxygen reduction in solid oxide fuel cell cathodes.
Chem. Rev. 104(10), 47914843. DOI: 10.1021/cr020724o
Agrawal R.C. and Gupta R.K. (1999) Superionic solids: Composite electrolyte phaseAn
overview. J. Mater. Sci. 34(6), 11311162. DOI: 10.1023/A:1004598902146
Arachi Y., Sakai H., Yamamoto O., Takeda Y., and Imanishi N. (1999) Electrical conduc-
tivity of the ZrO
2
-Ln
2
O
3
(Ln = lanthanides) system. Solid State Ionics 121(14), 133139.
DOI: 10.1016/S0167-2738(98)00540-2
Barsan N., Koziej D., and Weimar U. (2007) Metal oxide-based gas sensor research: How
to? Sens. Actuators B 121(1), 1835. DOI: 10.1016/j.snb.2006.09.047
Bhoga S.S. and Singh K. (2007) Electrochemical solid state g as sensors: An overview. Ionics
13(6), 417427. DOI: 10.1007/s11581-007-0150-7
Blakely J.M. and Danyluk S. (1973) Space charge regions at silver halide surfaces. Surface
Sci. 40(1), 3760. DOI: 10.1016/0039-6028(73)90050-2
Burggraaf A.J. and Winnubst A. (1988) Segregation in oxide s urfaces, solid electro lytes
and mixed conductors. In: Nowotny J. and Dufour L.-C. (eds.), Surface and Near
Surface Chemistry of Oxide Materials. Materials Science Monographs, Vol. 47. Elsevier,
Amsterdam, 449477.
Chandra S. (1981) Superionic Solids: Principles and Applicat ions. North-Holland, Amsterdam.
Chebotin V.N. and Perliev M.V. (1978) Electrochemistry of Solid Electrolytes. Khimiya,
Moscow (in Russian).
Chebotin V.N., Remez I.D., Solovieva L.M., and Karpachev S.V. (1984) Electrical double
layer in solid electrolytesTheory: Oxide electrolytes. Electrochim. Acta 29(10), 1380
1388. DOI: 10.1016/0013-4686(84)87016-4
Chen L. (1986) Composite solid electrolytes. In: Chowdhari B .V.R. and Radhakrishna S.
(eds.), Materials for Solid State Batteries. World Scientic, Singapore, 69.
Dudney N.J. (1989) Composite solid electrolytes. Annu. Rev. Mater. Sci. 19, 103120. DOI:
10.1146/annurev.ms.19.080189.000535
Fabry P. and Siebert E. (1997) Electrochemical sensors. In: Gellings P.J. and Bouwmeester
H.J.M. (eds.), The CRC Handbook of Solid State Electrochemistry. CRC Press, Boca
Raton, FL, chap.10.
Frenkel J. (1946) Kinetic Theory of Liquids. Oxford Unive rsity Press, Oxford, UK.
Goodenough J.B. (1993) Oxide-ion electrolytes. Annu. Rev. Ma ter. Res. 33, 91128. DOI:
10.1146/annurev.matsci.33.022802.091651
Gopel W. and Schierbaum K.D. (1995) SnO
2
sensors: Current st atus and future prospects.
Sens. Actuators B 2627(13), 112. DOI: 10.1016/0925-4005(94)01546-T
Grimley T.B. (1950) The contact between a solid and an electrolyte. Proc. R. Soc. (Lond.) A
201, 4061. DOI: 10.1098/rspa.1950.0043
Guo X. and Maier J. (2001) Grain boundary blocking effect in zirconia: A Schottky barrier
analysis. J. Electrochem. Soc. 148(3), E121E126. DOI: 10.1149/1.1348267
Guo X. and Waser R. (2006) Electrical properties of the grai n boundaries of oxygen ion
conductors: Acceptor-doped zirconia and ceria. Prog. Mater. Sci. 51(2), 151210. DOI:
10.1016/j.pmatsci.2005.07.001
Guo X., Sigle W., Fleig J., and Maier J. (2002) Role of spac e charge in the grain boundary
blocking effect in doped zirconia. Solid State Ionics 154155, 555561. DOI: 10.1016/
S0167-2738(02)00491-5
Guo X., Sigle W., and Maier J. (2003) Blocking brain boundar ies in yttria-doped and un-
doped ceria ceramics of high purity. J. Am. Ceram. Soc. 86(1), 7787. DOI: 10.1111/
j.1151-2916.2003.tb03281.x
Gurevich Y.Ya. and Pleskov Yu.V. (1983) Photoelectrochemistr y of Semiconductors. Nauka,
Moscow (in Russian).
Haile S.M., Boysen D.A., Chisholm C.R.I., and Merle R.B. (20 01) Solid acids as fuel cell
electrolytes. Nature 410, 910912. DOI: 10.1038/35073536
Hangenmuller P. and Van Gool W. (eds.) (1978) Solid Electrol ytes General Principles,
Characterization, Materials, Applications. Academic Press, New York.
Heitjans P. and Indris S. (2003) Diffusion and ionic conduction in nanocrystalline ceramics.
J. Phys.: Condens. Matter 15(30), R1257R1289. DOI: 10.1088/0953-8984/15/30/202
Heyne L. (1983) Interfacial effects in mass transport in ion ic solids. In: Beniere F. and
Catlow C.R.A. (eds.), Mass Transport in Solids. Plenum Press, New York, 425456.
Hull S. (2004) Superionics: Crystal structures and conductio n processes. Rep. Prog. Phys.
67(7), 12331314. DOI: 10/1088/0034-4885/67/7/R05
Ikuhara Y., Thavorniti P., and Sakuma T. (1997) Solute segregation at grain boun daries
in superplastic SiO
2
-doped TZP. Acta Mater. 45(12), 52755284. DOI: 10.1016/
S1359-6454(97)00152-3
Ishihara T. (ed.) (2009) Perovskite Oxide for Solid Oxide Fu el Cells. Springer-Verlag,
Dordrecht, The Netherlands.
Jamnik J. and Maier J. (2003) Nanocrystallinity effects in lithium battery materials.
Aspects of nano-ionics. Part IV. J. Phys. Chem. Chem. Phys. 5(23), 52155220. DOI:
10.1039/B309130A
Jamnik J., Maier J., and Pejovnik S. (1995) Interfaces in so lid ionic conductors: Equilibrium and
small signal picture. Solid State Ionics 75, 5158. DOI: 10.1016/0167-2738(94)00184-T
Jow T. and Wagner J.B. (1979) The effect of dispersed alumin a particles on the electri-
cal conductivity of cuprous chloride. J. Electrochem. Soc. 126(11), 1963-1972. DOI:
10.1149/1.2128835
Khandkar A.C. and Wagner J.B. (1986) Fast ion transport in c omposites. Solid State Ionics
18/19(2), 11001104. DOI: 10.1016/0167-2738(86)90316-4
Khaneft A.V., Zhogin I.L., and Kriger V.G. (1990a) The model of formation of Frenkel defects
on the surface of ionic crystals. Poverkhnost 6, 6571 (in Russian).
Khaneft A.V. and Kriger V.G. (1990b) Formation of vacancies on the surface and in the bulk
of the ionic crystal. Zh. Fiz. Khim. 64(9), 24242429 (in Russian).
Khaneft A.V. (1992) Inuence of the surface disordering on the formation of Frenkel defects
in the ionic crystal. Zh. Fiz. Khim. 66(11), 30373044 (in Russian).
Kim S. and Maier J. (2002) On the conductivity mechanism of nanocrystalline ceria.
J. Electrochem. Soc. 149(10), J73J83. DOI: 10.1149/1.1507597
Kingery W.D. (1984) The chemistry of ceramic grain boundarie s. Pure Appl. Chem. 56(12),
17031714. DOI: 10.1351/pac198456121703
Kliever K.L. and Koehler J.S. (1965) Space charge in ionic crystals. I. General approach
with application to NaCl. Phys. Rev. 140(4), A1226A1240. DOI: 10.1103/PhysRev.140.
A1226
Kliever K.L. (1965) Space charge in ionic crystalsIII. Silver halides containing divalent
cations. J. Phys. Chem. Solids 27(4), 705717. DOI: 10.1016/0022-3697(66)90221-6
Korotcenkov G. (2007) Metal oxides for solid-state gas sensors: What determines our
choice? Mater. Sci. Eng. B 139(1), 123. DOI: 0.1016/j.mseb.2007.01.044
Korotcenkov G. (2008) The role of morphology and crystallographic structure of metal
oxides in response of conductometric-type gas sensors. Mater. Sci. Eng. R 61(16),
139. DOI: 10.1016/j.mser.2008.02.001
Kreuer K.D. (2003) Proton-conducting oxides. Annu. Rev. Mater. Sci. 33(3), 3359. DOI:
10.1146/annurev.matsci.33.022802.091825
Krger F.A. (1964) The Chemistry of Imperfect Crystals. North Holland, Amsterdam.
L antto V. (1992) Semiconductor gas sensors based on SnO
2
thick lms. In: Sberveglieri G.
(ed.), Gas Sensors. Kluwer Academic, Dordrecht, The Netherlands, 117167.
Lauer U. and Maier J. (1990) Conductance effects of ammonia on silver chloride boundary
layers. Sens. Actuators B 2(2), 125131. DOI: 10.1016/0925-4005(90)80021-Q
L avrova G.V., Ponomareva V.G., and Uvarov N.F. (2000) Nanocomposite ionic conductors
in the system MeNO
3
SiO
2
(Me=Rb, Cs). Solid State Ionics 136137, 1285289. DOI:
10.1016/0167-2738(00)00590-7
L ee J.S., Adams S., and Maier J. (2000) Transport and phase transition characteristics in
AgI:Al
2
O
3
composite electrolytes evidence for a highly conducting 7-layer AgI polytype.
J. Electrochem. Soc. 147(6), 24072418. DOI: 10.1149/1.1393545
L ee J.S. and Maier J. (1997) Interfacial phase transitions in AgI:Al
2
O
3
composite electro-
lytes. In: Negro A. and Montanaro L. (eds.), Proceedings of the International Conference
EUROSOLID-97, Politechnico di Torino, 115122.
L ei Y., Ito Y., Browning N.D., and Mazanec T.J. (2004) Segregation effects at grain boun-
daries in uorite-structured ceramics. J. Am. Ceram. Soc. 85(9), 23592363. DOI:
10.1111/j.1151-2916.2002.tb00460.x
L iang C.C. (1973) Conduction characteristics of the lithium iodide-aluminum oxide solid
electrolytes. J. Electrochem. Soc. 120(10), 12891292. DOI: 10.1149/1.2403248
L ifschits I.M., Kossevich A.M., and Geguzin Ya.E. (1967) Surface phenomena and diffusion
mechanism of the movement of defects in ionic crystals. J. Phys. Chem. Solids 28(5),
783792, IN1IN2, 793798. DOI: 10.1016/0022-3697(67)90007-8
M acdonald J.R., Franceschetti D.R., and Lehnen A.R. (1980) Interfacial space charge and
capacitance in ionic crystals: Intrinsic conductors. J. Chem. Phys. 73(10), 52725294.
DOI: 10.1063/1.439956
M aier J. (1985) Space charge regions in solid two-phase systems and their conduction
contributionI. Conductance enhancement in the system ionic conductor-inert phase
and application on AgC1:Al
2
O
3
and AgC1:SiO
2
. J. Phys. Chem. Solids 46(3), 309320.
DOI: 10.1016/0022-3697(85)90172-6
M aier J. (1986) On the conductivity of polycrystalline materials. Ber. Bunsenges Phys.
Chem. 90(1), 2633. DOI: 10.1002/bbpc.19860900105
M aier J. (1987) Defect chemistry and conductivity effects in heterogeneous solid electro-
lytes. J. Electrochem. Soc. 134(6), 15241535. DOI: 10.1149/1.2100703
M aier J. (1989a) Heterogeneous solid electrolytes. In: Chandra S. and Laskar A. (eds.),
Superionic Solids and Solid Electrolytes: Recent Trends. Academic Press, New York, 137.
M aier J. (1989b) Krger-Vink diagrams for boundary regions. Solid State Ionics 32/33,
727733. D OI: 10.1016/0167-2738(89)90351-2
Ma ier J. (1995) Ionic conduction in space charge regions. Prog. Solid State Chem. 23(3),
171263. DO I: 10.1016/0079-6786(95)00004-E
Maier J. (2002) Thermodynamic aspects and morphology of nano-structured ion con-
ductors: Aspects of nano-ionics. Part I. Solid State Ionics 154155, 291301. DOI:
10.1016/0167-2738(02)00499-X
Ma ier J. (2003) Defect chemistry and ion transport in nanostructured materi-
als: Part II. Aspects of nanoionics. Solid State Ionics 157(14), 327334. DOI:
10.1016/0167-2738(02)00229-1
Ma ier J. (2004) Ionic transport in nano-sized systems. Solid State Ionics 175(14), 712.
DOI: 10.1016/j.ssi.2004.09.051
Ma ier J. (2005) Nanoionics: Ion transport and electrochemical storage in conned systems.
Nature Mater. 4(11), 805815. DOI: 10.1038/nmat1513
Mc Lean D. (1957) Grain Boundaries in Metals. Clarendon Press, Oxford, UK.
Na kamura O. and Saito Y. (1992) Interfacial conductivities in Na
4
Zr
2
Si
3
O
12
/insulator parti-
cle systems. In: Chowdary B.V.R. and Wang W. (eds.), Solid State Ionics: Materials and
Applications. World Scientic, Singapore, 101.
No rby T. (2009) Proton conductivity in perovskite oxides. In: Ishihara T. (ed.), Perovskite
Oxide for Solid Oxide Fuel Cells. Springer-Verlag, Dordrech, The Netherlands, 217242.
Norrman K., Vels Hansen K., and Mogensen M. (2006) Time-of-ight secondary ion mass
spectrometry as a tool for studying segregation phenomena at nickel-YSZ interfaces.
J. Eur. Ceram. Soc. 26(6), 967980. DOI: 10.1016/j.jeurceramsoc.2004.12.030
Pa rk C.O., Akbar S.A., and Weppner W. (2003) Ceramic electrolytes and electrochemical
sensors. J. Mater. Sci. 38(23), 46394660. DOI: 10.1023/A:1027454414224
Poeppel R.B. and Blakely J.M. (1969) Origin of equilibrium space charge potentials in ionic
crystals. Surface Sci. 15(3), 507523. DOI: 10.1016/0039-6028(69)90138-1
Po nomareva V.G., Lavrova G.V., and Simonova L.G. (1999) The inuence of heterogeneous
dopant porous structure on the properties of protonic solid electrolyte in the CsHSO
4
SiO
2

system. Solid State Ionics 118(34), 317323. DOI: 10.1016/S0167-2738(98)00429-9
Po nomareva V.G., Lavrova G.V., and Simonova L.G. (2000) Effect of silica porous structure
on the properties of composite electrolytes based on MeNO
3
(Me=Rb, Cs.). Solid State
Ionics 136137, 12791283. DOI: 10.1016/S0167-2738(00)00589-0
Po nomareva V.G., Lavrova G.V., Uvarov N.F., and Hairetdinov E.F. (1996) Composite pro-
tonic solid electrolytes in the CsHSO
4
-SiO
2
system. Solid State Ionics 90(14), 161166.
DOI: 10.1016/S0167-2738(96)00410-9
Po nomareva V.G. and Lavrova G.V. (2011) Controlling the proton transport properties of
solid acids via structural and microstructural modication. J. Solid State Electrochem.
15(2), 213221. DOI: 10.1007/s10008-010-1227-1
Saito Y., Asai T., Ado K., and Nakamura O. (1988) Ionic conductivity enhancement of
Na
4
Zr
2
Si
3
O
12
by dispersed ferroelectric PZT. Mater. Res. Bull. 23(11), 16611665. DOI:
10.1016/0025-5408(88)90256-5
Sa ito Y. and Maier J. (1995) Ionic conductivity enhancement of the uoride conductor
CaF
2
by grain boundary activation using Lewis acids. J. Electrochem. Soc. 142(9), 3078
3083. DOI: 10.1149/1.2048691
Sa lamon S.B. (ed.) 1979 Physics of Superionic Conductors. Topics in Current Physics, Vol.
15. Springer-Verlag, Berlin.
Sato N. (1998) Electrochemistry at Metal and Semiconductor Electrodes. Elsevier, Amsterdam.
Sc hoonman J. (1997) The defect chemistry in solid state electrochemistry. In: Gellings P.J.
and Bouwmeester H.J.M. (eds.), The CRC Handbook of Solid State Electrochemistry.
CRC Press, Boca Raton, FL, chap. 5.
Schoonman J. (2005) Nanostructured materials in solid state ionics. Solid State Ionics
135(14), 519. DOI: 10.1016/S0167-2738(00)00324-6
Sh ahi K. and Wagner J.B. (1981) Enhanced electrical transport in multiphase systems.
Solid State Ionics 3/4, 295299. DOI: 10.1016/0167-2738(81)90101-6
Sh astry M.C.R. and Rao K.J. (1992) Thermal and electrical properties of AgI-based compo-
sites. Solid State Ionics 51(34), 311316. DOI: 10.1016/0167-2738(92)90214-A
Sh ukla A.K., Manoharan R., and Goodenough J.B. (1988) Ehancement of ionic conducti-
vity by dispersed oxide inclusions: Inuence of oxide isoelectric point and cation size.
Solid State Ionics 26(1), 510. DOI: 10.1016/0167-2738(88)90238-X
Sh ukla A.K., Vaidehi N., and Jacob K.T. (1986) Ionic conduction in dispersed solid-
electrolytes. Proc. Indian Acad. Sci. (Chem. Sci.) 96(12), 533547.
Sh imizu Y. and Egashira M. (1999) Basic aspects and challenges of semiconductor gas
sensors. MRS Bull. 24, 1824.
Singh K., Lanje U.K., and Bhoga S.S. (1995) Ferroelectric and Al
2
O
3
dispersed Li
2
CO
3
com-
posite solid electrolyte systems. In: Extended Abstracts: 10th International Conference
on Solid State Ionics, December 38, Singapore, 112.
St ern O. (1924) Zur theorie der elektrolytischen doppelschicht. Z. fr Electrochem. Angew.
Phys. Chem. 30, 508516.
Sun arso J., Baumann S., Serra J.M., Meulenberg W.A., Liu S., Lin Y.S., and Diniz da Costa
J.C (2008) Mixed ionicelectronic conducting (MIEC) ceramic-based membranes for oxy-
gen separation. J. Membrane Sci. 320(12), 1341. DOI: 10.1016/j.memsci.2008.03.074
Tak ahashi T (ed) (1989) High Conducting Solid Ionics Conductors: Recent Trends and
Applications. World Scientic, Singapore.
Tallon J.L., Buckley R.G., Staines M.P., and Robinson W.H. (1985) Vacancy formation
parameters from isoelectric temperatures in calcium-doped potassium chloride. Phil.
Mag. B 51(6), 635649. DOI: 10.1080/13642818508243152
Tha ngadurai V. and Weppner W. (2006) Recent progress in solid oxide and lithium ion con-
ducting electrolytes research. Ionics 12(1), 8192. DOI: 10.1007/s11581-006-0013-7
Tschpe A. (2001) Grain size-dependent electrical conductivity of polycrystalline cerium
oxide II: Space charge model. Solid State Ionics 139(34), 267280. DOI: 10.1016/
S0167-2738(01)00677-4
Tschpe A., Sommer E., and Birringer R. (2001) Grain size-dependent electrical conduc-
tivity of polycrystalline cerium oxide: I. Experiments. Solid State Ionics 139(34), 255
265. DOI: 10.1016/S0167-2738(01)00678-6
Tsch pe A., Kilassonia S., and Birringer R. (2004) The grain boundary effect in heavily
doped cerium oxide. Solid State Ionics 173(14), 5761. DOI: 10.1016/j.ssi.2004.07.052
Tuller H.L. (2003) Defect engineering: design tools for solid state electrochemical devices.
Electrochim. Acta 48(2022), 28792887. DOI: 10.1016/S0013-4686(03)00352-9
Ulihin A.S., Slobodyuk A.B., Uvarov N.F., Kharlamova O.A., Isupov V.P., and Kavun V.Ya.
(2008) Conductivity and NMR study of composite solid electrolytes based on lithium
perchlorate. Solid State Ionics 179(2732), 17401744. DOI: 10.1016/j.ssi.2008.02.027
Ulihin A.S., Uvarov N.F., Mateyshina Yu.G., Brezhneva L.I., and Matvienko A.A. (2006)
Composite solid electrolytes. Solid State Ionics 177(2632), 27872790. DOI: 10.1016/j.
ssi.2006.03.018
Ulihin A.S. and Uvarov N.F. (2009) Electrochemical properties of composite solid electro-
lytes LiClO
4
MgO. ECS Trans. 16(51), 445448. DOI: 10.1149/1.3242260
Urusov skaya A.A. (1969) Electric effects associated with plastic deformation of ionic
crystals. Sov. Phys. Usp. 11(5), 631643. DOI: 10.1070/PU1969v011n05ABEH003738
Uvarov N.F., Shastry M.C.R., and Rao K.J. (1990) Structure and ionic transport in Al
2
O
3
-
containing composites. Rev. Solid State Sci. 4(1), 6167.
Uvarov N.F., Isupov V.P., Sharma V., and Shukla A.K. (1992) Effect of morphology and
particle size on the ionic conductivities of composite solid electrolytes. Solid State Ionics
51(12), 4152. DOI: 10.1016/0167-2738(92)90342-M
Uvarov N.F., Khairetdinov E.F., and Bratel N.B. (1993) Composite solid electrolytes in the
system AgI-Al
2
O
3
. Elektrokhimiya 29(11), 14061410 (in Russian).
Uvarov N.F., Bokhonov B.B., Isupov V.P., and Hairetdinov E.F. (1994) Nanocomposite
ionic conductors in the Li
2
SO
4
-Al
2
O
3
system. Solid State Ionics 74(12), 1527. DOI:
10.1016/0167-2738(94)90432-4
Uvarov N.F. (1996a) Unusual bulk properties of ionic conductors in nanocomposites. In:
Chowdary B.V.R. and Wang W. (eds.), Solid State Ionics: New Developments. World
Scientic, Singapore, 311.
Uvarov N.F., Hairetdinov E.F., and Bratel N.B. (1996b) High ionic conductivity and unu-
sual thermodynamic properties of silver iodide in AgI-Al
2
O
3
nanocomposites. Solid State
Ionics 8688(1), 573576. DOI: 10.1016/0167-2738(96)00207-X
Uvarov N.F., Hairetdinov E.F., and Skobelev I.V. (1996c) Composite solid electro-
lytes MeNO
3
-Al
2
O
3
(Me = Li, Na, K). Solid State Ionics 8688(1), 577580. DOI:
10.1016/0167-2738(96)00208-1
Uvarov N.F., Skobelev I.V., Bokhonov B.B., and Hairetdinov E.F. (1996d) Composite solid
electrolytes based on rubidium and cesium nitrates. J. Mater. Synthesis Processing 4(6),
391395.
Uvarov N.F., Vanek P., Yuzyuk Yu.I., Zelezny V., Studnicka V., Bokhonov B.B., Dulepov
V.E., and Petzelt J. (1996e) Properties of rubidium nitrate in ion-conducting
RbNO
3
-Al
2
O
3
nanocomposites. Solid State Ionics 90(14), 201207. DOI: 10.1016/
S0167-2738(96)00400-6
Uvarov N.F., Politov A.A., and Bokhonov B.B. (2000a) Amorphization of ionic salts in nano-
composites. In: Chowdary B.V.R. and Wang W. (eds.), Solid State Ionics: Materials and
Devices. World Scientic, Singapore, 113.
Uvarov N.F. and Vanek P. (2000b) Stabilization of new phases in ion-conducting nanocompo-
sites. J. Mater. Synthesis Processing 8(56), 319326. DOI: 10.1023/A:1011346528527
Uvarov N.F., Vanek P., Savinov M., Zelezny V., Studnicka V., and Petzelt J. (2000c) Percolation
effect, thermodynamic properties of AgI and interface phases in AgIAl
2
O
3
composites.
Solid State Ionics 127(34), 253267. DOI: 10.1016/S0167-2738(99)00288-X
Uvarov N.F., Brezhneva L.I., and Hairetdinov E.F. (2000d) Effect of nanocrystalline alumina
on ionic conductivity and phase transition in CsCl. Solid State Ionics 136137, 1273
1277. DOI: 10.1016/S0167-2738(00)00587-7
Uvarov N.F. and Boldyrev V.V. (2001) Size effects in chemistry of heterogeneous systems.
Russ. Chem. Rev. 70(4), 265284. DOI: 10.1070/RC2001v070n04ABEH000638
Uvarov N.F. (2007a) Ionics of nanoheterogeneous materials. Russ. Chem. Rev. 76(5), 415
434. DOI: 10.1070/RC2007v076n05ABEH003687
Uvarov N.F. (2007b) Surface disordering of classic and superionic crystals: A description in
the framework of the Stern model. Russ. J. Electrochem. 43(4), 368376. DOI: 10.1134/
S1023193507040027
Uvarov N.F. (2008a) Composite Solid Electrolytes. Nauka, Novosibirsk (in Russian).
Uvarov N.F. (2008b) Estimation of the surface potential in superionic oxide conduc-
tors using the Stern model. Solid State Ionics 179(216), 783787. DOI: 10.1016/j.
ssi.2008.01.043
Uvarov N.F. (2008c) Estimation of the surface potential and concentration of surface de-
fects in superionic oxides. ECS Trans. 11(33), 207216. DOI: 10.1149/1.3038924
Uvarov N.F. (2011) Composite solid electrolytes: Recent advances and design strategies.
J. Solid State Electrochem. 15(2), 367389. DOI: 10.1007/s10008-008-0739-4
Vashis hta P., Mundy J.N., and Shenoy G.K. (eds.) (1979) Fast Ion Transport in Solids.
Elsevier/North-Holland, New York.
Verker k M.J., Hammink M.W.J., and Burggraaf A.J. (1983) Oxygen transfer on substituted
ZrO
2
, Bi
2
O
3
and CeO
2
electrolytes with platinum electrodes, J. Electrochem. Soc. 130(1),
7078. DOI: 10.1149/1.2119686
Vinod M.P. and Bahnemann D. (2002) Materials for all-solid-state thin-lm rechargeable
lithium batteries by sol-gel processing. J. Solid State Electrochem. 6(6), 498501. DOI:
10.1007/s10008-001-0251-6
Wagner J.B. (1985) Composite materials as solid electrolytes. In: Sequeira C.A.C. and
Hooper A. (eds.), Solid State Batteries. NATO ASI Series E101. Martinus Nijhoff,
Dordrecht, The Netherlands, 77.
Wagner J.B. (1989) Composite solid electrolytes. In: Takahashi T. (ed.), High Conductivity
Conductors: Solid Ionic Conductors. World Scientic, Singapore, 102
Wang D.Y., Park D.S., Grifth J., and Nowick A.S. (1981) Oxygen-ion conductivity
and defect interactions in yttria-doped ceria. Solid State Ionics 2(2), 95105. DOI:
10.1016/0167-2738(81)90005-9
Wang C ., Yin L., Zhang L., Xiang D., and Gao R. (2010) Metal oxide gas sensors: Sensitivity
and inuencing factors. Sensors 10(3), 20882106. DOI: 10.3390/s100302088
Waser R. (1995) Electronic properties of grain boundaries in SrTiO
3
and BaTiO
3
ceramics.
Solid State Ionics 75, 8999. DOI: 10.1016/0167-2738(94)00152-I
Whitwo rth R.W. (1975) Charged dislocations in ionic crystals. Adv. Phys. 24(2), 203304.
DOI: 10.1080/00018737500101401
Wilkes M.F., Hayden P., and Bhattacharya A.K. J. (2003) Catalytic studies on ceria lan-
thana solid solutions III. Surface segregation and solid state studies. J. Catal. 219(2),
305309. DOI: 10.1016/S0021-9517(03)00046-0
Wonnel l S.K. and Slifkin L.M. (1995) Subsurface ionic space charges in silver chloride and
silver bromide. Solid State Ionics 75, 101106. DOI: 10.1016/0167-2738(94)00149-M
Yarost alvtsev A.B. (2000) Ion transport in heterogeneous solid systems. Russ. J. Inorg.
Chem. 45(suppl. 3), S249S267.
Zhu J. , Van Ommen J.G., Knoester A., and Lefferts L.J. (2005) Effect of surface compo-
sition of yttrium-stabilized zirconia on partial oxidation of methane to synthesis gas.
J. Catal. 230(2), 291300. DOI: 10.1016/j.jcat.2004.09.025

Koro 5 Chap One

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Koro 5 Chap One

Uploaded by

Copyright:

Available Formats

CHEMICAL SENSORS

SIMULATION AND MODELING

is the concentration of the defects in the

are the concentrations of the anion and cation vacancies, respec-

are values of the adsorption energy for anion vacancy and

are taken equal to zero. Theoretical temperature dependencies of

were taken to be equal to 1.0 and

but also depends on the defect formation energy g

= 0.1 eV, and g

are shown in Figures

= 0.51 eV; the energy

= 0.51 eV: (a) at constant concentration of dopant c = 1 ppm and

) have been obtained (in eV):

. These values may be evaluated with

are the adsorption energies of the oxy-

]; the sign of the

. For simplicity it was assumed that the entropy

) were also put to be zero, and

= 1 eV, and different values of g

values. Approximate relation (1.18a) gives more

, the absolute val-

= 0 eV. Figure 1.6 presents the

and depleted in negatively charged dopant cations Me

, is taken to be zero. It is seen that the surface

= 0 (in this case the surface potential is given by a negative value

]), on the surface ([ V

energies, and differs

(the adsorption energy

) and the surface should be enriched

= 0 a strong segregation of cations takes place on

for all the defects. Physically, each g

). In general, the surface can be considered as an independent

of the defect adsorption for one of the phases (the corre-

; (3) the number of active surface

. (Data from Uvarov 2011.)

) and formation of high positive charge at the interface. As a result, a

on the surface of the oxide. Therefore, the

. Reaction (1.30) competes with

due to reaction (1.30) shifts

(ad) or surface anionic defect O

You might also like