Defects, Diffusion and Conduction in Ceramics

 Point defects in metals are electrically neutral whereas in ionic oxides, these are
electrically charged. The migration of the ionic defects plays a crucial role in the
conduction in ceramic materials that happens via an atomistic process called as
diffusion. Diffusivity of species in the materials is also related to their physical properties
such as electrical conductivity and mobility.

 The conductivity in ceramics is a sum of ionic and electronic conductivity and the ratio
of the two determines the applicability of ceramic materials for applications. Understanding
the temperature dependence of conductivity via a simple atomistic model leading to
the same conclusions as predicted by the diffusivity model.

 The conducting glasses and fast ion conductors are important materials of for a variety
of applications. Presence of charged defects in ceramics also means the existence of
electrical potential gradients, in addition to the chemical gradient, which results in a
unified equation for electrochemical potential called Nernst Equation which is used to
design oxygen sensors in automobile exhaust.
MT4760/ SUA/2024 1
 DEFECTS
Materials in general consist of defects which can be divided into a variety of categories such as point
defects or 0-D defects, line defects or 1-D defects and 2-D or surface defects. These defects play
an important role in determining the properties of ceramic materials and in this context, the role of
point defects is extremely important.
Point Defects :
 Point defects are caused due to deviations from the perfect atomic arrangement or
stoichiometry. These could be missing lattice ions from their positions, interstitial ions or
substitutional ions (or impurities) and valence electrons and/or holes.
Usually, point defects in metals are electrically neutral whereas in ionic oxides, these are
electrically charged.
 Ionic defects
• Occupy lattice positions
• Can be either of vacancies, interstitial ions, impurities and substitutional ions
 Electronic defects
• Deviations from a ground state electron orbital configuration give rise to such defects when
valence electrons are excited into higher energy orbitals/ levels and lead to formation of
electron or holes. MT4760/ SUA/2024 2
Kroger–Vink notation in a metal oxide, (MO)

Kroger-vink notations are typically used to depict

the atomic defects with charges. Following tables
provide the most common notations

Regular Sites

Point Defects (a • (dot) means a positive

charge and a ' (prime) means a negative
charge)

MT4760/ SUA/2024 3
MT4760/ SUA/2024 4
Defect Reactions
 Rules for writing defect reactions
• Ratio of regular cation and anion sites is always constant.
• Mass balance to be preserved.
• Electrical neutrality is to be always preserved.
 Both ionic and electronic defect compensations are possible determined by the energetics.
 Assume complete ionization of defects.

Defect Structures in Stoichiometric Oxides

Charged point defect is a defect which is ready to be ionized and provides a complimentary
electronic charged defect. Various such combinations are possible such as,
• Cation and anion vacancies (VM and VO)
• Vacancies and interstitial ion of same kind i.e., VO and Oi or VM and Mi
• Misplaced atoms interchanged (MO and OM) - interchanged
• Interstitial and misplaced atoms i.e.,Oi and MO
• Interstitial atoms i.e., Mi and Oi

Among all of these, the first two are most important as these are regularly seen in many important
oxides. The first is called Schottky disorder while the second is called as Frenkel disorder.
MT4760/ SUA/2024 5
MT4760/ SUA/2024 6
 Schottky Disorder
This defect normally forms at the outer or inner surfaces or dislocations.
It eventually diffuses into the crystal unit as equilibrium is reached.
The defect reaction is written as 0 (or Null) VM‘’+ V0••
 This defect is preferred when cations and anions are of
comparable sizes.
 Examples are rocksalt structured compounds such as NaCl, MgO,
Corundum, Rutile etc..
Schottky Disorder
Frenkel Disorder
 This defect can form inside the crystal.
 It forms where cations are appreciably smaller than anions.
 Defect reaction is written as
0 VM'' + Mi••
 In cases where anions form the disorder, then it is called as Anti-
Frenkel. The corresponding defect reaction in that case would be
0 V0•• + Oi''
 Examples of compounds showing this defect are AgBr type
Frenkel Defect
compounds such as AgBr, AgI etc. MT4760/ SUA/2024 7
Extent of Non-Stoichiometry

 In highly stoichiometric pure oxides such as MgO, Al2O3, ZrO2, the extent of oxidation or
reduction is very small. These are often characterized by large energy for oxidation or reduction.
Changes in oxygen pressure have very little effect on the defect concentration. When cations are of
fixed valence, the tendency for retaining the stoichiometry is even larger.

 Oxides containing multivalent cations, such as transition elements, are much more prone to
be non-stoichiometric. Examples are TiO2+x, BaTiO3-x and SrTiO3-x , where Ti4+ ions can be easily
reduced to Ti3+ creating oxygen deficiency of order 1% within the limits of the stability of oxide i.e.
before decomposition and phase change.

 Transition metal mono-oxide series Ni1-xO, Co1-xO, Mn1-xO and Fe1-xO are the oxides in which a
fraction of the divalent cations is easily oxidized to the divalent state resulting in cation deficiency, x .
The deficiency is ~5x10-4 % for Ni1-xO, ~1% for Co1-xO, ~0.1% for Mn1-xO and ~0.15% for Fe1-xO.
FeO is seldom stoichiometric, and it has a minimum nonstoichiometric of 0.05%.
MT4760/ SUA/2024 8
Defect Structures in Non - Stoichiometric Oxides
Mainly of two types Metal Deficient Oxides
i. Oxygen deficient (or excess metal)  Formation of either metal vacancies or
ii. Metal deficient (or excess oxygen) oxygen interstitials (excess oxygen)
 Nonstoichiometry necessitates presence  Formation occurs typically at the surface.
of point defects and extent of non-
stoichiometry determines the
concentration of Defects.
 In such oxides, electrical neutrality is
preserved via the formation of point
defects and electronic changes.
 Intrinsic ionization is always a possibility.
Oxygen Deficient Oxides
 Formation of oxygen vacancies or
metal interstitials or both are In FeO crystal, some of the Fe2+ ions are missing
possible. from the crystal lattice. To maintain the electrical
 Formation occurs only at the surface. neutrality, twice the number of other Fe2+ ions in
the crystal is oxidized to Fe3+ ions. In such
cases, overall number of Fe2+ and Fe3+ ions is
less than the O2- ions.
MT4760/ SUA/2024 9
Oxygen Deficient Oxides
Formation of oxygen vacancies or metal interstitials or both are possible. Formation occurs only at the
surface.

If oxygen vacancies are the dominating defects :

 Depicted by MO2-x (x is the extent of non-stoichiometry) and overall reaction as MO2 ↔ MO2-x+ x/2 O2↑

 Due to loss of oxygen, possible defect reactions would be

• Electronic compensation leading to creation oxygen vacancies and of electrons
O0 ↔ VO•• + ½ O2+ 2e'
• Ionic compensation leads to formation of oxygen vacancies and reduction of metal ions
on their sites.
O0 ↔ VO•• + ½ O2+ 2M'M

MT4760/ SUA/2024 10
If metal interstitials are the dominating defects, then :

 Depicted as Mi+yO2 is the extent of non-stoichiometry)

 Possible defect reactions are

• Ionic compensation leading to the formation of metal interstitials and reduction of metal
ions on their sites

M ↔ M Mi •••• + 4 M'M OR

• Electronic compensation leading to the formation of metal interstitials and free electrons

M ↔ MMi•••• + 4e'
Creation of quasi-free electrons (extra charge is represented as M’)
Conduction occurs due to transport of electrons

Typically, n-type conductors.

Example: TiO2, ZrO2, CeO2, Nb2O5

MT4760/ SUA/2024 11
 Electronic Disorder
 Unlike intrinsic point defeats, intrinsic electronic defects are optically or thermally created.
 This occurs in materials having a forbidden energy gap between conduction and the valence band
and are categorized as semiconductors and insulators.
 Defect density is in number per unit volume of the crystal.
 Disorder implies elevation of electrons into higher energy levels creating vacant states in
lower energy bands which are called as holes.
• Excitation of electrons across the bandgap into conduction band
• Bandgap (Eg) for semiconductors is typically below 2.5 eV e.g. Si has bandgap of ~1.1
eV whereas for insulators it is typically above 2.5-3 eV.
• The band diagram for a semiconductor or insulator can be seen below.

Band gap energy

values for a few
selected materials
are shown in the
table:

Band diagram for an insulator

MT4760/ SUA/2024 12
 DIFFUSION
 Diffusion causes changes in the microstructures to take place in processes such as sintering,
creep deformation, grain growth etc.

 Diffusion is also related to transport of defects or electronic charge carriers giving rise to
electrical conduction in ceramics. Ionic conductors are used in variety of applications such as
chemical and gas sensors, solid electrolytes and fuel cell. For example, an oxygen sensor
made of ceramic ZrO2 is used in automobiles to optimize the fuel/air ratio in the engines.

 Atomic diffusion rates and electrical conductivity are largely governed by defect types and
their concentration where concentration is a function of temperature, partial pressure of
oxygen or pO2 and composition.

Typically diffusion is explained on the basis of compositional gradients in an alloy which act as driving
force for diffusion. Thermodynamically speaking, this amounts to gradient in the chemical potential
which drives the migration of species from regions of higher chemical potential to lower chemical
potential so that system reaches a chemical equilibrium. The atomic flux as a result of driving force is
expressed in terms of chemical composition gradient, also called as Fick’s law(s).
MT4760/ SUA/2024 13
Fick's First Law of Diffusion
It states that atomic flux, under steady-state conditions, is proportional to the concentration
gradient. It can be stated as,

 J is the diffusion flux with units moles/cm2/s, and basically means

the amount of material passing through a unit area per unit time;
 D is the proportionality constant, called as diffusion coefficient or
diffusivity in cm2/s;
 x is the position in cm; and
 c is the concentration in cm3 .

The negative sign on the right-hand side indicates that diffusion takes place from regions of
higher concentration to lower concentration i.e. down the concentration gradient.
Diffusivity is a temperature dependent parameter and is expressed as D = D0 exp (-Q/kT) where Q is the
activation energy, k is Boltzmann's constant and D0 is the pre-exponential factor in cm2 /s.

MT4760/ SUA/2024 14
Fick's Second Law of Diffusion :
Strictly speaking it is not a law, but rather a derivation of the first law itself. It predicts how the
concentration changes as a function of time under non-steady state conditions . It can be derived
from Fick's first law easily.

Diffusivity: A Simple Model

As shown in Figure, a schematic diagram shows atomic planes,

illustrating 1-D diffusion of species across the planes .
Concentration gradient and flux can be derived from the
basic principles.

Schematic of the planes of atoms with arrows

showing the cross movement of species

MT4760/ SUA/2024 15
 Temperature Dependence of Diffusivity

where ΔH* is the enthalpy of migration, k is Boltzmann's

constant and D0 is the pre-exponential factor in cm2 /s.

Equation explains the thermally activated nature of diffusivity showing an exponential temperature
dependence resulting in significant increase in the diffusivity upon increasing the temperature.

Examples of Diffusion in Ceramics e.g.: Diffusion in lightly doped NaCl

NaCl is often used a conducting electrolyte and it a good case for proving an example. Consider the
ease of NaCl containing small amounts of CdCl2. In such scenario, for each Cd ion, Cd occupying
Na site with an extra positive charge and a sodium vacancy with one negative charge is created
according to the following defect reaction:

In addition, NaCl will also have certain intrinsic sodium and chlorine vacancy concentration (VNa' and VCl• )
due to Schottky dissociation, depending on the temperature.
MT4760/ SUA/2024 16
In such a scenario, the diffusivity of sodium ions is governed by vacancy diffusion and can be worked
out as,

where ΔGNa* is the migration free energy for sodium

vacancies and VNa' is the sodium vacancy
concentration. This diffusivity of vacancy is
dependent on vacancy’s concentration which is
related to the dopant concentration. However, the
diffusivity dependence on temperature shows two
regimes as;
 low temperature extrinsic regime where vacancy
concentration is independent of temperature and is
determined by solute concentration.
 High temperature intrinsic regime where
Schematic diagram showing temperature
vacancy concentration is dominated by intrinsic dependence of diffusivity in ceramics
thermally degenerated vacancies.
MT4760/ SUA/2024 17
Mobility and Diffusivity
In addition to diffusivity, another useful term to describe conduction in ionic compounds and
ceramics is mobility which is defined as velocity (v) of an entity per unit driving force (F) and is
expressed as,
The Force F can be defined as either of chemical potential
gradient, electrical potential gradient, interfacial energy
gradient, elastic energy gradient or any other similar
parameter.
Absolute mobility

Where NA is Avogadro's Number, μi is chemical potential in

J/mol and x is the position in cm.

Electrical mobility
Chemical mobility

MT4760/ SUA/2024 18
For atomic transport, Einstein first pointed out that the most general driving force is the virtual force
that acts on a diffusing atom or species and is due to negative gradient of the chemical potential or
partial molar free energy. It is expressed as,
Where μi is the chemical potential of i and NA is the
Avogadro's Number.

To obtain the relation between mobility and diffusivity of species, i, we need to write the flux in a
general form as a product of concentration, ci, and velocity, vi, i.e.

If we compare the above equation with Fick's first law, diffusivity of species I can be written as,

This equation is called Nernst Einstein Equation.

As a result, conductivity in the ionic systems also follows a temperature dependent Arhenious
type behavior. MT4760/ SUA/2024 19
 Conduction in Ceramics vis-à-vis metallic conducators: General Information

 While it is possible to apply Nernst-Einstein's relation to express the motions of both ions and electrons to state the
similarity between electrons and ions, in reality the motion of electrons is quite different.

 In metals, semiconductors and high mobility ceramics, the electrons and holes are considered as quasi-free
particles with a drift velocity, vd , acquired under an applied electric field. The extent of this drift velocity is
governed by various scattering phenomenon such as scattering from the lattice.

Ionic Conduction: Basic Facts

 Conduction in ionic solids is not similar to the band transport in metals and semicoductors. Conduction in
these solids is often governed by concentration of impurities, dopants and point defects.
 Conduction in these materials tends to be of polaronic or hopping type which is migration of charges between
either two dissimilarly charged ions or counter migration of ions and defects such as vacancies.
 In ionic solids, proximity of the carriers to the ions results in the formation of a polarized region i.e., a polaron
which may be longer or smaller than size of the unit-cell.
 Effective mass, m* , depends on level of interactions between electrons or holes and periodic lattice
potential.

 Effective mass can be taken as equal to that of a free electron i.e. me* = mfree electron. Large polarons typically
depict weak interaction between carries and ions and hence electronic mobility expressions can be valid for the
polaron. MT4760/ SUA/2024 20
 In a number of ionic systems such as LiNbO3, COFe2O4, FeO etc., carrier-ion interaction is strong and these
are called as strongly-correlated-systems, In such cases, polarons are of smaller sizes and the effective
mass tends to be larger than that of a free electron.
 In such systems, motion of carriers is thermally activated i.e. via hopping mechanism.

 For example, Li2O doping in NiO under oxidizing conditions gives rise to oxidation of Ni2+ ions to Ni3+ ions (write
the defect reaction yourself). Mixed presence of Ni ions in +2 and +3 states leads to hopping type conduction of
electrons between two states.
 Ni3+ acts as a perturbing charge which polarizes the lattice surrounding it leading the formation of Polaron.
 The movement of carrier between these differently charged states can happen by hopping which is a
thermally activated phenomenon.
 Conduction in such cases is determined by dopant concentration and hence higher the dopant concentration
(within appropriate limits), higher the conductivity. In contrast, in the band model, carrier concentration is
determined by temperature.
 In ionic solids, while carrier concentration is independent of temperature (within extrinsic region), mobility
is strongly affected by temperature
 Typically carrier mobilities in strongly interacting ionic systems are of the order of ~0.1 cm2/Vs which are
about 3-4 orders of magnitude smaller as compared to the weakly interacting systems where carrier
mobilities can be between 100-1000 cm2/Vs.

MT4760/ SUA/2024 21
Characteristics of Ionic Conduction
 Long range migration of ionic charge carriers, the most mobile species, through the lattice under
application of an electric field for example migration of Na+ ions in soda-silicate glasses.
 Dependent on the presence of vacant sites in neighbourhood of mobile defects/ions.
 Can occur through grain boundaries such as in polycrystalline ceramics or through the lattice
as in fast ion conductors.
 When external field is absent, the thermal energy, kT, is required for counter migration of ions
and vacancies overcoming the migration energy Ea, as shown in Figure, which is nothing but
process of self diffusion.
Ionic conductivity is promoted by
 Small ionic size
 Small charge i.e. less Coulomb interaction between ions
 Favourable lattice geometry
 Cations are usually smaller than anions and hence,
they diffuse faster. For example, incase of NaCl,
smaller size of Na+ ion (102 pm) as compared to Cl- ions
(181 pm) makes them diffuse faster.
Schematic diagram showing the lattice ions and
vacancies in a potential well with and without
applied electric field MT4760/ SUA/2024 22
Conduction in Glasses

 Glasses have a random 3-D network and contain glass forming agents such as SiO2 , B2O3
,and Al2O3
 Interstitial positions are occupied by modifier ions such as Na+ , K+, and Li+ or blocking
ions such as Ca2+ and Mg2+.
 Conductivity increases as the temperature increased as barrier for migration for mobile
modifier ions is easily overcome by thermal activation.
 Usually as the size of the modifier ion decreases, the conductivity increases.
 As the concentration of blocking ions increases, the conductivity decreases.

 Refer conductivity molten Silicates

MT4760/ SUA/2024 23
 Fast Ion Conductors
 Materials showing very high conductivities, several orders of magnitude higher than normal
ionic solids are called Fast Ion Conductors.

 Conductivities greater than 10-2 (W-cm)-1 at temperatures much lower than melting points.

 Number of charge carriers is a very large fraction of potentially mobile ions which is significantly
larger than in a normal ionic compound hence ion diffusion is solely governed by the diffusion
energy and hence defect formation energy need not be provided resulting in overall low activation
energy for migration.

 Intrinsic FICs
• Halides and Chalcogenides of silver and copper, e.g. - AgI is a silver ion (Ag+ ) conductor
• Alkali metal conductors such as non-stoichiometric aluminates e.g. ß-Al2O3 or Na2O.11Al2O3.

 Extrinsic FICs
• Oxides with fluorite structure such as ZrO2 are doped with aliovalent oxides creating a large
number of oxygen vacancies and are called as fast ion conductors such asY2O3.ZrO2. In these
materials the enthalpy of migration, ΔHm , can be as low as ~0.01-0.2.
MT4760/ SUA/2024 24
 Ionic conduction in cubic ZrO2

Zirconia, ZrO2, is made stable in cubic form by doping with small amounts, ~8-10%, of aliovalent (the
valence of the substituting ion is different from that required by the site) CaO or type of impurities. Cubic ZrO2 is
considered as a very good and fast ion conductor. Oxygen vacancies, the prevailing defects in Cubic
ZrO2 are created according to,

Small energy of migration (Zi)e of oxygen vacancies results in an unusually high diffusion of oxygen
vacancies in ZrO2

Example: Take the case of Ca0.14 ,Zr0.86, O1.86 . Bandgap of ZrO2 is ~5.2 eV. At high temperatures, the
diffusion coefficient is given as,

From the defect reaction

You can show that 14% molar concentration of CaO in ZrO2 gives rise to an oxygen vacancy
concentration of ~ 3.9*1021 cm-3.
MT4760/ SUA/2024 25
MT4760/ SUA/2024 26
Now, ionic conductivity can be written as

Mobility can be derived by using Nernst-Einstein equation as the following

Hence, ionic conductivity is

Electronic conductivity is calculated as

At 1823 K, electron concentration, ne , using band model, is estimated to be 1.32*1013 cm-3 and electron
mobility, μe , is 24 cm2/V.s
Hence, the electronic conductivity is

From the above analysis it turns out that sionic >> selec
sionic increases only until about 12-13% of CaO doping
MT4760/and then decreases due to increased defect interactions
SUA/2024 27
Conductivity in SrTiO3
Perovskite oxides SrTiO3 and BaTiO3 have two cation sub-lattices onto which aliovalent ions can
substitute. These oxides can also be easily reduced in reducing atmospheres leading creation of
mobile defects. Typically, Ba/Ti or Sr/Ti is approximately very close to 1. Maximum deviation occurs at
temperatures lower than 1200°C and is of the order of ~100ppm.
To promote n-type conductivity in these materials, solutes such La3+, Nb5+, and Ta5+ are added to
BaTiO3, giving rise to shallow donor levels in the bandgap. The defect reaction upon doping with
La or Nb is expressed as,

Occupancies are determined by the relative atomic sizes.

On the other hand, undoped materials often contain impurities such as Fe3+ or Al 3+ substituting

Ti4+ acting as acceptors.

MT4760/ SUA/2024 28
Consider lightly acceptor doped SrTiO3 with Al2O3 as impurity with the
following defect reaction:

(3)

In such a situation, either SrO is accepted or TiO2 is rejected to maintain the stoichiometric cation ratio.
The above equation is ionically compensated reaction. It follows that [AlTi'] = 2[V0•] and thus

Electronic compensation is also possible according to

Experiments show that Equation (3) , i.e. ionic compensation is predominant

mechanism.
MT4760/ SUA/2024 29
Electrochemical Potential
 Usually in ionic systems, both the effects of gradients in chemical and electrical potential on
the mobilities are present.
 Non-uniform distribution of charged defects can give rise to internal electric field even in the
absence of external electric field.
 Total driving force for mass transport is defined as Electrochemical potential.

The electrochemical potential of spices i , denoted as ηi, is the sum of the chemical potential
μi, and electric potential f acting upon it and can be written as,

where Zi is the effective charge and F is Faraday's constant (=96500 Coulomb/mol). Now, virtual force
on the particle is negative gradient of ηi i.e.

Flux, J, can be expressed as

or

MT4760/ SUA/2024
This shows that flux, Ji , is a function of electric field, (df/dx). 30
 Nernst Equation and Application of Ionic Conductors

We have seen that the electrochemical potential is a combination of chemical potential and
electrical potential.
Applied chemical potential gradient gives rise to a voltage which is useful in the operation of devices
like batteries, fuel cells, sensors.
Ionic transport is essential for the imposed oxygen pressure gradient to establish a gradient in
the oxygen concentration. Simultaneously, electronic conduction must be avoided to prevent short
circuiting which requires ti ˜ 1. (ti -ionic transference number)

Consider two sides separated by an electrolyte. At equilibrium, the chemical potential of both the
sides is equal, i.e.

Electrical potential across the electrolyte is related to the chemical potential difference by,

MT4760/ SUA/2024 31
Chemical potential can be expressed as

The above equation is referred to as Nernst Equation

Example:
In automotive applications, usually on one side is air i.e., pO2 is 0.21 atm and the other side is at pO2
of 10-9 - 10-21atm. From the Nernst equation, one can calculate that at 350°C, a voltage
difference, Δf , between 0.25-0.6V will be developed which can be easily monitored and can be
used as the feedback in control system toMT4760/
regulate the fuel supply to the engine.
SUA/2024 32
Examples of Ionic Conductors in Engineering Applications

 Conducting ceramics are used in a variety of applications such as

• SiC and MoSi2 as heating elements and electrodes
• ZnO and SiC as varistors for circuit protection
• CSZ (cubic stabilized Zirconia with 8-10 mol% Y2O3), ß -Alumina as electrolytes in fuel
cells and batteries
• Materials like YSZ (Y2O3 stabilized ZrO2 ) in gas sensing applications

 While heating elements and varistors utilize the electronic conduction properties of the
ceramics, fuel cells and sensors typically utilize ionic conductivity characteristics.

 In the following sections, major applications of conducting ceramics will be discussed

viz . three applications: varistors, solid oxideMT4760/
fuel SUA/2024
cells and exhaust sensors in automobiles 33
Varistors or Voltage Dependent Resistors
 Devices offering high resistance at low voltages and low resistance at high voltages.

 Used to protect a circuit against sudden upsurges in the voltage or transients.

 Typically used materials are ZnO and SiC powders bonded in a matrix with an intergranular
layer that is highly resistive. The current-voltage characteristics are governed by the electrical
nature of grain and intergranular layers.

 The basic principle of operation is that the material's resistance is high at low voltages and
decreases when voltage is increased, allowing excess currents to pass through thereby
protecting the circuit.

Current (I) - Voltage (V) characteristics of a Schematic diagram showing use

MT4760/ SUA/2024 34
varistor of varistor in circuit protection
 For example, zinc oxide particles of size about 10-50 μm are mixed with other oxides
such as bismuth or manganese oxides which form a highly resistive intergranular
layer (IGL) of thickness between 1 nm to 1 μm between zinc oxide grains. These IGLs
contain defect states whose role is to trap free electrons from the neighbouring n-
type semiconducting zinc oxide grains.

 This results in the formation of a space charge depletion layer in the ZnO grains in
the region adjacent to the grain boundaries as shown in the figure. These boundaries
between each grain and its neighbor forms a junction showing diode like
characteristics and allowing flow of current in one direction only.

In the figure ( next slide), the left-hand grain is forward biased by a voltage VL, and the
right side is reverse biased to VR. The respective barrier heights are fL and fR while
the zero biased barrier height is f0. As the voltage bias is increased, fL is decreased
and fR increased, leading to a lowering of the barrier and an increase in conduction.

MT4760/ SUA/2024 35
 The whole mass of material resembles an
electrical circuit network with back to back
pair of diodes, with each pair of diode in
parallel with many others. Very small current
flows when a small voltage is applied across
the electrodes which is mainly caused by
reverse leakage through the diode
junctions. When voltages are high enough
to cause break down of the diode junction,
large currents flow due to thermionic
emission as well as tunneling, as seen in the
Schematic of the electronic above graph.
structure of ZnO grain and
intergranular layer (note the bias on
the grains)
MT4760/ SUA/2024 36
 Solid Oxide Fuel Cells (SOFC)
 SOFCs are useful as electrochemical conversion devices in which oxidation of fuel leads to
production of electricity.

 Schematic of such a device is shown in Figure where two electrodes, anode and cathode,
are separated by an electrolyte e.g. CSZ (cubic stabilized Zirconia with 8-10 mol% Y2O3).

 This device can usually work with an efficiency of ~60% at 1000 oC with an output power of
100 kWh.

 The device can be used in a wide range of oxygen partial pressures (~1 atm to 10-20 atm) and
can work at temperatures between 700-1000oC with areal resistance of ~ 15μΩm-2 .

 Anode material is typically porous Ni and cathode is a porous p-type semiconducting oxide
such as La1-x Srx MnO3+y .

 One needs to connect many of these devices together to obtain substantial power and for that
purpose, interconnects are used which are typically made of Ca, Mg or Sr doped LaCrO3 or
stainless steel depending upon the operation temperature.
MT4760/ SUA/2024 37
 Ceramic Electrolyte
 The electrolyte is ZrO2 (zirconia) doped with 8-10 mole% of Y2O3 (yittria).
 When yittria is mixed with zirconia, some Y 3+ replaces Zr 4+ in the fluorite type crystal
structure.
 Number of oxide ion sites become vacant because three O2- replace four O2- when two Zr4+
are replaced by two Y3+ in the lattice.
 The mobile ionic species within the electrolyte in the SOFC system are the O2- ions. At
high temperatures, the O 2- ions move across the electrolyte via vacant lattice sites.

Processing Techniques
• Dry pressing
• Tape Casting
• Screen printing
• Slurry Coating

MT4760/ SUA/2024 38
Schematic of (a) a SOFC
device and (b) operation
of a SOFC device
In case of H2 and CO gases
as fuels, the following
reactions take place at two
electrodes, as shown in
Figure.

It must be noted that high ionic

conduction through the electrolyte is
required to minimize the resistive
losses.
MT4760/ SUA/2024 39
 Oxygen Sensors in Automobile Exhaust
 Modern automobiles demands good fuel efficiencies and pollution control.
 Exhaust gases are NOx gases, CO, Hydrocarbons.
 One alternative is to use the catalytic converters, which however do not regulate the fuel
consumption.
 The above air/ fuel ratio of ~ 14.5, hydrocarbons and CO are decomposed efficiently but not
NOx, while below 14.5, the reverse is true.
 When there are deviations in the air /fuel ratio from the stoichiometric value of 14.5, pO2 in
the exhaust changes dramatically from ~ 10-9 to 10-26 atm. This differential provides a way to
use the activity of oxygen, i.e. to measure the combustion efficiency.
 The process is to first measure the pO2 of the exhaust by using an oxygen sensor. The
difference between the pO2 of the exhaust and air can be translated into an electrical signal
which is fed as an electrical input to the fuel injection system to optimize the air/fuel ratio.
 Oxygen sensors are typically conducting ceramics operating on two principles:
• First type is based on Nernst effect where an electrical potential is created when there is a pO2
gradient across the sensor.
• Second type is where one can measure the variation in the conductivity of the ceramic vs
pO2 .
MT4760/ SUA/2024 40
 The schematic of an exhaust gas sensor using Y2O3 stabilized ZrO2 which gives rise to sufficient
voltages under a gradient of oxygen partial pressures encountered in automobiles which can
be used to regulate the fuel/air supply.

Schematic of a automobile exhaust sensors and (b)potential generated by

different oxygen partial pressure on two
MT4760/sides
SUA/2024 of the electrolyte. 41
Oxygen sensor probes are used to reduce vehicle emissions by ensuring that engines burn their fuel
efficiently and cleanly. Robert Bosch GmbH introduced the first automotive lambda probe in 1976,[2] and
it was first used by Volvo and Saab in that year. The sensors were introduced in the US from about 1979
and were required on all models of cars in many countries in Europe in 1993.
By measuring the proportion of oxygen in the remaining exhaust gas, and by knowing the volume and
temperature of the air entering the cylinders amongst other things, an ECU can use look-up tables to
determine the amount of fuel required to burn at the stoichiometric ratio (14.7:1, air : fuel by mass for
MT4760/ SUA/2024 42
gasoline) to ensure complete combustion.
 Defects, Diffusion and Conduction in Ceramics
 Migration of defects plays a very crucial role in determining conduction in ceramic materials. Migration of
ionic defects such as vacancies or interstitials is basically a diffusive process and is governed by the law of
diffusion. The diffusivity of species is a strongly temperature dependent parameter with an exponential
dependence on the temperature. The diffusivity related to electrical conductivity is shown by Nernst-
Einstein relationship.

 Conductivity in the ionic systems also follows a temperature dependent Arhenious type behavior. In ionic
systems, to utilize the ionic conduction of materials, it is essential that ionic conductivity is a dominant
contribution towards the total conductivity which typically happens at either higher temperatures or at
reasonably high doping levels unless the electron concentration is extremely low. The examples of
ceramic materials with high ionic conductivity are materials like fast ion conductors and a few ceramic glasses.

 The presence of composition gradients of charged defects also leads to electrical potential gradients and
this when combined with the chemical potential gradients, gives rise to an expression for electro chemical-
potential, called Nernst equation. This equation is often employed in designing the sensors for use in
automotives where differential pO2 gives rise to a voltage which can be used as a feedback to regulate the
fuel/air supply in the engines. Ceramic conductors are used in a variety of applications, for example, ZnO as
varistors, doped ZrO2 as solid electrolytes in fuel cells and as oxygen sensors in the automobile exhaust.

MT4760/ SUA/2024 43