CHAPTER- 1
METAL OXIDE SEMICONDUCTORS AND THEIR
APPLICATIONS
1.0 INTRODUCTION
Recently, scientists and technologists have made tremendous and significant
advances in the field of materials science, especially in metal oxide semiconducting
materials. Metal oxides are probably the most diverse, rich and multifunctional materials,
with properties covering almost all aspects of materials science and solid-state physics
[1]. Due to their unique characteristics these metal oxides have found important
applications in physical science for examples ferri-, anti- and ferromagnetism,
ferroelectricity, piezoelectricity, superconductivity, magneto resistivity, photonics; in
chemical science just as separation, catalysis ; environmental engineering, etc. [1-6]. In
technological applications, oxides are used in the fabrication of microelectronic
circuits[7],
capacitors[8],
sensors[9],
piezoelectric
devices[10],
fuel
cells[11],
semiconductors[12, 13], oxygen generators[14], organic synthetics[15–19], the
manufacture of engineered ceramics[20], coatings for the passivation of surfaces against
corrosion[21] and as catalysts as both the support and active component[22–24].
All the metal oxides are the compound semiconductors which consist of more than two
elements, some show semiconductor properties. Metal oxides which show semiconductor
properties have the following features according to Wilson’s model [25].
(i)
The conductivity of the semiconductor is electronic. Ionic conductivity is
excluded.
(ii)
Conductivity is largely increased as a function of temperature.
(iii)
Conductivity is very dependent on the kind of impurities and their
concentrations.
Some of important metal oxide materials along with their chemical formulae, their
properties and applications have been summarized in Table 1.1.
1
Table-1.1. The Properties and applications of some exotic metal oxide materials.
Oxides
Property
Applications
Al2O3, CeO2
Hardness
Abrasive
TiO2, CeO2, Fe2O3
Catalysts
Air and water purification
Mo/Al2O3 (Mo = Cu, Ag, Au, Pt,and Pd) Redox catalyst
Ce1-xMxO2-δ (M = Cu, Ag, Au, Pt, Pd, Rh,
and Ru)
Three-way
catalyst
automobile exhausts
Ti1-xM x O2-δ (M = Cu, Ag, Pt,Pd, Mn, Fe, Photocatalyst
Co, Ni and W)
Oxidation of organic matter
for
ZnO, TiO2
UV–Vis sunlight Photocatalyst, sunscreen, and
Absorbing
paint
MTi/ZrO3 (M = Ca, Sr, Ba, and Pb); PZT
Dielectric
Sensors, MEMS
γ-Fe2O3, BaFe12O19, MFe2O4
Super-
Cancer
detection
remediation, sensors
memory devices
paramagnetic
TiO2, Fe2O 3, Cr2O3 M Al2O4, MCr2O4
(M = transition metal ions) M/Al2O3 (M =
Cr3+, Co2+, Ni2+ ),M/ZrO2, RE/M:ZrSiO4(M
= Fe 3+, Mn 2+, V 4+ ; RE = rare earth ion)
Eu3+/Y2O3 (red), Eu
aluminate
Al2O3, ZrO2,
tialite
2+
,Tb
Colors
3+
/Ba-hexa- Luminescence
ZTA, mullite, cordierite, Refractory
and
and
Ceramic pigments
Phosphors-CFL,
picture tube
color
TV
Toughened ceramics
MgO, CaO, and ZnO
Adsorbent
Defluoridation and COD from
paper mill effluents
YSZ(Y2O3–ZrO2), Ni/YSZ, La(Sr)MO3,
Electrolyte
Solid
M = Mn, Cr
Anode /Cathode materials
interconnect
2
oxide
fuel
cell
Cu2O
High absorption Electrode material for solar
SnO2
coefficient
cells
Transparent
Electrode material for solar
conductor
cells
The above mentioned metal oxides have two unique structural features: switchable and/or
mixed cation valences, and adjustable oxygen deficiency, which are the bases for creating
many novel oxides materials with unique electronic, optical, and chemical properties.
However, nanoscale metal oxides are particularly attractive to both pure and applied
researchers because of the great variety of structure and properties, especially those
related to intrinsic size-dependent properties [11, 24, 26-28]. These oxides are usually
made into nanoparticles or thin films in an effort to enhance their surface sensitivity and
they have recently been successfully synthesized into the various nanostructures.
Utilizing these high surface areas of nanostructures, it may be possible to fabricate nanoscale devices with superior performance and sensitivity.
A great variety of metal oxide nanoparticles have been readily synthesized by using alkali
metal oxides, M2O (M is Na or Li) and soluble metal salts (metal chlorides) in polar
organic solutions, for example, methanol and ethanol, at room temperature. The oxidation
states of the metals in the resulting metal oxides (Cu2O, CuO, ZnO, Al2O3, Fe2O3, Bi2O3,
TiO2, SnO2, CeO2, Nb2O5, WO3, and CoFe2O4) range from 1 to 6 and remain invariable
through the reactions where good control of oxygen stoichiometry is achieved. Metal
oxide nanoparticles are 1–30 nm and have good mono dispersivity and displayed
comparable optical spectra. These syntheses are based on a general ion reaction pathway
during which the precipitate occurs when O2- ions meet metal cations (Mn+) in anhydrous
solution and the reaction equation is Mn+n/2O2- → MO n/2 (n=1–6).
Currently considerable interest in nanocrystalline oxide materials exists owing to their
unusual properties. Decreasing particle size results in some remarkable phenomenon. It
has been found that smaller the particles, the
(i)
Higher the catalytic activity (Pt/Al2O3).
3
(ii)
Higher the mechanical reinforcement (carbon black in rubber).
(iii)
Higher the electrical conductivity of ceramics (CeO2).
(iv)
Higher the photocatalytic activity (TiO2).
(v)
Higher the luminescence of semiconductors.
(vi)
Lower the electrical conductivity of metals(Cu, Ni, Fe, Co, and Cu alloys).
(vii)
Higher the blue-shift of optical spectra of quantum dots.
(viii)
Higher the hardness and strength of metals and alloys.
(ix)
Superparamagnetic behavior of magnetic oxides.
(x)
High transmittance and conducting oxide (SnO2)
(xi)
High absorption coefficient at 2.17 eV and energy conversion efficiencies
greater than 10% (Cu2O)
However, preparation of dimension-controlled oxide nanomaterials is difficult because of
the unavoidable conglomeration trends of the nucleation and growth phase during
hydrothermal, calcination, and condensation processes. The precipitation of metal oxides
from both aqueous and nonaqueous solutions is less straightforward than the precipitation
of their metal sulfides or oxy salts. Reactions for the synthesis of metal oxides can
generally be divided into two categories: those that produce an oxide directly [29–33] and
those that produce what is best termed a precursor that must be subjected to further
processing (calcination, dehydration, condensation, etc.) [34–38]. Most of the metal
oxides can be prepared by precipitating the corresponding metal hydroxide, carbonate,
oxalate, and even nitrate products, followed by their subsequent calcination, The
preparation of metal oxide nanoparticles/nanocrystals with different sizes is important for
the continued development of many fields of application, such as catalysis, photonic
devices, electronic devices, and sensors, provided materials can be prepared at controlled
size and reasonable cost. High temperatures (calcination) or high pressures (hydrothermal
treatment) are usually necessary for the reactions, and as such they are usually highenergy processes and the stoichiometry is difficult to control. Among these synthetic
routes, the most promising one is the soft-chemistry route [35, 37, 38], especially
nonaqueous sol–gel preparation, in which good control from the molecular precursor to
4
the final product is achieved, offering high purity and homogeneity, and low processing
temperatures (200–3000C). In comparison to aqueous sol–gel chemistry, the synthesis of
metal oxide materials in organic solvents under the exclusion of water provides some
peculiar features, which allow better control over particle size, shape, crystallinity, and
surface properties [38]. However, those syntheses mentioned above are neither direct nor
simple routes due to the hydrolysis/condensation steps involved.
1.1 SEMICONDUCTOR MATERIALS AND THEIR CHARACTERIZATION
The most widely used type AIIBVI semiconductor materials are the compounds ZnO, ZnS,
CdS, CdSe, ZnSe, HgSe, CdTe, ZnTe, and HgTe. Materials of this type are obtained
principally by means of chemical reactions in the gas phase or by fusion of the
components. Their resistivity and type of conductivity are determined not so much by the
doping impurities as by the characteristic structural defects associated with the deviation
of the materials from stoichiometric composition. The use of semiconductor materials of
this type is connected primarily with their optical properties and photosensitivity. For this
reason, the materials are used in such devices as photoresistors, photocells, electron-beam
devices, infrared detectors, and light modulators. Chemical compounds of the type
AIIIBV—that is, compounds of group III elements with group V elements—make up a
large class of semiconductor materials. This class includes arsenides, phosphides,
antimonides, and nitrides, for example, GaAs, InAs, GaP, InP, InSb, AIN, and BN, The
compounds are obtained by various techniques for producing single crystals from the
liquid and gas phases. The synthesis and growth of single crystals are usually carried out
in closed vessels made of high-strength high-temperature chemically inert materials,
since the pressure of the saturated vapor over a melt of such elements as P and as is
comparatively great. Impurities of group II elements generally give this semiconductor
materials p-type conductivity, and group IV elements impart n-type conductivity. The
materials of this class are used primarily in semiconductor lasers, light-emitting diodes,
Gunn diodes, and photomultipliers. They are also used as film radiation detectors in the
X-ray, visible, and infrared regions of the electromagnetic spectrum. Other
semiconductor materials include some amorphous vitreous chalcogenide systems—such
as alloys of P, As, Sb, and Bi with Ge, S, Se, and Te—and oxide systems—such as
5
V2O5•P2O5•RxOy, where R are metals of groups I-IV, x is the number of atoms of the
metal, and y is the number of oxygen atoms in the oxide. Such materials are used
primarily as optical coatings in instrumentation [39].The properties of semiconductor
materials vary over a broad range with changing temperature. They also change under the
action of, for example, electric and magnetic fields, mechanical stresses, and irradiation.
This variation is made use of in designing various types of sensors. Semiconductor
materials are characterized by the following basic parameters: resistivity, type of
conductivity, width of energy gap, concentration of charge carriers, carrier mobility,
effective mass of carriers, and carrier lifetime. Some characteristics of semiconductors,
such as energy-gap width and effective mass of the carriers, are relatively independent of
the concentration of chemical impurities and the degree of perfection of the crystal
lattice. Many parameters, however, are determined almost entirely by the concentration
and nature of the chemical impurities and structural defects. Some physical properties of
important semiconductor materials are given in Table1-2. In electronic devices,
semiconductor materials are used both in the form of bulk single crystals and in the form
of thin single-crystal and polycrystalline films applied to different substrates, such as
insulating or semiconductor substrates [40]. Such films range in thickness from fractions
of a micron (μm) to several hundred μm. The semiconductor materials used in electronic
equipment must have certain electrical and physical properties that are stable over time
and resistant to ambient conditions during operation [41]. Of great importance are the
uniformity of the properties of semiconductor materials within a single crystal or a film
and the degree of perfection of the crystal structure—for example, the dislocation density
and the concentration of point defects. Because of the high requirements for purity and
structural perfection of semiconductor materials, the technology for the production of
such materials is extremely complicated. High stability of such production conditions as
constancy of temperature, flow rate of the gas mixture, and duration of the process is
necessary. In addition, special conditions must be observed; for example, a certain level
of cleanness must be maintained in the equipment and work area—no more than four dust
particles larger than 0.5 μm, are allowed per liter of air [42]. Depending on the size of the
single crystals and the type of the semiconductor material, the length of the process of
single-crystal growth ranges from a few tens of minutes to several days. The processing
6
of semiconductor materials under industrial conditions involves cutting the materials with
a diamond tool, grinding and polishing the surfaces of the materials with abrasives, heat
treatment, and etching with alkalies and acids. Control of the quality of semiconductor
materials is extremely complicated and varied and is accomplished by means of special
equipment. The main controllable parameters are chemical composition, type of
conductivity, resistivity, carrier lifetime, carrier mobility, and level of doping. Optical,
spectroscopic, mass-spectrometric, and activation techniques are normally used to
analyze the composition of semiconductor materials. Electrical and physical
characteristics are measured by probe methods or through the Hall Effect [43]. The
perfection of single-crystal structure is investigated by X-ray diffraction techniques and
optical microscopy. The thickness of layers is measured either by noncontact optical
methods or by layer abrasion methods.
Looking at the tremendous applications and to the main aim of this thesis work, a
common need is to work with a nanostructured material. In the recent decades,
nanostructure science and technology has become one of the most interesting, diverse and
fast growing research areas in materials science and engineering. Some emerging
multidisciplinary fields of applications have appeared such as nanoelectronics,
nanostructured sensors (nano-nose) and nanostructured solar cells. A clear classification
and discussion on the use of nanomaterials can be found in the pioneering work of Gleiter
and coworkers in the early 1980s [44]. The main idea behind the nanoscience is to control
and/or engineer the structural, physical, chemical or biological properties of materials on
the nanometer (atomic) scale. It is worth stating that, in some cases the properties of these
materials can be very different (most often superior) in comparison to the macroscale
(bulk) properties of the same material. Most of the properties of say, a homogeneous bulk
spherical solid material with macroscopic dimensions are related to its crystal structure
and chemical composition. The surface atoms of this bulk material comprise a negligible
proportion of the total number of its constituent atoms and hence play a negligible role in
the observed (bulk) properties of the material. However, that surface atoms may play a
predominant role in properties involving exchanges at the interface between the material
and the surrounding medium such as crystal growth, chemical reactivity and thermal
conductivity. When the size of the particles is reduced to the nanometric scale, the
7
proportion of atoms located at a surface area is considerably high in relation to the total
volume of atoms in the material. This has a strong effect on the materials properties. For
example, at 5nm (ca. 8000 atoms) the proportion of surface atoms is estimated to be
about 20 % whilst at 2nm (ca. 500 atoms), it stands at 50%.Assuming that the particles
are spherical in shape, then the surface area to volume ratio can be given as S/V = 3/r
where r is the radius of the particle. Decreasing the particle radius increases the surface
area to volume ratio. For instance, a 1 cm3 of 1 nm sized particles would have an active
surface area of about 100 m2 [45]. From the literature [46], it is known that the effective
thermal conductivity of suspensions containing spherical particles increases with the
volume fraction of the solid particles. Since heat transfer takes place at the surface of the
particles, it is preferable to use particles with a large surface area to volume ratio. Thus, if
nanometer-sized particles could be suspended in traditional heat transfer fluids, a new
class of engineered fluids with high thermal conductivity, called “nanofluids” [47-48],
and highly sensitive gas sensors such as the nano-nose could be fabricated.
Because of the extremely small size of the grains (domains) of nanomaterials, a large
fraction of the atoms in these materials is located in the grain boundaries which impede
movement of dislocations, thereby allowing the material to exhibit a superior physical,
mechanical, magnetic, electronic and biometric properties in comparison with coarsegrained/bulk (>1 μm) materials. This phenomenon is attributed to the fact that the grain
boundary energy of nanocrystalline powders (see Chapter-3) is larger than the grain
boundary energy of fully equilibrated grain boundary [49]. Furthermore, these materials
show increased strength, high hardness, extremely high diffusion rates, and consequently
reduced sintering times for powder compaction.
Because of the small size particles, semiconductor nanoparticles may show quantum
confinement, a phenomenon, which arises due to the fact that the electronic energy levels
do not form a continuous set but rather, discrete in nature [50]. Hence, emissions from
excited nanoparticles tend to show size-dependent vibrational frequencies: this property
makes most nanoparticles useful in memory storage, sensor and electronics technologies
[51, 52]. Most semiconductor and metallic nanoparticles show strong particle size, shape
and surrounding media dependent optical properties. The synthesis of nanomaterials and
8
the creation of nanostructures are achieved mainly through two complementary
approaches identified as top-down and bottom-up as shown in Figure 1.1.
The top-down approach involves whittling down the size of materials from the bulk
(macroscopic) to the nanometer scale. This approach generally relies on physical
processes, or a combination of physical and/or chemical, electrical or thermal processes
for their production. Usually the top-down approach is cost effective but the control over
the produced material is poor. The bottom-up approach, on the contrary, involves
assembling atom-by-atom, or molecule-by- molecule into structures on the nanometer
Table 1.2 Some physical properties of important semiconductor materials.
Type
of Material
material
Width
of Carrier mobility Crystalenergy gap at
300°K structure
electron
(cm2/volt-sec)
type
volts
Lattice
constant
(angstrom
units)
Melting Vapor
point
pressure
at
(°c)
melting point
(atmospheres)
At
At
Electrons Holes
300°K 0°K
Element
C
5.47
(diamond)
5.51 1,800
1,600 Diamond
3.56679
4027
Ge
0.803
0.89 3,900
1,900 Diamond
5.65748
937
Si
1.12
1.16 1,500
600
Diamond
5.43086
1420
Diamond
6.4892
–
0.08
a-Sn
AIVBIV
compound
a-SiC
3
3.1
AIIIBV
compound
AlSb
1.63
1.75 200
BP
GaN
400
10–9
10–6
50
Zinc blende 4.358
3100
420
Zinc blende 6.1355
1050
<0.02
6
Zinc blende 4.538
>1300
>24
3.5
Wurtzite
3.186 (a=axis) >1700
>200
5.176 (c=axis)
GaSb
0.67
0.80 4,000
1,400 Zinc blende 6.0955
706
<4 ×10–4
GaAs
1.43
1.52 8,500
400
Zinc blende 5.6534
1239
1
GaP
2.24
2.40 110
75
Zinc blende 5.4505
1467
35
InSb
0.16
0.26 78,000
750
Zinc blende 6.4788
525
<10–5
9
AIIBVI
compound
InAs
0.33
0.46 33,000
460
Zinc blende 6.0585
943
0.33
InP
1.29
1.34 4,600
150
Zinc blende 5.8688
1060
25
CdS
2.42
2.56 300
50
Wurtzite
1750
4.16 (a=axis)
6.756 (c=axis)
CdSe
1.7
1.85 800
ZnO
3.2
ZnS
3.6
3.7
Zinc blende 6.05
1258
200
Cubic
4.58
1975
165
Wurtzite
3.82 (a=axis)
1700
scale with properties varying according to the number of constituent entities/grain size.
Building the system atom by atom or molecule by molecule guarantees the best control
over all particles in the system. Colloidal dispersions such as micro-emulsions are a good
example of the bottom-up concept of nanoparticles synthesis.
1.2 FUNDAMENTALS OF TRANSPARENT CONDUCTING OXIDES (TCO’s)
In our daily lives transparent conducting oxide (TCO) materials are used in numerous
devices. These applications are mostly found in display technology, (organic) lightemitting-diodes (OLEDs), thin-film solar photovoltaics (PV) and ‘smart windows’. A
combination of transparency and conduction can be achieved in two classes of material.
The first group is formed by extremely thin (~10 nm) metal films, especially Ag, Au or
Cu. The luminous transmittance can go up to 50%, and even higher when the metal is
sandwiched between anti-reflecting layers [53]. The conductivity is strongly thickness
dependant, and therefore quite low for these very thin films. The second class of
materials consists of wide band gap semiconductors. As early as 1907, coexistence of
electrical conductivity and optical transmittance was first observed in cadmium oxide
[54]. However, the technological advances in transparent conducting oxides (TCO’s)
emerged only after 1940, as the potential applications in industry and research became
evident. In 1956 Thelen et al. [55] found transparency and conduction in indium oxide,
which was later used in applications for heated windows. Years of extensive research
finally led to SnO2 doped In2O3 (known as Indium Tin Oxide or (ITO) with excellent
electrical and optical properties [56]. Hence at present, the ITO is commonly used as
transparent electrodes, especially in large-area applications such as display devices. As
10
can be seen from figure 1.2 display devices have quickly grown to become the primary
end user of the world’s indium production. The supply of indium has become unreliable
during the last 20 years [57], whereas the display market is still expanding. Indium prices
are therefore subjected to major instabilities. Alternative materials are welcome to
manufacturers of flat panel displays, and much research is conducted on suitable
candidates. With the discovery of p-type TCO’s a new field in optoelectronics device
technology has opened up [58]. In contrast to passive electrodes, the use of transparent
conductors as active components became possible. This has led to new applications as
blue or ultra-violet (UV) light-emitting-diodes [59-61].
Figure 1.1 The two complementary approaches to nanoparticles synthesis.
11
Figure 1.2 World-wide indium consumption by end use in 2004. The majority of indium
used in thin film applications is the display industry.
Furthermore, it became possible to design all-transparent devices [62, 63]. Although the
fabrication of transparent circuits is nowadays limited to laboratory research [64], they
are believed to find revolutionary applications in the next decades. However, appropriate
p-type TCOs for this purpose are not available yet [65]. Moreover, most are far from
suitable for processing on polymeric materials.
It is evident from the above discussion that devices employing TCO’s become ever more
sophisticated and that the demand for implementation of polymer materials is increasing.
The process conditions of TCO’s need to be adjusted to allow coating of such heatsensitive substrates. Under these circumstances it is important to optimize the TCO
performance in order to meet the technological requirements. Understanding the
fundamental properties is an important for the improvement of existing materials.
Moreover, these insights are of great scientific importance for the design and synthesis of
new type of TCO’s. The use of a deposition technique that allows indirect, but precise
control of the fundamental properties can assist the research of high-performance TCO’s
on substrates.
Limitations of TCO have become more critical as passive and active devices based on
these materials get more sophisticated. For example, as displays become larger and
writing speeds get faster, it becomes increasingly important to decrease resistivity while
the transparency is maintained [66]. Simply increasing the thickness is not acceptable
12
since the optical absorption will increase. A profound understanding of the fundamental
aspects of transparent semiconductors is therefore required in order to improve either the
properties of existing materials, or design new type of TCO’s. These insights are of great
scientific importance whether the realization of high-performance TCO’s on polymer
substrates is possible.
A vast amount of literature is available on different TCOs deposited by various growth
techniques and their fundamental aspects. Reviews on TCO’s concerning these issues
have been reported earlier [67, 68]. The next sections will discuss the electronic and
optical properties of TCO’s, respectively. Most of the considerations refer to indium
oxide and indium tin oxide as an example. Still, they hold for all other types of n-type
transparent and conducting oxides if the correct material related numbers are used in the
calculations. In most cases this also holds for p-type semiconductors. The carrier density
then refers to the hole concentration in the material. However, mobility and carrier
density in p-type wide band gap semiconductors usually differ by a few orders of
magnitude. The conduction mechanism can therefore be quite different.
1.2.1 ELECTRICAL AND OPTICAL PROPERTIES OF TCO
The conductivity (σ) is a product of the number of charge carriers (n) in a material, and
the mobility (μ) of these charge carriers, times the elementary electron charge (e). The
resistivity (ρ) is defined as the inverse of the conductivity.
σ = e. n. μ =
𝟏
𝝆
………………(1.1)
For thin films of uniform thickness d, the electrical resistance is sometimes expressed as
the sheet resistance (Rs=ρ/d). Other than the thickness, the sheet resistance is independent
of the film dimensions. In order to promote conductivity, the number of charge carriers
can be increased by doping. Dependant on the material this can be done by substitutional
doping, creation of vacancies or implantation of interstitials. Dependant on the valence of
dopants or vacant sites, acceptor or donor states will induce p- or n-type conductivity.
Another possibility to enhance the conductivity is to increase the mobility. However, the
mobility is dependent on intrinsic scattering mechanisms, and can therefore not be
controlled directly. In general these mechanisms limit the mobility as the carrier density
increases. As a result, the mobility is the most important parameter influencing the total
conductivity.
13
The presence of a bandgap, providing low absorption in the visible range, is an essential
feature of TCO’s. Metal-oxide semiconductors having a bandgap of at least 3.0 eV meet
this condition. The top of the valence band is mostly formed by the oxygen 2p bands,
whereas the bottom of the conduction band is composed of a single and highly dispersed
metal s band. The O 2p orbitals are low in energy and it is therefore that a large band gap
can be obtained in oxides. In intrinsic stoichiometric oxides, coexistence of electrical
conductivity besides visible transparency is not possible. However, substitutional doping
by cationic donors or anion vacancies can create charge carriers, i.e. electrons. The donor
(or acceptor) states alter the electronic band structure of the material. For increased donor
density, the donor states merge with the conduction band at a certain critical density nc,
whose magnitude can be estimated by Mott’s criterion [69, 70].
𝟏/𝟑
𝐧𝐜 . 𝐚∗𝟎 ≈ 0.25
The effective Bohr radius a0* is given by:
𝒂∗𝟎
h 2ε εm
0
=
2
πe m*c
...…………………….(1.2)
……………………..(1.3)
Where εm is the static dielectric constant of the host lattice (equals 8.9 for In2O3 [71]), and
mc* is the effective mass of the electrons in the conduction band. After comparing
experimental data of several authors, Kostlin et al. [72] calculated the effective mass of
In2O3 to be about a third of the electron rest mass; i.e. mc*=0.35·me. Using these
numbers, one can obtain a0* ≈1.3 nm, and the critical density nc is calculated as 6.4 x1018
cm-3. In general, the carrier density in TCO’s is by far larger than this (up to 1021 cm-3 in
ITO). Above this Mott critical density, free electron behavior can be expected. The donor
states have merged with the conduction band and the material is said to be degenerate.
The Fermi energy EF is determined by the highest occupied state of the conduction band,
and one can write:
EF 
2 k 2
2 m c*
……………..……….(1.4)
The band structure of tin oxide is approximated by parabolic functions of k close to the
band edges. The schematic representation in figure 1.3 is valid for most binary metal
oxide wide band gap semiconductors. The valence band maximum and conduction band
14
minimum are both located at k=0, so the material is said to be a direct band
semiconductor. If optical transitions or carrier transport occur far from the conduction
band bottom or valence band top, the shape of the bands are slightly distorted. This band
distortion is caused by interactions with other bands and many-body effects. It is
therefore hard to determine the effective mass theoretically in these regions, since this
value is determined by the band curvature. For theoretical calculations on the band
structure, non-parabolic effects should be taken into consideration [73].
1.2.1(a) Scattering mechanisms
As said earlier, the conductivity of TCO’s is reflected from the mobility that can be
expressed as:-
e.
e .
mfp


m* m*. V
F
………..……………..(1.5)
The relaxation time τ depends on the drift velocity VF and mean free path λmfp of the
charge carriers. The parameters are affected by the different scattering mechanisms;
lattice scattering, ionized impurity scattering, neutral impurity scattering, electronelectron scattering, electron-impurity scattering and grain boundary scattering [74]. The
total mobility can therefore be written as:
1
1

tot i i
………………………….(1.6)
In crystalline TCO’s all scattering mechanisms have little effect, except the ionized
impurity scattering [71]. However, TCO’s deposited at lower temperatures posses a lower
crystalline nature, and high doping concentrations result in the formation of neutral
complexes [75]. In these cases grain boundary and neutral impurity scattering should also
be taken into consideration.
Ionized Impurity Scattering
The Coulomb interaction between the ionized (donor) impurities and the free electrons
provide a source of scattering that is intrinsic to the doped material, and can therefore set
a lower limit to the resistivity, regardless of other scattering mechanisms such as neutral
impurities, grain boundaries or structural disorder. The attainable resistivity due to
ionized impurity scattering was first calculated by Brooks (1955) [76] and Dingle (1955)
15
[77] using the Born approximation band Thomas- Fermi screening. The theory assumes
that the ionized impurities form a uniform background of immobile charges.
The mobile charges (electrons or holes) provide screening. The relaxation time of an
electron of wave vector k is then given by:
Ni Z 2e4m*
1

f (k )
 (k ) 8 (  ) 2 3k 3
r
0
……………….(1.7)
Here is Ni the number of scattering centers per unit volume and Z the charge on the
impurity, ε0 represents the permittivity of free space whereas εr is the low frequency
relative permittivity. f(k) is the screening function and is dependent on the Thomas-Fermi
screening wave vector. For a degenerate system it is only necessary to consider the
scattering of electrons at the Fermi surface (k=kF). The upper limit of the mobility can
now be expressed by:-
24 3 (  r ) 2 3n 1
0
iis 
2
f (k )
Ni Z e3m*2
F
……………….(1.8)
Later, Moore [78] made corrections to the original work of Dingle based on second order
terms in the Born approximation. This resulted in a slightly lower mobility limit.
Pisarkiewicz et al. [79] applied corrections to this model by taking into account the nonparabolic band shape. This caused a larger effect on the intrinsic limit. According to his
approximation, the effective mass at the Fermi-energy is dependent on the carrier
concentration as;
2
m*  mo* 1  2C
(3 2n)2/ 3
*
m
0
………………….(1.9)
where C is a constant used to fit experimental data. Equations 1.7 and 1.8 can be used to
calculate the intrinsic mobility limit as a function of carrier concentration (n).
Grain boundary scattering
The grain boundary is a complex structure, usually consisting of a few atomic layers of
disordered atoms. These incomplete atomic bonds induce a large number of defects. Here
charge carriers are trapped; i.e. immobilized. The traps become electrically charged,
creating a potential energy barrier, reducing the mobility of free carriers from moving
from one crystallite to the other. Seto [80] assumed a barrier with thickness d, which is
16
considerably smaller than grain size L. Further the grain boundary contains Qt traps per
cubic centimeter located at energy Et with respect to the intrinsic Fermi level. All charges
in the region with thickness d around the grain boundary are assumed to be trapped, and
form a depletion layer. The grain boundary can now be thought of as a potential barrier
for electrons characterized by its height Eb. The contribution μgb is thermally activated
and can be described by the Petritz [81] relation:
E
Lq
  0 exp( b ), with 0 
gb
k T
2m*k T
b
b ……………..(1.10)
The energy barrier height is:
q 2Qt2
Q
E 
, with N  n  t d
b 8  r N
d
0
d
…………………(1.11)
Where q is the charge of the trap, kb is Boltzmann constant and T is the absolute
temperature. Nd is the total number of ionized donors. Free fitting parameters are the
number of traps Qt and barrier thickness d. This model was later applied to TCO
materials [79], and data of polycrystalline ITO [82], F: SnO2 [83] and ZnO [84] films was
fitted to this relation. It has been shown that the energy barrier height is approximately
0.01 eV [74, 82].
The grain boundary scattering has an effect on the total mobility only if the grain size is
approximately of the same order as the mean free path of the charge carriers (L ~λmfp).
The mean free path can be calculated for known carrier density and mobility. Using a
highly degenerate electron gas model [85] the electron velocity VF is:
V  (3 2 )1/ 3( / m*) n1/3
F
……………………(1.12)
Substitution in equation 2.5 results in the following expression for the mean free path
λmfp:
m*
  VF   VF
 (3 2 )1/ 3 ( / e) n1/ 3
e
………………(1.13)
Neutral impurity scattering
The solubility of dopants in oxide semiconductors can be quite high. The solid solubility
limit of Sn in ITO for instance is found to be well over 60% [72]. At high doping
17
concentrations, an increasing amount of dopant material remains inactive, and can form a
variety of neutral complexes [86]. Therefore neutral impurity scattering might be taken
into consideration. The contribution to the mobility μN is given by [87, 88, 89].impurities
(dash dotted curve). μiis was calculated for double ionized donors (oxygen vacancies;
Ni•Z2=2n) resulting in larger contribution.
N 
m*e3
20  r  3
0 N
………………….(1.14)
Here μN is the concentration of neutral impurities. This value is often estimated from the
difference between measured carrier density and carrier concentration. The ionized
impurity and grain boundary scattering on the other hand, show a direct relation with the
carrier concentration.
Figure 1.3 Schematic representation of the band structure of undoped (a) and doped (b)
metal oxide wide band gap semiconductor in the vicinity of the top of the valence band and
bottom of the conduction band. The grey areas denote the occupied states.
18
Figure 1.4 The upper limit of mobility (solid line) influenced by the effect of scattering on
ionized impurities (dashed curve), grain boundaries (dotted curve) and neutral variety of
neutral complexes [86].
1.2.1(b) Intrinsic conductivity limit
According to the previous sections, the maximum conductivity of wide band gap
semiconductors is restricted by the theoretical upper limit of mobility. This intrinsic
mobility limit is dependent on carrier concentration for each scattering mechanism.
Modeling the upper limit of separate or combined mechanisms for different TCO
materials has been reported frequently [74-90]. In figure 1.2 the upper of mobility was
simulated for polycrystalline indium oxide with intermediate grain size (50 nm). The total
mobility limit is determined by the contributions of the scattering mechanisms from
relations 1.8, 1.10 and 1.11. The non-parabolicity of the conduction band has been taken
into account in the calculations. It can be seen that the importance of electron scattering
on neutral impurities has negligible effect on the total mobility. The amount of neutral
impurities in relation 1.11 was even overestimated since it was set to half the number of
free carriers. The values for the mobility reported in literature are all located in the
marked areas of figure 1.2. Best performing films are found in the region where the
ionized impurity scattering is dominant (horizontal lines). If the grain size is small
enough, TCO thin films up to intermediate carrier density are affected by grain boundary
scattering as well (vertical lines).
19
Although most authors agree on the mobility limit due to ionized impurity scattering, the
mobility due to grain boundary scattering is modeled differently in many cases.
Apparently the free parameters are used to fit the obtained data exactly, whereas in
general the average grain size is just a rough estimate. Therefore the influence of grain
boundary scattering in TCOs is still under debate [91]. However, it is commonly accepted
that the mobility is only affected if the grain size is considerably small. And moreover
this effect is only noticeable up to intermediate carrier concentrations, as for high electron
density the ionized impurity scattering dominates.
The average grain size for calculating μgb was taken as 50 nm, and the barrier height was
fixed at 0.01 eV. The number of neutral impurities was taken as 50% of the amount of
donors. Striped areas indicate the spread of data points in literature. The effect of ionized
impurities is dominant (horizontal lines) or grain boundaries can play a role (vertical
lines). From figure 1.5 it becomes evident that the structure of TCO thin films affects the
electrical properties on going from single- to polycrystalline phase. As the grain size
decreases further, the thin film is said to be nanocrystalline or even amorphous. Although
the grain boundary density increases to infinity in the amorphous state, the mobility does
not drop to zero. Namely in some amorphous oxide materials a reasonable mobility can
be obtained, resulting in amorphous conductivity. In most n-type transparent conductor,
the bottom of the conduction band consists of a highly dispersed single s-band. This high
dispersion and s-type character results in a rather uniform distribution of the electron
charge density. The scattering of these states is relatively low, and provides a large
mobility [92]. As mentioned earlier, a high mobility is critical in attaining a high
conductivity. Since the mobility is proportional to the width of the conduction bands, a
large overlap between the relevant orbitals is required [93]. The magnitude of this overlap
is also of great importance in case of polycrystalline or amorphous materials. A large
overlap is namely quite insensitive to the structural randomness of the amorphous state.
The spatial spreading of the spherical s-orbitals should be large enough to create
continuous conduction paths. In particular for post transition metal cations, a direct
overlap between neighboring metal s-orbitals is possible [92, 94]. The conduction paths
become insensitive to the distorted metal-oxygen-metal bonds in amorphous materials as
20
schematically illustrated in figure 1.6 [95]. For this reason, mobility in amorphous oxides
semiconductors can be similar to the corresponding crystalline phase (>10cm2s-1V-1).
1.2.2 OPTICAL PROPERTIES
An important feature of TCOs is the existence of a transmission window covering most
part of the visible spectrum. In literature, the optical transmission is defined as the ratio
between incoming light intensity and transmitted intensity averaged over all values in
between 400 nm and 700 nm. The typical spectral dependence of TOS’s (from [96]) is
schematically shown in figure 1.6. The transmission window is defined by two regions
where no light is transmitted due to different phenomena. At low wavelengths (λ< λgap)
the absorption due to the fundamental band gap dominates. The photon energy in this
near-and deep-UV part of the spectrum is high enough to equal the band gap energy (3-4
eV). This energy is absorbed and transformed to band to band transitions, and no light is
transmitted because of this quantum phenomenon. For longer wavelengths, in the (near)
infrared (IR) part of the spectrum no light is transmitted due to the plasma edge (λ>λp).
Here the light is electronically reflected which can be best described by the classical
Drude free electron theory [97].
In the free electron model, the electrons may be thought of as plasma whose density is
set into motion by the electric field component of the electromagnetic field. The plasma
oscillates at a natural frequency ωp’ the resonance or plasma frequency. This frequency
corresponds to the plasma wavelength λp and is of the order of 1-4 μm for TCO’s [98].
The interaction of free electrons with the electromagnetic field influences the relative
permittivity ε of the material, which is expressed as a complex number:
  ( N  ik ' )2
………………………….…….(1.15)
The real and imaginary part is the refractive index (N) and the extinction coefficient (k’)
respectively. These parameters determine the reflectance and absorptance of the material.
Close to the plasma frequency the properties of the material changes drastically.
In the infrared (IR) part of the spectrum, below this critical value (ω˂ωp’ or λ>λp) the
imaginary part of formula (2.12) is large, and the penetrating wave drops off
exponentially [99]. The real part is negative, and the material has nearunity reflectance.
For ω>ωp (or λ< λp) the imaginary part tends to zero, and absorption is small. The
refractive index is positive and almost constant with frequency according to:
21
N   1/ 2 1  ( p /  )2   
………………....(1.16)
Here ε∞ is the high frequency permittivityI. The TCO behaves like a dielectric and is
transparent in the region for ù>ùp [100]. In this transparent regime the film is weakly
absorbing (k’2< N2) and the transmission can be expressed as [101]:
T  (1  R) exp(d )
……………….(1.17)
R is the zero degree incidence reflectance; d is the film thickness and α is the absorption
coefficient and is dependent on the wavelength according to:

4k '
…………………..(1.18)

Close to λgap the reflectance is zero and the absorption coefficient as a function of
wavelength can be obtained easily from the transmission curve. The following relation
applies for direct allowed transitions:
  h  Eg
….…………………..(1.19)
where hν is the photon energy. The bandgap energy is calculated from the Tauc plot
[102], α2 versus the photon energy hν. If α2 is extrapolated to the x-axis intersection
(α2=0), formula 1.17 implies that the photon energy equals the band gap energy (hν α Eg).
This method is commonplace to extract bandgap energies from transmission data.
1.2.3 CORRELETION OF OPTICAL - ELECTRICAL PROPERTIES
The optical parameters of TCOs are affected by the electrical properties of the material.
The earlier mentioned plasma resonance frequency is not a fixed value, but varies with
the electron concentration. The plasma frequency in relation to the carrier concentration
is expressed by:
p 
n . e2
 0  mc*
………………………...…..…………….(1.20)
At this frequency the dielectric-like visible transmittance equals the metallic-like IR
reflectance (T=R). Thus the IR reflectivity of the material can be tuned, which is
Important for heat reflecting or low emissive window applications. For example, the
plasma wavelength of ITO films can be tuned from 1.5 μm to 4 μm due to the carrier
22
concentration by changing the composition and deposition parameters [103,104]. The
refractive index of commonly TCO materials varies in between 1.7 and 2.1 [68].
Moreover, a large spread for similar material is frequently reported. Main reason is that
also the refractive index is dependent on the carrier concentration according to:
N2 
4ne2
opt  * 2
m 
…….…………………..(1.21)
Here ω is the frequency of incoming light. Changing the refractive index is useful for
waveguide applications. For instance in ITO this value can be tuned between 1.70 and
2.05 [105,106]. For heavily doped oxide semiconductors a gradual shift of the bandgap
towards higher energy as the electron density increases is generally observed [71, and the
references therein]. This well known effect is attributed to the Burstein- Moss shift (BM
shift) [107]. In heavily doped semiconductors, the lowest states in the conduction band
are blocked. Hence transition can only take place to energies above EF (see figure 1.3),
enlarging the effective optical gap. The energy gap between the top of valence band and
Figure 1.5. Schematic illustration of the electron density in the atomic orbitals responsible
for the carrier transport paths in transparent conducting oxides consisting of light metal
cations (a), heavy metal cations in crystalline phase (b) and amorphous phase (c). In ‘ns
orbitals’, the n denotes the principal quantum number.
23
Figure 1.6. Spectral dependence of semiconducting transparent materials: λgap and λp are
the wavelengths at which the bandgap absorption and free electron plasma absorption takes
place.
lowest empty state in the conduction band (both assumed parabolic) can be given by:
Eg  E
g0
 EgBM   
…………………….(1.22)
In this formula Eg0 is the intrinsic bandgap and the BM shift given by:
2
E gBM 
(3 2n)2 / 3
*
2m
VC
.....…………………(1.23)
∗
Here 𝑚𝑉𝐶
is the reduced effective mass of the electron carriers given by:
1
1
1


m*
m* m*
V
VC
C
......………………………….(1.24)
Where 𝑚𝐶∗ and 𝑚𝑉∗ are effective mass of the carriers in the conduction and valence band
respectively. The term .Ó in the equation (1.20) represents self energies due to electronelectron and electron-impurity scattering, causing a band gap narrowing that counteracts
the BM shift. This effect is of importance at very high carrier concentrations (order 1021
cm-3).
24
Some n-type and p-type TCOs is given in table 1.3 and 1.4, respectively. The references
in these tables are not selected on best performance of the TCO, but are merely a
selection of the vast amount of literature on these materials. It illustrates the diversity of
materials and corresponding properties deposited by different deposition methods.
Among these are chemical vapor deposition (CVD), sputtering techniques (DC, RF,
magnetron), evaporation (reactive, thermal or e-beam), spray pyrolysis and pulsed laser
deposition (PLD). Particular deposition techniques for TCO manufacturing are chosen for
several reasons as thickness uniformity, low production costs or high throughput.
However, the electrical and optical performance is not directly related to the deposition
method. They are more dependent on the intrinsic properties as structure, morphology
and composition of the thin film. Each deposition method and conditions can influence
the intrinsic behavior differently. Although each technique has its own advantages or
limitations, they are all capable of turning the intrinsic properties within a specific range
in order to optimize the TCO’s performance.
1.2.4 CONCLUDING REMARKS ON TCO’s
Coexistence of transparency and conductivity in semiconductors is possible if the
bandgap is large enough to avoid visible light absorption (~3 eV). This gap is situated in
between parabolic O 2p and metal s bands, forming the valence and conduction band
respectively. Intentional doping creates free carriers (electrons or holes) which are
responsible for the conductivity. Electron densities are in general high for n-type TCOs
(order 1020 cm-3) resulting in degenerate electron systems. The electrical conductivity is
dominated by scattering mechanisms, which are strongly related to the electron
concentration. All electronic scattering contributions together influence the mobility, and
determine the upper limit of conductivity. For single-crystalline materials the ionized
impurity scattering is considered to be the most important. Though, as the structural
nature of the thin film decrease, the contribution of grain boundaries becomes ever more
important in polycrystalline films. The transparency window for electromagnetic waves
between UV and near-IR is typical for TCOs. This window is on the short wavelength
side determined by the bandgap and on long wavelength side by the plasma frequency.
Both values are dependent on the electron concentration, and can therefore be tuned to
serve specific applications. The electrical conductivity of common n-type materials is
25
around 10-4 Ωcm, whereas the transmission can be as high as 90% in the visible regime.
The conductivity of p-type TCOs is in general at least a factor of 1000 lower. Although
the figure of merit is a measure for the TCO electrical and optical performance, in
practice the properties are tailored for the different applications. Many deposition
methods can be used to grow TCO’s, and varying performances are reported. Each
deposition method has its own influence on the intrinsic properties as thin film structure
and composition. It is these properties and not the deposition method by itself
determining the electrical and optical performance of the TCO’s. For the purpose of
obtaining lower resistivities, various TCO semiconductor materials have been developed;
n-type TCO semiconductors now available for thin-film transparent electrodes are listed
in table 1.5, grouped by compound type. Nevertheless, there has been no report on the
preparation of a p-type TCO thin film suitable for use as a practical transparent electrode.
One advantage of using binary compounds as TCO materials is the relative ease of
controlling the chemical composition in film depositions compared to using ternary
compounds and multicomponent oxides. Up to now, various TCO thin films consisting of
binary compounds such as SnO2, In2O3, ZnO and CdO have been developed, with
impurity doped SnO2 (SnO2:Sb and SnO2:F), impurity-doped In2O3 (In2O3:Sn, or ITO)
and impurity-doped ZnO (ZnO:Al and ZnO:Ga) films in practical use. In addition, it is
well known that highly transparent and conducting thin films can also be prepared using
metal oxides without intentional impurity doping.
1.4 METAL OXIDE SEMICONDUCTOR AS GAS SENSOR
Another field in which oxides play a dominant role is in solid state gas sensors. A wide
variety of oxides exhibit sensitivity towards oxidizing and reducing gases by a variation
of their electrical properties. There is an obvious close relationship between the gas
sensitivity of oxides and their surface chemical activity and thus gas sensing applications
and catalytic properties should be considered jointly. The great diversity of oxide
materials could not be better demonstrated than in the variety of self-assembled nanoscale
materials that have been recently discovered. Thus the metal oxides offer strong and
26
Table 1.3 Reported properties of n-type transparent conducting oxides.
TCO
Deposition
ρ (Ωcm)
method
T
FOM
Eg
(%)
(10-3
Ω-1)
(eV)
N
n
Μobility
(cm-3)
(cm2/s V)
Ref.
SnO2
Spray
4.3x10-3
97
1.45
4.11
SnO2
Sputtering
6.1x10-3
95
56.4
4.13
1.3x1020 7.7
[109]
SnO2:F
Spray
5x10-4
>80
4.41
4.6x1020 28
[110]
SnO2:Mo
Reactive ev
3x10-3
>85 >0.0.7
4.10
8.0x1020 10
[111]
SnO2:Sb
Spray
10-3
85
3.75
7.0x1020 10
[112]
Cd2SnO4
Sputtering
5x10-4
>80
5.0x1020 40
[113]
Cd2SnO4
Sputtering
5x10-4
93
34.6
2.7
5.0x1020 22
[114]
CdIn2O4
Sputtering
2.7x10-4
90
69.7
3.24
4.0x1020 57
[115]
In4Sn3O12 Sputtering
3.5x10-4
>80
3.5
7.0x1020 11.5
[116]
In2O3
Thermal ev
2x10-4
>90
3.56
4.0x1020 70
[117]
In2O3
PLD
2x10-4
86
9.0x1020 37
[118]
In2O3:F
CVD
2.9x10-4
>85 3.9
GaInO3
Sputtering
2.5x10-3
90
ITO
e-beam ev.
2.4x10-4
90
ITO
CVD
1.7x10-4
90
183
3.9
ITO
Sputtering
2,4x10-4
95
70.4
4.0
ITO
PLD
8.5x10-5
85
72.9
ITO
Sol-gel
5.0x10-3
ITO:F
Sputtering
6.7x10-4
>80 3.5
In2O3:Mo
Sputtering
5.9x10-4
90
ZnO
Reactive ev
10-3
88
19.6
2.2
2.05
2.1
[108]
3.5
[119]
4x1020
14
3.85
7.7
27
2.0
2.0
10
[120]
8.0x1020 30
[121]
8.8x1020 43
[122]
1x1020
[123]
12
1.4x1021 53.5
[124]
1.9x1020 12
[125]
6.0x1020 16
[126]
4.3
5.2x1020 20.2
[127]
3.3
1x1020
[128]
10
ZnO
Sputtering
2x10-3
>80
ZnO:Al
Sputtering
10-2
90
ZnO:Al
CVD
3.3x10-4
85
49.2
ZnO:Al
PLD
3.7x10-4
90
28.3
ZnO:Ga
Sputtering
10-3
>85
ZnO:In
Sputtering
2x10-2
>80
Zn3In2O6
PLD
1.0x10-3
85
ZnSnO3
Sputtering
4x10-3
>80
4
3.52
3.8
1.98
3.59
3.29
1.85
3.4
1.2x1020
16
[129]
4.7x1020
1.47
[130]
8.0x1020
35
[131]
8.0x1020
18
[132]
10x1020
10
[133]
7x1019
1.9
[134]
4.0x1020
20
[135]
1x1020
10
[136]
Table 1.4 Reported properties of p-type transparent conducting oxides.
TCO
Deposition
σ (Scm-1)
method
n (cm-3)
µ (cm2/s V)
T
Eg
S
Ea
(%)
(eV)
μV/K
(eV)
+214
0.22
2.7x1019
0.13
[137]
0.12
1.8x1019
0.16
[138]
6.1x1017
0.46
[139]
>1.0
[140]
CuAlO2
PLD
0.34
70
3.5
CuAlO2
CVD
2
˂70
3.75
SrCu2O2
PLD
4.8x10-2
75
3.3
+260
0.10
CuYO2:Ca
Thermal ev.
1.0
50
3.5
+275
0.13
AgCoO2
Sputtering
0.2
50
4.15
+220
0.07
CuGaO2
PLD
6.3x10-2
80
3.2
+560
ZnO:P
Sputtering
1.7
ZnO:N
PLD
0.5
3.35
85
+
Ref.
mobility
[141]
1.7x1018
0.23
[142]
1.0x1017
0.53
[143]
6x1018
0.1
[144]
Table 1.5 TCO semiconductors for thin-film transparent electrodes.
Material
Dopant or compound
SnO2
Sb, F, As,Nb, Ta
In2O3
Sn, Ge,Mo, F, Ti,Zr, Hf,Nb, Ta,W,Te
ZnO
Al,Ga, B, In,Y,Sc, F, V, Si,Ge, Ti,Zr, Hf
28
CdO
In, Sn
ZnO–SnO2
Zn2SnO4, ZnSnO3
In2O3–SnO2 In4Sn3O12
ZnO–In2O3 Zn2In2O5, Zn3In2O6
CdO–SnO2
Cd2SnO4, CdSnO3
CdO–In2O3
CdIn2O4
MgIn2O4
GaInO3, (Ga, In)2O3
Sn, Ge
CdSb2O6
Y
ZnO–In2O3–SnO2
Zn2In2O5−In4Sn3O12
CdO–In2O3–SnO2
CdIn2O4−Cd2SnO4
ZnO–CdO–In2O3–SnO2
versatile base materials for development of novel magnetic, humidity, LPG, gas sensors,
[145]. A wide variety of metal oxides exhibit sensitivity towards oxidizing (O2, NO2, SO2
etc.) and reducing (H2, hydrocarbon, LPG etc.) gases by a variation of their electrical
properties. The gas sensor based on metal oxide semiconductor generally involves a
catalytic reaction (e.g. oxidation or reduction) of the gas or vapour on the surface of the
sensor. Different oxides like transition metal oxides (SnO2, ZnO, TiO2, Fe2O3), catalytic
oxides (V2O5, MoO3, CuO, NiO), metals deposited on oxide supports (Pt/SnO2, Pt/ZnO,
Pd/SnO2) and mixed (or complex) oxides (La2–xSrxCuO4) exhibit different physical
properties on exposure to different gas species [146]. These oxides are being used for
many applications due to their low price, robustness and simple measurement electronics.
In these devices, electrical conductivity of oxides varies with the composition of the gas
atmosphere surrounding them.Due to the recent advent of nanoscience , nanostructured
semiconductor metal oxides as gas-sensing materials have attracted great attention for a
long time due to their advantageous features, such as high sensitivity under ambient
conditions, low cost and simplicity in fabrication [147]. Among them zinc oxide (ZnO)
with its wide band gap energy (3.10–3.40 eV) and room temperature resistivity in terms
29
of several mega ohms is one of earliest discovered and most widely applied oxide gassensing materials. This is primarily due to the high mobility of conduction electrons in
the material and good chemical and thermal stability under operating conditions [148,
149]. However, most of the ZnO gas sensors are based on films, whose performance is
severely hampered by the limited surface-to-volume ratio. In addition, these thin film gas
sensors usually have to operate at relatively high temperatures of over 673K [150]. A
typical fabricated high-temperature gas sensor unit is shown in figure 1.9.
1.5 METAL OXIDE SEMICONDUCTOR AS DILUTE MAGNETIC
SEMICONDUCTORS (DMS)
Finally a new and most interesting field in which metal oxide semiconductors play a
crucial role is oxide-diluted magnetic semiconductors (O-DMS). Diluted magnetic
semiconductors (DMSs) combine a non-magnetic semiconductor with a transition metal
(TM) dopant, e.g. Co, Cr, Cu, Fe, Ni, or Mn, resulting in interactions not present in the
parent compounds. Among them are s−d exchange interaction between band carriers and
localized magnetic moments of the TM ions, and the d –d interaction between the ions
themselves. DMSs can be divided into two groups. In the first group are materials where
the magnetic ions concentrate in nanosized regions inside the host semiconductor
material, forming local condensed magnetic semiconductors (CMSs). Second group
consists of materials that have the magnetic ions randomly distributed throughout the
semiconductor matrix [151]. Current research interests in these DMS compounds are
mainly aimed towards
second-generation spintronics applications, where high spin
polarization of the charge carriers and compatibility with current semiconductor materials
used in industry are desirable. For use in industrial scale, these materials should also have
their magnetic Curie temperatures (TC) above room temperature, and they should be
available as both n- and p-type materials for direct use in semiconducting junction
applications like transistors [152]. Most DMS materials studied so far have low TC
values, so they are not suitable for spintronics applications [153, 154]. These include Mndoped III-V group semiconductors, which have been the most studied DMS materials
[155]; for example TC of only 110 K has been reported for Ga1−xMnxAs [156].
30
The O-DMS have attracted a great deal of interest in recent years due to the possibility
of inducing room temperature ferromagnetism (RTFM). These materials are of particular
interest for spintronic (spin transfer electronics) devices such as spin valves [157].
In 2000, room temperature ferromagnetism (RTFM) in Mn-doped p-type GaN and ZnO
was predicted by Dietl et al. using Zener model [158]. These predictions were soon
experimentally confirmed for both materials [159, 160]. Since then, RTFM has been
reported in several other materials, like Co- or Cr-doped AlN [161] and Mn-doped GaP
[162]. Especially dilute magnetic oxides like ZnO are a promising research field towards
novel applications like transparent spintronics. So far, a variety of semiconductor
materials, called diluted magnetic semiconductors (DMS) [157], combine the two
interesting properties: semiconducting and magnetic. Such a compound (figure 1.8(b)) is
an alloy between a non-magnetic semiconductor (figure 1.8(c)) and a magnetic element.
Many DMS of III–V (GaAs) or II–VI (CdTe) types have been obtained by doping
semiconductors with magnetic impurities (Mn, for example) but most of them have a low
Curie temperature (TC) which limited their interest [163]. A theoretical prediction by
Dietl et al [158] also demonstrates that the Curie temperature can be increased above
room temperature in p-type semiconductor-based DMS (see figure 1.9).
Figure 1.7 A typical fabricated high-temperature gas sensor unit.
31
Figure 1.8 The different types of semiconductors: (A) a magnetic semiconductor; (B) a
DMS and (C) a non-magnetic semiconductor.
Figure 1.9 Computed values of Curie temperature for various p-type semiconductors
containing 5% of Mn and 3.5 × 1020 hole cm−3 (from [158]).
The calculations also show that ferromagnetism (FM) is stable in a DMS which is based
on a wide bandgap semiconductor. In addition to that, FM of ZnO-based semiconductors
were investigated by ab initio calculations based on the local density approximation
[163]. Here, a brief review of the experimental work done over the last few years has
been presented.
1.5.1. ZnO-based materials: Zn1−xMxO (M = Co, Mn, V, Fe. . .)
ZnO is a II–VI compound semiconductor with a wide bandgap energy (Eg = 3.35 eV) and
is interesting from the viewpoint of forming a transparent ferromagnetic material. Also,
it has electron (n-type) conductivity naturally but p-type conductivity can also be induced
by using a co-doping technique [164]. Theoretical work on ZnO system done by Sato and
Katayama-Yoshida [163] showed, using first-principles calculations, that doping by 3d
transitionmetal atoms, such as V, Cr, Fe, Co and Ni, in the ZnO matrix exhibits a
ferromagnetic ordering without any additional carrier doping while Ti and Cu result in a
32
paramagnetic state. In the case of Mn doping, the calculations show a ferromagnetic state
is induced by hole doping. Keeping in mind that highly conductive p-type ZnO can be
obtained by the co-doping method [165], the theoretical predictions have opened the
route for experimentalists. Soon after, several doping elements (essentially 3d transition
elements) have been used, including Mg [166], Ni [1167, 168], S [169], V [170], Mn
[171, 172–176], Cr [171, 176], Fe [176, 177] and also Co [171, 176, 178–185].
Many studies were carried on Co:ZnO films which were deposited either by radiofrequency magnetron co-sputtering [175], pulsed laser deposition (PLD) using a KrF
laser [179, 180–182, 184] or a sol–gel method [185]. Due to the wurtzite (hexagonal)
structure of ZnO, which is conserved upon doping [160, 179], the substrates used are
sapphire (0001)-oriented corresponding to the lower mismatch between the film and the
substrate (2%). The films are grown using a ceramic target made by a standard solid-state
reaction technique (ZnO and CoO are mixed together, calcined at 500 ◦C for 7 h and
sintered at 950 ◦C for 12 h in air [184]) but it is also possible to synthesize the films
starting from pure metal targets of Zn and Co [168] or two oxide targets (this technique is
called alternating deposition, AD) of ZnO and Zn1−xCoxO [182]. This method allows a
homogeneous films’ repartition of the Co inside the structure and precise control of the
Co content [168, 182] but this also seems possible using a sol–gel process [185]. But
more importantly the AD method showed a better crystallinity than conventionalmethods
[182]. The typical deposition conditions of laser ablated films are a temperature of 600
◦C, apartial pressure of oxygen close to 10−5–5×10−5 Torr [160, 168, 182, 184] and a laser
energy [160, 168, 178, 184] of 1–3 Jcm−2. The films are sensitive to the deposition
conditions as observed in many oxides. In fact, when the deposition temperature is
relatively low (<600 0C), homogeneous alloy (predominantly paramagnetic) films with a
wurtzite ZnO structure are grown, whereas inhomogeneous films of a wurtzite ZnO phase
mixed with rock-salt CoO and hexagonal Co phases form when the temperature is
relatively high and the oxygen pressure is fairly low (<10−5 Torr) [184]. In the case of
Ni-doped ZnO [159], films with Ni content ranging from 3 to 25 at.% exhibit FM at 2 K
but a superparamagnetic behavior is observed between 30 and 300 K.
33
REFERENCES
1. Z. L. Wang, Functional and Smart Materials – Structural Evolution and Structure
Analysis, Plenum Press (1998).
2. A. F. Wells, Structural inorganic chemistry, 6th ed., Oxford University Press,
Oxford, UK, (1987).
3. W. A. Harrison, Electronic structure and the properties of solids: the physics of the
chemical bond, Dover Publications, New York, (1989).
4. H. H. Kung, Transition metal oxides: surface chemistry and catalysis, Elsevier,
Amsterdam, Netherlands, (1989).
5. V. E. Henrich, P. A. Cox, The Surface Science of Metal Oxides, Cambridge
University Press, Cambridge, UK, (1994).
6. C. Noguera, Physics and chemistry at oxide surfaces, Cambridge University Press,
(1996).
7. J. P. Colinge, K. Hashimoto, T. Kamins, S. Y. Chiang, E. D. Liu, S. S. Peng, P.
Rissman, IEEE Electron Device Lett. 7, 279 (1986).
8. V. Srinivasan, J. W. Weidner, J. Electrochem. Soc. 144, 210 (1997).
9. J. Watson, Sensor & Actuators. 5, 29(1984).
10. B. Noheda, Curr. Opin. Solid-State Mater. Sci. 6, 27 (2002).
11. K. Eguchi, T. Setoguchi, T. Inoue, H. Arai, Solid State Ionics. 52, 165 (1992).
12. C. C. Hsu, N. L. Wu, J. Photochem. Photobiol. A. 172, 269 (2005).
13. S. Lindroos, M. Leskela, Int. J. Inorg. Mater. 2, 197 (2000).
14. D. Golberg, Y. Bando, K. Fushimi, M. Mitome, L. Bourgeois, C. C. Tang, J.
Phys.Chem. B. 107, 8726 (2003).
15. K. Nakagawa, T. Nakata, R. Konaka, J. Org. Chem. 27, 1597 (1962).
16. K. Nakagawa, T. Tsuji, Chem. Pharm. Bull. 11, 296 (1963).
17. K. Nakagawa, J. Sugita, K. Igano, Chem. Pharm. Bull. 12, 1964, 403.
18. K. Nakagawa, J. Sugita, H. Onoue, Chem. Pharm. Bull. 12, 1135 (1964).
19. C. Pierlot, V. Nardello, J. Schrive, C. Mabille, J. Barbillat, B. Sombret, J. M. Aubry,
J. Org. Chem. 67, 2418 (2002).
34
20. C. G. Levi, J. Y. Yang, B. J. Dalgleish, F. W. Zok, A. G. Evans, J. Am. Ceram. Soc.
81, 2077 (1998).
21. A. Nazeri, P. P. Trzaskoma Paulette, D. Bauer, J. Sol-Gel Sci. Technol. 10, 317
(1997).
22. Z. R. Tian, W. Tong, J. Y. Wang, N. G. Duan, V. V. Krishnan, S. L. Suib, Science.
276, 926 (1997).
23. I. E. Wachs, Catal. Today. 27, 437 (1996).
24. G. C. Bond, S. F. Tahir, Appl. Catal. 71, 1 (1991).
25. A. H. Wilson, Proc. Roy. SOC. A133) 458. (1931).
26. Z. L. Wang, Annu. Rev. Phys. Chem. 55, 159 (2004).
27. M. Fernandez-Garcia, A. Martinez-Arias, J. C. Hanson, J. A. Rodriguez, Chem. Rev.
4, 063 (2004).
28. M. E. Franke, T. J. Koplin, U. Simon, Small. 2, 36 (2006).
29. F. Zhu, Z. X. Yang, W. M. Zhou, Y. F. Zhang, Solid State Commun. 137, 177
(2006).
30. W. J. Lee, W. H. Smyrl, Electrochem. Solid-State Lett. 8, 7 (2005).
31. K. Deshpande, A. Mukasyan, A. Varma, Chem. Mater. 16, 489 (2004).
32. H. Y. Dang, J. Wang, S. S. Fan, Nanotechnology. 14, 738 (2003).
33. M. Monge, M. L. Kahn, A. Maisonnat, B. Chaudret, Angew. Chem. 115, 5479
(2003).
34. M. Froba, A. Reller, Prog. Solid State Chem. 27, 1 (1999).
35. M. Kakihana, J. Sol-Gel Sci. Technol. 6, 7 (1996).
36. P. Cousin, R. A. Ross, Mater. Sci. Eng. A. 130, 119 (1990).
37. J. Livage, M. Henry, C. Sanchez, Prog. Solid State Chem. 18, 259 (1988).
38. M. Niederberger, G. Garnweitner, Chem. Eur. J. 12, 7282 (2006).
39. Tekhnologiia poluprovodnikovykh materialov. Moscow, (1961)(Transl. from
English)
40. M. Rodot,. Poluprovodnikovye materialy. Moscow, (1971). (Translated from French.)
41. S.M. Sze, S. M. Fizika poluprovodnikovykh priborov. Moscow, (1973).
(Translated from English.)
42. A.S. Palatnik, and V. K. Sorokin. Osnovy plenochnogo poluprovodnikovogo
35
materialovedeniia. Moscow, (1973).
43. Kristallokhimicheskie, fiziko-khimicheskie i fizicheskie svoistva poluprovodnikovykh veshchestv. Moscow, (1973).
44. H. Gleiter, Progress in Materials Science. 33, 223 (1989).
45. N. Ichinose, Y. Ozaki, and S. Kashu, Springer-Verlag, New York, (1988).
46. J. C. Maxwell, “A Treatise on Electricity and Magnetism,” 2nd ed., Oxford University
Press, Cambridge, (1904).
47. U. S Choi, “Development and Applications of Non-Newtonian Flows”. Ed. D. A.
Siginer and H. P. Wang. Vol. 231/MD-Vol. 66. New York: The ASME, (1995).
48. J. A. Eastman, S. Choi, S. Li, W. Yu, and L. J. Thompson, Applied Physics Letters.
78, 718 (2001).
49. A. Johnson, in: “New Materials by Mechanical Alloying Techniques”, (E. Arzt and L.
Schults, eds.), Informationsgesellschaft Verlag, Calw-Hirsau, 354, (1988).
50. D. Chakravorty, and A. K. Giri. “Chemistry of Advanced Materials”. Ed. Rao, C. N.
R. Boca Raton, FL: Blackwell Scientific Publication, (1993).
51. A.P. Alivisatos, Science. 271, 933(1996).
52. X .Wang, and Q. Gao. Solid State Phenomena. 121, 479 (2007).
53. C.G. Granqvist, A. Hultaker, Thin Solid films. 411, 1 (2002).
54. K. Badeker, Ann. Phys. 22, 749 (1907).
55. A. Thelen, H Konig, Naturwissenschaften. 43, 297(1956).
56. R. Groth, E Kauer, Philips Technical Review. 26, 105 (1965).
57. B. O’Neill, Indium: Supply, Demand & Flat Panel Displays, Presented at Minor
Metals 2004, London, June (2004).
58. A.N. Banerjee & K.K. Chattopadhyay, Prog. Cryst. Growth Charact. Mater. 50, 52
(2005).
59. H. Ohta, K.-I. Kawamura, M. Orita, M. Hirano, N. Sarukura, & H. Hosono, Appl.
Phys. Lett. 77, 475 (2000).
60. K. Tonooka, H. Bando & Y. Aiura, Thin Solid Films. 445, 327 (2003).
61. H. Hosono, H. Ohta, K. Hayashi, M. Orita & M. Hirano, J. Cryst. Growth. 496,
237 (2002).
62. G. Thomas, Nature. 389, 907 (1997).
36
63. J.F. Wager, Science. 300, 1245 (2003).
64. R.E. Presley, D. Hong, H.Q. Chiang, C.M. Hung, R.L. Hoffman & J.F. Wager, SolidState Electron. 50, 500 (2006).
65. H. Kawazoe, H. Yanagi, K. Ueda & H. Hosono, MRS Bull. 25, 28 (2000).
66. D.S. Ginley & C. Bright, MRS Bull. 25, 15 (2000).
67. K.L. Chopra, S. Major & D.K. Pandya, Thin Solid Flms 102, 1 (1983).
68. H.L. Hartnagel, A.L. Dawar, A.K. Jain & C. Jagadish, Semiconducting Transparent
Thin Films (Institute of Physics Publishing, Bristol and Philadelphia, 1995).
69. N.F. Mott, Can. J. Phys, 34, 1356 (1956).
70. N.F. Mott, Philos. Mag. 6, 284 (1961).
71. I. Hamberg & C.G. Granqvist, J. Appl. Phys. 60, R123 (1986).
72. H. Köstlin, R. Jost & W. Lems, Phys. Status Solidi A 29, 87 (1975).
73. J.S. Blakemore, J. Appl. Phys. 53, R123 (1982).
74. M. Chen, Z.L. Pei, X. Wang, Y.H. Yu, X.H. Liu, C. Sun & L.S. Wen, J. Phys. D:
Appl. Phys. 33, 2538 (2000).
75. G. Frank & H. Köstlin, Appl. Phys. A 27, 197 (1982).
76. H. Brooks, in Advances in electronics and electron physics, Chapter 7, p85, (1955).
77. R.B. Dingle, Philos. Mag. 46, 831 (1955).
78. E.J. Moore, Phys. Rev. 160, 618 (1967).
79. T. Pisarkiewicz, K. Zakrzewska & E. Leja. Thin Solid Films 174, 217 (1989)
80. J. Y. W. Seto, J. Appl. Phys. 46, 5247 (1975).
81. R.L. Petritz, Phys. Rev. 104, 1508 (1956).
82. A.K. Kulkami & S.A. Knickerbocker, J. Vac. Sci. Technol. A. 14, 1709 (1996).
83. J. Bruneaux, H. Cachet, M. Froment & A. Messad, Thin Solid Films. 197, 129 (1991).
84. K. Ellmer, J. Phys. D: Appl. Phys. 34, 3097 (2001).
85. C. Kittel, in Introduction to solid state physics, 5th edition, Maruzen,Tokyo,(1985),
86. G. Frank & H. Köstlin, Appl. Phys. A 27, 197 (1982).
87. C. Erginsoy, Phys. Rev. 79, 1013 (1950).
88. S. Noguchi & H. Sakata, J. Phys. D: Appl. Phys. 14, 1523 (1981).
89. Y. Shigesato & D.C. Paine, Appl. Phys. Lett. 62, 1288 (1993).
90. T. Minami, MRS Bull. 38, 15 (2000).
37
91. T.J. Coutts, J.D. Perkins, D.S. Ginley & T.O. Mason, Conference Paper, NREL/CP520-26640, (1999).
92. A.J. Freeman, K.R. Poeppelmeier, T.O. Mason, R.P.H. Chang & T.J. Marks, MRS
Bull. 25, 45 (2000).
93. H. Hosono, N. Kikuchi, N. Ueda & H. Kawazoe, J. Non-Cryst. Solids. 165,
198 (1996).
94. M. Orita, H. Ohta & M. Hirano, Phil. Mag. B. 81, 501 (2001).
95. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano & H. Hosono, Nature. 432,
488 (2004).
96. F. Simonis, M. van der Leij & C.J. Hoogendoorn, Solar Energy Mater. 1, 221 (1979).
97. P. Drude, Ann. Phys. 3, 369 (1900).
98. R.G. Gordon, MRS Bull. 25, 52 (2000).
99. E. Hecht, in Optics, 2nd edition, Addison-Wesley Publishing Company, Inc. (1990).
100. T.J. Coutts, D.L. Young & X. Li, MRS Bull. 25, 58 (2000).
101. J.C. Manifacier, M. De Murcia, J.P. Fillard, E. Vicario, Thin Solid Films 41, 127
(1977).
102. J. Tauc, R. Grigorovici & A. Vancu, Phys. Status Solidi 15, 627 (1966).
103. P.K. Biswas, A. De, N.C. Pramanik, P.K. Chakraborty, K. Ortner, V. Hock & S.
Korder, Mater. Lett. 57, 2326 (2003).
104. H. Kim, J.S. Horwitz, A. Piqué, C.M. Gilmore & D.B. Chrisey, Appl. Phys. A 69,
447 (1999).
105. R.T. Chen & D. Robinson, Appl. Phys. Lett. 60, 1541 (1992).
106. Y. Ohhata, F. Shinoki & S. Yoshida, Thin Solid Films. 59, 255 (1979).
107. E. Burstein, Phys. Rev. Lett. 93, 632 (1954).
108. H.H. Afify, R.S. Momtaz, W.A. Badawy & S.A. Nasser, J. Mater. Sci.: Mater.
Electron. 2, 40 (1991).
109. A. De & S. Ray, J. Phys. D: Appl. Phys. 24, 719 (1991).
110. E. Santhani, A. Banerjee, V. Dutta & K.L. Chopra, J. Appl. Phys. 53, 1615 (1982).
111. V. Casey & M.I. Stephenson, J. Phys. D: Appl. Phys. 23, 1212 (1990).
112. H. Kaneko & K. Miyake, J. Appl. Phys. 53, 3629 (1982).
113. S. Schiller, G. Beister, E. Buedke, H.J. Becker & H. Schicht, Thin Solid Films
38
96, 113 (1982).
114. N. Miyata, K. Miyake, K. Koga & T. Fukushima, J. Electrochem. Soc. 127, 918
(1980).
115. K. Budzylqska, E. Leja & S. Skrkzypek, Sol. Energy Mater. 12, 57 (1985).
116. T. Minami, T. Kakumu, K. Shimokawa & S. Takata, Thin Solid Films 317, 318
(1998).
117. C.A. Pan, T.P. Ma, J. Electron. Mater. 10, 43 (1981).
118. F.O. Adurodija, H. Izumi, T. Ishihara, H. Yoshioka & M. Motoyama, J. Vac. Sci.
Technol. A 18, 814 (2000).
119. T. Maruyama & K. Fukui, Jpn. J. Appl. Phys. 29, L1705 (1990).
120. J.M. Phillips, J. Kwo, G.A. Thomas, S.A. Carter, R.J. Cava, S.Y. Hou, J.J.
Krajewski, J.H.Marshall, W.F. Peck, D.H. Rapkine & R.D. van Dover, Appl. Phys.
Lett. 65, 115 (1994).
121.S.A. Agnihotry, K.K. Saini, T.K. Saxena, K.C. Napgal & S. Chandra, J. Phys. D:
Appl. Phys. 18, 2087 (1985).
122.T. Maruyama & K. Fukui, J. Appl. Phys. 70, 3848 (1991).
123. K. Screennivas, T. Sudersena Rao & A. Mansingh, J. Appl. Phys. 57, 384 (1985).
124. A. Suzuki, T. Matsushita, T. Aoki, A. Mori & M. Okuda, Thin Solid Films 411, 23
(2002).
125. P.K. Biswas, A. De, N.C. Pramanik, P.K. Chakraborty, K. Ortner, V. Hock, S.
Korder, Mater. Lett. 57, 2326 (2003).
126. C. Geoffroy, G. Campet, J. Portier, J. Salardenne, G. Couturier, M. Bourrel, J.M.
Chabagno, D. Ferry & C. Quet, Thin Solid Films 202, 77 (1991).
127. W-n. Miao, X-f. Li, Q. Zhang, L. Huang, Z.-j. Zhang, L. Zhang & X.-j. Yan, Thin
Solid Films 500, 70 (2006).
128.G.H.G. Swamy & P.J. Reddy, Semicond. Sci. Technol. 5, 980 (1990).
129. J.B. Webb, D.F. Williams & M. Buchanan, Appl. Phys. Lett. 39, 640 (1981).
130. S. Ghosh, A. Sarkar, S. Bhattacharya, S. Chaudhuri & A.K. Pal, J. Cryst. Growth
108, 534 (1991).
131. J. Hu & R.G. Gordon, J. Appl. Phys. 71, 880 (1991).
132. H. Kim, U,A. Pique, J.S. Horwitz, H. Murata, Z.H. Kafafi, C.M. Gilmore & D.B.
39
Chrisey, Thin Solid Films 377-378, 798 (2000).
133. B.H. Choi, H.B. Im, J.S. Song & K.H. Yoon, Thin Solid Films 193-194, 712 (1990).
134. A. Sarkar, S. Ghosh, S. Chaudhuri & A.K. Pal, Thin Solid Films 204, 255 (1991).
135. N. Naghavi, A. Rougier, C. Marcel, C. Guéry, J.B. Leriche & J.M. Tarascon, Thin
Solid Films 360, 233 (2000).
136. T. Minami, H. Sonohara, S. Takata & H. Sato, Jpn. J. Appl. Phys. 33, L 1693(1994).
137. H. Yanagi, S-i Inoue, K. Ueda, H. Kawazoe, H. Hosono & N. Hamada, J. Appl.
Phys. 88, 4159 (2000).
138. H. Gong, Y. Wang & Y. Luo, Appl. Phys. Lett. 76, 3959 (2000).
139. A. Kudo, H. Yanagi, H. Hosono & H. Kawazoe, Appl. Phys. Lett. 73, 220 (1998).
140. M. K. Jayaraj, A.D. Draeseke, J. Tate & A.W. Sleight, Thin Solid Films 397, 244
(2001).
141. J. Tate, M.K. Jayaraj, A.D. Draeseke, T. Ulbrich, A.W. Sleight, K.A. Vanaja, R.
Nagarajan, J.F. Wager & R.L. Hoffman, Thin Solid Films 411, 119 (2002).
142. K. Ueda T. Hase, H. Yanagi, H. Kawazoe, H. Hosono, H. Ohta, M. Orita, & M.
Hirano, J. Appl. Phys. 89, 1790 (2001)
143. K.-K. Kim, H.-S. Kim, D.-K. Hwang, J.-H. Lim & S.-J. Park, Appl. Phys. Lett, 83,
63 (2003).
144. X.-L. Guo, H. Tabata & T. Kawai, Opt. Mater. 19, 229 (2002).
145. P.T. Moseley, Solid state gas sensors. Meas. Sci. Technol. 8, 223(1997).
146. G. Eranna, B.C. Joshi, D.P. Runthala, and R.P. Gupta, Crit. Rev. Solid State Mater.
Sci. 29, 111 (2004).
147. M.A. Chougule a , S.R. Nalage a , Shashwati Sen b & V.B. Patil, Journal of
Experimental Nanoscience, 9, 482 (2014).
148. T. Seiyama, A. Kato, K. Fujiishi, and M. Nagatani, Anal. Chem. 34, 1502 (1962).
149. N. Yamazoe, G. Sakai, and K. Shimanoe, Catal. Surv. Asia 1, 63 (2003).
150. V.R. Shinde, T.P. Gujar, C.D. Lokhande, R.S. Mane, and S.H. Han, Mater. Sci. Eng.
B 137, 119 (2007).
151. A. Bonanni and T. Dietl, Chem. Soc. Rev. 39, 528 (2010).
152. J. M. D. Coey, Current Opinion in Solid State and Materials Science 10, 83 (2006).
153. S.A. Wolf , D.D. Awschalom, R.A. Buhrman, J.M. Daughton, S. von Molnar,
40
M. Roukes, A.Y. Chtchelkanova, and D.M. Treger Science 294, 1488 (2001).
154. J. M. D. Coey, M. Venkatesan, and C. B. Fitzgerald, Nature Materials 4, 173 (2005).
155. S.J. Pearton et al., J. Appl. Phys. 93, 1 (2003).
156. H. Ohno, J. Magn. and Magn. Mater. 200, 110 (1999).
157. J.K. Furdyna, J. Appl. Phys. 64, R29 (1998).
158. T. Dietl, H. Ohno, F. Matsukura, J. Cibert and D. Ferrand Science 287, 1019(2000).
159. G. T. Thaler et al., Appl. Phys. Lett. 80, 3964 (2002).
160. P. Sharma et al., Nature Mater. 2, 673 (2003).
161. R. M. Frazier et al., J. Appl. Phys. 94, 1592 (2003).
162. F. J. Owens, J. Phys. Chem. Solids 66, 793 (2005).
163. K. Sato and H. Katayama-Yoshida Japan. J. Appl. Phys. 40, L485 (2000).
164. M. Joseph, H. Tabata and T. Kawai Japan. J. Appl. Phys. 38, L1205 (1999).
165. S. Muthukumar, J. Zhong, Y. Chen, Y. Lu and T. Siegrist Appl. Phys. Lett. 82, 742
(2003).
166. C. Kim, S.J. Leem, I.K. Robinson, W.I. Park, D.H. Kim and G.C. Yi Phys. Rev. B
66, 113404 (2002).
167. T. Wakano, N. Fujimura, Y. Morinaga, N. Abe, A. Ashida and T. Ito Physica E 10,
260 (2001).
168. K. Ueda, H. Tabata and T. Kawai Appl. Phys. Lett. 79, 988 (2001).
169. Y.Z. Yoo, Z.W. Jin, T. Chikyow, T. Fukumara, M. Kawasaki and H. Koinuma
Appl. Phys. Lett. 81, 3798 (2002).
170. H. Saeki, H. Tabata and T. Kawai Solid State Commun. 120, 439 (2001).
171. T. Fukumura, Z. Jin, A. Ohtomo and H. Koinuma Appl. Phys. Lett. 75, 3366
(1999).
172. X. M. Cheng and C.L. Chien J. Appl. Phys. 93, 7876 (2003).
173. D.S. Kim, H.M. Kim, S.U. Yuldashev, S.J. Lee, T.W. Kang and D.Y. Kim J.
Korean Phys. Soc. 42 S333 (2003).
174. A. Tiwari, C. Jin, A. Kvit, D. Kumar, J.F. Muth and J. Narayan Solid State
Commun. 121, 371 (2002).
175. S.W. Jung, S.J. An, G.C. Yi, C.U. Jung, S.I. Lee and S. Cho Appl. Phys. Lett. 80,
4561 (2002).
41
176. Z. Jin et al Appl. Phys. Lett. 78, 3824 (2001).
177. S.J. Han, J.W. Song, C.H. Yang, S.H. Park, J.H. Park, Y.H. Jeong and K.W. Rhie
Appl. Phys. Lett. 81, 4212 (2002).
178. W. Prellier, A. Fouchet, B. Mercey, Ch. Simon and B. Raveau Appl. Phys. Lett. 82,
3490 (2003).
179. K. Rode, A. Anane , R. Mattana, J.P. Contour, O. Durand and R. LeBourgeois J.
Appl. Phys. 93, 7676 (2003).
180. J.H. Kim, H. Kim, D. Kim, Y.E. Ihm and W.K. Choo Physica B 327, 304 (2003).
181. J.H. Kim, W.K. Choo, H. Kim, D. Kim and Y.E. Ihm J. Korean Phys. Soc. 42, S258
(2003).
182. Y-Z. Yoo, T. Fukumura, Z. Jin, K. Hasegawa, M. Kawasaki, P. Ahmet, T. Chikyow
and H. Koinuma J. Appl. Phys. 90, 4246 (2001).
183. S.W. Lim, S.K. Hwang and J.M. Myoung Solid State Commun. 125, 231 (2003).
184. J.H. Kim, H. Kim, D. Kim, Y.E. Ihm and W.K. Choo J. Appl. Phys. 92, 6066
(2002).
185. H.J. Lee, S.Y. Jeong, C.R. Cho and C.H. Park Appl. Phys. Lett. 81, 4020 (2002).
42