plasma-deposited silicon-based coatings on polymer substrates. Thin Solid Films,
340, 33-39.
van Oss C.J., Good R.J., Chaudhurys M.K. (1988). Additive and Nonadditive Surface
Tension Components and the Interpretation of Contact Angles. Langmuir, 4, 884-
891.
Young T. (1805). An Essay on the Cohesion of Fluids. Philosophical transactions Royal Society
London. 95, 65-87.
Zhao Q., Liu Y., Abel E.W. (2004). Effect of temperature on the surface free energy of
amorphous carbon films. Journal of Colloid and Interface Science, 280, 174–183.
40
Optoelectronic Devices and Properties
Zheng S., Ashcroft I.A. (2005). A depth sensing indentation study of the hardness and
modulus of adhesives. International Journal of Adhesion & Adhesives, 25, 67–76.
3
Organic Semiconductor Based Heterostructures
for Optoelectronic Devices
Anca Stanculescu1 and Florin Stanculescu2
1National Institute of Materials Physics, Bucharest-Magurele,
2University of Bucharest, Bucharest-Magurele
Romania
1. Introduction
In the last decades organic materials are considered a very important alternative to the
inorganic semiconductors in the manufacturing of a large variety of devices because of the
great diversity of the organic compounds, their remarkable properties, low production cost,
compatibility and complementarity with the very well known silicon technology. Lately the
organic semiconductors started to represent a new class of materials very attractive for opto-
electronic applications (Huang, 2002; Tang, 1988; Forrest, 2000; Kalinowski, 2003) and
electronic applications (Dodabalapur, 1995; Karl, 2000; Katz, 2000 a; Horowitz, 1999; Torsi,
2000; Schön, 2000 a; Inoue, 2005) in particular light emitting devices and lighting sources
(Yersin, 2007; Van Slyke, 1996; Wang, 2005; Dini, 2005), solar cells (Lane, 2005; Ameri, 2009;
Troshin, 2008; Duggal, 2005), photodetectors (Hofmann, 2005; Troshin, 2008), field effect
transistors (Molinari, 2007; Mas-Torrent, 2008) and lasers (Duarte, 2009; Liu, 2009) being
considered candidates to replace the inorganic semiconductors.
The electroluminescence (EL) of the organic molecules is a well-known phenomenon for
more than 50 years, but only in the late ’80th has become utile for practical applications. The
successful application of the organic materials luminescence in Light Emitting Devices
(OLEDs) requires adequate device structures to overcome the problems associated with the
high resistivity of the organic materials and with the difficulty to obtain a good charge
injection from the electrodes in the organic layers. These devices have a thin film
architecture, which includes both organic/organic and organic/inorganic interfaces.
Starting with the first prototype of OLED (Vincett, 1982), lately has been proposed the first
efficient OLED based on heterostructures containing thin films from small molecule organic
compounds (Tang, 1988) that have shown an attractive efficiency (1% external quantum
efficiency, 1.5 lm/W luminous efficiency, higher than 1000 cd/m2 brightness) and driving
voltage below 10 V.
From this moment, different types of OLEDs have been realised, based on both fluorescence
and phosphorescence, that have started to be used in mobile, small dimension displays.
Probably, the most spectacular application of the Organic Light Emitting Devices is in the
flat panel displays for TV and monitor technology because they are characterized by high
brightness and wide viewing angle, which are the most important advantages over the
liquid crystals. Significant effort is devoted now to increase the quantum efficiency, lifetime
and thermal stability of these devices. Devices with high efficiencies, low working voltages
42
Optoelectronic Devices and Properties
and long lifetime involve low electrical resistivity and good chemical stability of the contact
organic semiconductor/metal to assure a good charge carriers injection and transport. On
the other hand the particularities of the charge injection and transport across the
inorganic/organic and organic/organic heterojunctions are determinant for the properties
of the devices. The improvement of the organic electroluminescent diode performances
involves the separate optimisation of the two processes implied in the radiative emission
phenomena: 1. the injection and transportation of the charge carriers and 2. their radiative
recombination in different layers. The injection of the charge carriers must be not only
efficient but also stable under operation conditions.
In this chapter we summarize some results of our work in the field of applied research
concerning the charge injection and transport through inorganic/single and multilayer
organic heterostructures.
2. OLED structure. General considerations
The basic p-n structure is formed from two organic thin layers, one a holes transporting
layer (HTL) and the other an electrons transporting layer (ETL), which can be also the
emitting layer where the recombination takes place. The organic layers are disposed
between an anode and a cathode and are offering an adequate medium for the charge
transfer through the interface situated between the two organic layers. In the OLED
structures the relevant mechanism of electroluminescence involves several steps: injection,
transport, capture and radiative recombination of the positive and negative charge carriers
inside an organic semiconductor characterized by an energy band gap suitable for yielding
visible light.
metal contact
n-organic
emissive layer
p-organic
ITO
glass
Fig. 2.1. Basic p-n junction for OLED application
The OLED structure is delimited by the metallic layer and the transparent conductor such as
indium tin oxide (ITO)/glass interfaces. While the electrical properties are controlled by the
mobility of the charge carriers and the energy levels offset between the layers that compose
the heterostructure, the optical properties are controlled by the refractive index mismatches
at the interfaces glass/air and organic layer/transparent conductor. These mismatches
generate the trapping of a large fraction of the incident light by the total internal reflection
into glass and transparent conductor (ITO) (Lu, 2002). An improved electroluminescence
can be obtained by the individual optimization of the steps mentioned above (injection and
recombination), in the configuration of organic multilayer light emitting devices based on
hetero-interfaces between different organic materials.
Multiple layers can be used to increase the quantum efficiency or avoid the degradation of
the devices. This way can be assured a balanced injection of the charge carriers. Since
Organic Semiconductor Based Heterostructures for Optoelectronic Devices
43
recombination of the electrons and holes within an organic layer gives rise to luminescence,
it is necessary that both charge carriers are simultaneously injected in equal numbers from
both electrodes, in the organic layer. This can be achieved by the optimization of both
electrons and holes transporting layers or by the use of intermediate organic layers acting as
blocking layers that assure a good separation of the charge carriers.
The first critical factor in determining the OLED efficiency is the charge injection being
correlated with the lifetime of the device. The electrodes are used to inject charge into or to
extract charge from the organic semiconductor layers, more precisely to inject electrons and
holes into opposite sides of the emissive organic layers: (1) the electrons are injected from
the Fermi level of the cathode into the lowest unoccupied molecular orbital (LUMO) of the
organic, overcoming the contact barrier; (2) the hole are injected from the Fermi level of the
anode into the highest occupied molecular orbitals (HOMO) of the organic. To improve the
carrier injection, the energy level for the charge carriers in organic should match the
electrode work function. On the other hand the HTL and the ETL should be characterized by
high mobility carriers to assure the transport of the injected carriers to the emission zone.
The second critical factor is the radiative recombination of the charge carriers in the bulk of
the organic layer. The electrons and holes accumulated at the HTL and ETL interface are
creating pairs, known as excitons, that release the energy as light. Beside the improvement
in the fluorescence yield of the emitting material, which can be obtained by doping, a well-
balanced injection of positive and negative charge carriers is necessary.
The inorganic semiconductor/organic and/or metal/organic junction, as the key element(s)
of any organic device, can offer the possibility to overcome the disadvantages of the
conventional inorganic heterostructures, but the process of injection of the charge carriers
across the interfaces of the structure and the transport of the hole and electrons inside the
structure influence the performances of the organic devices.
Other important processes dominating the electrical properties of the interfaces and
affecting the performance and lifetime of the devices, are the diffusion and chemical reaction
of the metal deeper into the organic layer at the metal/organic interface and the molecular
interdiffusion at the organic/organic interface.
In some molecular compounds, the intermolecular coupling is supposed to be stronger
because of the significant overlapping between the π electrons clouds supporting the charge
carriers generation and transport, leading to a significant delocalization of electrons and, as
a consequence, a relatively significant intrinsic conductivity.
The limitation in performance of the device realized with organic semiconductor
characterized by low charge mobility (µ<<1 cm2/Vs) is determined by the current limitation
due to the space charge. This process is favoured if one contact is able to inject more carriers
than would be present in semiconductor at thermal equilibrium. The electrical transport
properties are determined by the low mobility of the charge carriers in organic solids (2 or 3
orders of magnitude lower than in silicon) that induces a limitation associated with the
space charge and by the grain boundaries, defects and imperfections in the organic films
acting as trap centres that induce a limitation associated with the trap charge.
Another source for the generation of interfacial space charge in hetero or multilayer devices
can be the energy level offset at the organic/organic interfaces. Even the energy barriers are
not present, the differences in the charge carriers mobility between the layers can generate a
mobility barrier at hetero-interfaces, which can also be a source for the generation of
interfacial space charge. Trapped and interfacial charges have an important effect on the
performances of the OLEDs (Riess, 2001).
44
Optoelectronic Devices and Properties
A stronger emission can be obtained introducing a supplementary luminescent layer
between the holes transporting layer and the electrons transporting layer, as it is presented
in Figure 2.1, but this way the manufacturing of the OLED more is complicated.
If we use doping to improve the fluorescence, the electrical properties of the resulted “host-
guest” material can be controlled through the selection of the dopant (Makinen, 2002). The
electronic transfer direction and the doping polarity are strongly depending of the relative
position between HOMO and LUMO for dopant (guest) and matrix (host) molecules. So, a
good concordance between the ionisation energy (IE) corresponding to HOMO for the host
matrix and the electron affinity (EA) corresponding to LUMO of the dopant molecule is
necessary to produce a „p” type doping while the alignment between the LUMO level of the
host matrix and the HOMO level of the dopant molecule leads to a „n” type doping. Fermi
level will freely move inside the band gap HOMO-LUMO of the host matrix and dopant,
supposing that this movement is not limited by the band edges (no pinning phenomenon).
The role of the dopants is very complex. On one hand the doping of the “p” and “n”
transport layers has two important transport effects: 1. generates narrow space charge
regions at the contacts and an improved charge carriers’ injection through an efficient
tunnelling; 2. assures a low voltages across the transport layers because of the high
conductivity. On the other hand, the “p” doping of the “n” conduction layer has a
stabilization effect by trapping the positive charge carriers partially responsible for the
degradation of the device (Zhi-Lin, 1998; Cheng, G, 2006).
3. Theoretical approach
Many theories have been proposed in the last years to describe charge injection into
materials characterised by a hopping transport, such as organic compounds. The electrical
transport theories in organics try to describe the charge injection process as thermally-
assisted tunnelling from the metal to localized states (Abkowitz, 1995), tunnelling into
polaron levels in polymers (Conwell, 1997), thermally-assisted injection into an energetically
disordered dielectric (Arkhipov, 1998), or as diffusion-limited thermionic emission (Emtage,
1966; Scott, 1999). The most important factors playing a role in the injection of the charge
carriers from metal to organic are the charge mobility in the organic layer (Emtage, 1966;
Scott, 1999), the dependence of the mobility of the electric field intensity (Borsenberger,
1998) and of charge density (Roichman, 2002), the trapping of injected charge at the interface
due to the image potential (Gartstein, 1996), the interface dipoles arising from the charge
transfer (Crispin, 2002) or the interfacial chemistry (Abkowitz, 1998) and disorder in these
interface dipoles (Baldo, 2001).
The simplest case to analyse can be a metal electrode that injects electrons in a trap free
organic semiconductor. At lower voltages and neglecting the diffusion, the current is
determined by the motion of the free electrons that are present in the semiconductor and the
current density is given by the Ohm’s law (Lambert, 1970):
V
J = qμ
Ω
N
(3.1)
0
0 d
where q is charge of the electron, N0 is the number of free electrons per unit volume, µ is the
mobility of electron, V is the applied voltage and d is the length of the sample (film thickness).
In the case of purely injection-limited currents, neglecting the specific limiting mechanisms,
the current at constant intensity field, E, has no explicit thickness dependence and is given by:
Organic Semiconductor Based Heterostructures for Optoelectronic Devices
45
J=J( E) (3.2)
The transport over an interfacial barrier at a metal/semiconductor interface can be analysed
with one of the following theories:
1. For the process thermally activated at room temperature, is used the thermionic,
Richardson-Schottky (RS) emission (Sze, 1981; Ashcroft, 1976) over the barriers to
describe the transport characteristics.
2. For tight barriers at low temperatures, electron tunnelling or Fowler-Nordheim (FN)
tunnelling (Sze, 1981) dominates the transport process.
The (RS) model is based on the lowering of the energetic barrier height by the image charge
potential under the effect of an external electric field of intensity E=V/d. In this model, the
current density JRS, as a function of field intensity, is given by the equation:
⎛ φ − β
⎞
* 2
E
B
RS
J = A T exp⎜ −
⎟ (3.3)
RS
⎜
k T
⎟
⎝
B
⎠
where: *
* 2
3
A = 4π qm k h [A*=120 A/cm2K2 for m*=m
B
0 (electron mass)] is the Richardson
constant (q=electron charge, kB=Boltzmann constant; h=Planck constant, T= Thermodynamic
temperature);
3
β = q
πεε (ε is the relative permittivity and ε
RS
4
0
0 is the vacuum
permittivity) and ΦB is the zero-field injection barrier.
In the case of FN tunnelling model, the Columbic effects are neglected. It is considered the
tunnelling through a triangular barrier into a continuum of states. In this model the current
density is given by the following equation (Gao, 2002):
* 2 2
3/2
A q E
⎛ 2αΦ ⎞
J
=
exp
B
⎜ −
(3.4)
FN
⎟
2 2
α k
⎜
3 qE ⎟
Φ B
B
⎝
⎠
Under certain conditions, the (RS) and (FN) models can be applied to explain the
conduction also in inorganic semiconductors characterized by extended band states and
large mean free path.
At low voltages, and neglecting the diffusion, the current is determined by the motion of the
free electrons and the current density is given by the Ohm’s law. As the voltage increases,
electrons injected from the contact begin to exceed the electrons that are initially present
inside the semiconductor. This happens when the charge corresponding to the number of
electrons becomes approximately equal to C ⋅ V , where C is the capacitance of the sample
and V is the applied voltage. This is the space charge limited (SCL) regime and the Mott-
Gurney (Mott, 1940; Shen, 2004) equation can be applied for SCLC for field-independent
mobility and the current density expression becomes:
2
9
V
J
= εε μ
(3.6)
SCLC
0
3
8
d
where εε0 is the permittivity of the organic. The threshold voltage, V0, at which the current
turns from ohmic to SCL can be found by the relation:
2
8
d
V = eN
(3.7)
0
0
9
εε0
46
Optoelectronic Devices and Properties
The SCL conduction relies on a phenomenological theory developed in the 1950s and
subsequently improved with more specific details (Lampert, 1970; Kao, 2004).
Considering a trap free semiconductor with a limitation of the current imposed only by the
space charge (SCLC), with the mobility depending or being independent of the field
intensity, the current at constant intensity field becomes
J J*
= ( E) d (3.8)
where d is the thickness of the layer as mentioned above.
Space charge limited transport appears in undoped, wide-gap semiconductors in which the
density of charge carriers at equilibrium is very small (de Boer, 2005). In these materials the
current is generated by the charge injected from the contacts. The density of the charge
carriers is determined (limited) by the electrostatic phenomena.
Especially in amorphous molecular materials, molecularly doped polymers and most of the
conjugated polymers, the mobility is characterized by a field-dependence and an
approximate analytical expression has been deduced (Murgatroyd, 1970). This is the Poole-
Frenkel (PF) field dependence of the mobility:
μ( E) = μ exp β E (3.9)
0
( )
where μ0 is the zero-field mobility, E is the electric field strength and β is associated with the
Poole-Frenkel-like field dependence of the mobility in the organic (Silveira, 2004). For field
dependent mobility the SCL current density becomes:
2
9
V
J
= εε μ
exp 0.89β V d (3.10)
SCLC
0 0
3
(
)
8
d
Measuring the current as a function of voltage it was remarked a transition from „ohmic”
conduction JΩ at low voltage to a SCL conduction at high voltages, with the corresponding,
low voltage, current densities JΩ given by:
V
J = μ N
exp(β
Ω
V d ) (3.11)
0
0 d
where N0 is the density of free carriers and JSCL is given by Eq. ( 3.10).
The conduction in organic semiconductors can be injection-limited or transport-limited
(Boer, 2005). For injection-limited conduction the models for bulk-limited transport, such as
trap charge limited transport (Burrows, 1996; Lampert, 1970; Campbell, 1998; Campbell,
1997; Hung, 1997; Parker, 1994; Sze, 1981) and space charge limited transport with a field
and temperature dependence of the mobility (Bloom, 1997) can not be applied.
For a bulk-limited transport such as the trap charge limited conduction (TCLC), with an
exponential trap distribution and a mobility independent of field, the current at constant
field varies with d-l with l>1 (Lampert, 1970; Kao, 1981):
*
= ( ) l
J J E d (3.12)
The bulk-limited transport explains both the thickness and temperature dependences of
current-voltage characteristics over many order of magnitude and the variations in injection
Organic Semiconductor Based Heterostructures for Optoelectronic Devices
47
for different cathode materials. This model can not explain the dependence of current-
voltage characteristics on the composition of the injection contact (Baldo, 2001).
The shape of the I-V curves can be the result of the conjugated effect of the surface and bulk
traps and not of different physical mechanism (de Boer, 2005).
In the case of Poole-Frenkel field dependence of the mobility, SCLC can show exactly like
TCLC, for the current versus voltage curve, a power law behaviour, but the power
coefficient is higher than 2. The distinction between these two mechanisms can be done by
the difference in the thickness dependence of these mechanisms given by Eq. (3.8) and Eq.
(3.12) respectively.
4. Experimental
4.1 Types of substrates
To investigate the electrical conduction properties of the interface inorganic/organic and
organic/organic have been prepared sandwich type heterostructures on different substrates:
ITO with different sheet resista