Organic Semiconductors

Fang-Chung Chen, National Chiao Tung University, Hsinchu, Taiwan

r 2018 Elsevier Ltd. All rights reserved.

Introduction

Organic semiconductors (OSCs) are receiving increasing attention these days because they have many attractive properties –
including light weight, low-cost production, low-temperature processing, mechanical flexibility, and abundant availability – that
distinguish them from their conventional inorganic counterparts. The recent successful commercialization of active-matrix organic
light-emitting diodes (OLEDs) for high-end smartphone displays suggests that organic compounds have promising prospects for
use in various other electrical and optoelectric devices. For example, the fabrication of solar cells, transistors, photodetectors, and
lasers with OSCs can be highly feasible and highly efficient. Another attractive property of OSCs is that they can be processed using
simple solution processing techniques (e.g., ink-jet printing, reel-to-reel fabrication), making the fabrication of electronic devices
much easier and cheaper. As Professor Heeger said in his Nobel lecture, “I’m convinced that we are on the verge of a revolution in
‘Plastic Electronics’.” We foresee rapid advances in the field of OSCs bringing increasingly more inexpensive, workable, and smart
organic devices into our daily lives.
Organic materials are substances comprising mostly carbon and hydrogen atoms; they had been considered for many years to
be exclusively electrically insulating. The discovery of electrical conductivity in organic compounds can be traced back to the
1950s; it has opened up a new fascinating field of research. Electroluminescence (EL) from anthracene single crystals was reported
by Pope et al. in 1963. In the 1970s, Shirakawa, Heeger, and MacDiarmid found that the conductivity of polyacetylene could be
increased tremendously – up to a level close to that of a typical metal – after exposure to vapors of chlorine, bromine, or iodine.
Such high conductivity from a traditional “insulator” of electricity drew much attention and encouraged the rapid growth of the
field of organic electronics. The Nobel Prize in Chemistry in 2000 was awarded to Heeger, MacDiarmid, and Shirakawa for “the
discovery and development of conducting polymers.”
Among the various organic electronic devices known today, the technology of OLEDs has definitely been under the spotlight.
Initially, the EL observed from organic single crystals was not energy efficient and required a large voltage to drive the devices.
In 1987, Tang and Van Slyke developed light-emitting devices based on amorphous organic thin films and achieved a quantum
efficiency of 1% under an electrical bias of less than 10 V. In 1990, OLEDs incorporating conjugated polymers were also reported
by Professor Friend at Cavendish Laboratory. Since then, relevant technologies – including new materials, device structures, and
device engineering approaches – evolved rapidly, gradually establishing a new industry.
As organic-based devices have matured from scientific concepts in the laboratory to applications in the real market, there is a
requirement to understand the unique properties of OSCs and determine the opportunities that this new generation of electronic
materials might offer. In this article, we discuss the electronic structures of OSCs and several fundamental charge storage scenarios
in OSCs. We also review the concept of doping in OSCs. We conclude with a brief overview of interesting device applications,
including electrophotographic devices, thin film transistors, and light-emitting transistors.

Electronic Structures

OSCs are usually categorized into two main groups, depending on the molecular weights of their organic components: con-
jugated polymers (Fig. 1) and small molecules (Fig. 2). Polymers are constituted by repeated structural units (monomers)
connected by covalent bonds. Because the number of repeating units usually cannot be controlled precisely, they are prepared
with a distribution of molecular weights. On the other hand, small molecules have precise and much smaller molecular weights.
The other distinction between the two OSC categories is in their methods of fabrication. Most conjugated polymers are soluble
in organic solvents and can be processed using various solution-processing approaches. In contrast, small molecules are usually
deposited through thermal evaporation under high vacuum – a deposition method that allows good control over the thickness,
morphology, and other properties of the organic films. Accordingly, very thin (even down to a few nanometers), high-
quality films of small molecules can be obtained quite readily. The chemical structures of the small molecules can be also
tailored so that they become soluble in organic solvents, facilitating the use of solution fabrication processes. For example,
pentacene (Fig. 2(d)) can be deposited only through thermal evaporation, whereas 6,13-bis[(triisopropylsilyl)ethynyl]pentacene
(TIPS-pentacene) (Fig. 2(f)) features side chains that improve its solubility, simplify its processing while maintaining high
crystallinity.
To understand the electronic structures of OSCs, we begin with the atomic orbitals (AOs) of carbon, the most important
element of any organic materials. A carbon atom has six electrons; its electronic configuration is (1s)2(2s)2(2p)2. It can be bonded
to other atoms using various hybridization orbitals, including sp, sp2, and sp3 hybrid orbitals (Fig. 3). For example, one s and two
p orbitals are superposed to form three sp2 hybrid orbitals; the orbitals are distributed in the same plane with an angle of 120
degree between them. In a simple model molecule, ethylene, each carbon atom offers three sp2 hybrid orbitals to bond with other
atoms (Fig. 4). The 1s orbital of a hydrogen atom overlaps with one of the sp2 hybrid orbitals to form a molecular orbital (MO),

220 Encyclopedia of Modern Optics II, Volume 5 doi:10.1016/B978-0-12-803581-8.09538-2

 Organic Semiconductors 221

Fig. 1 Chemical structures of polymer-based OSCs: (a) poly[1-methoxy-4-(2-ethylhexyloxy)-p-phenylene vinylene] (MEH-PPV), (b) poly(9,9-
dioctylfluorene) (PFO), (c) poly(3-hexylthiophene-2,5-diyl) (P3HT), (d) poly{2,5-bis(3-hexadecylthiophen-2-yl)thieno[3,2-b]thiophene} (PBTTT), and
(e) poly[3-(4-octylphenyl)thiophene] (POPT).

a so-called s-orbital; the resulting bonding is a s-bond. The two carbon atoms are also bonded through sp2 hybrid orbitals. Each
carbon atom, however, retains one p atomic orbital, aligned orthogonal to the plane of the sp2 hybrid orbitals. The two pz atomic
orbitals overlap side-by-side with the electron density distributed above and below the internuclear axis, forming a different type
of MO called a p-orbital. Note that the p-electron cloud is delocalized between the two carbon atoms. Consequently, the carbon
atoms are bonded by two bonds: a s-bond and a p-bond; this arrangement is termed a “double bond.” In view of the electron
wave functions, the overlapping of two AOs produces two MOs: a bonding state and an anti-bonding state, after constructive and
destructive interference, respectively. Because destructive interference results in lower electron density between the atoms, the
energy of the anti-bonding state is higher than that of the bonding state. For p-orbitals, the bonding and anti-bonding states are
denoted as p- and p*-MOs, respectively.
Benzene, a larger molecule (Fig. 5), has six carbon atoms that are also bonded using sp2 hybrid orbitals. Similarly, the six
pz AOs that remain form six p-bonding MOs: three bonding and three anti-bonding orbitals. As a consequence of interference
among the six pz orbitals, the lower bonding and upper anti-bonding states are degenerate (Fig. 5(b)). Six pz electrons occupy the
three bonding states in a neutral benzene molecule. Similar to ethylene, the p-electron cloud is delocalized over the six carbon
atoms. In other words, the electrons can move from one carbon atom to another; such a delocalized structure is said to feature
“conjugation.”
The MO of highest energy that electrons occupy is known as the highest occupied molecular orbital (HOMO). Meanwhile, the
lowest unoccupied molecular orbital (LUMO) is the MO of lowest energy that electrons do not occupy. The difference in energy
between the HOMO and LUMO is the energy required to excite an electron in the molecule. In benzene, for example, Fig. 5(b), the
upper bonding MOs (p2, p3) are the HOMOs and the lower anti-bonding MOs (p 
4 , p5 ) are the LUMOs. For OSCs, because the
energy splitting of p-bonds is usually smaller than that of s-bonds, electronic processes (e.g., photon absorption and emission)
occur energetically favorably on p-orbitals. Similarly, the charges injected from the metal contacts of organic electronic devices
would tend to occupy p-MOs. In other words, the electronic functions of OSCs originate from the smaller and favorable energy
levels of the frontier orbitals of the p-electron system; the s-electrons are hardly excited because much higher energy is required for
electronic transitions between s-orbitals. Furthermore, the HOMO–LUMO energy gap decreases upon increasing the size of the
p-conjugation system, because of the greater extent of p-electron delocalization.
222 Organic Semiconductors

Fig. 2 Chemical structures of small-molecule semiconductors: (a) 4,4-N,N0 -dicarbazolyl-biphenyl (CBP), (b) N,N0 -diphenyl-N,N0 -bis(3-
methylphenyl)-1,10 -biphenyl-4,40 -diamine (TPD), (c) 5-(4-biphenyl)-2-(4-tert-butylphenyl)-1,3,4-oxadiazole (t-PBD), (d) pentacene, (e) [6,6]-phenyl-
C61-butyric acid methyl ester (PCBM), and (f) 6,13-bis[(triisopropylsilyl)ethynyl]pentacene ((TIPS-pentacene) soluble pentacene).

Fig. 3 Schematic representation of (a) sp, (b) sp2, and (c) sp3 hybrid orbitals.

Fig. 4 (a) Chemical structure of ethylene and (b) schematic orbital diagram.
 Organic Semiconductors 223

Fig. 5 (a) Chemical structure of benzene, it consists of two resonance forms and (b) energy diagram of the p-MOs in benzene.

Trans-polyacetylene was the first conducting polymer to be discovered (Fig. 6). For a single chain of neutral polyacetylene,
the p-band, which can be considered to be composed of an almost infinite number of pz orbitals, would be half-filled, resulting
theoretically in metallic behavior (Fig. 6(a)). Experimental results have strongly indicated, however, that polyacetylene is a
semiconductor having an energy gap of approximately 1.5 eV. In reality, the one-dimensional (1D) “metal” is unstable
and the structure tends to distort, as displayed in Fig. 6(b). Accordingly, the main chain of polyacetylene actually has alternately
long and short bonds. The so-called Peierls distortion converts the repeating unit of trans-polyacetylene to trans-(–HC¼ CH–)n,
instead of trans-(CH)n. As displayed in Fig. 6(g), dimerization of the molecule opens an energy band gap. The p-band is fully
occupied and the upper p*-band is empty; this band structure is similar to that of a typical inorganic semiconductor, which
features a full-filled valence band and an empty conduction band. While a 1D p-electron metal is not stable, higher-dimensional
p-electron systems are not subject to such Peierls instability. For example, carbon nanotubes and graphene can be metallic or
semi-metallic.

Solitons, Polarons, and Bipolarons

The electronic structure of a polymer can be described using the Su–Schrieffer–Heeger (SSH) model, based on a quasi-one-
dimensional tight-binding model. It assumes that the p-electrons are strongly coupled to the polymer backbone through the
electron–phonon interactions. Charge storage occurs through formation of solitons, polarons, and bipolarons. Solitons are
structural defects on the polymer chains. As illustrated in Fig. 6(b) and (c), polyacetylene has two resonance forms. The soliton is a
domain boundary of these two degenerate resonance configurations (Fig. 6(d)); the domain wall can actually extend over several
carbon atoms (Fig. 6(e)). A soliton may be neutral or carry a charge. The energy of the solition lies between the p–p* gap, because
of its nonbonding nature. This defect state is mobile because the system energy does not depend on the position of the soliton. For
non-degenerate ground-state polymers – for example, poly(phenylene vinylene) (PPV, Fig. 7(a)) and polyparaphenylene (PPP,
Fig. 7(b)) – in which the interchange of single and double bonds leads to changes in energy, the solitons cannot be stable because
the structure is not symmetric. Instead, double defects are required for such non-degenerate polymers. Such double defects are
called polarons; they can be considered as a bound state of a charged soliton and a neutral soliton (Fig. 7(c)). Consequently, its
energy levels hybridize the two states, which are split into a bonding, less energetic level and an anti-bonding, more energetic one
(Fig. 7(d)). The two energy levels can be occupied by zero, one, or two electrons, resulting in the formation of positive or negative
polarons. A bipolaron is a bound state of two charged solitons of equal charge; it also consists of a split pair of energy levels
between the energy gap. Furthermore, solitons, polarons, and bipolarons are coupled with structural relaxation, implying strong
electron–phonon interactions of the charge storage in conjugation polymers. In other words, the charges are surrounded by lattice
distortions that can stabilize the system.

Excitons

Excitons are states of electronic excitations in OSCs. An exciton is a particle composed of an electron/hole pair bounded by
coulombic force. The total charge of an exciton is zero. An exciton is, however, capable to move through the organic materials
carrying the excitation energy. There are three types of excitons (Fig. 8): Frenkel, Wannier–Mott, and charge-transfer (CT) excitons.
The main difference among them is the binding energy and/or the distance between the electron/hole pair. When the radius of the
electron/hole pair is comparable with the size of a single molecule or the crystalline constant (aL), the exciton is of Frenkel type.
The small radius also implies a strong coulombic interaction. In other words, it is very difficult to overcome the binding energy and
dissociate the exciton into free charges. If the radius of the exciton is much larger than aL, it is a Wannier–Mott exciton. Therefore,
the binding energy of this type of exciton is small. The intermediate case, in which the exciton extends over several molecular units,
it is termed a CT exciton. The Wannier–Mott exciton is the most common one found in inorganic semiconductors. Usually it can
224 Organic Semiconductors

Fig. 6 (a–e) Polyacetylene: (a) undimerized structure; (b, c) two resonance forms resulting from Peierls effects; (d) soliton in trans-polyacetylene;
(e) domain wall extension of the soliton. (f, g) Electronic band dispersion E(k) plotted with respect to k for (f) non-dimerized and (g) dimerized
systems. Ef is the Fermi energy; a gap appears in the dimerized system, the so-called “Peierls gap.”

be detected only at low temperature, because the thermal energy at room temperature is sufficiently high to overcome its binding
energy. In OSCs, most excitons are Frenkel-type because of their strong binding energy, which originates from the low dielectric
constants of organic materials and their relatively localized electronic clouds. A typical binding energy for excitons in OSCs is
approximately 0.5 eV.
The generation of excitons is a very important process in organic optoelectronic devices. In OLEDs, the recombination of
electrons and holes results in excitons that subsequently decay to produce EL. In organic photovoltaic devices (e.g., organic solar
cells), photon absorption generates the excitons, which move freely within the semiconductors, due to the strong binding energy,
but are subject to effective charge dissociation at the interface between strong organic electron-donors and organic electron-
acceptors. Such exciton dissociation generates free charges, eventually leading to a photocurrent after the charges have been
collected by the electrodes.
Because of the strong binding energy, the directions of spin of the excitons are not likely to change readily when
generated in common OSCs. Therefore, the excitons can be categorized into two different types depending on their combinations
of spin directions. As displayed in Fig. 9(a), the total spin angular momentum of a molecule in the ground state is a sum of
the two spin vectors, equal to zero because of the opposite spin directions. In other words, the overall spin quantum number (S) is
zero
S ¼ js1 þ s2j ¼ jðþ1=2Þ þ ð1=2Þj ¼ 0 ð1Þ
 Organic Semiconductors 225

Fig. 7 Chemical structure of (a) poly(phenylene vinylene) (PPV) and (b) polyparaphenylene (PPP). (c) Schematic representation of a negative
polaron in PPP. (d) Band diagram of various polarons.

Fig. 8 Schematic representation of three different types of excitons in solid state materials; aL is the lattice constant.

where s1 and s2 are the spin quantum numbers of the individual electron. If one electron is excited to the p-orbital, the resulting
exciton is called a singlet because no net spin angular momentum exists in the molecule (Fig. 9(b)). On the other hand, if the two
spin vectors have the same direction, the total spin angular momentum is now equal to 1, as revealed in the following equation
S ¼ js1 þ s2j ¼ jðþ1=2Þ þ ðþ1=2Þj ¼ 1 ð2Þ
Note that if s1 and s2 were both equal to –1/2, the sum would remain the same. Accordingly, the excited state (exciton) now
exhibits a net angular moment. Assuming the presence of a magnetic field, the energy of the exciton could be split into three
different energy levels. Therefore, it is called a triplet (Fig. 9(c)).
Fig. 9(d) displays possible decay processes of a typical organic molecule, where S0, S1, and T1 represent the ground state, first
0 0
singlet state, and first triplet state, respectively; knr and knr are the rate constants of non-radiative decay processes and kr and kr are
the rate constants of radiative decay processes. The singlet exciton can flip its spin direction and transform into a triplet exciton
226 Organic Semiconductors

Fig. 9 Schematic representation of electron spins in the (a) ground state, (b) singlet exciton, and (c) triplet exciton of an organic molecule.
(d) Excitation and possible decay processes in an OSC.

through a process called inter-system crossing (ISC); its rate constant is kisc. As indicated in Fig. 9(d), the relaxation of a singlet exciton
(S1) to the ground state (S0) is possible because the spin angular momentum is conserved. In quantum mechanics, such a transition
is allowed; its radiative decay may lead to a highly efficient emission, called fluorescence (kr). On the other hand, the transition of a
triplet exciton to its ground state is forbidden. Consequently, the efficiency of the radiative decay is very low and, for most organic
0
molecules, can be observed only at a very low temperature. The emission of triplet excitons is called phosphorescence (kr ).

Concept of Doping and p- and n-Type OSCs

Like inorganic semiconductors, OSCs can also be classified as p- and n-type semiconductors. The doping concept and process are,
however, substantially different from those of inorganic materials, even though the same terminology of “doping” is used. For
example, the inorganic semiconductor silicon is doped through ion implantation, in which impurities (phosphorus (P) or boron
(B) atoms) are introduced. These foreign atoms substitute some of the silicon atoms and lead to new energetic levels in the band
gap. Silicon becomes an n- or p-type semiconductor after the introduction of P or B atoms, respectively.
On the other hand, many different methods can be used to “dope” OSCs, including chemical doping, electrochemical doping,
photo-doping, and charge injection from metal contacts. Chemical doping involves CT redox chemistry, as illustrated in the
following examples for polymers
p-type: ðp-polymerÞn þ 3=2n yðI2 Þ-½ðp-polymerÞþy ðI3  Þy n ð3Þ

n-type: ðp-polymerÞn þ ½Naþ ðC10 H8 Þ2 y -½ðNaþ Þy ðp-polymerÞ2y n þ ðC10 H8 Þ0 ð4Þ

From a chemical point of view, p- and n-type doping is analogous to electrochemical oxidation and reduction, respectively.
The doping process generates charge carriers in the OSCs. The electrical conductivity increases upon increasing the doping level
and can reach a level comparable with that of a metal at a very high doping concentration. The drawback of this chemical method
is that the doping level is hard to control, especially for intermediate doping levels. Fortunately, electrochemical doping can solve
this problem. An electrode supplies redox charges to the semiconductor, and ions simultaneously diffuse in or out of the polymer
chains from the electrolyte to compensate the charge mismatch. Through control of the voltage of the electrode, the doping level
can be determined precisely. Electrochemical doping can be illustrated using the following equations:
p-type: ðp-polymerÞn þ ½Liþ ðBF4  Þsoln -½ðp-polymerÞþy ðBF4  Þy n þLielectrode ð5Þ
 Organic Semiconductors 227

h i
n-type: ðp-polymerÞn þLielectrode - ðLiþ Þy ðp-polymerÞy þ ½Liþ ðBF4  Þsoln ð6Þ
n

The third method, photo-doping, is a process in which photon absorption generates an electron/hole pair, which could lead to
locally oxidized (p-type doping) and nearby reduced (n-type doping) states of the material. The unique mechanism is illustrated
by the following equation:
ðp-polymerÞn þ hv-½ðp-polymerÞþy þðp-polymerÞ2y n ð7Þ
After photon absorption, recombination to the ground state can be either radiative or nonradiative. The value of y in the
equation depends on the excitation rate in competition with the recombination rate.
The last scenario involves charge injection from metallic contacts into the OSCs. Holes and electrons are introduced into the
p- and p*-bands, respectively, as illustrated by the following equations:
p-type: ðp-polymerÞn 2yðe2 Þ-½ðp-polymerÞþy n ð8Þ

n-type: ðp-polymerÞn þ yðe2 Þ-½ðp-polymerÞ2y n ð9Þ

Eq. (8) describes the injection of a hole into the filled p-band, corresponding to an oxidation reaction (removal of an electron).
Eq. (9) describes the injection of an electron into an empty p*-band, analogous to a reduction reaction. The major distinction of
doping through charge injection from other doping processes (e.g., chemical doping) is the absence of counter ions to compensate
the charges. Therefore, the doping process is only temporary; the states are relatively unstable. On the other hand, for chemical or
electrochemical doping, the doping is permanent, unless the carriers are removed by an “undoping” process. As for photo-doping,
the resulting photoconductivity is transient and limited by recombination and decay processes.
Even though OSCs can indeed be “doped” as described above and become p- or n-type semiconductors, many reports define
OSCs as p- or n-type by judging whether they are good conductors for electrons or holes. For instance, pentacene is usually referred
to as an excellent p-type semiconductor (Fig. 2(d)) because it exhibits a very high hole mobility. TPD (Fig. 2(d)) is often employed
as a hole-transporting material in OLEDs. As a result, TPD is also considered a p-type OSC because it transports holes effectively
and because holes are injected readily from the metal electrodes. Note that these OSCs are not intentionally doped and they are
usually purified to as high a degree as possible to remove impurities before they are processed for device fabrication. Therefore, we
cannot assume extra energy levels appear near the HOMOs, as they do for inorganic semiconductors. Another way of defining
OSCs is to determine whether they function as electron donors (p-type) or electron acceptors (n-type) during device operation. For
example, fullerene derivatives (Fig. 2(e)) are defined as n-type OSCs because they accept electrons upon photoexcitation and
transport the photogenerated electrons in organic solar cells. Meanwhile, many conjugated polymers blended with fullerene
derivatives are considered as p-type OSCs because they donate electrons upon photoexcitation in solar cells.

Device Applications

In this section, we review some of the most important organic electronic devices, including electrophotographic devices, OLEDs,
organic thin-film transistors (OTFTs), and organic light-emitting transistors (OLETs). For practical applications, organic photo-
conductors were the first technology to use OSCs as a key component in commercially available products; they are often employed
in offices in copy machines and laser printers (Fig. 10). In such an electrophotographic device, the image of a document is
transferred to a photoconductive plate (photoreceptor), previously charged with static electricity, through a set of optical com-
ponents. Because only parts of the photoconductive plate are discharged, as a result of its high photoconductivity, upon exposure
to light, an image duplicates the original document on the surface of the photoconductive plate. The next development step
involves the placement of the electrostatically charged toner particles over the partially charged surface; some of the particles are
adhered to the image by the electric field. An electrostatic field then transfers the image composed of toner particles to a sheet of
paper. Finally, the image is fixed through pressure and heating. After swiping out the toner particles (clean) and exposing to light
to remove the remaining static electricity (erase), the photoconductive plate is ready for the next copy process.
The simplest device structure of an OLED is displayed in Fig. 11(a). It typically consists of two organic layers: an electron-
transport layer (ETL) and a hole-transport layer (HTL). This design obviates the need to find an OSC possessing balanced injection
and transport between electrons and holes. Unlike an individual single crystal (e.g., anthracene) that allows only effective hole
injection, the bilayer structure facilitates hole injection from an indium tin oxide (ITO) electrode and electron injection from a
metal alloy (MgAg) electrode. The charge balance can be further finely tuned through control over the thicknesses of the HTLs and
ETLs. The excitons formed after recombination of the injected electrons and holes relax to the ground states, resulting in photon
emission. Because the ITO anode is transparent, a strong green emission is observed through the ITO side of this device.
In solid state organic materials, only very weak van der Waals forces exist between the molecules. Many experimental results
have indicated, however, that p-electrons can be transported effectively within OSCs. Organic field effect transistors (OFETs)
provide a useful platform for examining the charge transport properties of OSCs. Transistors are fundamental building blocks in
modern electrical circuits; they function as either signal amplifiers or on/off switches. The phenomenon of the conductivity of a
semiconductor being modulated by applying an electric field normal to its surface is called the “field effect.” Most OFETs are
OTFTs, implying that the active (semiconducting) layer is nearly two-dimensional. Fig. 11(b–d) presents the three common device
structures of OTFTs; Fig. 11(e) illustrates the operation mechanism. The charge carriers travel in a channel at the dielectric–OSC
228 Organic Semiconductors

Fig. 10 Working principle of an electrophotographic device: (1) charge, (2) expose, (3) develop, (4) transfer, (5) fix, (6) clean, and (7) erase.
Reproduced from Pope, M., Swenberg, C., 1999. Electronic Processes in Organic Crystals and Polymers, with permission from Oxford University
Press, Copyright (1999).

interface. The source and drain electrodes provide contacts to the OSCs, and can inject and collect, respectively, the charges in the
channel. The conductance of the channel can be modulated by a third gate electrode, which is separated by a layer of gate
insulator. In a p-channel OTFT, for example, a negative voltage is applied to the gate (VG), inducing holes in the channel through
field effects. A negative bias is also applied between the source and drain electrodes (VD), resulting in a driving force to flow the
field-induced holes from the source to drain electrodes. The device behaves as a resistor, with the drain current (ID) increasing
linearly with respect to the increasing value of VD. From the relationship between ID and VD (Fig. 11(f)), this operation region is
called the linear region. While a substantial value of VD is supplied, the magnitude of the field inducing the charges is partially
canceled near the drain. When the effective voltage is lower than the threshold voltage (Vt), the concentration of induced charges
drops to almost zero, forming a high resistance regime in the channel close to the drain end. The channel now pinches at one side
and the value of ID starts to saturate. This phenomenon is known as “pinch-off” and this operation region is the saturation region
 Organic Semiconductors 229

Fig. 11 (a) Device structure of a typical organic light-emitting diode (OLED); the layers of Alq3 and diamine are the electron-transport layer (ETL)
and hole-transport layer (HTL), respectively. Common structures of organic thin-film transistors (OTFTs): (b) top-contact, bottom-gate, (c) bottom-
contact, bottom-gate, and (d) bottom-contact, top-gate. (e) Illustration of the operating mechanism of a typical OTFT. (f) Output characteristics
(ID–VD) of a typical OTFT. ITO, indium tin oxide.

(Fig. 11(f)). Notably, the magnitude of the saturation current depends on the applied gate voltage. For n-channel OTFTs, the
electrical behavior is similar, but the current–voltage curves are located in the first quadrant, instead of the third. The charge
carriers are now electrons.
In the linear region, the drain current is described by:
 
ID ¼ Cox mðW=LÞ ðVG 2Vth ÞVD 2ð1=2ÞVD 2 ð9Þ
where Cox is the capacitance per unit area of the dielectric layer, m is the mobility of the OSC, and W and L are the channel length
and width, respectively. In the saturation region, the value of ID can be expressed as

ID ¼ Cox mðW=2LÞðVG 2Vth Þ2 ð10Þ

Therefore, fitting the curves with the above equations can allow determination of the mobilities of the OSCs in either the linear
or saturation regions. The other important parameter for evaluating the performance of an OTFT is the on/off ratio, defined as
Ion/Ioff, where Ion and Ioff are the drain currents when the channel is turn on and off, respectively. A noticeable leakage current
usually exists because charges can leak through the gate and/or because some charges are stored within the semiconductors. For
state-of-the-art OTFTs, the mobility can readily reach up to 1 cm2 V–1 s–1. Mobilities greater than 10 cm2 V–1 s–1 have also been
demonstrated for solution-processed OTFTs. For example, a highly aligned metastable structure of 2,7-dioctyl[1]benzothieno
[3,2-b][1]benzothiophene (C8-BTBT), grown from a blended solution of C8-BTBT and polystyrene, has displayed a mobility of
43 cm2 V–1 s–1 (25 cm2 V–1 s–1 on average), suggesting great potential for future electronic applications.
Because OLEDs are typically two-terminal devices, field-effect transistors are essential to control their luminescence levels and
to switch individual OLED pixels when they serve as the elements of active-matrix display panels. In contrast, no transistors are
needed when OLETs are employed for similar applications; a single OLET can perform both light-generation and electrical
switching functionalities. For an unipolar OTFT, only a single type of charge carrier, electron or hole, is injected and transported
along the conducting channel. Two carriers, however, are required to form excitons for light emission in OLETs. Therefore, an ideal
OLET is an ambipolar device, although unipolar OLETs have also been reported. Light emission in unipolar OLETs occurs close to
one of the electrodes, and the excitons are readily quenched by the metals, potentially decreasing the luminescence efficiencies.
230 Organic Semiconductors

Fig. 12 (a) Schematic representation of a typical organic light-emitting transistor (OLET) device structure and its working principle. (b) Device
structure of the first ambipolar OLET; the Au and Ca metal sources are positioned off-center, opposite each other, leading to two electrodes for
hole and electron injection, respectively; the chemical structure of MDMO-PPV is also depicted. (c) Movement of the recombination zone as a
function of drain bias: VG ¼–90 V; VD ¼–79, –86, –93, and –100 V (from left to right). (d) Transfer characteristics at VD ¼–80 V. (e) Output
characteristics. Reproduced from Cicoira, F., Santato, C., 2007. Organic light emitting field effect transistors: Advances and perspectives. Advanced
Functional Materials 17, 3421–3434 with permission from WILEY-VCH Verlag GmbH & Co, Copyright (2007); and Zaumseil, J., Friend, R.H.,
Sirringhaus, H., 2006. Spatial control of the recombination zone in an ambipolar light-emitting organic transistor. Nature Materials 5, 69–74 with
permission from Nature Publishing Group, Copyright (2007).

Fig. 12(a) displays the structure of a typical OLET having a bottom contact geometry; it is similar to that of a conventional OTFT.
In other words, other structures, like the two other geometries in Fig. 11, are also possible choices. The source and drain electrodes
are hole-injecting and electron-injecting electrodes, and the OSC should be emissive. Upon application of a reasonable gate bias
(VG), holes and electrons are injected from the source and drain contacts, respectively. The two opposite charges form excitons in
the channel and, subsequently, they decay radiatively to produce light.
Ideal performance of an ambipolar device is, in practice, very difficult to achieve. First, the source and drain should be capable
for injecting both holes and electrons efficiently. The OSC should also exhibit balanced hole and electrode mobilities. In 2006,
Zaumseil et al. reported the first ambipolar OLETs incorporating poly(2-methoxy-5-(3,7-dimethyloctoxy)-p-phenylene-vinylene)
(MDMO-PPV), a conjugated polymer displaying high photoluminescence efficiency (Fig. 12(b)). On the dielectric surface, they
deposited a layer of crosslinked benzocyclobutene derivative (BCB) as a buffer dielectric to prevent the trapping of electrons at the
dielectric–semiconductor interface. This approach is one of the keys to significantly improve electron (n-channel) transport on
oxide dielectric surfaces. Au and Ca electrodes were then deposited for effective hole and electron injection, respectively. The use of
Au electrodes only for both contacts would have led to high electron injection barrier at one side. Similarly, the use of Ca
electrodes only would have led to a high hole injection barrier at one of the contacts. Fig. 12(c) illustrates the light emission from
the OLET, and the movement of the light emitting zone between the two electrodes. In fact, no light emission was observed when
the value of |VD| was below a certain threshold, because of limited electron injection. A narrow line of light appeared once the
value of |VD| exceeded the threshold voltage. The light emission line moved toward the Au electrode upon increasing the value of
|VD|. The transfer characteristics of the device (Fig. 12(d)) revealed ambipolar behavior; the minimum point, at a value of VG of
approximately –20 V, indicates the transition from the electron- to hole-dominated current in the OLET. The output characteristics
in Fig. 12(e) also reveal ambipolar currents. High-performance OLETs can also be fabricated by using mixed p- and n-type OSCs as
the semiconducting layer. This approach can be accomplished through a bilayer structure, or co-deposition of hole and electron
 Organic Semiconductors 231

transporting materials. More recent results have indicated that OLETs can outperform equivalent light-emitting diodes when using
a trilayer semiconducting heterostructure.

Summary

The properties of OSCs are much different from those of inorganic semiconductors in many aspects. For example, p-electrons are
responsible for the charge transport in OSCs, and the concept of doping is fundamentally different. Nevertheless, OSCs are very
attractive for use in electronic devices because they are formed from inexpensive starting materials and have many useful prop-
erties, including solution-processability and flexibility. These unique properties offer many possibilities for various applications.
For example, high-performance inorganic light-emitting diodes are usually fabricated through delicate crystal growth, and large-
area devices are very difficult to implement. In contrast, OLEDs are readily pixelated and, therefore, can be used for flat-panel
displays. We must emphasize, however, that organic electronic devices are not designed intentionally to compete with those
containing inorganic counterparts. Rather, we believe that there will be many new applications in which inorganic semiconductors
will not be useful; therefore, the exploration of organic devices should continue in the future.

See also: Organic Lasers. Organic Photodetectors

Further Reading

Capelli, R., Toffanin, S., Generali, G., et al., 2010. Organic light-emitting transistors with an efficiency that outperforms the equivalent light-emitting diodes. Nature Materials 9,
496–503.
Cicoira, F., Santato, C., 2007. Organic light emitting field effect transistors: Advances and perspectives. Advanced Functional Materials 17, 3421–3434.
Heerger, A.J., 2001. Semiconducting and metallic polymers: The fourth generation of polymeric materials (Nobel lecture). Angewandte Chemie International Edition 40,
2591–2611.
Helfrich, W., Schneider, W.G., 1965. Recombination radiation in anthracene crystals. Physical Review Letters 14, 229–231.
Pope, M., Swenberg, C.E., 1999. Electronic processes in organic crystals and polymers. New York, NY: Oxford University Press.
Tang, C.W., VanSlyke, S.A., 1987. Organic electroluminescent diodes. Applied Physics Letters 51, 913–915.
Turro, N.J., 1991. Modern molecular photochemistry. Sousalito, CA: University Science Book.
Yuan, Y., Giri, G., Alexander, L., et al., 2014. Ultra-high mobility transparent organic thin film transistors grown by an off-centre spin-coating method. Nature Communications
5, 3005.
Zaumseil, J., Friend, R.H., Sirringhaus, H., 2006. Spatial control of the recombination zone in an ambipolar light-emitting organic transistor. Nature Materials 5, 69–74.