This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Abstract - Organic Field Effect Transistors (OFETs) were first reported in 1987 by Koezuka and co-workers and since then OFETs have undergone remarkable improvements in terms of device performance owing to new materials research. This paper presents an overview of different key aspects pertaining to OFETs, namely the charge carrier transport in organic semiconductors, materials used and properties desired for better performance, device operation and electrical characteristics.
RGANIC Field Effect Transistors (or OFETs) are MOSFETs (metal-oxide-semiconductor FETs) in which the semiconductor layer is an organic material with a conjugated structure. Like any other MOSFET, OFETs operate as voltage controlled current devices and have three terminals namely the source, drain and gate. The operation of FETs is based upon the modulation of the charge carrier density (in the semiconductor channel) by the voltage applied to the gate electrode. The other two metallic contacts (source and the drain) inject electric current that can be controlled efficiently by varying the bias on the gate electrode.
OFETs may not be an alternative to the silicon technology but are commercially important devices due to their niche applications such as Organic Light Emitting Transistors (OLETs) and others, where the key optimization parameter is cost.
This paper analyzes OFETs in terms of their use and operation. At several points, principles associated with OFETs are compared with silicon MOSFETs to understand the inherent differences and similarities between the two.
MOSFETs form the very basis of digital technology today; they are omnipresent in the form of microprocessor chips, memories and displays, amongst others. The commercial significance and dominance of silicon MOSFETs can be estimated from the fact that in 2009 alone, 1019 transistors were produced worldwide.1 For the past two decades or so, research has enabled remarkable improvements in the performance of organic semiconductor (OSC) based devices. Despite the fact that the characteristic field effect mobility in OSCs is generally lower than their inorganic counterparts, mobility values of greater than 10cm2 /Vs have been reported for new organic materials such as rubrene.3 In order to realize low-cost and environment friendly devices, research on OSCs is of high significance. 8 Another interesting feature that makes organic FETs and other organic electronic devices a commercial curiosity is the fact that they can be manufactured close to room temperature and deposited on foldable and unbreakable surfaces such as paper.1 This has spurred the conception of novel electronic products that form the family of "bendable electronics".
Charge Transport in Organic Semiconductors (OSCs)
Charge transport in organic semiconductors is governed by the property of conjugation, i.e. the overlap of p-orbitals in an arrangement of alternating single and double bonds amongst carbon atoms. In other words, the pi-electrons are delocalized and this enables charge carrier transport in organic compounds. Greater the overlap of orbitals (as in the case of ordered molecular structures), more efficient is the charge transport due to delocalized states. Solution process polymeric semiconductors usually form complex microstructures that are highly disordered; this poses a limitation on the charge transport in these materials, hence resulting in low field-effect mobilities.2 Research on organic materials for devices, therefore tends to focus on small conjugate molecules which have higher field-effect mobilities.
Several models have been proposed to explain charge transport in organic semiconductors; however the exact nature of the same is still not completely understood. Each organic compound behaves differently in the presence of an applied electric field depending on its preparation and chemical arrangement. Two of the more widely accepted models for charge carrier in OSCs are presented below.
Variable Range Hopping (VRH): This model is used to describe charge transport in amorphous semiconductors and hence can be applied to low conductivity organic semiconductors. The VRH model assumes that charge carriers are transported by hopping between localized states. This occurs via quantum tunneling, a process that is thermally assisted. As a result, unlike inorganic semiconductors, mobility in an organic semiconductor increases with increasing temperature. The VRH model is unable to satisfactorily explain mobility values observed for highly ordered molecular arrangements such as the thin-film polymorph of pentacene.1
Multiple Trapping and Release (MTR): The MTR model proposed by Horowitz et al. assumes that charge carriers get trapped into localized states and each such state has a narrow delocalized band associated with it in which transport occurs. This model is able to explain the transport mechanism and the relatively larger mobility values (~0.1 cm2/Vs) reported for ordered microstructures.
Organic semiconductors can be broadly classified into two categories, polymers and small conjugates molecules. Recent improvements in the field effect mobility can be attributed to breakthroughs in making highly ordered films.
p-type OSCs: The majority charge carriers in p-type OSCs are holes. Among the small molecule OSCs, pentacene gives large charge carrier mobilities owing to its crystal structure which facilitates pi orbital overlap. However, a major concern with using pentacene for OFETs is its rather short-lived air stability. When exposed to air, oxidation disrupts the conjugated Ï€-system of the molecule and hence carrier mobility decreases irreversibly.
Fig. 1. Conjugated small-molecule organic semiconductors. (a)Pentacene11, (b) 2,6-di[2-(4-phenyl)vinyl]anthracene (DPVAnt)1,
(c)dinaphtho-[2,3-b:20,30-f]thieno[3,2-b]thiophene (DNTT) and
(d) Fullerene (C60).1
Two conjugated molecules worth mentioning are 2,6-di[2-(4-phenyl)vinyl]anthracene (DPVAnt) and dinaphtho-[2,3-b:20,30-f]thieno[3,2-b]thiophene (DNTT). Both of these molecules have a larger ionization potential as compared to pentacene and are therefore more air stable. The crystal structure of both DPVAnt and DNTT allows excellent orbital overlap, giving mobilities similar to that of pentacene.
n-type OSCs: n-type OSCs have electrons as the majority charge carriers. It is however found that most organic FETs show p-channel rather than n-channel behavior i.e. in OSCs, current due to hole transport is the dominant current. This behavior can be better explained by the energy level diagram shown in Figure (2). Compared to the terminology used for an inorganic semiconductor, the LUMO (or the lowest unoccupied molecular orbital) and HOMO (highest occupied molecular orbital) correspond to the conduction band and valence band in an OSC. Hole current in OSCs is due to the injection of holes into the HOMO (Highest Occupied Molecular Orbital) level from the metal contact. This kind of hole transport requires the energy difference between Fermi level of the ohmic contact and the HOMO level to be small. This is usually favored in OSCs. Electron current and transport, on the other hand requires that the energy difference between LUMO (Lowest Unoccupied Molecular Orbital) and the Fermi level of the contact to be small in order to allow injection of electrons. Typically, this is not favored in organic materials. For example, pentacene has HOMO energy of 4.5eV (with respect to vacuum level) and LUMO energy of 2.5eV. When contacted with gold (having a work-function of 5eV), transport of positive charge carriers between contact and HOMO is favored. Hence, current in the device is dominated by holes.
n-type operation can be obtained in two ways- by employing a metal contact with a lower work function or by having an organic material with a high enough electron affinity to allow efficient injection of electrons into the LUMO of the molecules. By virtue of the first method, calcium contacts (with a work-function of 2.8eV) used in conjunction with pentacene have been reported to operate as n-channel OFETs.1 Fullerene (C60) and its derivatives have comparatively higher electron affinity and have been reported to show mobilities as high as 6 cm2/ Vs.2 All n-type OFETs discussed so far are prone to performance degradation when exposed to air and are only operational in an inert environment. Reports of air stable n-channel OFETs are rare, posing a severe limitation to the application of OFETs in CMOS technology.
Fig. 2. Illustration of (a) p-channel operation (b) n-channel operation in an OFET, with respect to the HOMO and LUMO levels in an OSC.10
Another reason for the hole transporting behavior seen in OSCs is that electrons are frequently trapped into localized states at the semiconductor-dielectric interface and are not able to contribute to conduction as much as holes. The performance of n-type devices can then be substantially improved by making use of gate dielectrics that suppress the influence of electron trap sites. Simultaneous studies on the soluble fullerene derivative PCBM ( [6,6]-phenyl C61-butyric acid methyl ester) have shown the influence of different dielectric materials on the device performance. For instance, mobility of 0.2 cm2/ Vs has been observed with PVA 60 as the dielectric material.2
OFETs usually adopt the standard Thin-Film Transistor (TFT) structure. As shown in figure (3), two common TFT configurations for OFETs are top and bottom contact. In the former configuration, source and drain electrodes are evaporated on the top of the organic semiconducting layer while in the latter, they are evaporated on the dielectric material before depositing the organic semiconductor.
Fig. 3. TFTs can be fabricated in the above four basic structures. These are (a) Bottom-gate (inverted) staggered TFT, (b) Bottom-gate (inverted) co-planar TFT, (c) Top-gate staggered TFT and (d) Top-gate co-planar TFT.1
The device physics governing the operation of OFETs is markedly different from their commercially indispensible silicon counterparts. Silicon MOSFETs operate in inversion mode i.e. an applied gate voltage forms a conducting channel of minority charge carriers between the heavily doped source and drain regions. Organic FETs, on the other hand operate in the intrinsic regime and in the accumulation mode. Additionally, silicon MOSFETs are typically doped to operate as n-channel or p-channel MOSFETs, both of which can be grown on the same substrate; a feature that is advantageously exploited by the CMOS technology. However, in case of OFETs, a number of factors like the choice of the organic semiconductor and the work-function of metal contacts decide whether its behavior is hole or electron transporting (i.e. p-channel or n-channel OFET). 1
Fig. 4. Energy-level diagram across the semiconductor-dielectric interface of OFETs (a) For a negative gate voltage, positive charge carriers (holes) accumulate at the interface, this represents the situation in a p-channel OFET, (b) For a positive gate voltage, negative charge carriers (or electrons) accumulate at the interface, this represents the situation in an n-channel OFET. WF is the work-function of the gate contact, EA is the electron affinity of the OSC, IP is the ionization potential of the OSC, LUMO is the Lowest Unoccupied Molecular Orbital and HOMO is the Highest Occupier Molecular Orbital in the OSC.1
Figure (4) shows the energy-level diagram of the MOS structure in an organic FET. In the absence of an applied gate voltage, no charge carriers are injected and hence the device is in the OFF state. When a gate voltage (greater than the threshold voltage value) is applied, mobile charge carriers accumulate at the semiconductor-dielectric interface and form a conducting channel. If a bias is now applied to the drain-source terminal, charge transport occurs and the device is switched ON.
Unlike silicon MOSFETs, the drain and source regions in an OFET are not doped and directly contacted via metal. The work function of these metal contacts determine whether the dominant current flow is due to positive or negative charge carriers.
Fig. 5. Energy-level diagram showing charge transport along the semiconducting channel in an OFET, (a) p-channel OFETs: injection of negative charge carriers from the drain contact into the semiconductor is blocked due to the large energy difference between the Fermi level of the metal and the LUMO of the OSC (b) n-channel OFETs: injection of positive charge carriers from the drain contact into the semiconductor is blocked due to the large energy barrier between the Fermi level of the metal and the HOMO of the OSC.1
Figure (5) shows how charge transport occurs along the carrier channel between the source and drain terminals in an OFET.10 When a negative gate-source voltage (VGS) is applied, positive charge carriers are induced near the semiconductor-dielectric interface and a p-channel OFET is obtained. If the Fermi level of the metal contact is close to the HOMO energy level of the semiconductor then it is possible to extract these positive charge carriers by applying a negative bias between the drain and source contacts i.e. for VDS<0.
Conversely, when a positive gate-source voltage is applied, negative charge carriers accumulate close to the semiconductor-dielectric interface and an n-channel OFET is formed. If the Fermi level of the source and drain metal contacts lies close to the LUMO level of the organic semiconductor then negative charge carriers can be extracted by applying a positive bias between the drain-source terminals. Therefore, the proximity of the metal Fermi level to the HOMO and LUMO of the organic semiconductor is a key parameter in the device theory of OFETs. In the presence of a positive gate-source voltage, negative charge carriers accumulate and form a n-type channel, however, if the LUMO level lies away from the metal Fermi level, injection of electrons becomes difficult and a low drain current is obtained.1
An important device parameter of FETs is the on-off current ratio which is dependent upon the field effect mobility and hence the conductivity of the OSC used. As FETs are widely used in digital logic circuits as switches, a high on-off current ratio is desired to ensure that the OFF state leakage current is minimal. High ON current can be obtained by a large value of field-effect mobility.
Mobility variation in OSCs: An important distinction between silicon and organic FETs arises from the fact that OFETs have a field-effect mobility that is dependent on the gate bias and is thermally activated. This can be explained by taking into account the role of trap states in an organic semiconductor. As the gate bias voltage is increased in an OFET, trap states which are initially empty get occupied by charge carriers. At high bias all trap states are filled and further trapping is limited. This facilitates charge carrier transport and hence high field effect mobility is observed in this regime. It is also important to note that during channel formation each trap has a different effect on the carrier mobility due to a different time constant associated with each trap state.6,7 Temperature variation of trap time constant (slow, intermediate or fast traps) and the filling fraction of the trap states in the semiconductor also affect the overall mobility in an OFET.
In OSCs, it is commonly found that the hole mobility is higher and hence, p-type OFETs are more common and practically viable. Hole mobility of the range of 0.1 to 1.0 cm2/Vs have been reported for OSCs.8 As of now, amongst the high mobility materials for p-channel and n-channel OFETs are pentacene and fullerenes respectively.10
Threshold voltage: Threshold voltage for silicon FETs is described as the minimum applied gate voltage at which strong inversion occurs. OFETs on the other hand operate in the accumulation mode and hence the threshold voltage is defined differently for these devices. The threshold gate voltage in an OFET is characteristic of the point where majority of the trap states in the semiconductor layer are occupied and appreciable drain current is obtained. This minimum gate voltage required to turn ON the device is the threshold voltage of an OFET.
I-V characteristics: The drain current in an organic FET can be described by the same relations used for silicon FETs. In the linear region i.e. for, the drain current is given by:
In the saturation region i.e. for >0,
An important parameter related to FET operation is the transconductance 'gm' of the device. It is a measure of the modulation of the drain current ID with respect to the gate voltage. At constant drain-source voltage VDS the transconductance can be defined as:
Following from equations (1) and (2), in the linear operation region:
And in the saturation region:
Figure (6) shows the various DC characteristics associated with a p-channel TFT-type OFET employing DNTT as the organic semiconductor. The OFET has a channel length (L) of 10Âµm and a channel width (W) of 100Âµm.1
Sub-threshold conduction in OSCs: Ideally when the applied gate-source voltage is less than the threshold voltage, the current through the device should be zero. However during operation, the transition from the ON to OFF state is rather gradual due to the thermally excited carriers that diffuse across from the source to the drain terminal.
Fig. 6. (a) Output characteristics (b) Input and transverse characteristics (c) Transconductance 'gm' versus VGS (d) Carrier field-effect 'Âµ' versus VGS1
This phenomenon is known as sub-threshold conduction and is undesirable as it causes deviations from the desired ideal switch-like behavior of FETs.9 Sub-threshold conduction also contributes to OFF state leakage current, giving rise to static power consumption. Hence, it is favorable that the drain current should drop quickly as soon as VGS falls below the threshold voltage. This is measured by the sub-threshold swing S and is given by the expression:
In the above equation, 'n' denotes the ideality factor and is related to the semiconductor-dielectric interfacial trap density and Cdiel by the following expression:1
Silicon MOSFETs enjoy the advantage of having a high quality Si/SiO2 interface and hence the trap density contribution to the ideality factor is negligible. For n=1 (as in the case of silicon FETs), value of the sub-threshold swing is 60mV/dec. Organic FETs on the other hand, have a semiconductor-dielectric interface of lower quality and hence a higher ideality factor as compared to silicon FETs. The p-channel OFET of Figure (6) has a sub-threshold swing of 80mV/dec.1
Present limitations and future trends
The question of carrier transport in organic semiconductors still remains open and needs to be addressed in order to make further progress in the performance of OFETs. A better understanding and modeling of the charge carrier transport in OSCs, especially small molecules could hint at reasons for inherent limitations in mobility and their possible solutions.
As of now, it seems unlikely that OFETs would replace silicon transistors especially in applications that require large transistor counts, small chip-size and high frequency operation.1 Silicon MOSFETs are commercially popular owing to their efficient application to CMOS technology. P-channel and n-channel silicon MOSFETs are fabricated on the same substrate and logic circuits are implemented in a complementary fashion. This ensures low static power consumption and high frequency operation.
Organic FETs do not enjoy these benefits. For application of OFETs to CMOS logic, n-channel OSCs that remain stable upon exposure to environmental quenchers (like oxygen) are necessary. As earlier mentioned, such n-type OSCs are rarely reported and finding such materials will be fundamental for the design of organic complementary circuits.
. Also, the same conjugated semiconductor cannot be used to prepare n-channel and p-channel OFETs. Therefore, organic complementary logic circuits can only be fabricated by using two different dedicated OSCs to form n-channel and p-channel OFETs. An alternative to this is to implement logic circuits by using only p-channel OFETs. Unipolar circuits however have a static power consumption that is several orders of magnitude higher than that of silicon CMOS circuits.
Fig. 7. Flexible active matrix displays have been made possible by the deposition of organic materials on inexpensive substrates such as paper.2
By far, the most appealing property of OFETs is the ease with which they can be deposited on virtually any substrate including very low-cost substrates such as plastic foil, paper and glass. An emerging class of devices being realized with OFETs is active-matrix electronic paper display, which is very likely to be commercialized. These devices consist of a backplane of organic transistors, each one of which operates as a switch to control the display of microencapsulated electrophoretic ''inks'' formed by charged pigments.2 When the OFET is switched ON, an electric field is generated causing movement of the pigment within the microcapsules, which changes the color of the pixel.
Another promising application of OFETs is in the field of optoelectronics. Thin film OFETs provide a useful device structure for more efficient light generation as compared to the vertical device geometry of Organic Light Emitting Diodes (OLEDs).5 Organic Light Emitting Transistors (OLETs) use a planar device structure whereby the charge transport occurs in the plane at the semiconductor/dielectric interface. Fig() schematically shows the structure of an OFET and OLED and gives a comparison of the optoelectronic processes occurring inside the two device geometries.
Fig. 8. Schematic representation of the OLET and OLED structure, (a) Side-view of an OLET: Charge carrier transport in an OLET is in-plane at the semi-dielectric interface; (b) Top view of an OLET: Mobile charge carriers are injected from the source (S) and drain (D) contacts while the position at which recombination occurs in the channel is controlled by the bias on the gate (G) contact; (c) In an OLED, charge carriers move vertically across the organic layers.5
The device geometry of an OLET gives the advantage of a higher electroluminescence quantum efficient as compared to OLEDs. Moreover, the presence of an additional electrode in an OLET structure facilitates reduction in the exciton-electron quenching, giving improved electroluminescence. Hence, the charge carrier mobility and the expected current density in OLETs are typically 3-4 orders of magnitude higher than OLEDs.
Devices based on organic materials can be fabricated with low-cost technologies such as direct printing, ink-jet printing, spin-coating and other solution-based processes. Hence, OFETs have great potential for applications where low-cost, flexible and large-area coverage devices are needed. Intense research work on the fabrication techniques and organic materials has enabled device performance of the same order as amorphous silicon. However OFETs are still hampered by inherently limited performance for certain applications and can presently only compete in application markets where cost is the key factor and performance requirements do not make the use of inorganic semiconductors such as silicon absolutely necessary. In this regard, OFETs are particularly promising in the field of optoelectronics, electronic displays and biological sensing.