A wide range of organic semiconductors (OS) are currently being explored extensively for their applications in organic electronics (OE) primarily due to the growing need of substituting silicon (Si) with some alternative cost-effective materials offering relatively simpler and commercially viable technologies [1–7] at least in some niche areas. Based on the recent developments in the area of OS, processes involving printing of semiconductor, conductor, and dielectric patterns on a variety of hard as well as flexible substrates are fast emerging as part of a commercially viable technology for the upcoming OE devices. In contrast to OS (primarily Si), the additional possibility of chemically modifying the organic molecules by incorporating a number of functional groups during synthesis resulting in a number of chemical functionalities along with significant influence on the charge carrier transport properties  is adding further impetus to make this kind of search more attractive and meaningful. The examples, cited later, highlight this concept further. For instance, it is now easy to design a molecule to have the desired features in terms of its solubility in a specific solvent, color of the light emission and a typical crystalline molecular packing, to name a few. Some such features realized from modifying the molecular designs have already been put to use in a number of newer applications as mentioned here. A good example is that of nonvolatile memory elements , wherein polyvinylidene fluoride copolymers and tri/tetra-fluoroethylene radicals  were used in preparing useful devices that were found especially appropriate for flexible electronic circuits. In another case, the end groups of an organic sensor array were modified such that they not only responded to chemical and biological species but also measured pH levels , food freshness, toxic compounds, stress, and pressure in apparels . Rapid growth witnessed in OE is offering several newer areas of applications with better performances, improved reliability and stability, long lifetime, good control, and reproducibility as demanded by the different industry sectors. For instance, organic light-emitting device (OLED)-based displays, marketed by Philips and Organic Electronic Components, in cell phones  and other mobile devices along with car radios and digital cameras are some of the applications that are expected to multiply in the near future.
The quest for exploring the electronic applications of OS has already grown manifold in recent past [8, 14–23] while attempting to replace the use of inorganic materials in some relevant areas. In addition, efforts are also being made to explore newer concepts and theoretical models including molecular-level designs for controlling the structural, physical, and chemical properties of organic molecules to meet the system requirements better in the future. This kind of approach is currently more relevant, as the earlier attempts made in realizing OS devices faced unavoidable problems of material instabilities during processing besides the large density of structural defects introduced during material growth that prevented the fuller exploitation of the intrinsic charge carrier transport properties.
OS are currently getting almost practically poised for supplementing and/or replacing the conventional OS, especially in some niche areas, because of the associated distinct advantages. From the 1980s onward, OS have gone through various stages of development meant for improving the material quality that led to and even surpassed the performance of amorphous Si (α-Si)  in terms of the charge carrier mobilities in organic thin-film transistors (OTFTs). Easier solution printing of the organic materials at low temperatures on even fairly larger area substrates without using a high vacuum system [24–28] is well worth considering, where various processes including screen , ink-jet [30–33], and microcontact [34, 35] printings have been especially advanced for flexible and transparent device fabrications [36–39] employing plastic substrates offering practically feasible integration of LEDs and organic photovoltaic (OPV) solar cells on the same substrate.
Knowing that a single-crystal semiconductor is an ideal platform for studying and exploiting the intrinsic charge carrier transport properties, extensive efforts were put in to grow crystalline OS [40–43] and successive improvements achieved in material crystallinity did show significant improvement in the charge carrier mobility. For example, very low values of the mobilities, measured in earlier samples, in the range of 0.001 to 0.1 cm2/Vs  were pushed to as high as 20 cm2/Vs in crystalline ruberene and pentacene samples at room temperature. Still higher mobility of 400 cm2/Vs was measured in naphthalene samples at 10 K . This kind of improvement in charge carrier mobility provided a real push for using them in device fabrications [8, 46].
A number of OS-based devices, especially OLEDs [2, 47–49], OPV solar cells [50–59], and organic field-effect transistors (OFETs) [1, 60–64], are fastly being developed involving conjugated organic molecules that offer special features in terms of tunable energy band-gap, redox potentials, and charge carrier transport properties, besides being easy to process. These materials and processes are certainly going to offer the lightweight, low-cost, thin-film, large-area, and flexible electronics of tomorrow.
1.1 Organic molecules for device applications
From the point of view of device applications, the organic molecules are conveniently classified into groups according to the nature of the charge carrier transport they support, for instance, hole and electron-transporting (HT/ET) materials, besides their architecture involving small-molecule oligomers or macromolecular polymers and dendrimers.
Commonly used definition of p- or n-type material does not necessarily reflect the intrinsic ability of an organic material to transport holes or electrons as mentioned in the published literature. Rather, it only specifies how easily holes and electrons are injected from their contact electrodes. This subtle deviation from the classic definition was further elaborated in subsequent studies [65, 66], where it was shown that, in many organic materials, although the intrinsic electron and hole mobilities may be comparable, their drastically reduced values, measured experimentally, may be the result of external influences of the traps or instabilities introduced by water, hydroxyl groups, or oxygen exposures [67, 68]. This external influence on intrinsic charge carrier mobilities was further illustrated in the case of SiO2 gate dielectric that carried a large density of hydroxyl groups on its surface, which acted as traps for injected electrons into the channel. However, the process of covering the dielectric with BCB resulted in good electron transport in materials such as polythiophene, polyfluorene, and polyparaphenylene-vinylene with charge carrier mobilities in the range of 10-2 to 10-3 cm2/Vs .
Small organic molecules for device applications include polycyclic aromatic hydrocarbons, fused heterocyclic aromatic compounds, oligothiophenes, oligoarylenes, and macrocycles such as phthalocyanines, fullerenes, and perylene pigments along with perylene and violanthrone as electron donor and conducting materials, respectively [69, 70]. Crystalline pentacene, tetracene, and ruberene are, so far, the best molecules for OE. Ever since the realization of sexithienyl-based organic transistors, the interest in oligothiophenes has been growing quickly . Although most oligothiophenes and oligoacenes are p-type materials, their fluorinated backbones do show n-type behavior . Vacuum-deposited TPD from triphenylamines family were extensively used as hole transport layers (HTLs) in OLEDs. Perylene exhibited a herringbone-like molecular packing, where attaching perylenetetracarboxylic dianhydride (PTCDA) or perylenetetracarboxylic diimide (PTCDI) moieties produced n-type materials . Pentacene and numerous other aromatic hydrocarbons are frequently used in fabricating OFETs . Well-ordered and crystalline oligothiophenes are currently being investigated for optoelectronic applications especially in the form of molecular wires [73–76] using raw as well as alkyl-substituted forms including 16-, 20-, and 27-mers. Besides crystalline oligothiophenes, amorphous oligothiophenes [76, 77] have also been used in several devices as reported in published literature [73, 75, 78–83].
Although the significant influence of the material crystallinity on device performance was already demonstrated while comparing the mobility of 35 and 1.5 cm2/Vs  in devices prepared from crystalline and thin-film pentacene, respectively, still it is interesting to note that the polycrystalline organic materials make better OFETs and OPV solar cells despite having lower mobilities compared to their crystalline counterparts . Nevertheless, using amorphous materials in device fabrication has its own merits over the crystalline ones by considering the ease in processing, transparency, and homogeneously isotropic material properties.
During thin-film deposition, smaller organic molecules easily convert into glasses  possessing no grain boundaries but are otherwise replete with disorders in intermolecular distances and orientations, which turn out to be helpful in preparing uniform films by vacuum evaporation or spin coating with well-defined ordered molecular structures. Besides, nonplanar molecules such as triarylamines, spiro-compounds, and 1,3,5-triphenylbenzene or 2,4,6-triphenyltriazine exhibit mobilities in the range of 10-2 to 10-5 cm2/Vs . Compounds such as quinoxaline, benzimidazole, pyridine, and oxadiazole having electron-withdrawing properties [85, 86] are found to be useful for device applications, for instance, materials such as 1,3,5-tris(9,9-dimethylfluoren-2-yl) benzene  and 2,4,6-tris[di(2-pyridyl)amino]-1,3,5-triazine offer better form of electron transport layers (ETLs) for OLED applications .
In contrast to general polymers, where significant variations are seen in the average molecular weight and in their batch-to-batch distributions causing varying device properties accordingly, well-structured dendrimers are monodisperse macromolecules with well-defined functionalities according to their generation number, size, and molecular weight. Additionally, in dendrimers, a regular correlation is found to exist between the structure and the charge carrier transport properties. For instance, the dependence of charge carrier transport properties on the nature of the core, the total generation number, and the presence of functional groups in each layer of a dendrimer was carefully studied  using spin-coated amorphous dendrimer thin films. The experimental results indicated that, in the case of a phenylene-vinylene encapsulated triphenylamine core , there is a drastic reduction in mobility from 5.1×10-6 to 5.5×10-9 cm2/Vs after changing the generation number from 0 to 3 . The PAMAM decorated [Ru(bpy)3]+2 dendrimers exhibited similar trend of mobility reductions with increasing generation number . In contrast, iridium core dendrimers with carbazole in each layer  exhibited twofold increase in the mobility from (3–6)×10-5 cm2/Vs for generation 1 to (7.3–12)×10-5 cm2/Vs for generation 3.
Conjugated polymers have been studied extensively for their applications in flexible electronics [54, 59, 92–99] and optoelectronic devices [51, 93, 100] as reported in the published literature. Several methods were subsequently developed to take care of the polydispersion-induced disorders in such thin films. For example, the inherent nature of P3HT leading to self-assembly was found to be helpful in enhancing the mobility in the range of 10-4 to 10-1 cm2/Vs depending on the processing conditions including solvent, annealing time, and temperature [60, 62, 101]. The lamellar assembly of P3HT molecules along the plane perpendicular to the substrate with π-π interactions among the 3-hexylthiophene rings  showed higher mobility measured in a plane parallel to the substrate, while that along the plane normal to the substrate showed reduced values  due to poor carrier hopping.
Because of higher degree of disorders in organic materials especially in thin films, significant contributions are required from other considerations than the charge carrier hopping to show moderate mobilities in the range of 10-6 to 10-1 cm2/Vs. For example, poly(thienylenevinylene) molecules with lower band-gap of 1.6 to 1.8 eV exhibited higher hole mobilities of 10-1 cm2/Vs, making them useful in OPV applications.
Mobility measurement in naphthalene and anthracene crystals  clearly established that the intrinsic behavior of charge carrier transport was realizable only in the bulk crystals. It further demonstrated enhanced mobility at lower temperatures with pronounced anisotropy [104, 105]. However, there are applications where surface transport is more relevant, for instance, in the case of OFETs. Field-induced charge carriers, moving along the interface between the OS and the gate, are essentially confined to only a few monolayers [1, 106, 107], where the charge carrier density is especially several orders of magnitude higher than that encountered in the time-of-flight (TOF) mobility measurements besides facing strong electron-polaron interactions [108, 109]. Additionally, movement of the charge carriers in the channel is not only affected by the polarization of the gate dielectric  but also by the molecular arrangements present at the organic surfaces, which may be very different from that in the bulk. It is thus clear that the study of polaronic transport on organic surfaces is necessary to understand the processes that determine the overall performances of these organic devices. Despite having limited knowledge of transport properties of OS, the family of all-organic devices of the active matrix displays involving OLEDs and OFETs is already entering into commercialization. This is, incidentally, very different from the situation that existed in the case of the inorganic electronics in the mid-1960s, when the first Si transistor was developed .
1.2 Material selection criteria for OS devices
Based on the concept of a conventional FET, simple replacement of the inorganic channel by an organic one was adapted in early developments of OFETs. In actual device fabrication, it begins with the coating of the organic layer onto the patterned source and drain contacts on a SiO2-coated heavily doped Si substrate, wherein the Si gate induces charge carriers in the channel region according to the applied gate bias by modulating the channel conductivity. Although the OFET channel is certainly different from that of conventional Si, still the basic transistor architecture and the functioning remain identical to that of an inorganic counterpart. In subsequent developments, all organic devices were also developed using organic gate electrode and dielectric layers, but such devices were found to be comparatively inferior.
The charge carrier mobility in the channel region is an important FET parameter as the higher the charge carrier mobility, the faster it switches the device. Thus, the primary target during device design and fabrication is to get as high mobility and on/off ratio as it is practically possible. For example, a typical mobility and on/off current ratio for an α-Si FET is approximately 1 cm2/Vs and 108, respectively, as a reference for comparing the emerging OS devices.
Despite sharing the same device architecture and electrical operations of an inorganic device, the charge carriers in an organic device “hop” from one π-orbital to another on the molecular backbone, for which several ways have been developed to maximize the mobility and on/off ratio as briefly discussed below. For meeting the above target, it is, first of all, necessary to ensure that the energy levels of the contact electrodes and the active layers align well. In the case of a gap between the two energy levels, there would be an equally large potential barrier responsible for reduced charge carrier injection. For having proper energy-level alignment, it is necessary to have either the right kind of electrode material with the desired value of its work function or chemically functionalize the electrode surfaces to shift its energy level to the matching position with respect to the selected electrode material as discussed later. Next, the organic molecules in the channel region must be appropriately organized to ensure the optimal charge carrier transport. Here, proper molecular design, improved self-alignment through surface modification, and thermal annealing all help to enhance the orbital overlap leading to improved charge carrier hopping [112–116]. Subsequently, the dielectric film used must provide the maximum capacitance, which is possible either using high dielectric constant material or reducing the dielectric film thickness. The thinner the dielectric layer, the lesser is the operating voltage required to turn the device “on”. While depositing thin dielectric layer, it is essential to ensure the film uniformity so that the leakages through relatively thinner regions or defects left within are minimized. In addition, the dielectric layer should also facilitate promoting self-organization of the semiconducting layer deposited on top. Finally, the device dimensions should be decided carefully in optimizing the OFET performance. For instance, reducing the channel length results in not only higher mobility but also provides lesser chances of meeting defects in the path of charge carriers or amorphous regions inhibiting charge carrier flow.
Keeping these factors in mind is certainly helpful in improving the OFET performance as briefly mentioned above, it is useful to know what all have been attempted in this context as mentioned here. In recent years, OFETs were explored for flexible display backplane applications. Charge carrier mobilities in the evaporated- and solution-processed materials, in the mean time, have already exceeded the best value of the α-Si devices. In a recent review , it was noted that, by mid-2011, there were about 40 different OS possessing mobilities above 1 cm2/Vs. However, despite having access to such a large number of OS, there were only a few that could go up to the stage of large-scale integration necessary for display backplane arrays possibly due to serious reliability issues.
OS are either aromatic ring-based small molecules or the conjugated polymers. Generally, lower solubility of small molecules makes it difficult to use them in low-cost fabrication processes involving solution-coating techniques. Instead, they are better vacuum evaporated to get the crystalline films. Taking pentacene as an example of a small-molecule, better-quality pentacene film, OFETs have already surpassed the performance of the best α-Si counterpart after exhibiting mobilities in the range of ∼3 cm2/Vs, three times that of the highest mobility of α-Si [118, 119]. However, due to a higher degree of conjugation and low ionization potential (IP), pentacene molecules oxidize very fast in normal ambience degrading the device performance leading to serious reliability problems. Because of such a drawback, the search for a soluble precursor for pentacene deposition after fabrication has been a subject to investigate in the future. On the contrary, the conjugated polymers exhibit better environmental stability and tunable electronic properties due to extensive π-conjugation in addition to better solubility due to appropriate side chains. Although these materials are easy to coat as thin films, perhaps a simple coating of complex molecules generally results in inferior molecular order showing relatively poor performance compared to small-molecule films as observed in the case of P3HT, which has the mobility of 10-3 cm2/Vs for an unannealed film on a plain substrate. However, by activating the dielectric surface to promote molecular self-organization [3, 120] and thermal annealing, it could improve the molecular ordering leading to the mobility up to 0.1 cm2/Vs. It is thus concluded that although polymers are poor semiconductors from a conductivity point of view, they are otherwise better in stability and ease of processing than small-molecule materials. It was further noted that not only is the π-conjugation along the main chain important, but so are the nature of the side chains and narrow molecular weight distribution which both help in improving the crystallinity and reducing the band gap leading to higher mobilities . Knowing well the importance of reducing the intermolecular spacing through improved crystallinity, methods based on surface modification to promote “edge-on” orientation and regioregularity-based molecular designs to promote lamellar ordering were successfully developed. While trying to improve the charge carrier transport by conjugation enhancement using closely packed π-orbitals, it was also noted that the presence of higher energy levels and smaller band-gaps imparted a better mobility but otherwise made these materials more susceptible to the oxidation and photodegradation under ultraviolet (UV) light exposure. To alleviate these problems, side chains were, very often, added to improve solubility by distorting the molecular conjugation as a result. For example, in PQT-12, the presence of solubilizing side chains on the first and fourth thiophene rings [121–123] leaving two thiophenes without side chains created a slight twist in the molecular chain that did not disturb the conjugation much but otherwise created a larger band-gap, making it very stable under ambient conditions while still retaining higher mobility. Such stable PQT-12 films exhibited mobility of 0.07 to 0.18 cm2/Vs.
A dielectric layer with structural perfection and low trap density is required for the OFET gate insulator [124–126]. For easier movement of the charge carriers across the channel surface along the dielectric semiconductor interface, it needs lower surface energy as well as polarity. Moreover, the dielectric layer deposition must ensure a very smooth surface, especially in the case of top contact devices where the surface roughness is noted to affect the molecular ordering adversely. In the case of very rough surfaces, molecules are not able to orient into lamellar structures resulting in crystal domain defects, which behave as traps, reducing the overall device performance. For instance, a surface roughness of even a few nanometers disturbs the charge carrier conduction across the channel in the proximity; consequently, a much larger electric field is required to create a conduction channel to overcome such a disturbance , which defeats the basic purpose of achieving the target of the low-voltage applications of OFETs.
Dielectrics have been chosen from both inorganic and organic materials, where the most common ones are SiO2 and SiN used in test devices with a heavily n-doped Si as the gate electrode. Besides being expensive to deposit, using complex methods such as physical and chemical vapor deposition, most of the inorganic dielectric films are brittle and as such are not appropriate for flexible substrates. To avail of the solution-based processing methods, polymeric dielectrics were thus preferred, which were usually insulating, having a large band-gap that made them good dielectrics. Defect-free and a few tens of nanometer thin dielectric films are possible now to deposit easily from solution. Electrophoresis and in situ polymerization methods have also been used to deposit very smooth and very thin dielectric films. For instance, the common organic dielectric materials include polymethyl-methacrylate (PMMA) , polyimide , and polyvinyl alcohol (PVA)  for their device applications.
The performance comparison of OFETs using Si/SiO2 and flexible polymer substrates was made  in terms of mobility and VT. PEN substrate, polyimide dielectric along with pentacene and PTCDI-C13 as p and n-type semiconductors were used in these experimental OFET devices. It was noted in this study that the n-type devices on flexible substrates exhibited better mobility and lower VT in contrast to those on Si/SiO2. On the contrary, p-type devices on Si/SiO2 substrates were better than those on the flexible substrates .
While studying the influence of annealing on structure and electronic properties of pentacene OFETs with polyimide dielectric , it was noted that 140°C annealing improved the mobility by almost doubling from 0.07 to 0.12 cm2/Vs along with substantial reduction in trap density. Further measurements confirmed that the crystal structure of the pentacene film on polyimide did not change with annealing up to 140°C, whereas pentacene on Si/SiO2 substrate exhibited transitions from the (001) thin-film phase to the bulk phase . Gate dielectrics prepared from methylated poly(melamine-coformaldehyde) (MMF) and polyvinyl phenol (PVP) for flexible pentacene OFETs  on polyestersulfone (PES) substrate exhibited mobility of 0.88 cm2/Vs along with reduced hysteresis and leakage current .
For OFET contact electrodes, the foremost important parameter is the proper matching of the work functions of the electrode and the active region materials as mentioned already. Gold (Au) is the preferred electrode material for organic devices, as it has a work function of 5.1 eV, which is fairly close to the highest occupied molecular orbitals (HOMOs) of many conjugated polymers in the range of ∼5 eV. In case a conducting material is not available with a work function close to the energy level of the semiconductor, surface treatments are available to modify the work function for minimizing the detrimental effect of contact resistance as discussed later separately. Having chosen the material with aligned work functions, the next parameter is the material conductivity, which should be as high as possible. For these reasons, the natural choice falls mostly on metals. Unfortunately, in the case of patterned metal contacts, prepared by vacuum deposition and shadow masking, it is rather difficult to keep the cost low. Consequently, the recent developments have been looking for alternate options such as either using metal precursors in solution form to deposit continuous metal films after printing followed by reduction and annealing [126, 134] or employing metal nanoparticle (NP) suspensions to form a continuous film after printing and sintering [126, 135]. The solution-processed metal films do not possess as high conductivity as that from the vacuum-deposited parent conductor but are otherwise reasonably adequate for organic devices. The metal NP films require sintering to render the printed electrodes conducting, but this step can be easily converted into a roll-to-roll manufacturing, as the sintering temperature is only approx. 120°C to 200°C depending on the size and capping agents of the metal NPs. Using this method, electrodes are printed with reasonable resolution and high conductivities, for instance, using Au and silver (Ag) NP suspensions, where conductivities have been realized in the range of 4–10×106 and 2–4×106 S/m, respectively. On the contrary, an organic conductor was reported using poly(3,4- ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) combination as a transparent conducting film with conductivity up to 103 S/m . Although this value is enough for organic conductors, it is still low compared to the solution-processed metal electrodes, and for this reason, PEDOT:PSS films are generally preferred where a transparent electrode is required, such as a cathode for OPV solar cells.
In order to modify the conductivity of polymeric materials beyond the limit set by crystallization and annealing, possibilities of preparing composites were examined. For example, a low band-gap semiconductor  PDPP/TNT was prepared by combining naphthalene and diketopyrrolopyrrole (DPP) as the acceptor and donor component, respectively, in a polymer backbone that exhibited hole mobility of 0.65 and 0.98 cm2/Vs in the bottom gate and dual gate OTFT device configurations, respectively, besides being well suited for OPV solar cell applications due to its band gap of 1.5 eV and HOMO of 5.29 eV. While using this semiconductor with PC71BM as D-A pair in OPV solar cells, an efficiency of 4.7% was reported . In a more recent publication , the structural design, optoelectronic properties, molecular organization, morphology, as well as the device performance in OFETs and OPV solar cells using high-mobility DPP-based materials were examined in detail. By extending the basic concept of forming highly ordered crystalline domains while associating the DPP unit with an electron donating polymer, the charge carrier transport was further improved due to intermolecule and interdomain interactions besides energy-level adjustments, resulting in p/n-type/ambipolar OS. In contrast to the observed mobility of 1 cm2/Vs and efficiency of 4.7%, as noted earlier , more recent semiconductor formulations exhibited still higher hole, electron, and ambipolar (hole/electron) mobilities of 10.5, 3, and 1.18/1.86 cm2/Vs, respectively , and efficiency of 6.05% in the OPV solar cells .
The influence of the starting purity of the polymer on the performance of OFETs was specifically studied using in situ measurement of electrical characteristics. Starting with purified and nonpurified pentacene samples, it was noted  that the field effect started appearing at 1.5 nm thickness and above in purified samples in contrast to that starting at 3.0 nm and above for the nonpurified counterpart. Moreover, the hole mobility improved from 0.13 to 0.23 cm2/Vs after sublimation purification of the pentacene, which also exhibited larger grain size and film coverage, resulting in better crystallinity of the thin-film structure due to the absence of the impurities .
1.3 Relevance of fullerenes, nanotubes, and graphene in OS devices
Despite considerable advances made in developing high-quality organic materials, the mobility has still been relatively lower as the most polymeric materials exhibited mobilities in the range of 0.1 to 0.6 cm2/Vs. One way to improve the mobility without synthesizing new materials was to introduce an additive within the polymeric matrix that resulted in something imbibing the properties of either of the components involved. From a large number of additives available for OS-based composites, carbon allotropes were found to be the best in this context. This originally started with carbon fullerenes and nanotubes (CNT) with P3HT in OPV solar cells , where fullerene became an obvious choice due to its better electrical properties and the ease with which it could be dispersed in the organic matrix in bulk heterojunction solar cells. Although CNTs did follow as the next additive their one-dimensional (1D) structure did not improve the solar cell performance; rather, their applications in OFETs appeared more promising [141, 142].
The advantages of using CNTs include ultrahigh conductivity/mobility values, arising from very efficient conjugation present along the CNT axis that imparts very efficient charge carrier transport along the length as well as low band-gap and very low threshold voltages. Unlike multiwalled CNTs (MWCNTs), where each unit is always conducting, single-walled CNTs (SWCNTs) behave both as semiconducting and conducting entities, and due to this nature, they are not used alone as they will short circuit the transistor units. It is therefore advisable to disperse mixed CNTs in an organic matrix for device applications. The large surface-to-volume ratio and an extended π-conjugation make MWCNTs form bundles very easily, which can be checked by using stabilizer having functional groups that can covalently bond to the outermost layers. In one way, this appears helpful, but finally it disrupts the electron delocalization causing formation of charge trap sites leading to impaired electrical conduction . Further, these functional groups also introduce additional energy barriers within CNT composites by interrupting the interaction between CNT and the polymeric matrix. This is also true for graphene sheets, which are SWCNTs cut open to form a monolayer thin sheet. The graphene being a two-dimensional (2D) C-allotrope, the same extensive network of π-orbitals is present there on its surface leading to similar aggregation as observed in CNT bundle formations, and a large amount of energy is required to break such aggregations.
The process of physically reducing the channel length for increasing mobility by bringing source and drain contacts closer is useful in device fabrication, but it has its own limitations. Alternatively, blending conductive additives, CNTs, and graphenes with the host polymer or stabilizer molecules also reduces the effective channel length causing mobility improvements without asking for physically bringing the electrodes closer. In addition, CNT and graphene composites exhibit higher mobility by enhancing the charge carrier transfer between two disordered regions of the amorphous phase that are invariably there in polymeric films mixed with the regions of ordered crystalline structures. Proper dispersion of conducting additives helps in reducing the influence of barriers created by disordered regions by providing additional bridging regions . For instance, a small amount of CNT/graphene helps in linking the ordered regions and improves the charge carrier mobility by reducing the possibility of charge carrier transport across the disordered regions. However, there is a typical percolation problem associated with CNTs, as an additive concentration is in the range of 10 to 20 wt% causing shorts in OFETs.
Reducing the energy band-gap  is yet another possibility of increasing the polymer conductivity. In general, most OS have an energy level above -5 eV. The SWCNTs have a work function in the range of -5 to -5.2 eV, which places them somewhere between the semiconductor and Au electrode. Thus, a CNT composite, with such a semiconductor, would have an overall energy level slightly lower than the pure polymer, which would reduce the energy barrier for efficient injection of charge carriers. In this context, P3HT has been widely used as the matrix polymer. From among several methods developed to prepare CNTs and graphene composites in stabilized forms, the one based on chemical functionalization is preferred over the others [146–155]. MWCNTs are better mobility-enhancing additives compared to SWCNTs, as the outer layer of MWCNTs can be easily functionalized without disturbing the conjugations of the inner shells. Such is not the case in a single sheet of graphene, but using few stacked graphene layers known as graphene nanoplatelets (GNPs) can be used as an alternative in the same context.
Because carbon fullerenes, CNTs, and graphenes have all been extremely useful additives for preparing nanocomposites, it will be helpful to quickly look into the ways of preparing CNTs and graphenes for realizing the right kind of composites for device applications.
For preparing a CNT-polymer nanocomposite, the nature of the interactions between CNTs and the matrix and the degree of dispersion would decide the overall behavior of the end product. The CNTs are, as such, inert additives facilitating load transfer across the CNT-matrix interface through van der Waals interactions with chemically stable constituent carbon atoms. Therefore, for realizing a good composite with an objective of improving the charge carrier transport, methods would be needed to modify the surface properties of the CNTs for their improved dispersions besides their active participation in carrier transport.
CNT functionalization employs either chemical or physical methods based on the interactions between the active organic molecules and carbon atoms of the CNTs. In the chemical method, the functional groups attached to the CNT sidewall change sp2 to sp3 hybridization, resulting in reduction of π-conjugation. This was clearly evidenced in reversible fluorination of SWCNT fluorination, where anhydrous hydrazine could easily remove the fluorine atoms . Additionally, C-F bonds, involved in fluorination, were relatively weaker than those of alkyl fluorides . They, in turn, offered better substitution sites for additional functionalization  involving amino, alkyl, and hydroxyl groups [159, 160]. Besides, other conjugation methods including the Diels-Alder reaction, carbene and nitrene addition [161–163], chlorination, bromination , hydrogenation , and azomethineylides  reactions were also explored successfully in this context.
The defect sites present on the opening ends and/or holes in the sidewalls as well as the pentagons/heptagons in the hexagon network including oxygenated sites on CNTs all were used in the “defect functionalization”. For instance, oxidation induced defects created on the sidewalls and the open ends of CNTs by treating with strong acids such as HNO3, H2SO4, or mixture  or with a strong oxidizing agent as KMnO4 , ozone [169, 170], and reactive plasma [171, 172]. It was possible to implement the defect functionalization. In addition, attaching (-COOH) or (-OH) groups stabilized the defects on CNTs created by the above-said oxidants, as these functional groups, once attached, imparted a rich chemistry to the CNTs that could also be used as a precursor for further chemical reactions such as silanation , polymer grafting , esterification , thiolation , and alkylation and arylation . These functionalized CNTs dissolve easily in organic solvents because the attached polar groups change them from hydrophobic to hydrophilic forms. Thus, it is anticipated that the chemically functionalized CNTs participate well in forming interfacial bonds with many polymers, enabling CNT nanocomposites to possess improved mechanical and functional properties.
Although a number of covalent functionalization schemes were successfully developed to attach functional groups onto the surface of CNTs, in general, these techniques suffered from two major drawbacks. First, the ultrasonic treatment required for initiating functionalization reactions created a large number of defects on the sidewalls, and in some extreme cases, even CNTs were fragmented into smaller pieces, which not only degraded the mechanical properties but also disrupted the π-electrons of the nanotubes (NTs). Such disturbances ultimately affected the transport properties of CNTs, as they caused additional electron and phonon scatterings impeding the electrical and thermal conductions, respectively. Second, the concentrated acids or strong oxidants, often used in CNT functionalization, were highly toxic in nature, and due to these reasons, efforts were made to develop methods that were easy to use at low costs besides causing less damage to the CNT structure.
Keeping these above-mentioned problems in view, one process of noncovalent functionalization was alternatively developed using suspension of CNTs in poly(phenylene-vinylene)  or polystyrene  resulting in polymer wrapping around the CNTs to form supermolecular complexes employing van der Waals interactions and π-π stacking between CNTs and polymer chains containing aromatic rings.
Besides polymers, appropriate surfactant molecules were also found useful in functionalizing CNTs by studying the influence of the surfactant on dispersion and other properties of CNTs [143, 179–187]. Lowering of surface tension due to physical adsorption of the surfactant on the CNT surfaces very effectively prevented the aggregate formations. Further, the surfactant-treated CNTs overcame the van der Waals attractions by electrostatic/steric repulsive forces as a function of the surfactant, medium chemistry, and polymer matrix. It was also noted that, in water-soluble polymers such as polyethylene glycol, cationic surfactants have some advantages, whereas, in water-insoluble polymers such as polypropylene, CNT dispersion is promoted by a nonionic surfactant [184, 185]. The nonionic surfactants treatment deploys a strong hydrophobic attraction between the solid surface and the tail group of surfactant, and once the surfactant is adsorbed onto the additive surface, the surfactant molecules self-assemble into micelle forms .
In addition, endohedral confinement of atomic and molecular species inside the hollow cavity of the NTs is another noncovalent method for CNT functionalization, where they enter via defect sites, localized at the ends or on the sidewalls involving capillary action. The insertion of inorganic NPs such as Ag, Au, Pt, and C60  is a typical example in this category. Even small biomolecules such as proteins and DNA were easily entrapped in the inner cavity of CNTs by adsorption [189, 190]. Such a combination of CNTs and the guest molecules has been examined for their uses in catalysis, energy storage, nanotechnology, and molecular-scale devices .
Compared to the traditional polymer microcomposites, improved CNT dispersion leads to shorter interparticle distances, which influence the properties even at low additive concentrations. For example, the electrical conductivity of CNT/epoxy composite was enhanced by several orders of magnitude with <0.5 wt% of CNTs in one study . The mechanical properties such as strength and stiffness, combined with other physical properties of CNTs, offer many potential applications as already seen in the current experimental studies, although the commercial-level success is still awaited for its realization in the near future.
Following the first report  of preparing a CNT/polymer nanocomposite in 1994, numerous efforts were already made to understand their structure-property relationship to explore their applications in different fields and such efforts became more pronounced after the industrial-scale production of CNTs at lower costs at the beginning of the 21st century . Application-wise, CNT/polymer nanocomposites are classified as “structural” or “functional composites” . In “structural composites”, the mechanical properties of CNTs, including the high values of modulus, tensile strength, and strain to fracture, are put to use for much improved mechanical properties. Similarly, in “functional composites”, the CNT features such as electrical, thermal, optical, and damping properties along with their excellent mechanical strength are invoked in multifunctional composites for thermal insulation, chemical sensing, electrical and thermal management, photoemission, electromagnetic absorbing, and energy storage, to name a few .
It has been experienced that solution mixing is the commonest method of preparing CNT/polymer nanocomposites [185, 194, 195] involving processes such as solvent dispersion by mechanical, magnetic, or ultrasonic treatment at room/elevated temperature followed by precipitation or casting at the end.
Melt blending is another method, where thermoplastics such as polypropylene , polystyrene , and poly(ethylene 2,6-naphthalate)  are used without any solvent to disperse CNTs. In melt blending, a high shear force at high temperature disperses CNTs in a polymer that is, as such, very compatible with current industrial processes employing high-temperature extruder/injection machines. Compared to the solution mixing, however, the melt blending has limited scope to handle only low additive concentrations in thermoplastic matrices .
On the contrary, in situ polymerization offers uniform CNT dispersion in a thermosetting polymer by blending with monomers either in the presence/absence of a solvent followed by its polymerization via addition/condensation reactions with hardener and curing agents at an elevated temperature. The possibility of covalent bond formation between the functionalized CNTs and the matrix offers an additional advantage for improving mechanical properties of the composites through strong interfacial bonds.
In latex method, SWCNTs/MWCNTs are blended with [185, 194, 195] colloidal aqueous dispersions of polymer particles either produced by emulsion polymerization or converted into the emulsion form. Compared to the in situ polymerization, the CNTs are added after the polymer synthesis is complete. Thus, the first step of the process begins with the exfoliation of SWCNT bundles or dispersion/stabilization of entangled MWCNTs in an aqueous surfactant solution followed by mixing the stable dispersion of surfactant-treated CNTs with polymer latex. After freeze-drying and subsequent processing, CNT nanocomposite dispersed in a polymer matrix is obtained. This technique [185, 194] enjoys flexibility, reproducibility, and reliability in processing the two aqueous components into a viscous polymer matrix at low cost and in environment-friendly manner.
For explaining the conduction properties of the nanocomposites, the “percolation theory” has invariably been used. In this context, it is generally noted that while slowly increasing the additive concentration, the composite undergoes a transition from insulator-to-conductor state. The additive concentration at which this transition occurs is termed as the “percolation threshold” when the electrical conductivity sharply increases by several orders of magnitude due to the formation of continuous electron conducting pathways. Prior to this transition, electron pathways are not there and the matrix material dominates the electrical properties while above the transition, and multiple electron pathways facilitate conduction, which finally ends up in saturation [191, 197, 198].
In conventional conducting additives, micron-sized metal/carbonaceous materials, such as carbon black (CB), exfoliated graphite, and carbon fibers were used in high concentrations from 10 to 50 wt% [191, 197–200], resulting in poor mechanical strength. In contrast to these, the blending of CNTs into polymer minimized such problems by forming conducting pathways in the insulating polymer by converting them into conducting composites at very low additive concentrations. For most polymers, they change from an insulator to a conductor as the conducting additive concentration reaches approximately 5 wt%, but there is no fixed percolation threshold for CNTs, as this may spread over a range of 0.002 to 7 wt% [191, 201–204] primarily due to the variations introduced by the type of the CNTs used and the preparation methods involved. This large spread in percolation threshold confirms that the CNT dispersion is affected by functionalization along with the processing conditions in determining the ultimate conduction property of the nanocomposites. For instance, in one of the related studies , a 50 times increase in conductivity was measured in well-dispersed CNTs compared to that of the entangled ones. In another experiment , CNT/epoxy nanocomposites were prepared exhibiting percolation thresholds in the range from 0.1 to 1 wt% decided by the dispersion state and aspect ratio of CNTs. It is thus concluded that aspect ratio, disentanglement, and uniform distribution of individual/agglomerates of CNTs finally decided the percolation threshold.
Despite experiencing all the advantages mentioned earlier, however, such functionalizations do disturb the π-electrons and degrade the intrinsic electrical properties of CNTs besides the additional damages introduced during processing as discussed already. It is interesting to note that silane functionalized CNTs exhibited wrapping of insulating material onto CNTs , which showed no percolation behavior. On the contrary, amino-functionalized CNT nanocomposites  did show a typical percolation behavior but with conductivities lower than those containing pristine or untreated CNTs.
The electrical conductivity of CNT/polymer composites is typically in the range of 10-5 to 10-3 S/cm above the percolation threshold [201–204], but further increase in additive concentration increases the conductivity only very marginally. Additionally, the solution viscosity becomes too high to produce void-free composites for the CNT content higher than 1.0 wt%.
It is noted from the above discussions that the processing techniques for improving the electrical conductivity of nanocomposites around the percolation threshold becomes critical to produce highly conducting composites. The critical role of contact resistance of CNTs in enhancing the conductivity of nanocomposites was further substantiated by the simulation studies [207, 208]. This was experimentally confirmed , where the use of silver-decorated CNTs exhibited a significantly higher conductivity above the percolation threshold than those containing pristine CNTs alone; for instance, a conductivity of 0.81 S/cm was measured with 0.5 wt% of Ag-decorated CNTs.
Higher cost of CNTs, especially of SWCNTs, compared with other additives such as graphite, CB, and carbon fibers puts serious constraints on the large-scale use of CNT composites. Therefore, nanocomposites of hybrid additives were purposely developed as a compromise solution [209–215]. In such composites, the CNT network serves as backbone and the other conducting additives facilitate the charge carrier transport enhancing the overall conduction. For example, nanocomposites  blended with hybrid additives of 1% Au NPs and 1% CNTs showed a conductivity of 4.7×10-3 S/cm, which was 2 orders of magnitude higher than that with 2 wt% Au NPs alone. In yet another study , hybrid CNTs/CB additives were mixed for enhancing the electrical conductivity with balanced mechanical strength while reducing the cost, which not only reduced the percolation threshold but also enhanced the ductility and fracture toughness of the nanocomposites. Hybrid additives with varying geometrical shapes and dispersion characteristics offer unique ways to lower the final cost of CNT-based nanocomposites with multifunctional properties.
The next category of conducting additive involves functionalized graphene with numerous functional groups including dispersion in organic solvents after attaching certain organic groups, which is necessary for preparing nanocomposites. In addition, organic functional groups offering new properties could also be combined with the conduction properties of the graphene. In most cases, when organic molecules are covalently attached to the graphene surface, although the π-conjugation present there is disturbed, still it enables the manipulation of its conduction properties. For instance, arriving at a suitable band-gap through chemical doping is a powerful method for using graphene in nanoelectronic devices .
The covalent functionalization of graphene follows two routes: the one forming bonds between free radicals/dienophiles and C=C bonds of pristine graphene and the other forming bonds between functional groups and the oxygen groups of graphene oxide (GO). The most useful radicals for reaction with sp2-hybridized C-atoms are organic free radicals/dienophiles, which are implemented by heating a diazonium salt to produce free radical that attacks the sp2-hybridized C-atoms, which in turn is used for attaching nitrophenyls [217, 218]. In situ measurement of the conductivity of a graphene sheet during chemical functionalization with diazonium salts showed noteworthy decrease in conductivity due to disruption caused by the transformation from sp2 to sp3 hybridization as also observed in the case of CNTs. Similarly, the covalent attachment of nitrophenyls to graphene sheets  introduces a band gap, which can be controlled, making the functionalized graphenes a semiconducting nanomaterial. The reaction with diazonium salts is used for functionalizing different types of graphenes including chemically/thermally converted graphenes, single sheets from cleavage of bulk graphite, as well as epitaxial graphenes [218–223]. Another alternative method involving the reaction of benzoyl peroxide with graphene sheets , where Ar-ion laser irradiation initiated reaction on graphene deposited on a Si substrate put in a benzoyl peroxide/toluene solution, is used to modify a graphene sheet placed on an OFET device, where, apart from significant decrease in conductivity due to the increase of sp3 carbon atoms after the covalent addition of phenyl groups, an increase in the level of hole doping is also observed.
Azomethineylide is one of the commonest dienophiles that is used in the functionalization of fullerenes, NTs, nano-onions, and nanohorns, offering numerous applications in areas including polymer composites, biotechnology, nanoelectronic devices, drug delivery, and solar cells [225–230]. Graphene sheets dispersed in organic solvents [231, 232] are substituted with pyrrolidine rings via azomethineylide cycloaddition. The introduction of hydroxyl groups increases the dispersion in ethanol and dimethylformamide (DMF). This procedure allows choosing from several aldehydes or substituted α-amino acids as precursors, resulting in a variety of desirable functional groups. For example, conjugating graphene with tetraphenylporphyrin (TPP) or palladium-TPP, TPP aldehyde, Pd analogue, and sarcosin are used as useful precursors  in this context.
In another attempt, for selective binding of Au nanorods (NRs) to graphene sheets , paraformaldehyde precursors with NH2-terminated α-amino acid are used, where phenyl and alkyl azides react with the C-C bonds via intermediate formation of nitrene resulting in a variety of graphene derivatives. The aziridine ring is used  to attach several functionalities resulting in combinations of varying solubility-dispersion and surface energy combinations. Nitrene addition is also used to functionalize graphene sheets with phenylalanine. A variety of hexyl, dodecyl, hydroxyl-undecanyl, and carboxy-undecanyl radicals were also attached to graphene . For instance, attaching carboxylate group facilitated attaching Au NPs, which were used as markers to investigate the reactive site distribution.
A novel structure comprising a 2D macromolecular brush of a large number of chemical moieties on a graphene sheet was prepared  via nitrene cycloaddition possessing enhanced thermal and chemical stabilities. The functionalized graphene was further modified by different chemical reactions, including amidation, surface-initiated polymerization, and reduction of metal ions resulting in electrically conducting materials with excellent dispersion and processing features in various solvents. Covalent bond formation between thermally generated nitrene and epitaxial graphene was successfully explored  for controlling the band gap of the functionalized graphene. Similarly, aryne cycloaddition  to the graphene surface using 2-(trimethylsilyl) aryl triate as a precursor resulted in enhanced dispersion in DMF, o-DCB, ethanol, chloroform, and water.
Functionalized graphene-based flexible OFETs were fabricated for the first time  by preparing conducting source/drain/gate electrodes, with device characteristics very much similar to those with Au source/drain electrodes on SiO2/Si substrates with Si gate, with excellent flexibility without performance degradation over severe bending cycles. Furthermore, successful demonstration of inverter circuits using all-graphene-electrode OFETs confirmed that the dream of all-carbon flexible electronics is fairly close to a practical reality .
Another useful variant of graphene is GO, a single monolayer with randomly distributed sp2-hybridized C-atoms and oxygenated aliphatic regions of sp3-hybridized C-atoms containing hydroxyl, epoxy, carbonyl, and carboxyl functional groups. It was noted that the epoxy and hydroxyl groups were placed above and below each graphene layer and the carboxylic groups were usually attached along the edges. The oxygen groups existing on GO surface provided hydrophilicity and analogous chemical reactivity. GO were prepared either by oxidation of graphite with strong acidic media  or ozone  or the chemical/thermal exfoliation of graphite oxide [242–247]. Further, it was noted that GO formed unstable dispersions in water and ethylene glycol, DMF, N-methylpyrrolidone (NMP), and tetrahydrofuran (THF) solvents, having tendency to aggregate through π-π stacking forming large particles known as GO platelets. Compounds such as octadecylamine , 1-octyl-3-methyl-imidazolium , large aromatic molecules , didodecyldimethyl-ammonium bromide , polystyrene , and poly(sodium 4-styrenesulfonate)  and also elastomeric Si foams  and DNA  were explored for stabilizing GO nanoplatelets in the solution.
It was rather difficult to remove all the oxygen groups and defects even after full conversion of sp3 into sp2-hybridized C-atoms during reduction of GO. In this context, several methods of GO reduction were attempted from hydrazine [242, 243] to bacterial treatments  besides a number of other methods [171, 247, 256–258]. The conductivity of the reduced GO was invariably influenced by the residual oxygen and defects .
Superior optical limiting property was shown by amine-terminated oligothiophenes attached GO nanoplatelets compared to standard C60 and control sample consisting of GO and oligothiophene . The solubility of GO functionalized with -CH2OH-terminated regioregular P3HT  was improved facilitating its use in photovoltaic devices along with C60, which recorded twice the power conversion efficiency of a pure P3HT/C60 system.
GNPs grafted with porphyrins, phthalocyanines, and azobenzene [224, 260–262] exhibited higher visible light extinction coefficients necessary for OPV applications with better dispersion solvents. Similarly, GO grafted with polymeric chains of hydroxyls and amines including polyethylene glycol, polylysine, polyallylamine, and PVA showed improved dispersion solvents along with mechanical strength, electrical conductivity, chemical reactivity, and reinforcement of the mechanical properties.
2 Historical development perspectives
Although the photoconductivity of anthracene crystals was studied during the early 20th century, the phenomenon of electroluminescence (EL) in molecular crystals, observed in the 1960s, triggered renewed interest by identifying and studying the basic processes involved in optical excitation and charge carrier transport [263–265]. Even after successfully demonstrating organic electroluminescent diodes, there were few drawbacks that prevented the use of these devices. For example, it was difficult to sustain sufficient current to produce adequate light output mainly due to material stability problems. Using relatively thicker materials in the range of microns to millimeters resulted in very high operating voltages besides poor-quality contacts. However, this situation changed considerably over a period of another decade, resulting in improved material synthesis and controlled doping of conjugated polymers that produced a very important class of OS. Simultaneously, with the availability of the organic photoconductors and conducting polymers, the first few applications of organic materials started in the form of conductive coatings  or photoreceptors in electrophotography .
In the 1980s, undoped OS started drawing more attention of the researchers with the demonstration of two important classes of devices, namely, photovoltaic solar cells involving p- and n-type materials in a heterojunction configuration  and OFETs from conjugated polymers and oligomers [269–271]. The progress was thus accelerated much faster after fabricating high-performance electroluminescent devices from vacuum-evaporated organic thin films  and OFETs employing conjugated polymers . Consequent upon the consistently improving performances achieved in preparing better-quality materials and fabricating high-performance devices over the last few decades, OLEDs have progressed to the level of a commercial product incorporating OLED displays.
Having gone through various stages of the process development including synthesis of altogether new organic materials, optimizing thin-film depositions to have best-suited ordered structures and morphologies and optimizing the device architectures for improving the overall performances, it finally led to a very impressive device development involving OS for their specific applications in various fields . It started with a very low mobility of 1.5×10-5 cm2/Vs in mercocyanine, which was reported in 1984; the mobility figures for various organic materials remained confined to the range of 10-5 to 10-3 cm2/Vs up to 1992; it started improving slowly and reached up to 4×10-2 cm2/Vs over another 3 years; and in 1996, the technology almost matured to a level where mobility went up to 0.6 cm2/Vs, but thereafter the improvement has been rather slow and only incremental .
The inspection of the compiled highest mobility data for the most widely used OS such as pentacene, thiophene oligomers, and regioregular poly(3-alkyl-thiophene) points toward a performance maturity attained by now as individual improvements vs. time has reached almost saturation . It is interesting to note that room temperature hole mobility of 2.4 cm2/Vs of vapor-grown pentacene films [273–277] containing the grain sizes larger than the channel length reached fairly close to the mobility of 3.2 and 2.7 cm2/Vs [274, 276, 277] reported in vapor-grown pentacene single crystals. This also explains the upper limit of charge carrier mobility in OS crystals determined by room temperature TOF experiment  falling in the range of 1 to 10 cm2/Vs. It could thus be concluded that the molecular vibrational energies being fairly close to that of the intermolecular bonding energies at or above room temperature ultimately decide the highest mobility values instead of other scattering mechanisms . In the case of phonon-assisted charge carrier hopping among the localized states causing scatterings in the disordered OS, the mobility was found to be temperature dependent. It is also interesting to note that the mobility of ∼1 cm2/Vs coincides with the boundary between band-assisted and hopping charge carrier-based transports [61, 265, 279], which were measured in ordered OS such as pentacene and other derivatives [279–281]. The observation of temperature-independent mobility in some cases  including polycrystalline pentacene films  could be explained by taking into account the charge carrier trapping at the grain boundaries and the dependence of trap concentration on the film deposition conditions .
At low temperatures, coherent band-like transport of delocalized charge carriers in single-crystal pentacene, tetracene, and other related materials was confirmed by measuring TOF hole mobility of 400 cm2/Vs in naphthalene at 4.2 K [45, 282] and field-effect hole mobility of 105 cm2/Vs in tetracene and pentacene at 1.7 K . The mobility increase from its room temperature value of 3 to 105 cm2/Vs at 1.7 K followed a power law type of temperature dependence [103, 277, 282, 283]. Furthermore, band-assisted charge carrier transport at low temperatures was confirmed by quantum Hall and cyclotron resonance experiments in acene single crystals [277, 284, 285]. The temperature dependence of the electron mobility in pentacene and tetracene single crystals was shown to follow the same power law that described the hole mobility temperature dependence in the same materials from 1.7 to 300 K. Similarly, the temperature dependence of the electron mobility in single-crystal naphthalene, below 100 K, also followed a power law type of temperature dependence along the three principal directions, consistent with the band model transport [45, 103, 286]. However, in the temperature range of 100 to 300 K, the electron mobility along the c-axis was found almost temperature independent [103, 263, 286], which could be interpreted as a result of the superposition of two independent carrier transport mechanisms involving (i) molecular polarons  that are formed due to interaction of the carriers with intramolecular vibrations of the local lattice environment  and (ii) involving polarons that participate in thermally activated hopping resulting in an exponential dependence of mobility with temperature. The superposition of these two mechanisms could thus explain the temperature-independent mobility from just a few Kelvin to room temperature . All these experimental results, measured in the case of the different materials, proved that ultrapure single-crystal OS were essential for higher mobilities, as it was not possible in the polycrystalline films where the charge carrier transport through two or more grains, in terms of their associated traps due to structural defects, dominated the transport.
Based on above-mentioned observations, it seems possible to remove the upper limit of ∼1 to 10 cm2/Vs of the room temperature mobility for OFETs by further strengthening the nearest-neighbor interactions without breaking the molecular conjugation and reducing the intermolecular overlaps. These stronger interactions would ultimately push the organic molecular arrangements toward rigid crystalline structures, which would generate substantial scattering of highly delocalized carriers by lattice vibrations at higher temperatures. Employing this kind of strategy might improve the room temperature mobility either very comparable to that observed at low temperatures in acene series crystals or otherwise much higher than the room temperature values. The second strategy may involve an array of single molecules such as NTs or polymer chains bridging between the source and drain contacts to augment the conduction mechanism. For availing of the intramolecular transport of the charge carriers, the channel length should be reduced from microns to nanometers, so that it is less than the length of the single molecule. Based on these considerations, CNTs were used in OFETs, showing impressive mobility of 100 cm2/Vs [287, 288]. The successful realization of these strategies would prove that the experimentally observed device performance limitations were due to nonoptimal design architecture but certainly not due to the intrinsic nature of the organic materials involved.
3 High-mobility OS thin films
Due to inherent technological problems with the use of monocrystalline organic materials in device fabrications, it is natural to consider thin films prepared out of such materials as a second option even at the cost of sacrificing the associated mobilities to a certain extent. There is roughly an order of magnitude reduction in the mobility in employing thin films in place of crystals, which still appears a reasonable price to pay for availing the additional advantages of the flexible devices on plastic substrates covering larger areas of applications.
High-quality organic films are deposited in a vacuum system, where the base pressure decides the mean free path of the target molecules in the presence of the combined population of impurity atoms and target molecules present at the substrate during film formation, especially at the initial stages of the film deposition. In this context, various kinds of pumps are used to produce ultrahigh vacuum (UHV), high vacuum (HV), and low vacuum (LV) in the range of 10-9, 10-6, and 10-3 torr, respectively, during thin-film deposition using molecular beams , bell jar deposition, and glass-wall vacuum sublimation systems , respectively. The efficacy of removing unwanted impurities is accordingly very high, medium, and low in UHV, HV, and LV environments, respectively. Besides, carrier gas is also used in transporting the organic molecules from the source to the substrate [273, 275, 277], where the substrate temperature and deposition rate are the two parameters that critically influence the thin-film morphology and the transport characteristics of OFETs [291, 292] prepared from the material so grown. Ultrahigh purity precursors and ultraclean substrates ultimately ensure the overall quality of an OFET, as they directly affect the charge carrier accumulation, which are generally confined to first few monolayers of the OS at the interface with the insulator [1, 278]. The impurities at the growing interface thus affect the mobility, the on/off ratio, and sometimes even the polarity of the OFETs. For example, iodine-doped pentacene is, as such, a p-type material , whereas alkaline metal-doped pentacene shows n-type behavior . During the early 1990s, vapor-grown polycrystalline 6T and α-ω-di-hexyl-sexi-thiophene [271, 295, 296] film-based OFETs played a key role in further evolutions of these devices by not only proving that relatively higher mobilities were feasible in polycrystalline organic films but also helped in setting guidelines for choosing the right kind of methodologies that turned out to be necessary for improving the device performance. For instance, in the case of chain or rod-like molecules, such as thiophene oligomers, π-conjugations along the long axis of the molecule as well as the close molecular packing along at least one of the short molecular axes (π-stacking) were noted as two critical conditions for achieving high carrier mobility even valid in the case of OFETs based on vapor-deposited polycrystalline pentacene thin films [277, 291, 295, 297].
Growing amorphous pentacene films at the relatively much lower temperature of -196°C produced insulating films, as the overlap of the molecular orbitals of nearest-neighbor molecules was insufficient because of the prevailing disorders frozen in the solid at such a low temperature. In contrast, the room temperature depositions resulted in highly ordered films [291, 292, 298] with reasonably high mobility of 0.6 cm2/Vs. When a mixture of amorphous and monocrystalline phases was employed to grow thin films , the associated carrier mobility was found to be very low possibly due to the presence of high defect concentrations resulting from the coexistence of the two phases.
The detailed examination of room temperature-grown pentacene films on SiO2 using shadow masks  revealed single-crystal island formations during the initial stage of the growth followed by the subsequent layers growing on top of these islands that were smaller in size leading to a terrace-and-step type of morphology. Polycrystalline pentacene films with grain sizes approaching 100 μm were fabricated on clean Si (001) surfaces passivated with a cyclohexene layer , where the larger-sized grain growth was attributed to the relatively low nucleation density of pentacene grains on such surfaces and the absence of heterogeneous nucleation on the clean Si substrates . The nucleation density on SiO2 surface was 100 times more than that on Si (001) and cyclohexene-modified Si (001) surfaces under the same deposition conditions . It was further noted that polycrystalline pentacene films were grown on polyimide substrates with grain sizes of 100 μm using a substrate temperature below 200°C .
4 Organic single crystals
Most polymers possess a mixture of small crystals and amorphous material melting over a range of temperature than at a single temperature. In crystalline polymers, some sort of molecular order is frequently formed involving folding and stacking of the polymer chains in contrast to the amorphous or glass-like structures, where the chains are very randomly tangled. Although there are polymers that are completely amorphous, most of them have a combination of the disordered regions enclosing the crystalline areas.
In a polymer crystallization process, it has been observed that relatively short-chain molecules organize into crystalline structures more readily than larger molecules. Polymers with a higher degree of polymerization get tangled very easily. During slow cooling of molten polymer, the process of reorganization of molecular structures gets sufficient time to turn into a better crystalline structure, whereas rapid quenching produces highly amorphous materials. In addition, subsequent annealing produces a significant improvement in the crystallinity of most polymers primarily due to substantial stress relief resulting in reorientations.
Small-molecule polymers have very weak intermolecular bond strength due to van der Waals forces holding them together, which allow the crystalline layers to slip past one another causing a break in the material. Large-molecule polymers, in contrast, are held together relatively more strongly because the molecular species are fairly tangled between the layers. In most polymers, the combination of crystalline and amorphous structures offers a material composition with useful features of strength and stiffness. Besides, the size and shape of the monomer groups affect the polymer properties, as the larger and irregularly shaped monomers do not allow the molecular chains to arrange themselves in an ordered form resulting in amorphous solids in contrast to the smaller monomers possessing rod-like structures that easily form crystalline compounds.
For more than two decades in the recent past, considerable efforts were made to improve OFETs by carrying out extensive theoretical investigations and technological developments  to the extent that the organic OFETs outperformed the best α-Si (α-Si:H) transistors. However, even in the best OFETs, the charge carrier transport was still found to be dominated by the presence of structural defects and chemical impurities, which indicated that the current OFET structures were not appropriate for studying the basic transport mechanisms in organic materials . In contrast, the recently developed single-crystal organic transistors with significantly reduced disorders [18, 19, 21, 40, 302–305] provided sufficient push to explore the fundamental processes that ultimately control the operation and reliability of the organic devices. For the first time, the intrinsic transport properties of the field-induced charge carriers at organic surfaces were observed in these single-crystal OFETs [21, 306, 307] in the form of an order of magnitude higher carrier mobility compared to the best observed in conventional OFETs . Furthermore, the fabrication of single-crystal FETs turned out to be highly reproducible, that is, the devices fabricated even in different laboratories exhibited almost identical characteristics. This reproducibility, which is essential for the fundamental investigations of electronic properties of OS, has never been achieved in thin-film OFET devices, whose electrical characteristics were strongly dependent on the details of fabrication processes and handling environment. These observations necessitated exploring the single-crystal devices in more detail as described in the following paragraphs.
Single-crystal OS can be, in principle, prepared involving the transport of the constituent species from any one of the solid, liquid, or vapor phases. Based on these transformations, the organic crystal growth is put into solid, melt/solution, or the vapor growth category. It is by now well established that, during crystal growth, a polycrystalline/amorphous material is converted into a single crystal by pushing the defects in the form of grain boundaries out of the sample during the solid-phase growth of the crystal . A current survey of the crystal growth technology observed that almost 80% of the single crystals are melt grown compared to approx. 5% vapor grown, 5% each of low- and high-temperature solution grown, 3% of the solid, and only 2% of hydrothermal grown.
Organic materials are, in general, high-resistivity insulators at room temperature , but the molecules with sp2 hybridization and delocalized δ-electrons, such as in conjugated hydrocarbons, metal phthalocyanines, and oligothiophenes, exhibit semiconducting properties; besides, the organic charge-transfer compounds based on donors and acceptors  become conducting or superconducting. Organic thin films are currently used in fabricating OFETs [310, 311], LEDs , and photovoltaic organic solar cells (PVOSCs)  because of their light weight, flexibility, solubility, low-temperature processing, higher yields, and low cost of production based on various kinds of printing technologies that are especially tailored for large-scale productions . In this context, high purity and low defect density organic single crystals offer an ideal platform [40, 43, 314, 315] for studying structure-performance relationships , intrinsic  and anisotropic  charge carrier transport, and photoconductivities [317, 318] that are put to use in device simulations and modeling studies [104, 319]. Weak intermolecular interactions in OS make them very different from conventional in OS, as they possess relatively lower melting points and low sublimation temperatures, which make their crystal growth substantially different from those of inorganic semiconductors [308, 309].
Despite having synthesized a number of semiconducting organic crystals so far , only a few of them were successfully processed into samples fit for transport measurements and still fewer were available into few micron-sized freestanding single crystals [320–327]. This clearly explains why there is a need for going into thin-film-based device development and not on large-sized wafers as in the case of Si and other inorganic semiconductors.
4.1 Melt-grown OS crystals
After having access to a rich expertise in growing zero-defect large size crystal ingots of inorganic elemental and compound semiconductors , it was only natural to follow the same way in the case of organic materials. However, the organic crystal growth involving Czochralski, Bridgman, or float zone methods has not been that easy to extend due to high vapor pressure and the material instabilities present near the melting point of the starting organic samples . Of course, only in the case of some OS, available in quantity such as naphthalene , anthracene , phenanthrene , pyrene , tetracene , and stilbene , it was possible to melt-grow them, as these molecules were relatively stable at their melting points, but still they were noted to be susceptible to polymerization and/or decomposition during long heat treatments or under intense light irradiation. Mostly, the melt growth of OS is a complex phenomenon , as the material having concurrent sublimation/evaporation tendency during melting requires elaborate modifications to have a useful crystal puller. For instance, fast evaporation from the melt in the Czochralski method  during crystal pulling restricts the applicability of this method to only a few materials such as benzophenone and benzyl [334, 335], where organic powders form melt in a crucible and a rotating crystal seed is slowly immersed and then pulled out of the melt at slow rate, making the crystal grow along the orientation of the seed. However, the Czochralski method has not been sufficiently investigated for growing a large variety of OS; therefore, knowing the precise behavior and purity of the grown crystals needs further studies in the related areas.
In the Bridgman method, large-sized single crystals are grown inside quartz ampoules  sealed under vacuum or inert gas environment and by moving the ampoule through a properly designed temperature gradient  furnace. Once the crystal nucleation is induced at the tip of the ampoule, the crystallization front moves progressively through the melt. During the slow movement of the ampoule across the temperature gradient, the impurities in the melt having different solubilities segregate from the growing crystal [308, 337]. Further, as the Bridgman method uses a sealed ampoule, adequate chemical purification is not possible; therefore, the purification needs to be carried out separately before starting the crystal growth. Following rigorous material purification and setting optimal growth conditions, the electrical properties of the large-sized Bridgman-grown organic crystals matched well with those of the gas-phase-grown small crystals. The quality of Bridgman-grown DPA single crystals exhibited room temperature electron and hole mobility of 13 and 3.7 cm2/Vs, respectively , which were extremely useful for semiconductor devices. The Bridgman method produced large-sized crystals limited to the ampoule dimensions used, but the strain at the boundary between the crystal and the quartz walls of the ampoule did induce cracks, stress, or small-angle grain boundaries in the crystals.
In a zone melting process, a series of short-heating elements produce melt in several narrow zones and force purer crystalline material to accumulate outside the molten zone . This method is useful for purifying large amount of OS, such as anthracene, naphthalene, and perylene [339, 340]. Although very high purity crystals are prepared using this technique, because of the repeated melting of the material, this time-consuming method has been used only for few materials [339, 340].
4.2 Solution-grown OS crystals
Most OS are soluble in a number of solvents over a wide temperature and pressure ranges, making it feasible to use their solubility for growing crystals of large molecule compounds, which otherwise decompose easily into smaller molecules during high-temperature sublimation process. Based on the properties of a given organic molecule, several solution-grown methods  were developed in the past, as discussed very briefly in the following paragraphs.
The simplest but most effective method of growing organic single crystal is that of solvent evaporation employing solvents such as dichloromethane, chloroform, toluene, benzene, and chlorobenzene . The solvent evaporation from such a saturated solution causes supersaturation, where spontaneously formed nuclei grow finally into larger crystals. In the case of higher solubility of organic material in a solvent, the drop casting is another better option widely used for the fabrication of single-crystal films. For instance, single-crystal TTF films and its derivatives were prepared by pouring n-heptane or chlorobenzene saturated solution onto prepatterned substrates leading to crystallization after the solvent evaporation, where higher values of mobilities (∼1 cm2/Vs) were measured [322, 325, 327] in such samples. The drop-cast millimeter-sized single-crystal monolayers of HTEB with mobilities up to 1 cm2/Vs  were found very useful besides P3HT crystals .
For moderately soluble semiconductors, a slow cooling method was found better, especially when solubility changed with temperature . In such cases, raising the solution temperature dissolved more material leading to saturation, which made the solution spontaneously form nuclei, leading to large-sized crystals by consuming additional material to deposit onto the existing seeds. This process was repeated few times, raising and lowering the temperature around the saturation temperature in such a way that only a small portion of material dissolved and crystallized in the following temperature cycle, resulting in selective dissolution of the smallest seeds and further growth of the larger seeds leading to millimeter-sized organic crystals. For instance, rubrene crystals were grown using solvent evaporation method using 1-propanol, demonstrating a mobility of 1.6 cm2/Vs .
The vapor diffusion method  is yet another variant of growing OS, where a saturated solution is prepared by dissolving the molecules in a “good” solvent. Here, a container with such saturated solution is placed inside another larger container with a volatile solvent that dissolves the organic compound only slightly. When the larger container is sealed, the volatile solvent diffuses into the solution of the OS, making it supersaturated causing nucleation and crystallization . Although this method was initially used to grow smaller-sized crystals, by selecting only a few seeds, millimeter-sized tetrathiafulvalene (TTF)-tetracyanoquinodimethane (TCNQ) crystals were grown by adjusting the growth conditions appropriately .
A combination of two distinct solvent layers, having different solubility of the compound, was used in a liquid-liquid diffusion method , where the low-solubility layer diffused into the high-solubility layer, causing saturation at the interface between the two layers. For instance, triisopropylsilylethynyl pentacene (TIPS-PEN) microribbons were grown using toluene and acetonitrile as the “good” solvent and the low-solubility solvent, respectively . High-quality micron-sized crystals were grown at the interface without any stress from the substrate .
In the process of growing crystals from those compounds, which are insoluble in most common solvents and have melting points so high that they start dissociating , a solid solvent is found helpful. For instance, when a mixture of anthracene and CuPc was heated, the anthracene melt dissolved the CuPc , and after slowly reducing the melt temperature, black needle-shaped crystals were separated from the solid anthracene by adding toluene .
4.3 Vapor-grown OS crystals
Most organic compounds, having relatively lower melting points and low sublimation temperatures, are purified by high-vacuum evaporation , where even lightweight impurities sublime and condense together with the desired material on the substrate. The method of physical vapor transport (PVT) helps in separating the organic compounds from such impurities [274, 276, 344] by combining crystal growth and material purification in different configurations including open, closed, and semiclosed systems, where open and closed systems employ vacuum or inert gas environment. For instance, in an open PVT system, the inert carrier gas flow not only controls the sublimation speed, deposition, and crystal growth of the organic molecules but also prevents oxidation. However, the impurities deposit in a location other than that of the semiconductor because of their different molecular weights.
For starting the process of crystallization in the presence of weak intermolecular interaction forces operating on the molecules, the involved individual molecules must get extra energy to reconstruct, link to each other, and form crystal surfaces. It is an interesting observation that some OS do not crystallize in high vacuum but form good crystals in the presence of an inert gas environment. For instance, during high vacuum, evaporation of rubrene produces quasi-amorphous films involving free noninteracting twisted rubrene molecules  resulting in low mobility . In contrast, under inert gas flow, the same rubrene molecules form several centimeter-sized single crystals with maximum reproducible mobility ranging from 5 to 20 cm2/Vs depending on the dielectric constant of the material used as a gate insulator in OFETs . Hydrogen, argon, and nitrogen are generally used as carrier gas along with precisely controlled temperature for growing single crystals .
In the next variant  of the closed PVT system, the source material is sealed in a quartz ampoule under a vacuum/inert gas environment, where neither the reactants nor the products escape during crystal growth. Anthracene, pyrene, naphthalene, and C60 crystals were grown successfully in a closed PVT system . However, despite using very pure source materials, some new molecules got mixed up because of decomposition, photoreaction, polymerization, or chemical reactions that unavoidably occurred during growth in a closed PVT system. To remove such contaminants, a semiclosed system was thus advanced in which one end of the furnace tube was completely sealed and the other end was connected to a HV pump to continuously pump out the volatile impurities. For instance, urea crystals were successfully prepared in a semiclosed PVT system .
4.4 Supercritical fluid-grown OS crystals
Supercritical fluids (SCFs) are excellent solvents for precipitating large organic molecules. For example, SC-CO2 behaves like an antisolvent to control the precipitation of C60 crystals from the solution. In one of the experiments, when a saturated solution of C60 in toluene was injected into SC-CO2, it reduced the solvent power of toluene precipitating C60 . Various combinations of these kinds of interactions were possible to work out for preparing organic crystals by adjusting the operating parameters and selecting suitable solvents.
4.5 Binary OS crystals
A binary mixture of acceptor-donor-type charge-transfer compounds is certainly a useful combination for device fabrications. Fortunately, crystal growth of a binary compound is noted to be very much similar to that of the individual ones. For instance, a combination of parylene and TCNQ was used for preparing three sets of binary compounds, namely, P1T1, P2T1, and P3T1 [349, 350], from the gas-phase PVT or from solvents, but the stoichiometry control in such compounds was not clearly understood.
4.6 Patterned deposition OS crystals
OS device fabrication, in general, differs from those of the conventional microelectronic devices particularly because the OS surfaces are easily damaged during photolithography, patterning, wet and dry etching, sputtering, and similar others processes  involved therein. Instead, the organic crystals are manually placed or randomly cast on a prepatterned metallized substrate for realizing the device structure. This kind of handling is more liable to contaminate the device surfaces and cause damages resulting in impaired device performance with lower yields . In actual practice, it is known that manually placing a micron-sized organic crystal is an extremely difficult task with very low yield. On the contrary, unwanted deposition of smaller-sized debris during solution and/or solvent-based processing is equally troublesome. In order to mitigate these problems, a number of alternative schemes were developed including fabrication of free-standing single-crystal devices [18, 352], electrostatic bonding of a single crystal onto prefabricated device electrodes [19, 302, 323, 353], formation of single crystals by drop-casting from organic solutions [320, 354], and solvent-grown organic single-crystal nanowires  as reported in the published literature. Keeping various types of the associated problems in view, several alternate strategies were developed to deposit organic single crystals directly onto the substrate: either using PVT or solution-based processing techniques .
Most of the organic crystals used in OFETs are grown using PVT , where the starting material is placed in the high-temperature reactor zone producing gaseous precursors for growing crystals somewhere downstream, within a narrow temperature range in the furnace [40, 41, 274]. In 2006, patterned growth of large-area organic single crystal was successfully attempted for OFET applications , where octadecyltriethoxysilane (OTS) film was microcontact printed onto a clean SiO2/Si substrate using a polydimethylsiloxane (PDMS) stamp prior to deposition . This scheme was equally applicable in the case of a large variety of materials, including rubrene, pentacene, tetracene, C60, F16CuPc, and TCNQ. It was noted that material sublimation and crystal growth took only 5 min to complete in the case of pentacene; in contrast, it took 2 h for growing C60 film . In this experiment, the nucleation was specifically confined to OTS-stamped regions, leaving the rest of the SiO2 background unaffected. The single-crystal growth in patterned regions was thus realized by choosing the pattern size, such that it supported only one nucleation site at the most  and for determining this optimum size; a detailed study was carried to correlate the number of nucleation sites vs. OTS domain sizes. An average of one nucleus per 7 μm2 area was observed in this study . Site-selective pentacene thin-film growth on OTS-treated Au surface was attributed to different types of surface interactions . For instance, selective pentacene growth on SiO2-coated substrate with OTS and perfluorodecyltrichlorosilane (FDTS) stamps resulted  only within a small window of growth conditions. However, the selectivity of the above-referred growth was due more to the roughness of the thick OTS-stamped film. Further, the OTS films were directly stamped onto the OFET source-drain electrodes for fabricating large arrays of high-performance single-crystal devices demonstrating, for instance, a high mobility of 2.4 cm2/Vs in rubrene, comparing well with those of the best α-Si transistors and other high-performance devices reported in the literature [118, 358, 359]. Pentacene, C60, and TCNQ single-crystal OFETs were also fabricated using the same technique exhibiting mobilities of 0.3, 0.03, and 10-4 cm2/Vs, respectively. In addition to using SiO2/Si substrates, this technique was also tried on a flexible kapton sheet with Au gate and a spin-coated poly-4-vinylphenol film as the dielectric. In these experiments, flexible rubrene and pentacene devices showed mobilities as high as 0.9 and 0.1 cm2/Vs, respectively, besides showing no change in performance even with a 6-mm-diameter bend confirming the superior endurance of the patterned single-crystal devices to sufficiently large bending stress without performance degradation . In further extended studies , it was noted that the stamped OTS films were particularly not as smooth as a self-assembled OTS monolayers, as the stamped domains were populated with large numbers of approx. 100-nm-tall OTS pillars that facilitated the selective nucleation [356, 357]. This, in fact, was further clarified by observing that nucleation started from the base of the rough OTS surface and the crystal grew on the surface of OTS pillars . Having identified this mode of crystal growth, patterning of OS on other rough surfaces was subsequently realized, as expected. For example, a variety of OS exhibited preferential nucleation on CNT bundles, which contained many surface steps wherein it was noted that pentacene molecules nucleated directly from the surface of CNT bundles, and by stamping CNT bundles on the device channel region, pentacene single-crystal FETs with mobilities as high as 0.4 cm2/Vs were realized  successfully.
Another route for patterned growth of organic crystals involved selective growth from nanocrystalline seeds  prepared separately in the form of a suspension, where dipping of a SiO2/Si substrate led to predeposition of the nanocrystals (NCs). Such predeposited substrates transferred to a PVT system exhibited selective deposition of CuPc nanoribbons in the case of predeposited CuPc NCs. It was further noted that the predeposited NCs were aligned during dip coating, which resulted in parallel nanoribbons. OFETs fabricated with patterned CuPc  and F16CuPc  ribbons showed mobilities of 0.1 to 0.5 and 0.2 cm2/Vs, respectively, performing much better than those devices fabricated by the manual process of crystal placement described earlier .
Patterned deposition techniques based on other mechanisms were also developed, for example, CuTCNQ nanoribbons were grown via the selective reaction between TCNQ vapor and the patterned Cu surfaces [125, 361, 362] by exposing to TCNQ vapor for device fabrication. CuTCNQ nanoribbons were found to grow from two neighboring electrodes to stretch toward each other and finally bridge the interelectrode gap . In another example, using Au NPs as template, F16CuPc-based 1D nanostructures were grown using HV vapor deposition resulting in controlled assembly of F16CuPc into uniform 15- to 30-nm-diameter 1D nanostructures .
For availing of the low-cost processing advantages of OE, solution-processed organic single crystals directly deposited onto the device structures would be preferred ultimately. In the past few years, a variety of materials such as inorganic crystals , organic-inorganic hybrid materials , conjugated polymers [319, 365, 366], and oligomers [319, 367] were pattern deposited using solution processing with self-assembled monolayers (SAMs) as templates. It is noted that SAMs provide selective nucleation sites for patterned growth of organic single crystals by serving as a template  or wetting/dewetting sites [319, 360]. For instance, in a study of nucleation activity of anthracene on Au surfaces functionalized with different SAMs , it was noted that the nucleation density was highest on SAMs of terphenylthiol (-3P) and lowest on those of alkanethiols (-CH3). By dipping the patterned substrates into a saturated anthracene solution containing these two types of SAMs and allowing the solvent to evaporate, highly localized crystals were found onto -3P patterned regions . With SAMs serving as wetting/dewetting patterns, two strategies for preparing pregrown organic single crystals were demonstrated subsequently , employing suspensions prepared from pregrown organic single crystals. In one case, an organic crystal suspension was first cast on a substrate with patterned wettability and then removed. Upon receding of the three-phase contact line of the suspension, organic crystals were transported to the wetting regions and were deposited. Using this process, CuPc crystals with different sizes and shapes were prepared successfully. One advantage of this approach is its suitability to any OS crystal, soluble or insoluble in organic solvents. Another notable advantage of this method is in growing 1D crystals, which had otherwise been a challenge in most patterning techniques. In the other method, crystals are patterned into solvent dewetting regions instead of wetting regions and the organic single crystals are first deposited uniformly on surfaces with patterned wettability followed by rinsing with a solvent. The crystals on the solvent wetting regions are selectively removed by lift-off process, resulting in patterned crystals on the dewetting regions. OFET arrays are thus fabricated by patterning organic single crystals directly onto transistor source-drain electrodes. However, these devices show relatively poor performance possibly due to weak lamination of the pregrown crystals to the dielectric interface .
In the solvent dewetting method of patterned crystals growth, the device containing substrate, with lyophilic electrode regions and the lyophobic background, is dipped in an emulsion of a saturated aqueous semiconductor solution. While withdrawing the substrate, droplets of the molecular solution adhere to the Au electrode pads exclusively. After evaporation of the solvent, patterned crystals are formed only on the electrode regions. It was noted that the presence of a liquid vortex generated during continuous stirring significantly improved the selectivity of the molecular solution droplets. The shearing force exerted on the substrate surface by the vortex stripped off the small droplets that were adsorbed on the lyophobic surface regions and trim larger droplets that covered the individual electrode pads. Using this technique, large arrays of CH-4T and TMS-4T single-crystal OFETs were fabricated, exhibiting the highest mobility of 0.11 cm2/Vs in TMS-4T and 0.02 cm2/Vs in CH-4T transistors. Besides excellent device performance, this technique also offered several advantages over existing techniques, such as the cleanliness and selectivity of the patterns, the patterning speed, and the extremely small amount of material needed. This method is useful in depositing crystals of small OS molecules that are otherwise difficult to form films. Furthermore, it is possible to have fine control over the crystallization conditions by choosing the initial concentration and host-liquid temperature [360, 369].
4.7 OS epitaxy
Molecular beam epitaxy (MBE) along with its several variants has been extremely successful in producing high-quality epitaxial inorganic semiconductor thin films for quantum-well heterojunctions along with clarifying a number of epitaxial growth-related fundamental issues employing ultraclean and UHV environment loaded with high-precision diagnostic tools for in/post-processing material characterizations. Having access to such ultrahigh precision material growth technology for elemental and compound semiconductor thin films for device applications with monolayer precision [308, 370], it was very natural to consider a similar route for the OS after witnessing the successful development of a wide variety of multiple quantum well (MQW) devices, including low current threshold laser diodes, low-noise avalanche photodetectors, and high-bandwidth optical modulators [371–375]. The unique success of these MQW devices stemmed from two factors: ultrahigh precision-controlled growth of the epitaxial layers with required density of states in conduction and valence bands that lowered the laser threshold currents  and the quantum-confined stark effect in optical modulators and the heterojunction energy band offsets between two materials of different band structures.
Encouraged by the success of MBE in the past, work started to grow epitaxial organic films aiming for similar goals by tailoring the material properties in terms of controlling the density of states and the energy band offsets as mentioned above. Organic materials were especially examined for their varied applications in nonlinear optics [377–381] and optoelectronic devices [382–385]. Indeed, the progress of growing organic thin-film nanostructures by MBE using MQWs [386, 387] evoked theoretical interests in predicting completely newer optical phenomena in organic/organic and organic/inorganic MQW combinations [388–392]. Thus, the successful realization of entirely new class of materials, offering unique opportunity for extending the understanding of the basic properties of a large class of organic crystals, is undoubtedly to be explored in the very near future. At this early stage of development, MBE-deposited “small-molecule”-based organic LEDs are being commercially produced as light sources for flat-panel displays [272, 393, 394] with several advantages. Exploring the areas of applications of such MBE-deposited organic materials is only limited by the ultimate ability to control composition and structure of the resulting thin films.
From among a fairly large variety of organic molecular systems studied so far, the optoelectronic properties arising due to planar stacking of molecules such as phthalocyanines [380, 383, 395, 396] and polycyclic aromatic compounds using naphthalene and parylene have prominently been explored. In particular, the planar stacking of PTCDA molecules has attracted attention of various research groups [397–404].
The MBE system comprises an UHV chamber fitted with a number of temperature-controlled effusion cells to produce collimated beams of the desired species using a series of orifices, after passing through which they fall on a perpendicularly oriented substrate held at 10 to 20 cm away from where epitaxial growth takes place at a background pressure of 10-9 to 10-11 torr . The molecular beam flux is controlled by effusion cell temperature and mechanical shutters, which switch the specific beam “on/off” as desired. Using a number of effusion cells along with sequential shuttering of the beams, it is possible to grow multilayer structures of alternate layers of different compounds, required for the device applications, as reported [386, 387] in growing MQW structures.
Although the typical deposition rates in MBE system may be somewhere in the range of 0.0001 to 10 nm/s, it is, however, necessary to note that lower deposition rates allow contaminant absorption onto the substrate, whereas, during higher rate depositions, the precise control of layer thickness is rather difficult. Thus, the most useful growth rates were experimentally found in the range of 0.01 to 0.5 nm/s as a compromise between two extremes. Maintaining the effusion cell below sublimation temperature during the idle periods keeps them continuously degassed for storing high-purity source materials over long durations by eliminating the impurities and moisture completely.
The substrate temperature in organic thin-film depositions is chosen anywhere in the range of 80 to 400 K [386, 405–408], as the lower substrate temperature absorbs the impurities fast, which may be particularly critical in the case of films grown at modest background pressures. For the large-sized organic molecules, the sticking coefficients are almost in unity throughout the above range of substrate temperature, provided that it is maintained well below the desorption temperature. However, substrate temperature control can also be put to use with the advantage of changing the desorption rates during controlled growth of monolayers of organic materials , where loosely bound species are eliminated automatically due to their easy desorption resulting in more compact films.
The most significant advantage of MBE system is to avail of several in situ thin-film diagnostics tools such as residual gas analysis (RGA) meant for detecting the presence of all species inside the chamber prior to as well as during growth, reflection high/low-energy electron diffraction (LEED/RHEED) for determining crystal structure [410, 411], and thermal desorption spectroscopy (TDS) for identifying absorbed species by measuring their binding energies during as well as immediately after growth , besides many other analytical tools such as scanning tunneling microscopy (STM) [412, 413], Auger electron spectroscopy, and in situ ellipsometry that are very helpful for material characterization .
The crystalline order and symmetry of the inorganic single-crystal substrates can be used as a template for the growth of organic crystals, especially when the electrostatic interaction between the ionic lattice of halide crystals and π-electronic molecular orbitals of organic molecules promoted the epitaxial growth of organic crystals on inorganic substrates . This kind of study was extended further, for instance, in several cases including 6P on mica , CuPc on indium tin oxide (ITO) , α-6P on TiO2 , α-6T on TiO2 , α-quinquethiophene on Cu , hexabenzocoronene (HBC) on Au , α-6P on KCl , and PTCDA on KCl . However, many limited investigations were conducted on the fabrication of OFETs based on these crystals or thin films, which showed otherwise interesting morphologies and molecular alignments. Currently, aligned crystals/thin-film-based OFETs are mostly fabricated using a wet transfer method , where the organic single-crystal/thin film is transferred from the inorganic template to the device substrate by dissolving the inorganic template. A family of BP2T single crystals, grown on KCl substrates in hot-wall epitaxy  with aligned crystalline needles, showed high mobility of 0.66 cm2/Vs  as well as biphenyl-capped thiophene oligomers crystals realized in the same way . Besides, epitaxially grown thin-film OFETs were also reported , for instance, in the case of PtOEP deposited on KBr substrates exhibiting mobilities of 2.2×10-4 cm2/Vs . One problem of using epitaxially grown crystals/films for OFETs is that the in-plane mobilities are usually very low. Due to the strong interaction between organic molecules and the inorganic single crystals, organic molecules tend to lie/adsorb flat to these substrates; consequently, the direction of higher conductivity between molecules is not parallel to the film plane, leading to low in-plane mobility . To overcome this problem, a method known as weak epitaxy growth (WEG) was advanced, which allowed the molecules to stand up on the substrate by reducing the interaction between organic molecules and inorganic substrates. In this manner, a thin layer of rod-like p-6P oligomer was deposited on a substrate forming large domains followed by deposition of ZnPc, a disk-like molecule on the p-6P surface that resulted in oriented domains of ZnPc. Compared with ZnPc films on SiO2, much larger single-crystal ZnPc grains were observed on p-6P surfaces and the fabricated OFETs exhibited reasonably higher mobilities of 0.32 cm2/Vs, which was similar to that in the case of single-crystal OFETs with metal phthalocyanines.
Organic molecules having orientational and vibrational degrees of freedom are certainly different  from atomic species while considering their participation in epitaxial growth. For instance, the orientational degree of freedom may change the molecular orientations such as the films with “lying-down” and “standing-up” molecules. Similarly, the vibrational degrees of freedom may influence the interaction with the surface as well as the thermalization upon adsorption and the diffusion behavior of the molecules arriving upon the surface.
The organic molecules, belonging to soft materials, behave differently during intermolecular and molecule-substrate interactions compared to the atomic adsorbates. This certainly modifies the capacity of the growing thin film in accommodating larger strains, in terms of very large critical thickness, above which the growth mode is compelled to change to contain the strain in heteroepitaxy. Further, due to the presence of van der Waals interactions, the temperature needed for crucible evaporation as well as diffusion on the substrate is much lower. Although the integrated molecular interaction energy over the “contact area” may be comparable to that of strongly interacting atomic adsorbates, still the interaction energy per atom of the organic molecules is usually weaker. It is also useful to note that the closed-shell structure of organic molecules has no dangling bonds, and as a consequence, the surface energies are usually lower than those in the case of the inorganic substrates.
Fairly larger size of the molecules and the unit cells practically smear out the lateral potential variations resulting in weaker surface interactions than for atomic adsorbates. Moreover, the size difference between the adsorbates unit cells and the substrate provides more translational domains. Organic materials frequently crystallize in low-symmetry structures, which further lead to multiple translational and orientational domains behaving like disorders, in addition to vacancies known in inorganic systems.
Although a fairly large number of organic thin films were studied and employed in device fabrication, only a few of them are included here to illustrate their features responsible for their frequent uses.
Several research groups have used PTCDA, a parylene derivative-based red dye as a model system for studying the molecular beam deposition [430–439]. Although the bulk has layered molecular planes, even films grown on SiO2 and many other substrates are similar. In one of the studies, it was observed that PTCDA on SiO2 exhibited smooth surfaces only for low-temperature growth at <50°C although with very poor crystal quality . In contrast, higher-temperature films exhibited better crystallinity, but with a tendency toward island growth leading to surface roughness. Epitaxial growth of PTCDA was carried out [431, 436–438] on Ag(111) with a herringbone structure of the flat-lying molecules with a vertical PTCDA-Ag spacing of 0.285±0.01 nm , whereas the growth extending beyond a monolayer resulted in a complex azimuthal distribution as a function of temperature. The comparison of PTCDA on Au(111) showed a similar behavior [432, 434, 441], although the epitaxy details differed considerably. For instance, on Ag(110) surface, an entirely different structure comprising a “brick-stone” kind of layout was observed . Growth on Cu(110) differed from those known from other substrates; for instance, thicker films were similar to the case on Ag(111).
A red dye, named diindenoperylene (DIP) of perylene family, exhibited out-of-plane ordering [103, 442, 443] that gave good charge carrier transport properties. DIP films, prepared on SiO2/Si substrates at 145±5°C, had the upright-standing molecules on well-ordered film throughout , displaying rapid roughening that increases with thickness more than the random deposition limit. It is very possible that when certain regions of the surface grew faster than others, the surface turned out to be uneven. It was speculated that such spatial inhomogeneities cropped up from the different tilt domains of the film and inevitably present grain boundaries in between . In a DIP monolayer, the molecules are noted lying flat on the substrate  with a physisorptive  interaction with Au. In contrast to film on SiO2, due to the stronger interaction with the Au substrate, the lying-down configuration tends to prevail upon in multilayers. As the standing-up configuration appeared to have the more favorable surface energy, there is obviously a competition between the two configurations as they are found to coexist .
Phthalocyanines have been very popular materials [383, 445–449] as they exhibit tunability due to the central metal ion, which can be changed within a broad range in addition to having a number of side groups [383, 447]. For instance, F16CuPc is a good conducting polymer  with interesting optoelectronic applications as a blue dye [383, 447, 449]. Phthalocyanines grow in a stand-up configuration in thick films on “inert” substrates. F16CuPc films exhibit very good out-of-plane crystalline order . The in-plane structure is, of course, azimuthally disordered because of the isotropic substrate. However, due to these growth complications, F16CuPc and H16CuPc  both end up in needle-like crystals as a result. The molecules form a 2D structure in a phthalocyanine monolayer, whereas, in thicker films, there is a competition between the lying-down orientations in the first layer followed by the tendency to having stand-up configuration in the following layers onward. The pentacene has attracted considerable attention due to excellent charge transport properties as reported [24, 103]. It is noted that pentacene grows well in thin films in spite of the competition between “bulk” and “thin film” as the dominant structure. Pentacene on SiO2 was studied for OFETs . The pentacene grains as large as 0.1 mm  were grown successfully. In thicker films, complications were noticed in terms of simultaneous existence of “thin film” and “bulk structure” depending on the growth conditions. The effect of film morphology and dielectric surface preparation was examined by measuring the OFET characteristics including surface modification using SAMs [452, 453]. Pentacene monolayers on Cu(110) exhibited long-range order having molecules lying flat on the surface, but in thicker films an orientational transition from a lying-down configuration to a standing-up configuration was clearly seen . It was also noted that films grown at low temperatures (200 K) were ordered because of hyperthermal deposition, whereas thermal deposition resulted in disordered films.
4.8 Special OS crystals
While attempting oriented growth of organic crystals, a process known as polymer-assisted solvent vapor annealing (PASVA) was studied and developed further for preparing organic crystals in special forms [235, 454]. For instance, C8-BTBT was prepared as a rod-like structure with good air-stability and high mobility [235, 454, 455] using PASVA, an all-solution process for growing quality organic crystals at room temperature. The PASVA process helps in reorganizing the molecules into several hundred micron-sized single crystals in length. This method was carried out on a patterned substrate, limiting the growth direction of the crystals and allowing the fine control of the crystal location and orientation over a wide area. In a subsequent experiment , an SiO2-coated Si substrate was patterned into wetting/dewetting regions using OS solution [456–459] as mentioned earlier. The wetting regions were defined by standard photolithography followed by Cytop™ spin-coating and annealing at 90°C. The photoresist lift-off thus defined the dewetting Cytop™ regions on the substrate enclosing a pattern of wetting trenches. An anisole solution of C8-BTBT and PMMA was then applied to the substrate, resulting in trace amounts of the semiconductor solution entering the wetting trenches. While the anisole evaporated, a polycrystalline C8-BTBT thin film was formed on an underlying PMMA film by vertical-phase separation in the trenches. The chloroform PASVA was carried out for 10 h, causing the polycrystalline C8-BTBT films in the trenches to recrystallize into rod-like single crystals oriented along the trench directions. Besides, without PMMA, the crystals were not grown in the trenches, confirming the role of PMMA to absorb the chloroform, enabling C8-BTBT molecules to reorient easily for crystallization. An organic single-crystal array with single-crystal orientation over a wide area was thus achieved by using PASVA , with the evidence that the crystals were formed only in the trenches and they were uniformly similar, indicating good control of both the crystal positions and orientations. The X-ray diffraction (XRD) analysis confirmed that the C8-BTBT molecules were stacked in the same direction . A batch of 32 OFETs were successfully fabricated with bottom-gate top-contact configuration using highly doped Si substrate as a gate and SiO2 and PMMA layers underneath the organic crystals as gate insulator. With C8-BTBT HOMO of 5.7 eV, an efficient carrier injection was achieved for Au source/drain electrodes, while an amorphous FeCl3 layer was inserted at the metal/organic interface that improved the charge injection efficiency still further. The average mobility assessed from the measurements of 32 FETs was 1.1 cm2/Vs with maximum mobility of 3.8 cm2/Vs. These devices showed good device action with an average “on/off” ratio of 104, indicating an attractive performance for plastic electronics. This simple method is possible to extend to other solution-processed materials having 1D structure such as zinc oxide [460, 461] and lanthanide hydroxide NRs .
5 OS homo, hetero, and Schottky junctions
Two dissimilar materials used in an electronic device fabrication communicate through the interface formed between the two. In order to have minimum influence of such an interface on device functioning, it is ideal to have minimum hindrance to the flow of charge carriers across such an interface. In actual practice, it is almost impossible to get rid of this interface effect as stipulated above. Due to this unavoidable situation of interfaces, it is better to modify them electrically such that they start aiding the flow of charge carriers as far as it is possible. Therefore, instead of depending on the basic limitations alone posed by the real-life interfaces formed during device fabrication, it is better to modify them electronically to behave favorably. Some of these concepts tried out in OS devices are discussed below.
5.1 Impurity-doped OS p-n junctions
The phenomena of doping OS means that the n-type dopants should donate electrons to the lowest unoccupied molecular orbital (LUMO) states, while the p-type dopants extract electrons from the HOMO states, creating holes.
Although species such as iodine, bromine, lithium, cesium, strontium, and Lewis acids were used earlier in phthalocyanine exhibiting high conductivities, they did not produce stable layers due to their fast diffusing nature arising from their smaller sizes. In contrast, aromatic molecules such as ortho-chloranil, TCNQ, and dicyano-p-benzoquinone (DDQ) were used in doping phthalocyanines, stacked phthalocyanines, and oligothiophenes [463–467] but less effectively. In the last few years, systematic investigations were carried out to understand the underlying doping physics, which led to the development of OLEDs [468–481].
Although doping in inorganic semiconductors is explained well by the hydrogen model , where the electron is released from a dopant having the binding energy equal to that of hydrogen renormalized by the dielectric constant and the electron/hole effective masses, this is not applicable in organic hosts, as lower dielectric constant and higher effective masses of holes and electrons result in an order of magnitude higher binding and two orders of magnitude higher Coulomb binding energies, respectively, besides lacking the symmetry of a single atom. Consequently, attempts were made to study doping of organic hosts using UV photoelectron spectroscopy (UPS) and theoretical modeling . It was noted that the energy difference between hybridized and transport states of the host was found to reduce the doping efficiency. In another study, the interaction of the dopants was found to reduce the energy barrier of separating dopant and charge carrier .
The p-type doping of polycrystalline and amorphous hosts exhibited conductivities above 10-4 S/cm sufficient for OSC and OLED applications . Knowing the fact that increase in free charge carrier concentration shifts the Fermi level, it was possible to detect by Seebeck effect measurement. Further, using UPS, it was possible to resolve the transport state (HOMO) distributions and their positions with respect to the Fermi level in a metal/p-doped OS junction.
The alignment of the Fermi level (EF) and metal work function (Wfm) during the metal-semiconductor contact formation, along with IP and interface dipole, form injection barrier at the interface supported by a depletion region, causing HOMO states bend upward resulting in gradual decrease in the gap between HOMO and EF, which finally merges with the bulk Fermi level at the end of depletion layer .
In intrinsic MeO-TPD, with increase in doping level, there is a gradual decrease in EF until it saturates at approximately EFmin∼0.35 eV. In addition, F4-TCNQ doping of MeO-TPD, ZnPc, and PV-TPD shows that the host has a distinct influence on the saturation behavior; for instance, molecules such as MeO-TPD, PV-TPD, and ZnPc saturate at 0.35, 0.74, and 0.2 eV, respectively .
Taking the absorption due to C-N stretching mode of F4-TCNQ depending on the dopant charging state as signature, the efficiency of charge carrier transfer between the dopant and the host is estimated by Fourier transform infrared (FTIR) spectroscopy .
The efficient charge transfer during doping of phthalocyanines and TDATA derivatives was explained by the electron affinity (EA) being larger than the ionization energy of the host, whereas, in a higher IP host like TPD, the charge transfer is less efficient . Charge transfer in TPD derivatives can be enhanced in case the IP is lowered by electron-pushing methoxy groups at the outer benzene rings. However, the unity charge-transfer ratio does not necessarily mean that one free electron is generated per dopant molecule . For instance, the hole transferred to the host still experiences the attractive force of the negatively charged dopant due to a lower dielectric constant producing stronger attractive force, making the charge carrier not completely free  as mentioned earlier. The doping efficiency, defined as the ratio of the number of free charge carriers to the number of dopants, can be effectively below unity, even if all dopants transfer charge to the host. The method of UPS  estimates the free charge carrier density in m-i-p junctions with the Ag anode, an intrinsic layer of MeO-TPD, and p-doped layer of MeO-TPD.
Although it is difficult to find suitable materials for n-type doping, it is essential for realizing p-i-n devices needed at times. In addition, these dopants must be large enough to discourage their migrations in thin films and not to act as traps; they must add electrons to the LUMO of a wide variety of hosts. For a direct electron transfer from a dopant, its HOMO has to lie above the LUMO of the host, making it fundamentally more difficult, because a high HOMO reduces the stability against oxidation .
There are three ways  of n-type doping involving (i) alkali metals, (ii) compounds with very high HOMO levels, and (iii) air-stable electron-donating precursor molecules. The n-type doping by K, Na, and Li was reported way back  in the 1970s and 1990s. Li is still used in OLEDs based on controlled diffusion from Li-containing interface layer inside the bulk of BCP, CuPc, and Alq3 layers that are commonly used in OLEDs . High doping ratio coevaporation of Li with the host is another method of bulk doping, where the thickness of the initially doped layer is more controllable. Efficient OLEDs were produced with a Li-doped Alq3 or BPhen as an electron injection layer . Both methods of interface and bulk doping with alkali metals are frequently used for OLEDs and the beneficial effect of LiF interlayer between the organic material and the Al electrode is also useful in OSCs . Cs and their salts/alloy compounds are used along with organic ETLs leading to highly efficient OLEDs .
The first attempt  of organic molecule-based n-type doping of NTCDA involved electron-donating molecule BEDT-TTF. In this attempt, it was noted that the TTN molecules doped F16ZnPc but not the Alq3 . In another attempt, a compound CoCp2 was found to be a good dopant for tris(thieno)hex-aazatriphenylene derivative , where a shift of 0.5 eV in the position of EF was noted, confirming the phenomena of n-type doping.
The electron-donating nature of metal complexes  such as [Ru(terpy)2]0, [Cr(bpy)3]0, and [Cr(TMB)3]0 was explored for n-doping  for their applications in OSCs but was not useful as n-dopant in OLEDs because of the higher lying LUMO values of the electron-transporting materials.
Chromium and tungsten complexes such as Cr2(hpp)4 and W2(hpp)4  formed with 1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]pyrimidine (hpp) were found to be effective in doping C60 with conductivities well above 1 S/cm. Both dopants were found to produce stable doping in C60 up to 100°C. However, one general drawback of this approach was the increased instability of the dopants with respect to oxygen for higher HOMO values, requiring handling of the materials under an inert environment.
Another alternative suggested was to use precursors that donated an electron to the matrix by being activated by heat or illumination such as pyronin B chloride, which was found to be a strong molecular donor [489, 490].
Doping can also be used to optimize injection and to reduce the losses at the metal-semiconductor interface. Ohmic contacts are thus realized for a wide range of electrode/OS combinations .
For various reasons, already mentioned in brief earlier, it took several decades until 2005 to realize the first stable and reproducible p-i-n homojunction with organic materials . The material choice for organic p-n homojunction  very restricted by energy-level requirements. For instance, p-type dopant molecules must have LUMO similar to or even below the HOMO of the host, and even more challenging is finding suitable n-type dopants. Given that organic molecules are typically wide-gap materials, doping the same matrix with p- and n-type dopants requires dopants with an extraordinaryly low LUMO and with a very high HOMO, respectively. Thus, in many cases, formation of heterojunctions was preferred for device applications .
Further, employing doped wide-gap transport layers, the donor/acceptor layer thickness could be reduced without shorts and recombination at the organic-metal interfaces. The corresponding layer structure for a typical p-i-n stack would thus contain a transparent hole contact, a p-doped wide-gap HTL, a photovoltaic active layer, an n-doped wide-gap ETL, and an electron contact layer . This typical p-i-n structure turns out to be flexible for optimizing the absorption of the active layers, choosing the right kind of contact materials as well as the order of the layers. In recent years, this kind of p-i-n configuration is proving to be a versatile concept for investigating the fundamental processes as well as optimization of OSC devices .
5.2 Metal-OS Schottky junctions
The source and drain contacts in an FET structure are modeled as a Mott-Schottky (MS) junction , which states that such interfaces would behave as ohmic when the work function of the metal is close to the HOMO/LUMO level of the p- or n-type semiconductor, respectively. In the absence of this condition, formation of a barrier layer is inevitable at the interface, leading to a significantly reduced charge injection. From this point of view, Au-pentacene interface behaves as a low-resistance contact, as the IP of pentacene is close to the Au work function, but in actual practice the contact resistance of these devices is noted to be somewhat higher than the expected. Employing UPS to study the locations of the energy levels on both sides of the interface , it was clearly resolved that these interfaces strongly deviated from the MS model, confirming the presence of an additional “dipole” barrier that shifted the vacuum level upward by more than 1 eV, hence increasing the barrier height by the same amount. Such large interface dipoles were considered arising from the electron density tail extending from the metal free surface into vacuum, thus forming a dipole pointing at the metal bulk. Molecules, deposited on such a surface, tried to push back this tail, thus reducing the surface dipole and decreasing the metal work function. Additionally, in an actual device having both Schottky contacts, the voltage drop at the source is found to be somewhat higher than that at the drain in “bad” contacts, whereas “good” contacts show comparable drops and this difference was assigned  to the material quality in the regions just beneath the electrodes from that of the rest of the conducting channel. It was further observed that top contacts in OFETs usually offered lower contact resistance than bottom contacts. The asymmetry of the organic-metal contact, depending on whether the organic film was deposited on the metal or the metal on the organic layer, was studied from both the theoretical  and experimental  angles. In detailed experimental investigations, UPS/X-ray photoelectron spectroscopy (XPS) measurements clarified that the deposition of Au film on pentacene showed signs of metal penetration coupled with the formation of metal clusters, leading to a substantial reduction of the interface barrier from 1 to 0.3 eV.
Despite improving the charge carrier transport in each layer used in an organic device, the interface properties of the metal-semiconductor [498–512] and insulator-semiconductor [8, 513–521] are not only important from the point of view of influencing the device performance significantly but also assuming the role of a limiting factor while trying to extract the best out of a given configuration. For example, the quality of the interfaces between the OS on one side and source/drain electrode and the insulator on the other ultimately limits the overall performance of the OFETs realized experimentally.
In a FET structure, just next to the insulator-semiconductor interface, a conduction channel is formed on the semiconductor side through which the charge carriers flow supported by the source drain bias [24, 280, 522–524]. The associated parameters of such an interface in terms of roughness [515–517] and the presence of defects and charges [518–521] influence the charge carrier lifetime and the mobility significantly. Similarly, in the case of the source/drain electrode-semiconductor interfaces, actual interfacial contact resistance affects the carrier movements, and for easier flow, naturally the lowest barrier height is essential [494, 501–503, 506, 509, 510, 525–536]. Especially, in case the device dimensions are reduced, the contact resistance ultimately limits the total device resistance as it starts dominating over the channel resistance; therefore, for deciding the final speed of the organic integrated circuits, the intrinsic carrier mobility becomes secondary in importance .
It is a fact that improving the source and drain contacts in OFETs by the conventional processes such as semiconductor doping or metal alloying at elevated temperatures is not that easy. In this context, it is much better to search for a metal having the right kind of work function and low contact resistance from the group of different possible electrode materials. In addition, making use of the profound impact of surface functionalizations on the interface characteristics appears more promising in this context.
5.3 Interface engineering for OS devices
Taking the example of p-type OS pentacene, the barrier height between the metal and the pentacene layer is fixed by the difference between the metal work function and the HOMO of pentacene [528, 534, 537]. In this context, it is more helpful to have an idea of the work functions of various contact materials estimated by UPS as given here in each bracket as follows: Al (3.7 eV), Ni (4.6 eV), Mo (4.6 eV), Cu (4.6 eV), ITO (4.7 eV), NiOx (5.2 eV), and Au (4.85 eV). These values may differ from those reported in the literature possibly due to the fact that these measurements were carried out on thin films instead of the bulk materials; besides, there might have been some surface contaminations [538–541], considering the approximate nature of UPS values for the metal work function [538, 542, 543].
The contact resistances offered by the source and drain contacts influence the experimental values of the charge carrier field-effect mobility in FET devices [544, 545], as there is a finite difference between externally applied source-drain voltage and that effectively appearing across the channel. Under such conditions, a reduced channel current reflects lower mobility rather than the actual value. The metal-semiconductor interface studies generally attribute the contact resistance, mentioned above, to the presence of interface dipoles [546, 547] and the phenomena of Fermi-level pinning  examined separately.
In order to remove the influence of contact resistance on the channel characteristics, several methods were developed as reported in the published literature. The most popular method is of transmission line measurement (TLM), in which a number of test structures are fabricated under the same conditions only varying the channel length in a regular step. As the incremental change in the total device resistance is proportional to the additional channel length, included in each case, the contact resistance contribution remains the same throughout. It is thus very easy to estimate the contact resistance from the current-voltage characteristics of such a device configuration [15, 544]. For instance, a TLM scheme [106, 549–553] was implemented using an electrode material pattern deposited on a glass substrate with a gap between adjacent electrodes from 50 to 90 μm with an increment of 5 μm  and depositing the OS directly on this pattern. Total contact resistance was thus determined from the measured I-V characteristics of such OFETs. In the case of experimentally realized Al and Ni contacts, rectifying features were noted at low currents due to larger barrier heights; in contrast, Au electrode exhibited the lowest ohmic contact resistance. Observation of low current nonlinear I-V characteristics as mentioned above, especially at low voltages, was possibly due to the presence of native oxide formed on the metal surface.
Besides being relatively time-consuming and inaccurate due to the large number of samples involved, the serious problem of the TLM method is its inability to distinguish between source and drain contacts. For characterizing each contact separately, an alternate gated four-probe measurement was developed using thin ohmic electrodes extended into the source-drain channel region to monitor the potential drop along the channel and relating that to the potential drop in the contact regions. For an externally applied source-drain voltage, the effective voltage appearing across channel and the losses due to contact resistances are possible to estimate and correct for using the channel electrodes. The gated four-probe method was already used in pentacene OFETs , where it was noted that, for suitable device geometry, the channel probe fingers did not influence the potential distribution in the channel region. While using the four-probe geometry for P3HT devices, the entire contact geometry of source/drain contacts and the channel electrodes was patterned directly onto the gate insulator for minimizing process-induced damages to the active layer. The field-effect mobilities in P3HT devices, without correcting for the effects of contact resistances, were in the lower part of the 10-3 cm2/Vs range, whereas, after taking into account the contact resistances, the mobility was found to be in the higher part of the 10-3 cm2/Vs range .
Because of the matching Au work function with HOMO of pentacene (for instance, 5.1 vs. 5.2 eV, respectively), Au is the most widely used S/D contact material in pentacene [537, 538]. Further, during deposition, the Au surface is found to be relatively cleaner and stable without a native oxide. The only drawback with the usage of Au is its higher cost; therefore, other contact materials such as Ag, Ca, Pd, and ITO were also explored [553, 555–557] for similar uses.
Invented in 1991, Kelvin probe force microscopy (KPFM) is essentially a non-contact-type of scanning probe method, where the local surface work function is measured and mapped at an atomic/molecular level for analyzing the composition and electronic states of the local structures of the solid. This method measures potential offset between a probe tip and a surface using the same principle as the Kelvin probe, where the cantilever is a reference electrode forming a capacitor with the surface, over which it is scanned laterally at a constant separation. KPFM, being a noncontact technique, has been extensively used in exploring the electronic and morphological behavior of working devices based on molecular organic materials or polymers. Understanding and controlling the relation between the electronic transport properties and the morphology of conducting layers is important for optimizing the uses of these materials. In this context, suitable modification of the metal-organic interface for improving the device performance is a basic issue involved in the optimization of the working devices such as LEDs, FETs, and OPVSCs. Investigations using KPFM revealed that the contact resistance at the metal-organic interface depends on many factors such as metal work function, IP of the organic surface, the effect of the diffusion layer, and the built-in potential at the interface . Moreover, the metal-semiconductor interface also involves many controversial parameters such as vacuum-level shift, band-bending, and interface-dipole formations; therefore, interpretation of the results should be made with proper care. KPFM measures the potential difference map of the potential drops due to low-conductivity areas, short circuits, electrical defects, and phase separations providing valuable analytical information. Especially, in situ KPFM measurements are carried out for quantitative determination of the potential drop along source/channel/drain interfaces in different organic OFETs, while current is flowing across the operating device [502, 558, 559]. The measured potential decay in P3HT FETs allows the estimation of the resistances of the source and drain contacts, the charge carrier mobility and the resistance of the polymer, as well as the linear and nonlinear transistor operating ranges [558, 559].
A detailed study of the importance of source-drain contacts in OFET devices was recently reported  by mapping the surface potentials of the operating pentacene OFET with two different contact geometries, deposited above or beneath the active organic layer. The surface-potential distribution enabled quantitative estimations of the potential drops at the source and drain contacts; in this study, the bottom-contact devices were observed to be contact limited at large gate voltages, while the top-contact FETs were not. In both geometries, the contact and the channel resistances decreased notably with the increasing (negative) gate bias but did not depend strongly on the drain bias. This study highlighted the use of KPFM in monitoring the charge carrier-transport bottlenecks in operating pentacene devices and more generally to correlate the electrical behavior with the structure of the devices by comparing the surface potential to the topographic maps. It is necessary to go for a right kind of contact metal, as metal-semiconductor contact resistance is highly sensitive to the choice of the metal electrode. For instance, in pentacene devices, Pt contacts showed a very small contact resistance compared with Pd and Ni  and a similar behavior was noted in the case of methyl-quinquethiophene  FETs.
The electrical characterization of NT-based devices is very complex, as the semiconducting NTs behave as a p- or n-type when they are measured in air or a vacuum, respectively. KPFM measurements on Au-NT-Au systems, using only semiconducting single-walled NTs, clarified that the p-type behavior in air was due to oxygen adsorption on the Au surface, creating a dipole layer and thus changing the Au work function. It was easy to suppress the dipole formation by simple “passivation” of the Au surface using H2S or alkanethiols .
In order to adjust the electrical conditions at different interfaces encountered in practice, various types of SAMs were developed in the past and used in different contexts such as patterning of nanoscale devices, corrosion prevention, and controlling surface properties [518, 552–554, 562–574]. SAM modification of a metal surface offers a better possibility to improve the performance of OE devices as well [518, 562–574]. The formation of SAMs on a metal surface, in the present context of OFETs, is basically meant for improving the contact properties across a semiconductor-metal interface. In reality, it is not only the work function of a metal that is modified with SAMs on its surface [537, 570, 574–576] but also the morphology of the pentacene layer subsequently grown on SAMs is influenced compared to the layer grown directly on a metal surface [577–580]. The formation of SAMs employs the chemical reaction between the functional groups of the molecule and the surface, such as in chemisorption of alkanethiols on Au, hydrolysis of alkyl-trichlorosilanes on hydrated surfaces, and adsorption of carboxylic acids on metal oxides .
Consequent upon using the above criteria of interface engineering in OS device fabrications, the influence of SAMs on oxides such as SiO2, Al2O3, and TiO2 [563, 565–569], metals such as Au and Pt [562, 570, 572, 573, 579], and native oxides of metals such as Cu, Ag, Ti, Al, and Fe were investigated in detail [562, 571]. For instance, in the case of Mo thin films, although it has a low resistivity of 5 μΩ cm, its work function is in the range of 4.6 to 4.9 eV , resulting in a larger barrier height to hole injection in pentacene than that with Au. With the formation of SAMs on its surface, work function of Mo was increased resulting in improved organic-metal interface accordingly [575, 576]. In addition, annealing of as-deposited Mo thin film also increased the work function by another 0.2 eV , which was probably due to defect density reduction caused by heating. Following on the same lines, introducing OTS SAMs on Mo surface increased its work function by 0.1 eV. With ITO, the work function was modified using polar molecules such as 4-bromopropyl-trichlorosilane and trichloro(4-chlorophenyl) silane [553, 578, 583, 584] resulting in reduced contact resistance.
5.4 SAM-modified OS devices
SAMs have been extensively used for surface modifications of metals and oxides for diverse applications in the areas of molecular electronics [557, 585–587] and OE [23, 321, 588, 589] as partly described in brief above. In one such study, the growth of organosilane SAMs at the surface of OS was found to significantly increase the surface conductivity of organic materials approaching 10-5 S/square, which was close to the highest conductivity realized in OFETs at ultrahigh densities of charge carriers [108, 109, 590]. This kind of SAM-functionalized organic surface was found very useful for sensing applications, as the interaction between polar molecules in gaseous form and SAM-functionalized semiconductor resulted in a fast and reversible change of the conductivity, proportional to the analyte’s vapor pressure.
The concept of using single crystals in polymeric device fabrication was primarily motivated by successfully assessing the intrinsic charge carrier transport properties in the absence of grain boundaries and molecular disorders , which are invariably there in polycrystalline thin films [591, 592], degrading mobility and impairing device performance. Although improvements made in this direction could produce ruberene single-crystal OFETs exhibiting charge carrier mobilities as high as 20 cm2/Vs , there were problems such as poor-quality electrical contacts and difficulty in handling the fragile crystals that limited their uses in device fabrications. In the absence of a suitable replacement of a manual method of placing devices, which was highly impractical for fabricating devices over a large area, a method for inducing site-specific growth of large organic crystals using micropatterned SAMs as nucleation templates was definitely needed .
Although a number of challenging problems were faced in controlling nucleation sites, sizes, and orientation of the OS single crystals, some recent developments have been found helpful in controlling such crystallizations [593–596] as presented here. Although SAMs of functionalized thiols supported on metal films were successful in promoting nucleation and growth of inorganic crystals with finely tuned crystal sizes, crystallographic orientation, and crystal micropatterning [362, 597–599], their uses in organic crystallization were rather limited due to prevailing van der Waals interactions against the ionic ones that were critical at the organic-inorganic interfaces. It was found that patterned SAM templates could produce site-specific crystallization of organic charge transferring salts  and pentacene films [356, 357]. Successful growth of 100 μm size pentacene single crystals was induced by Si surface modified with a monolayer of cyclohexene molecules . For furthering this quest, oligoacene molecules were put to a detailed investigation in this direction. Substrates modified with alkanethiols such as methyl, amine, and carboxylic acid end-groups as well as with monolayers of thiophenol, biphenylthiol, and terphenylthiol SAMs were immersed in a saturated anthracene/THF solution and allowed to evaporate . Detailed analysis of the nucleation density and surface coverage of the OS crystals clearly indicated that the nucleation density of anthracene crystals on different SAMs increased in the following order: -CH3<-NH2<-COOH<Au<-P<-2P<-3P . Next, micropatterned substrates with regions of -3P and -CH3 having different geometries and relative sizes were microcontact printed  and placed in a saturated anthracene/THF solution, where subsequent crystal growth occurred by solvent evaporation . Highly localized crystal growth was observed onto oligophenylene thiol-patterned regions. Similarly, dip coating was used  for selective crystallization of OS molecules, where the coverage of the -3P region was three times that on the nonpatterned -3P substrates resulting in 100×100 μm size single crystals occupying the whole -3P region.
6 OS-NPs and NP thin films
Quantum confinement of charge carriers in nanosized entities including NCs, NPs, NRs, NTs, superlattices, nanofilms, and many other variants gives rise to additional energy levels superimposed upon the conduction and valence band edges of the parent material imparting tunable optical and electronic transport properties based on the size and shape of the sample. Various nanosized metal, semiconductor, and insulator species are currently being investigated and developed for their use as fundamental building blocks for realizing new type of metamaterials in which the required physical, chemical, and biological features are taken as input parameters to their design and subsequent synthesis. As OS, in their present form, are poised for their diverse applications in the area of flexible electronics of tomorrow, it is equally important to know about the changes brought out in their properties due to quantum confinement, especially from the point of view of their use as material building blocks for a newer class of materials. The prevailing situation in this context is examined in brief in the following paragraphs.
Solubility-based methods of preparing OS-NPs are particularly important because of their simplicity and meeting low-cost requirement of resources involved besides producing materials somewhat closer to those prepared by sophisticated self-controlled growth. In this context, the procedures developed such as precipitation, emulsion, and condensation are discussed here in brief by pointing out their positive and negative aspects for comparative evaluation of each method.
The precipitation method  of producing polymeric NPs, first introduced in 1992, employs higher and lower solubility of the target compound in two different solvents that are used in realizing a hydrophobic interaction while mixing the solution of the OS material dissolved in a “good” solvent and pouring it into a “poor” solvent such as water and stirring the mixture vigorously. After the NPs are formed, the organic solvent is removed either by vacuum evaporation or by repeated dialysis to produce water-dispersed OS-NPs. The driving force behind this process is the hydrophobic effect realized during mixing of the two solutions, where the subject molecules try to avoid contact with water. Forced by this tendency of hydrophobic interactions, they try to fold or pack into spherical shapes to achieve minimum exposure without involving any surfactant; especially, this process is equally applicable in a wide variety of OS including both polymers and small molecules given that they are soluble in some “good” organic solvents. Moreover, it is very clear that there is ample scope of varying the size of NPs by adjusting the solution concentration and the temperature appropriately.
Alternatively, in the emulsion method of preparing OS-NPs , the target compound is first dissolved in an organic solvent that is immiscible with water and then the solution is injected into an aqueous solution of an appropriate surfactant. The mixture is stirred vigorously to form a stable emulsion comprising smaller droplets of the polymer solution. The organic solvent is subsequently evaporated to obtain a stable dispersion of polymer NPs in water. The size of the NPs, in this process, can be varied in the range of 30 to 500 nm depending on the concentration of the polymer solution used. However, the processes such as Ostwald ripening or flocculation by coalescence start destabilizing such droplets, which can easily be prevented by adding either hydrophobic agent to the dispersed phase or suitable surfactant accordingly.
In the earlier two methods of precipitation or emulsion, a millimolar concentration of target material is added to a large volume of nonsolvent resulting in very dilute OS-NP dispersions, which are not applicable in the case of organic materials that are poorly soluble in organic solvents. A method employing condensation of the vapor of the target material into a liquid dispersion medium was therefore developed next  to overcome the problems associated with the methods of precipitation or emulsion. For concentrated dispersions of OS-NPs, temperatures in different zones of a properly designed tube furnace are adjusted according to the target material ensuring that no condensation of the organic vapor occurs in the tube. Rather, the evaporated target material is carried out by an inert gas flowing to the vapor injection tube, which guides the organic vapor into a liquid condensation medium consisting of an aqueous solution of surfactants or polymeric stabilizers, where the rapid cooling finally leads to condensation of the organic vapor into OS-NPs. These NPs are subsequently in situ stabilized by the surfactant or polymeric additives at the bubble/liquid interface to form a stable dispersion. The size of OS-NPs prepared by this method is, in general, in the range of 100 to 200 nm for fused aromatic hydrocarbons such as pentacene, rubrene, and tetracene.
In an entirely new concept of material preparation, laser ablation-based preparation of OS-NPs, several micron-sized organic crystals are pulse laser irradiated while suspended in a liquid medium [604, 605]. For instance, the target material is added and sonicated to a sodium dodecyl sulfate aqueous solution, which is put into a quartz cuvette, stirred vigorously, and simultaneously exposed to the second harmonic of a nanosecond YAG laser with a fairly large spot size. The organic crystals, present there in the solution, absorb the laser radiation leading to a local increase in temperature resulting in the evaporation of the material from the crystal surfaces, which is immediately cooled by the surrounding liquid forming OS-NPs. The competition between transient heating and liquid-assisted cooling that gives rise to a finite transient temperature rise ultimately determines the OS-NP size. For instance, higher fluency gives higher effective transient temperature rise, leading to efficient fragmentation to smaller particles. One advantage of this method is its superior control of size and phase of NPs by tuning laser pulse width, wavelength, fluency, and the shot number. However, this method is limited to fabricating OS-NPs based on small molecules only. It is further noted that NP formation occurs within the narrow laser beam resulting in only a small volume of dispersions. Possibly, the intense laser light may cause severe photochemical damages especially in the case of rather sensitive organic materials.
While solidifying the target material from a fluid to a solid phase, if the space available is confined to nanosize, it is very possible to have nanosized solid particles as a result. While confining the overall volume of the vacant space by some means, as explained later, it is practically possible to assign certain shapes and sizes to such a template. It is interesting to note that, in template growth, there is an automatic control of morphology provided by the template involved. There are a number of ways in which these templates are realized as discussed below in brief.
Micelles have been used as the soft templates for polymerization in an aqueous heterophase system by dispersing the appropriate monomers, surfactant, solvent, and catalysts in an aqueous medium, where the coupling reaction takes place exclusively within the confines of the hydrophobic interior of the surfactant-controlled micelles to produce, for instance, the poly(arylene diethynylenes)  and poly(p-phenyleneethynylene) (PPE) NPs using this method . The molecular structure of the surfactant used in the aqueous heterophase has significant influence on the shape of the NPs so formed, as using dodecyl-benzene sulfonic acid surfactant with PEDOT doping agent results in amorphous and polydisperse NPs having diameter dispersed in the 35 to 100 nm range . In a similar experiment, short-chain alcohol ethoxylate surfactants yielded more spherical NPs, but significant amounts of surfactant residue were also found trapped on the PEDOT latex causing secondary nucleation. These examples clearly show that the soft template approach is a versatile method for preparing conjugated polymer NPs. However, control of parameters such as diameter and polydispersion is often not so simple.
Shape retaining features of hard templates, such as monodisperse silica and polystyrene NPs, offer a more appropriate method to control the morphology of the conjugated polymer NPs, especially in core-shell configurations, which are finding valuable applications in optoelectronic devices. Conjugated polymers such as polypyrrole, polyaniline (PANI), PEDOT, and PPE are attached to the surface of colloidal NPs and the conjugated polymers are either polymerized in situ from the monomers absorbed on the surface of the particle templates or deposited from a layer-by-layer technique through electrostatic interactions .
Growing awareness to the environmental concerns in current times has been compelling to reorient the investigations toward developing NP syntheses without using organic solvents. In this context, SCF-based processes offer the possibility to design and prepare NPs without the limitations of traditional methods [610–615]. Working on these lines, two procedures were developed for NP preparations using SCF. The first one is based on a rapid expansion of a supercritical solution (RESS) and the second one uses an extension of the first one, where RESS is released into a liquid solvent (RESSLS).
In a RESS process, organic macromolecules are dissolved in an SCF solution, which subsequently undergoes a rapid expansion through a nozzle into ambient air forming well-dispersed particles from a homogeneous nucleation imposed by the high supersaturation conditions combined with the rapid pressure reduction . Generally, SCF-CO2 is used in the majority of cases. The polymer is first dissolved in a CO2 solution at ambient temperature in the mixing cell. Subsequently, the solution moves in the pre-expansion unit with the help of the syringe pump and heated to the pre-expansion temperature until it expands through the nozzle at ambient pressure. Poly(hepta-deca-fluoro-decyl-acrylate)  or poly(L-lactic acid)  NPs were prepared by this technique, wherein the factors such as concentration and degree of saturation of the polymer, the processing conditions, the molecular mass, and the melting point of the polymer mattered the most. Although the method is performed without organic solvents and produces a majority of nanosized particles, the major drawback is the simultaneous generation of micron-sized particles or agglomerates.
To overcome the above-said problem of agglomeration encountered in the RESS process, another modified process was developed, where the SCF solution is allowed to expand into a liquid solvent instead of ambient air . The liquid solvent suppresses the agglomeration of primary NPs. For instance, PHDFDA  NPs were produced using water as the solvent in which the SCF solution expanded and precipitated the polymer. It was shown that the particle formation resulted from the aggregation of initially formed NPs. In addition, the presence of NaCl in the water phase helped in stabilizing the NPs due to an increase in the ionic strength. PMMA and poly(L-lactic acid) NPs were synthesized by this method using a CO2 cosolvent as the SCF. The cosolvent allows a better dissolution of the polymers in the SCF solution and the presence of NaCl in the water solution generates only NPs . Despite the availability of a number of fluids such as CO2, n-pentane, ammonia, and many others, the poor solubility of polymers in these SCFs remains a major limitation of this process.
Incidentally, the spray-drying process has been in use, for the past several years, for preparing micron-sized organic particles or converting NP suspensions in dry powder form mainly for biomedical and pharmaceutical applications, especially with reference to drug delivery [622–624]. A typical spray-drying consists of first atomizing a liquid into a spray of fine droplets, which is subsequently brought in contact with a hot gas to remove the moisture and help in forming the solid product that is recovered via a cyclone unit. The basic form of spray-drying technology has gone through several stages of improvements continuously in the past and the synthesis of OS-NPs was recently realized in a one-step procedure by spray drying of a polymer solution. For example, preparations of NPs out of gum Arabic, whey protein, PVA, modified starch, and maltodextrin  were successfully carried out using a “nanospray dryer”. A similar technique was extended to produce bovine serum albumin NPs , where the “nanospray dryer” was modified to a vibrating mesh spray that created still finer droplets. The generation of these tiny droplets uses a 60 kHz piezoelectric actuator-fed perforated membrane with micron-sized holes, varying from 4 to 7 μm in diameter. The membrane vibration causes ejection of millions of nanodroplets per second with a very narrow size distribution. The final NP size and its standard deviation depends on parameters such as the nature and concentration of the polymer, the spray mesh size, the operating conditions including drying temperature, feed rate, drying gas flow rate, or the concentration of surfactant, in case used in the formulation. Finally, another advantage of this novel technology is its very high yield of NP production in the 70% to 95% range.
6.2 OS-NP thin films
Despite preparing, synthesizing, and characterizing a large variety of NPs, NCs, quantum dots (QDs), quantum wires (QWs), and many more nanostructured species in colloidal form or in isolation, so far, their further applications in the form of a useable piece of material is not that straightforward. For exploiting the unique properties of these nanostructured species, it is necessary to put them together in the form of a useable thin film or a piece of bulk material on a substrate or in free-standing form, where individual nanostructured species still retain their identities while participating in offering the typical properties of the assembly as such. In this kind of arrangement, it is easy to realize a large variety of superlattices for having access to the band structure engineering aspects of such assemblies for using their electronic and optoelectronic properties especially for realizing useful devices accordingly. How to assemble these nanostructured species in an ordered structure in two and three dimensions is currently an area of considerable interest current. It has been observed that, by controlling the interparticle separations in such assemblies, the delocalization of electron states are possible to maneuver the conductivity of the material so synthesized from being an insulator to conducting via the semiconducting state. It is therefore very important to explore the phenomena of thin-film formation using NP dispersion in different solvents as can be seen from the brief descriptions of the processes involved and their possible applications.
As OS-NPs are usually synthesized in solution with a low concentration, thin film-forming processes such as spin coating or dip casting are not suitable for preparing OS-NP thin films. Alternate methods were therefore developed as discussed here briefly.
In one of the methods of electrophoretic deposition of NP thin films, it involves field-assisted separation of small charged particles dispersed in dielectric liquids such as in producing phosphors for cathode ray tubes, oxide superconductors , and carbon NTs for cold cathodes . Colloidal solution of OS-NPs carry charged entities according to Coehn’s empirical rule, which states that the electrostatic charge separations occur when two dielectrics are put in intimate contact. The substance with the higher dielectric constant acquires the positive charges, while the other one receives the negative ones. A DC bias of a few hundreds of volts applied between two ITO-coated glass electrodes soaked in the NP suspension forces the particles to move toward the corresponding electrodes under the influence of the electric field .
The SCF-based technique of preparing OS-NP film is based on a rapid expansion of SCF-CO2 solution containing dissolved OS , where the dissolved compound dispersed in SCF is sprayed on the substrate through a capillary. After rapid evaporation of CO2, OS-NPs precipitate on the substrate surface, and by adjusting the process parameters, it is easy to produce OS-NPs with tunable sizes and optical properties opening up newer avenues to create functional films and devices using OS-NPs as building blocks with additional possibility of mixing different building blocks of OS-NPs or combining different molecules within each building block. However, the only limitation of this technique is that it is primarily applicable in the case of smaller molecule-based OS-NP films. Adding Fluorolink 7004 surfactant to the SCF solution helps in adjusting the size of the OS-NPs  experimentally.
In the solvent evaporation-based technique of NP thin-film preparation, the phenomenon of evaporation during the drying of a solution can be advantageously put to use in controlling the film morphology and the distribution of the solute in the final films so produced. It is known that, when a liquid drop, containing dispersed solid contents, evaporates on a flat surface, it leaves a ring-like deposit along the perimeter as the contact line is pinned during the drying process, leading to a fixed contact area on the substrate. In this process, a capillary flow of the solvent takes place from the center to the contact line of the drop to replenish the evaporation loss, and this flow thus transports the solute to its periphery . In the case of OS-NP solution, such a phenomenon results in an uneven distribution across the deposited films. However, in case another counterbalancing flow is introduced in the direction opposing the above-said capillary flow, it will be very possible to balance the transportation of NPs toward the contact line by the said capillary flow. In this context, a process based on the Marangoni effect can be put to use where the mass transfer along an interface between two fluids due to surface tension gradient, usually observed in a solution containing two solvents with different surface tensions and boiling points, gives rise to the liquid flow induced by the surface tension gradient in the solution during solvent evaporation. Such a flow and its direction can be adjusted in the form of either outward or inward spreading of a drop on a solid surface depending on the boiling points and surface tensions of the two solvents mixed together. Consequently, by proper mixing of a second solvent into the NP solution, a Marangoni flow with an opposite direction to the capillary flow is realized accordingly. The solvent evaporation-induced self-assembly method for preparing the NP films from their OS-NP solutions uses ethylene glycol as the second solvent having a higher boiling point but a low surface tension, and in this manner, the capillary flow in the solution is counterbalanced by the Marangoni flow. The final self-assembly of NPs on the substrate is achieved through the NP-substrate and NP-NP van der Waals interactions.
Alternatively, the vapor-driven self-assembly process employs selective phase demixing and self-assembled aggregate formation from a molecularly dispersed solid solution of a specific molecule in a polymer matrix when it is exposed to volatile organic solvent vapors . The supramolecular self-assembly of OS material finally leads to spherical NP formation after solvent evaporation. The advantage of this method is to form OS-NP films in situ on the substrate. Nevertheless, this method is, once again, only appropriate for small molecules with certain structures and is not applicable in the case of most of the polymer semiconducting materials.
Besides preparing NP thin films based on manipulation of NPs in liquid and solid phases, another area has emerged in this context using jet printing technology, which is especially well suited for cost-effective preparation of OS devices. In this context, it is useful to recollect that when a drop of OS-NP solution is placed on the surface of a substrate, the OS-NPs form a coffee-stain-like structure after evaporation of the solvent as described above; therefore, simple inkjet printing of OS-NP solutions cannot provide good film morphology. To take care of this problem, an aqueous dispersion of OS-NPs is deposited by inkjet printing onto a polymer surface patterned by soft embossing , wherein, due to interaction between the OS-NPs and the undulated surface, self-assembly is triggered, resulting in the formation of OS-NP-based nanostructures determined by the embossed template. This method is very appropriate for incorporating the OS-NPs into a device structure.
Yet, in another alternative, even spin coating of thin films of NPs was attempted to simplify the processing further. Although without using an appropriate additive, it is rather difficult to spin coat very low concentration of OS-NP solutions in a thin-film form. Consequently, additives such as surfactants or polymer matrices are added to assist the deposition of NP films. Using the charge-carrying properties of NPs on their surfaces when they are dispersed in solutions, negatively charged NPs are collected on polycationic films with the help of electrostatic interactions via spin coating and vice versa. For instance, layers of LPPP NPs were spin coated on polycyclic aromatic hydrocarbons (PAH) exhibiting a homogeneous fluorescence over large areas . Similarly, conjugated polymer NPs such as polyfluorene derivatives and LPPP spin-coated on PEDOT:PSS films also exhibited good film morphology . Besides auxiliary underlayers, a polymer matrix was also used as a binder to improve the film quality deposited from OS-NP solutions, as it assisted in reducing the electric field singularities around the NPs that might result in pinholes in electronic devices. PVA , hexadecyl-modified poly(ethylene oxide) , and PEDOT:PSS [635–637] polymer matrices were used, in this context, successfully. Although, in this way, the film quality is improved, the additives staying in the NP films are not at all good for the optical and optoelectronic properties of OS-NPs. For example, while using PEDOT:PSS additive to the OS-NP aqueous solutions for preparing thin films by spin coating, the acidity of PEDOT:PSS degraded the luminescent properties of the conjugated compounds significantly.
6.3 SAMs in OS devices
Carbon fullerene (C60)  is a material that is being targeted for its device applications such as in PVSCs [639–641] and FETs [642–644] especially because of its higher electron mobilities [645, 646]. C60 SAMs  offer unique and attractive features for their use in advanced electronic device applications , for example, monolayers of C60 derivatives on Au, ITO, or other substrates [648–655] have been investigated extensively in the recent past. Besides C60 monolayer on Au, there are only a few reports about them on SiO2 surfaces [346, 656–663] that are especially important for OFETs. While conducting these experiments, it would have been very possible that reactions with a nitrogen functional groups on the surface [346, 657, 658] resulted in incomplete reactions with C60 itself or prefunctionalized C60 with the surface using a coupling reagent [656, 660–663], leading to significant surface contamination, meaning it would be rather difficult to have a SiO2 surface with a uniform electronic structure under these circumstances. Such problems of unwanted reactions on SiO2 were taken care of at a later stage by growing monolayers of C60 derivatives on SiO2 surfaces by using C60 SAMs functionalized with carboxylic acids. Experimentally, C60 SAMs were deposited by soaking SiO2 substrate into CH2Cl2 at room temperature for 15 min followed by 20 min boiling in a 1:1:5=NH4OH/30% H2O2/deionized water (DI H2O) at 70°C. Immediately after rinsing in DI H2O and drying in N2 gas, the samples were immersed in a 2% (v/v) solution of 3-aminopropyltriethoxysilane (APTES) in ethanol for 25 min at room temperature. Post-baking at 120°C for 5 min gave monolayer of APTES and then these covered samples were immersed in 0.1 mM solution of CPC or CPCFe [655, 664] in THF for 24 h and 30 min, respectively, at room temperature. After the monolayer assembly, the samples were soaked in clean THF for 30 min at room temperature for washing out the unbound molecules. These CPC and CPCFe monolayers were very stable, as the UV-visible spectrum did not show any significant changes over several weeks under air or more than 15 h under argon at 300°C. Compared to monolayer of C60, pentabiphenyl derivative C60 (C6H4C6H4-COOH)5Me on Au  is very reasonable, as these derivatives stood upright on bare Au surface without any adhesion layer under in situ STM conditions.
Fullerene-based OS devices have been explored using SAM-based technique. In one of the typical experimental device fabrications of OFETs, the following sequences were followed as described here. OFETs were fabricated with bottom-contact geometry involving a highly doped n-type Si wafer with 300 nm thermally grown SiO2 for electrical measurements. The wafer was piranha cleaned  before deposition of Cr and Au films for source and drain contacts using shadow mask resulting in 115×10 μm and 2 mm×85 μm channels for C60 and HBC transistors, respectively. After the deposition of monolayers according to the procedure mentioned above, C60 monolayer-based FETs were fabricated by thermal evaporation of 50 nm C60 on to the substrate. HBC transistors were fabricated by spin coating from a solution of tetradodecyloxy HBC in CHCl3 or (CH2Cl)2. The devices were tested at room temperature, in Ar atmosphere in the case of C60 transistors, or in normal ambient atmosphere in the case of HBC transistors.
Field-effect mobility of 0.02 cm2/Vs was measured in CPC monolayer-based devices, which was very typical for C60 FETs , whereas, in CPCFe-based devices, the mobility was 0.04 cm2/Vs and this difference was attributed to the presence of ferrocene moiety. Compared to the mobility of a normal C60 transistor on bare Si as 0.01 cm2/Vs, both C60 transistors with CPC and CPCFe showed higher current possibly because of shielding of the surface hydroxyl group and higher crystalline packing of C60. An aligned dipole layer, generated by the salt of amine and carboxylic acid at the interface of CPC or CPCFe and APTES monolayer, could be the cause of the experimentally observed shift in VT [588, 665]. While investigating the effect of CPC and CPCFe monolayers in HBC OFETs, good performance of spin-coated HBC devices were reported from which it is concluded that the intermolecular interactions from these compounds are not only useful because C60 and HBC are electron acceptor and donors, respectively, but also their shapes are complementary . These two devices showed similar characteristics, but when the measurements were carried out in the ambient light, the CPC transistor had higher current than the CPCFe one. OFETs employing an acetic acid layer  instead of CPC or CPCFe showed similar characteristics. The behavior of the monolayer/HBC devices was explained by the photo-induced charge transfer between C60 moiety of the monolayer and the HBC [346, 668]. These experiments demonstrated that reliable CPC and CPCFe monolayers on SiO2 surface could be successfully used for the surface modification of the insulating layer of OFETs . When C60 is used as a semiconducting layer in an OFET, the mobility is 0.02 cm2/Vs and this value doubles in the presence of CPCFe monolayer, indicating the efficient channel formation by the electron-donating ferrocene moiety. The generation of photocurrent in the presence of contorted HBC is possibly due to donor-acceptor and the geometrical ball-socket interactions responsible for the photocurrent generation .
7 Ion implanted OS device stabilization
Improvement in the performance of OFETs  was realized with ion-implanted passivation of the superficial layer by eliminating the instability due to the atmospheric interactions while leaving a fresh layer at the dielectric interface besides enhancing the adhesion of the film to the substrate. In the case of inkjet printed organic OFETs or the organic electrochemical transistors, it is necessary to have good adhesion at the substrate, as the OS starts as a liquid drop that solidifies on touching the substrate and is then in contact with a liquid electrolyte . Ion implantation did render the interface strong with stable adhesion by modifying the interfacial atomic structures as reported .
Ion implantation-induced variations in the transport properties of OFETs, such as charge carrier mobility and VT, were measured in several samples implanted with different types of ions, doses, and beam energies including single, double, and coimplant. For instance, in the case of F+ and S+ implanted devices, it was noted that F+ implants not only improved the mobility but also stabilized the device performance, whereas S+ implants stabilized VT but with a huge initial shift at low voltage compared to VT0 before implantation .
The changes taking place in the transport properties with time in ion-implanted devices were recorded  to know the influence of the irradiation process on aging and device degradation. These measurements  showed clear stabilization of carrier mobility and/or VT for at least up to 2000 h after the implantation. Low-dose implant of 35 keV Ne+ ions stabilized the carrier mobility, whereas N+ ions induced VT stabilization. For instance, VT stabilization was optimized for 25 keV ions up to 1×1015 cm-2 dose and it was further noted that higher doses and energies also induced stabilization of VT, although it occurred at values that were largely shifted with respect to the reference of the nonimplanted device .
The optimal way  to have stable electrical parameters of ion-implanted pentacene OFETs was to coimplant N2+ ions with a combination of beam energy/dose of 25 keV/5×1014 ions/cm2 and then Ne+ ions with beam energy/dose combination of 25 keV/2×1014 ions/cm2. It was further noted that the implantation damage necessary to encapsulate the OFETs did not penetrate deeper but remained within a superficial layer between 50 and 300 nm of thickness. N+ and Ne+ implantations  resulted in harder superficial pentacene layers with higher conductivity due to the damage created by ion implantation in both cases. Subsequently, working pentacene OFETs were implanted where a controlled damage induced by the ions in the superficial part of the pentacene layer created an encapsulation layer; under that layer, a fresh layer of pentacene was available for device action . The superficial implanted layer of 50 nm was sufficient to keep the electrical parameters of the OFET, including mobility and VT stable, while protecting them from the atmospheric interactions for at least 2000 h . From among several combinations of ion species studied, the coimplantation N2+ and Ne+ was found to be the most suited without damaging the good features of OS as transparency, flexibility, and biocompatibility.
Plasma source-based ion implantation  was noted as having the potential for commercial production of large-area surfaces due to immersion of the sample inside the plasma of accelerated ions. This technique ensured the lower costs of the OS, employed in large-area configurations such as in large-area displays and photovoltaic solar cells.
8 Printed OS devices
Producing a variety of electronic components and devices using different kinds of raw materials employing a set of printing technologies  falls under the printed electronics (PE) category. The printing process consists of patterning of thin/thick film structures on flexible or rigid substrates, depositing single/multiple ink layers. The printing technologies, including flexography, soft lithography, screen, gravure, and inkjet printing, have all been growing very fast, especially keeping in view their applications in electronic manufacturing, which is strongly motivated by various factors such as reduced cost derived from high-throughput/volume production of lightweight/small, flexible, and disposable, inexpensive electronic components and devices.
In today’s printing technology, both inorganic and organics inks are used. The inorganic inks are prepared generally from metallic dispersions of Cu, Au, Ag, and Al NPs in a matrix primarily used for passive components and interconnects, while liquid-phase organic conductors, semiconductors, and dielectrics are especially used in fabricating active devices of OE. For instance, conducting polymer inks find their applications in fabricating batteries, electromagnetic shields, capacitors, resistors, and inductors and OS inks are used in printing active layers of organic photodiodes, LEDs, FETs, PVSCs, sensors, and electrochromic displays .
It is significant to note that PE is not considered as a substitute for conventional Si electronics that has been offering ultra-large-scale integration of ultrafast switching devices and systems; rather, it is especially poised for developing low-cost/high-volume production markets, which do not require the high-performance conventional electronic components and devices. Particularly, PE is supposed to offer viable options of low cost of production based on mostly low-temperature processes, which are carried out on flexible substrates such as plastics foils or paper with the capability of handling larger area substrates. The minimization of the material waste by employing an additive process, where the inks are site selectively deposited consuming only a few picoliters of material in each drop, is the most significant feature of PE against the subtractive processes, which involve blanket depositions of the material and photoresist over the whole substrate for patterning features using conventional photolithography and etching followed by removing the material from the unwanted areas and the residual photoresist layer – all adding to cost .
PE is generally divided into contact and noncontact types, in which contact printing includes processes such as flexography, offset lithography, gravure, and screen printing, where the printing plate comes in direct contact with the substrate, whereas in the noncontact printing, as in inkjet printing, the substrate receives the drops of inks form of printheads located at a distance.
8.1 Contact printing
Various types of contact printings have been dominating the PE industries  because of their associated high throughput and low production cost capability employing roll-to-roll printing configurations that are suitable for mass customization. The direct contact with the surface in such printing has, indeed, limitations in terms of the pattern resolution and the range of materials used including the substrates, inks, and solvents.
In flexography, a cylinder transfers the positive pattern onto the substrate surface after applying an adequate quantity of ink on the printing plate from an engraved cylinder that releases the ink according to the engravings and a doctor blade removes the excess ink before printing. During printing, the plate rolls over the substrate and the ink is transferred to the substrate in the desired pattern by applying pressure on the impression cylinder [674, 675]. This process is suitable for printing on plastics and foils [674, 676], employing a large variety of fast-drying inks and soft-printed plates, despite low precision and resolution resulting from the ink spread outside the image area, caused by the compressive pressure between the printing plate and the impression cylinder.
In gravure printing , laser or photolithographically etched target pattern on the rotogravure holds the ink inside the grooves from the ink supply before transferring to the substrate through the pressure between the printing plate and the impression cylinder enabling very high printing speed and quality using wide ranges of materials and inks.
The contact electrodes of flexible 5 μm channel OFETs on PEN substrates were reported to be fabricated  using a h-PDMS stamp and Ag ink resulting in a field-effect mobility of 0.06 cm2/Vs. Using PDMS stamp in gravure printing, roll-printed OFETs were fabricated  using Ag paste, flexible 150×150 mm2 substrates, PVP dielectric and TIPS-PEN exhibiting mobility approx. 0.08 to 0.1 cm2/Vs. This PDMS-based process produced 12 to 74 μm channel OFETs otherwise not possible in traditional printing techniques.
8.2 Noncontact printing
In a noncontact printing method, process-induced damage and contamination are practically nonexistent. The additional capability of having accurate alignment with the existing patterns on the substrate offers a unique opportunity of fabricating multilayered component structures and devices . Furthermore, no direct contact between the printhead and the substrate at low temperatures makes this technique print patterns on surfaces such as glass, Si, metals, rubber, plastics, and many other contact-sensitive materials. The process of noncontact printing, being digital in nature, does not require a physical master of the target image, as the information regarding the image is in digital form where a variety of modifications as well as a number of prints can be prepared without an additional cost .
Direct laser writing -based printing combines a number of noncontact printing techniques, which permits the realization of 1D to three-dimensional (3D) structures by laser-induced deposition of metals, semiconductors, polymers, and ceramics, without using lithography. Materials are deposited straight from the corresponding precursors. For instance, laser chemical vapor deposition uses focused argon ion laser with a micron-sized spot onto a substrate placed in a reaction chamber filled with a gaseous precursor of the target element where the heat generated by laser absorption dissociates the precursor leading to the deposition of a thin solid layer onto the substrate surface. Herein, the repetitive scans enable the deposition of multilayered structures. This technique is expensive and limited by the specific requirement of volatile metal-organic and inorganic precursor materials not suitable for organic substrates. Similarly, in laser-enhanced electroless plating, the substrate is submerged in a chemical solution of the metallic ions required for the deposition, while a laser beam irradiates the substrate surface causing instantaneous temperature rise, which decomposes the liquid leading to the deposition of a metallic layer on the substrate. A subsequent electroplating can be used to further increase the thickness of the deposited layer. This technique does not allow the creation of 3D structures in the real sense. In another variant, termed as laser-induced forward transfer, a laser beam causes controlled evaporation of the target material put on an optically transparent support plate placed about 100 μm below the substrate, wherein the laser beam evaporated material on the support recondenses on the substrate. Although a variety of materials such as metals, oxides, and polymers can be used for forming both 2D and 3D structures, the limitation of printing is only onto a flat substrate kept parallel to the material support.
The process of aerosol/spray printing  employs the following steps: first the ink is transformed into liquid particles with a diameter ranging from 20 nm to 5 μm based on the viscosity using an atomizer. The inks contain NP suspensions of metals, alloys, ceramics, polymers, adhesives, or even biomaterials. This dense aerosol from the atomizer is transported into the deposition head by a nitrogen flow. The aerosol is further focused by flow guidance in the deposition head before depositing on the substrate. This is a versatile technique capable of writing features with dimensions varying over three orders of magnitude. This technique has a wide range of applications in displays, TFTs, dielectric passivation layers, and solar cells [679, 680].
8.3 Inkjet printing
The process of inkjet printing  involves the transfer of a fixed quantity of liquid-phase ink in the form of droplets from a chamber through a nozzle. The ejected drops fall onto the substrate surface under the force of gravity and air resistance forming a shape and pattern. The solidification of the liquid occurs through evaporation of the solvent and the chemical changes such as the cross-linking of polymers or crystallization. The printed patterns are annealed or sintered in case they are needed [679, 681]. Inkjet printers work in two modes, namely, continuous or drop-on-demand (DOD). It is noted that the material throughput in a single nozzle is different from that in a continuous inkjet system with a higher printing speed. On the contrary, DOD technology allows higher placement accuracy and smaller drop size leading to better resolutions of the printed patterns.
In a continuous-mode inkjet , the ink is pumped through a nozzle forming a continuous jet where a piezoelectric transducer causes the breakdown of the ink flow in single drops at regular intervals, but a stream of liquid emerging from an orifice is intrinsically unstable with a tendency to break into drops also under surface tension alone . Once formed, the drops cross an electrostatic field acquiring charges as they leave the stream. These charged drops are then directed to the desired location on the substrate by an electrostatic deflection system and neutral drops are collected in an ink recirculation system [682, 683]. The pattern is thus created on a moving substrate. In contrast, in a DOD system, an ink droplet is ejected from a reservoir through a nozzle only when a voltage pulse is applied to the transducer. There are thermal and piezoelectric transducers employed in forming droplets. In a thermal DOD inkjet printer, the ink droplet is ejected by means of a heating resistor located inside the nozzle . In piezoelectric DOD systems, the mechanical force generated in a piezoelectric material due to electric field ejects the droplets. Before the printing starts, the ink chamber is influenced by a bias to the piezoelectric transducer to prevent the ink from falling from the nozzle. A zero voltage is next applied to the piezo transducer to bring back the chamber in its relaxed position, leading to a flow of fluid in the ink chamber from the reservoir. Subsequently, the chamber is strongly compressed causing the drop to be ejected from the nozzle. Finally, the chamber is brought back to the initial decompressed condition to pull back the ink in the chamber and to prepare the system for the next ejection (DMP-2800, 2008). As thermal DOD involves the vaporization of the ink inside the chamber, it requires the volatile solvent-based inks. This restriction is not there in piezoelectric DOD that is therefore more versatile than the thermal one.
The inkjet technology has recently proliferated into the area of mass production  primarily because of increasing printing speeds by increasing both jetting frequency higher than 50 kHz and the number of nozzles on a printing head. High nozzle density of 200 nozzles per inch has been achieved due to the development of micro-electro-mechanical systems (MEMS) [38, 685]. Furthermore, it embodies all the advantages, already mentioned for noncontact printing techniques, with the additional plus point of lower cost of the equipment compared to those involving other direct writing processes. For these reasons, it is certainly considered as the most promising technique in PE technology. Nevertheless, like every kind of microfabrication technique, the inkjet printing does have some limitations and critical issues, which must be taken care of to achieve the best performance. The poor resolution of conventional inkjet printers, which is lower than the resolution required for high-performance electronic devices such as OFET, is the first limitation. The minimum achievable resolution is primarily a function of the droplet size that is limited by the nozzle diameter and ink surface tension. In most of the commercial inkjet printers for electronic applications, it is difficult to reduce the droplet volume below 1 pL . Another feature that strongly affects the print resolution is the interaction between the ink and the substrate, especially the spreading phenomena leading to the formation of coffee-ring-like structures as a result of the surface-tension-driven solution transport along a surface with an evaporating solvent [341, 681]. The choice of the ink is therefore critical, as its physical properties such as density, viscosity, surface tension, volatility, and shelf life strongly affect the quality of printed films. Additionally, there is a practical problem of nozzle clogging, which needs frequent cleaning causing material and time waste besides the damage of the nozzles occasionally. In addition, different inks require different postprocessing treatments as sintering, annealing, or simply drying in air, which could change the morphology and the uniformity of the printed pattern and extend the manufacturing time . The optimization of the inks and of the substrate treatment processes thus constitutes the main research challenge in order to achieve improvements in resolution and repeatability of inkjet printed patterns and devices.
Pentacene, being poorly soluble in many solvents, poses severe limitations in terms of its inkjet printed depositions. On the contrary, the pentacene transport properties are strongly dependent on its crystal packing. Thus, in order to overcome the insolubility problem for preparing more ordered thin films, several pentacene derivatives were developed by substitution with different functional groups with the additional possibility of having access to electronic properties tuning, for instance, in terms of charge carrier injection, HOMO-LUMO gaps, and charge carrier transfer rates. Among the pentacene derivatives, TIPS-PEN is the most sought after for OFETs , possessing good solubility in solvents such as toluene, chlorobenzene, THF, and chloroform. Additionally, the bulky functional groups in TIPS-PEN maximize π-orbital overlap resulting in enhanced carrier mobility.
TIPS-PEN can be spin coated, drop cast, and inkjet printed for the active layer of OFETs, in which the electrical characteristics are especially linked to the processing conditions including the solvent used, the postprocessing treatment, and the deposition method, as they determine the morphology of the deposited film. The highest-performance devices have been fabricated with single-crystal TIPS-PEN semiconductor  because they are free of grain boundaries and molecular disorders, which otherwise degrade the charge carrier transport through the material. Generally, using high boiling point solvent and drop casting allowed slower solvent evaporation leading to highly ordered films. Mobilities up to 1.8 cm2/Vs in drop-cast TIPS-PEN OTFTs were reported in literature . Although the process optimization of inkjet printed TIPS-PEN OFETs is still in progress , trial runs demonstrated good device mobility by varying the parameters such as changing the TIPS-PEN concentration, the drop spacing, the channel length, the frequency of injection, the material used for the electrodes, the chemical treatments of the substrate, and the real number of drops that cover the channel .
There are two methods of inkjet printing in practice, namely, single-droplet printing and multiple-drop printing [459, 688, 689]. In single-drop printing, each droplet acts as an individual functional deposit, whereas, in the multiple droplets, a continuous film covering the whole area between the source and the drain electrodes forms the active channel. With the first method, high ordering is easily achievable controlling the hydrodynamic flow in the drying droplet, which forms the channel using a mixed solvent system or varying the surface energy of the substrate, leading to hole mobilities up to 0.2 cm2/Vs for pure TIPS-PEN inks and up to 1 cm2/Vs for inks containing TIPS-PEN blended with other polymers. On the contrary, printing multiple droplets is more effective in forming a uniform thin film over a large area, especially in case the spot produced by one single droplet does not cover the channel region. In the case of printing using several separate single drops, it is very difficult to avoid overlaps between droplets affecting the uniformity of the printed active layer . Using multiple-drop printing, device mobilities up to 0.24 cm2/Vs were estimated in the case of TIPS-PEN  and up to 0.11 cm2/Vs in devices where TIPS blended with polystyrene were used .
For preparing inkjet printing experiment , it starts with 5 wt% TIPS-PEN solution in toluene, which is stirred for 1 h at 90°C and subjected to ultrasonic treatment for 15 min before filling a DMP-11610 cartridge and printing on parylene C substrate kept at 60°C for fast solvent evaporation to obtain a better crystallization . The single-drop patterns were printed with 150 μm drop spacing using single nozzle at 1 kHz frequency, as increasing frequency led to chaotic and uncontrolled droplet depositions. The characterization of inkjet printed TIPS-PEN devices  showed that the multiple droplet approach did not result in high-mobility devices, whereas the single drop in a single row showed the highest mobility. Based on the experimental observations, the mobility trend seems strictly related to the inkjet process; in particular, a more ordered deposition of the droplets gives better performance [346, 691].
8.4 Soft lithography
The soft lithography  is based on a different principle utilizing an elastomeric stamp, molded from a master template, to transfer the desired pattern to the substrate. Mixing a polymeric material out of polyurethane, polyimide, or PDMS with a catalyst/curing agent and pouring the mixture over the master for preparing the stamp. Heating of mixture on master helps in cross-linking the compound in solid form in few hours. Thereafter, the stamp is pealed off from the master in the form of a flexible and transparent elastic negative of the master pattern [692, 693], which is used for transferring the material of interest on the substrate. The success of preparing high-definition microstructures and nanostructures in this process depends on the intrinsic property of the elastomer to form a conformal contact with the master surface . Soft lithography printing has provided a cost-effective solution for patterning over large areas with a resolution of about 20 nm. However, this process cannot be converted into a large throughput of the roll-to-roll type of printing as in flexography and gravure printing described earlier.
9 Recent OS device developments
Recently developed materials such as BDXs, BXBXs and DNXXs  produced air-stable and high-performance OFETs for different kinds of applications. The DNTT and alkylated BTBT-based devices showed very high mobility of 3 and 2.8 cm2/Vs, respectively . In continuation, C10-DNTT-based OFETs involving polyimide dielectric on polycarbonate substrate were developed with very high mobility of 2.4 cm2/Vs, which turned out to be five times of the parent DNTT-based devices placed on the same substrate .
Compared to pentacene, a new pyrene-based organic liquid crystal (LC) semiconductor was prepared by varying the substrate temperature and SAM-modified gate dielectric deposition that resulted in the mobility of 2.1 cm2/Vs in the case of BOBTP  with enhanced long-term device stability.
An OS compound was prepared  having mobility of 4 cm2/Vs at <10 V of operation with added advantage of being appropriate for solution coating process. The devices prepared from this compound especially appear to be of much use in flexible display backplane or low-cost, low-frequency printed logic applications.
Practical realization of flexible high-gain inverters and functional ring oscillators was demonstrated  recently to represent a significant progress leading to dispel the prevalent uncertainty in connection with the practical realization of high-performance microelectronic devices from OS, opening up the newer possibilities hitherto unknown in Si domain. Microcircuits produced by printing instead of photolithography offered the long sought after opportunity of having low-cost and large-area flexible electronics using solution-processed high-performance OTFTs and other devices and components.
Vapor-phase deposited ZnO nanowires and tetrapods  derived from zinc compound reduction in a novel approach were used for preparing an n-type OS – [6,6]-phenyl-acid methyl ester-based nanocomposite showing electron mobility in the range of 0.3 to 0.6 cm2/Vs representing a factor of 40 enhancement from the pristine state, with a promising enhancement in the case of n-type OFETs.
Using soluble polyacene in a proprietary binder, The Center for Process Innovation UK developed a compound named FlexOS™ for OFET applications. The devices prepared using FlexOS™ in top-gate-bottom-contact configuration with Au source/drain contacts defined by photolithography and chemical etching along with CytopCTL-809M fluoropolymer dielectric showed high mobility with <5% standard deviation across a 100-mm-diameter substrate. In contrast to earlier use of low permittivity binders in preparing small-molecule semiconductor thin films , in the present work, a series of binders having permittivity in the range of 2.8 to 5.8 were used , which enabled easy control of the phase separation of small-molecule polymer and binder in the deposited films.
10 Encapsulation issues of OE
Various reliability issues of OS devices were systematically studied in a doctoral program  from where some of the salient observations and experimental results are taken in the following discussion.
OS are generally sensitive to moisture and oxygen and therefore need encapsulation for proper protection [704–706] from further deterioration during their use in devices. Consequently, device reliability issues arise due to the environmental instability of both the active materials and low work function electrodes. Low work function cathode metals oxidize very fast after exposure to oxygen and moisture resulting in insulating oxide barrier layers that impair charge carrier injections. For protecting devices fabricated on transparent flexible substrates, which are not impermeable to moisture and oxygen, it is necessary that the barrier also meets the device requirements such as in the case of an OLED, the barrier should be flexible and transparent to give the full light output. For devices fabricated on glass substrates, combinations of epoxy/getter are found effective but are not useable for transparent flexible organic devices. The purpose of a barrier layer in encapsulation is to improve storage lifetime and enhance device-operating lifetime. The half-life is estimated by measuring the time after which, for instance, initial brightness of an OLED is reduced to 50%. This is a common method of determining the OLED life that has been used to evaluate the barrier properties [707, 708].
Several methods including epoxy and glass seal  were proposed for encapsulating OLEDs and other organic semiconductor-based devices that use a glass substrate. In this method, the adhesive is applied using a syringe, which may cause device damage in case the epoxy touches the semiconductor layers. Alternatively, Al-Li and high-density polyethylene (HDPE) multilayers were used with a half-life of 63 h on glass substrate , while elsewhere a combination of oxide, polymer, and epoxy seals  was employed for OLEDs on glass with a half-life of over 1000 h. Accelerated testing at 65°C and 85% relative humidity (RH) of PECVD SiOx/Si combinations on glass  gave a half-life of 7500 h. For devices prepared on flexible substrates, chemical vapor deposition (CVD) parylene-N/C, individually or in combination, showed four times enhanced lifetime over that of unencapsulated  devices. A proprietary barrier formulation on PET gave over 2000 h of half-life , whereas a lamination-based encapsulation of OLEDs on plastic substrate showed a half-life of 230 h .
Despite attempting several formulations for improving the reliability of OS devices as mentioned above in brief, it is still a challenging task to find a satisfactory encapsulation solution, which can be deployed using commercially available resources. Most of the commercial formulations typically require specialized equipment and intellectual property licenses.
Besides sensitivity to moisture and oxygen, delamination is another phenomenon , which may crop up due to environmentally induced oxidation in which moisture permeates through defects into the interface between cathode and the active layer, where the ensuing chemical reactions give rise to outgassing or volumetric expansion culminating into delamination. In addition, exposure of any active layer during device fabrication is equally harmful . For example, C60 in OFETs undergo rapid degradation after exposure to the ambient air when compared to a transistor with alumina .
From among various flexible encapsulation approaches, thin films have attracted the most attention due to their light weight, transparency, and high level of mechanical flexibility. Besides offering good barrier features to the encapsulant under consideration, other critical aspects such as compatibility with the OE, changes introduced into the active layers and substrates having low glass transition temperatures and thermal stability, and the limitations due to barrier layer deposition and processing temperatures are equally important to take care of . Consequently, all inorganic layers involved in encapsulation must be deposited at low temperature for being compatible with the device. However, the processing at low temperature is invariably associated with more defects, limiting the barrier performance. Thus, the entire process of developing and integrating high-barrier encapsulation films with OE remains a compromise between two extremes .
The barrier performances of inorganic materials are highly promising for thin-film encapsulation for improving the lifetime and reliability of OE, as they simplify the manufacturing process compared to other multilayer encapsulation methods. However, inorganic films have a serious drawback due to large defect density in the films, which enable moisture and oxygen to permeate through the barrier layers. Methods such as physical vapor deposition (PVD), plasma-enhanced CVD (PECVD), atomic layer deposition (ALD), and plasma-enhanced ALD are used in depositing thin films with improved barrier properties . Currently, the deposition of inorganic oxides such as Si-O-N and Al-O-N using PECVD and ALD are being considered most favorably in this context. Comparatively, PECVD is more viable for manufacturing due to higher deposition rates [714, 715] compared to ALD, whereas ALD provides defect-free compact thin films of very high quality. PECVD SiOx and SiNx films show useful barrier properties, except containing large defect densities, which may severely limit the utility of these films in meeting the target requirement of OE. The encapsulation quality of the thin film involved is directly correlated with improved lifetimes by integrating it with the actual device. For instance, the lifetime of encapsulated OLEDs with PECVD SiNx was 600 h compared to 6 h for a bare device . Recently, PECVD SiOx:H films exhibited better protection of OLEDs . Using an appropriate combination of HDMSO and O2 precursors, it was possible to deposit a single composite layer consisting of SiO2 and Si, which protected OLEDs up to 7500 h, when stored at 65°C in 85% RH. Alternatively, ALD-based Al2O3 has shown promising encapsulation properties as reported [710, 717, 718] due to high density, low number of defects, and conformal coating on the uneven surfaces at relatively low temperatures.
ALD is a layer-by-layer deposition of materials based on chemisorption of molecular precursors, which are introduced into the deposition chamber with an inert carrier gas to form monolayer coverage on the surface of the sample. Excess precursor is purging with an inert gas followed by the introduction of the next precursor. These processes are repeated to build high-quality films with featureless microstructure and conformal coatings. In addition, laminated structure of Al2O3 and ZrO2 and dual-layer structure of SiOx and Al2O3 using ALD have shown excellent barrier performance [719, 720]. In another attempt , nanolaminate comprising alternate layers of 2.6 nm Al2O3 and 3.6 nm ZrO2 film with total thickness of 100 nm deposited at 80°C was compared with single Al2O3 layer with same thickness.
In order to address various reliability issues related to OS device developments, a detailed study was carried out recently  to study the degradation mechanisms involved in OLEDs. Some of the salient points arrived at during this study have been included in the present discussion to highlight the importance of this kind of reliability investigations, which are necessary for improving the performance of these as well as other semiconductor devices needed in OE.
Degradation of the performance of an OLED over time can be attributed to three major categories, namely, dark spots formation, catastrophic degradation, and intrinsic degradation , which are briefly described here.
Formation of nonemissive spots within the emissive area of a device causing device degradation is termed as a dark spot in OLEDs, wherein the total device luminance is reduced . This phenomenon is accelerated by the presence of moisture and oxygen, which can be controlled via proper device encapsulation. As rigid encapsulation cannot be used in flexible device configurations and the existing thin-film encapsulation techniques lack the high impermeability requirement of glass encapsulation, encapsulation of flexible OLEDs still remains a challenge. In catastrophic degradation , a sudden decrease or total loss of the luminance is observed due to shorts appearing across the device  caused by preexisting defects in the active layers or the electrodes of the OLED. It is possible to contain this kind of degradation by minimizing the defects during material deposition resulting in very uniform and homogenous films. In intrinsic degradation , a progressive decrease in the luminance of the OLED occurs during device operation , leading to an intrinsic decrease in the EL quantum efficiency of the OLED.
Dark spots are generally attributed  to the external contaminants, pinholes, bubble formation, gas evolution, and crystallizations following causative phenomena in an OLED.
Substrate cleaning and removal of the loosely attached material chunk from the previous run sticking inside the chamber before starting thin-film deposition are very critical, as the foreign material contaminants either present on the substrate or getting dislodged from the chamber inner surface, during deposition, are noted to create direct pathways in the active layer and cathode-organic interface for permeation of oxygen/moisture . In the case of the foreign particles that are larger in size than the thickness of the organic layers, it may lead to incomplete coverage during organic film deposition offering easier conduits for moisture/oxygen propagation.
Rigor of the substrate cleaning involved in OLED fabrication can be appreciated from the brief description of a typical fabrication  sequence, where ITO-coated substrate is spin cleaned first with acetone, isopropanol, and methanol followed by HMDS and AZ4620 photoresist coating. The patterned resist is postbaked at 110°C before etching in aqua regia. The cleaning of ITO pattern in acetone, isopropanol, and methanol is finally flushed with O2 plasma for 2 min. Just after the plasma treatment, the ITO pattern is coated with PEDOT:PSS and baked at 110°C inside a glove box with access to the evaporation and CVD chambers. The OS comprises 50 nm spiro-TPD and 50 nm Alq3 and cathode of 50 nm of 10:1 Mg:Ag alloy along with 50 nm Ag. Although glass substrate is an excellent moisture/oxygen barrier, the cathode area, however, is permeable, and to protect the organics against attack from the cathode side, the sample is coated with 2 μm Parylene C and then 100 nm Al with proper masking so that the device is not shorted. This Parylene-Al stack is repeated twice. The measured half-life of 600 h at 196 cd/m2  was measured for devices prepared using the above-said sequence. In the case of OLEDs on PEN substrate , it uses ITO-coated PEN, which is cleaned using the same solvent-cleaning procedure as in the case of glass substrate, UV/ozone treatment for 5 min, and aqua regia etching of pattern. The OLED fabrication is also identical to that in glass substrate-based devices.
On the contrary, pinholes are formed during film depositions especially when deposition rates are faster, causing not only uneven layers but also trap gases. Such defects in the cathode and/or protective layers easily provide pathways for oxygen and/or moisture to permeate through the sensitive metal layers and reach to the active layers. Cathode oxidation, for instance, leads to reduced electron injection and the degradation of the active layers causes overall device performance impairment. The growth rate of the dark spots is related to the pinhole size . The larger the pinhole size, the faster the growth rate, causing still shorter device lifespans .
Gas evolution produced by electrochemical processes going on at the cathode/organic interface in the presence of moisture/oxygen results in bubble formation at the interface, which ultimately causes cathode delamination from the organic layers underneath [111, 725–729]. This causes disruptions in the electron injection at the cathode-organic interface; consequently, dark spots are seen at such sites.
Organic coordination compound tris(8-hydroxyquinolinato) aluminum, commonly abbreviated as Alq3, is employed in fabricating OLEDs and its exposure to humidity through pinholes induces the formation of Alq3 crystallites in the originally amorphous films , resulting in protruding structures that are several times thicker than the original film with water content higher than that in the amorphous Alq3 regions . The moisture diffusion through the microscopic defects in the cathode causes crystallization of Alq3 in the OLED structure itself, forming Alq3 crystals, which ultimately cause surface irregularities affecting the adhesion between the organic and metal layers. Delamination occurs due to the crystalline Alq3 clusters, being thicker than the surrounding amorphous regions, lifting the cathode, leading to loss of contact between Alq3 and the cathode. The areas with no contact appear as nonemissive dark spots .
Either heating of the HTL/ETL directly or Joule’s heating due to current flow is noted to induce crystallization in amorphous organic materials. Although the crystallites, produced in this manner, are comparatively less affected by moisture and hence do not cause chemical decomposition [730, 731], a higher degree of crystallization indeed induces irregular surfaces that lead to intergrain gaps or pinholes. Hence, understanding and controlling the thermal effect on the ETL/Alq3 is crucial for keeping the cathode-organic interface integrity within acceptable limits, which in turn controls device lifetime and performance. For instance, for good cathode-organic interface quality, Alq3 films should not exhibit small grain morphology as seen in amorphous organic materials; however, neither should they exhibit highly irregular crystalline morphology. When Alq3 is deposited on top of the HTL at room temperature, it is amorphous exhibiting rough surface morphology, comprising small grains with a high grain boundary density providing for a high number of percolation paths for moisture and oxygen. Studies have shown  that annealing Alq3 deposited on NPB at 100°C renders the surface of Alq3 very smooth and featureless , exhibiting fewer grain boundaries, which mean fewer percolation paths for moisture and oxygen. Further studies have shown  that Alq3 deposited at 100°C on top of ITO/NPB leads to improved device lifetime than annealed ITO/NPB/Alq3 stack. Atomic force microscopy (AFM) inspection revealed that substrate heating, while depositing Alq3 at 100°C, led to a slightly rougher surface when compared to the annealed Alq3/NPB/ITO stack . This minimal roughness might be beneficial in regards to increasing cathode adhesion by increasing surface-adhesion area. Moreover, larger-scale surface roughness due to larger grains might result in lesser moisture permeation. Hence, substrate heating while depositing Alq3 or annealing Alq3 at a temperature below glass temperature can result in large grain films more useful for improving device lifetime.
Traditionally, many OE devices employ glass/metal lids as barrier layers to protect against moisture/oxygen. An inert metal or glass layer is used as a “lid” that is sealed with an UV-cured epoxy resin. Getter materials are also used in such structures besides using an inert gas that fills the voids to make the device quite stable. Although this approach does provide protection against moisture/oxygen permeability, it has flexibility as well as robustness issues that need to addressed. A device encapsulated with glass is prone to accidental breakage compromising the barrier features causing overall device deterioration. Further, glass encapsulation is impossible to integrate into roll-to-roll OLED fabrication. Rollable display screens, foldable maps, and flexible lighting options are just few of the attractive market applications for OLEDs. However, flexible OLEDs require flexible encapsulation structures. Furthermore, flexible OLEDs also require low-temperature encapsulation procedures so as not to damage the flexible substrate and the organic layers. Thus, for roll-to-roll manufacturing of OLED, it is essential to go for flexible encapsulation.
On the contrary, using flexible polymer lids may seem to be better than glass in the aspect of flexibility, but polymer lids can be subjected to delamination while folded leading to device degradation. Hence, the only option for reliable encapsulation for flexible OLEDs is thin-film encapsulation.
Thin-film encapsulation layers provide easy integration into flexible OLEDs. In the case of thin-film encapsulation utilizing low-temperature deposition processes, it can overcome all the obstacles that glass encapsulation encounter for flexible OLEDs. Moreover, thin-film barrier layers are extremely thin and are transparent unlike metal lids, and they are not as bulky as glass substrates and are easy to integrate into roll-to-roll production. Despite many advantages, thin-film encapsulation layers have to maintain a high level of impermeability for being a viable option due to the sensitive nature of OLEDs. Certain multilayer thin-film encapsulations utilizing different inorganic and/or organic layers have been shown to possess high impermeability to moisture/oxygen.
In spite of possessing flexibility and transparency, thin-film barriers still lag behind the stringent requirements for moisture and oxygen impermeability requirement of OE. A water vapor transmission rate (WVTR) of 10-6 g/m2/day and an oxygen transmission rate (OTR) in the range of 10-3 to 10-5 cm3/m2/day are desirable for OLEDs.
Multilayer encapsulations offer promising features for protection from ambient conditions for flexible OLEDs. Oxygen permeation rates of 10-6 cm3/m2/day were achieved via Barix encapsulation on a plastic substrate . Barix multilayer encapsulation, comprising alternating organic and inorganic layers, prolongs the permeation path of oxygen and/or moisture through the barrier layers wherein the organic layers act as a planarization layer that helps in a smoother surface, whereas the inorganic layers such as aluminum oxide act like a permeation barrier to moisture and oxygen . Barix encapsulation is also transparent and hence is suitable for top-emitter OLEDs too. Most importantly, the encapsulation deposition process is a low temperature one, making the process extremely suitable for flexible OLEDs.
The organic materials in Barix method are deposited using a nonconformal deposition technique – the organic material starts as a liquid, which is then vaporized and UV-cured to produce a solid layer . The inorganic Al2O3 is deposited via DC sputtering. Four to five polymer (organic)/inorganic dyads are used for encapsulation . The drawbacks of multilayer structures such as Barix are that they require complex post-OLED fabrication deposition with the need to have separate deposition techniques for different layers.
Longer processing time associated with multilayer thin-film encapsulation makes it prone to complications/contamination. Multiple constituent materials may also increase the cost of fabrication and thus make the fabrication of the encapsulated OLED potentially expensive.
Besides lengthy processing, it was also shown that Barix thin-film encapsulation does not acquire the integrity of glass barrier layers for flexible or rigid substrate OLEDs . Therefore, a need exists for alternative means of suppressing dark spots growth – means that would not introduce complexity or potential for cost increase. We know that the cathode-organic interface is the main dark spots formation site as a result of moisture/O2 propagation through the pinholes in the metal cathode in the OLED. In case the interfacial adhesion between the cathode and the organic layers is strong enough, the facile formation of metal hydroxide sites at the interface is potentially hindered. By finding ways to strengthen the cathode-organic interface, and consequently the inherent resistance of the OLED structure to moisture and oxygen, suppression of dark spots can potentially be achieved.
Despite superior thin-film encapsulation method available such as Barix, it produced OLEDs with higher dark spots growth when compared to glass encapsulated devices . Another study conducted in this context  demonstrated that the use of Barix encapsulation on plastic substrates showed one fourth of the half-life of a glass encapsulated and glass substrate devices. For comparison, Barix encapsulation resulted in 3700 h of half-life for a glass substrate OLED vs. 2500 h for a plastic substrate OLED . This study clarified that even superior commercial thin-film barrier layers did not achieve the high-quality encapsulation, which was seen with glass barriers.
The nonencapsulation approach of dark spots suppression were studied by examining the influence of thermal annealing and heating flexible encapsulated OLEDs in terms of their lifetime and controlling dark spots growth [63, 730, 731]. Furthermore, using graded cathode, a mixed organic metal layer (MOML) was also noted to prolong device lifetime .
A study was conducted  to examine the influence of annealing of OLEDs especially in terms of device lifetimes under carefully selected temperature ranges by heating the individual layers. For instance, HTL and ETL were deposited at 140°C and the Al cathode at 60°C, which enabled OLEDs to show 30% better luminance at 20 mA/cm2 current density when compared to a non-heat-treated devices with the same organic and cathode layers . This improvement was mainly attributed to the crystallization of HTL, such that further operation and/or storage would not affect the device adversely.
When these devices were stored for 2 weeks in 40% RH environment, the heat-treated devices continued to possess EL at 9 V , and on the contrary, the non-heat-treated devices exhibited no emission at all. Another set of OLEDs with only the HTL deposited at 140°C was also prepared, which showed EL at 9 V after 2 weeks of exposure to 40% RH conditions, but they only possessed a slight increase in luminance over the non-heat-treated OLEDs when compared to the luminance increase seen with the individually heated layer OLEDs .
In another subsequent study, conducted later in this context , it was concluded that 160°C deposition of the organic layer retarded dark spots growth, and after 20 h storage in air, almost 80% of the emissive area of a non-heat-treated devices became nonemissive. On the contrary, dark spots on the 160°C deposited organic layer OLEDs were still smaller in size .
In yet another subsequent study , it was shown that 120°C annealing of OLED structure comprising ITO/NPB/Alq3/Al for 1 h resulted in marked improvement in luminescent efficiency, brightness, and stability when compared to nonannealed devices. For instance, maximum luminance of a 120°C annealed device was 6240 Cd/m2 compared to a mere 3650 cd/m2 for an unannealed device . Furthermore, OLEDs were selectively annealed layer by layer to focus on the effect of heating on progressively deposited layers. It was noted that, by annealing a device, which includes the Al cathode at 120°C, the best luminance (670 cd/m2) and efficiency (3.5 cd/A) were reported . Furthermore, new OLEDs with the configuration of ITO/NPB/Alq3/LiF/Al were fabricated with individual layers selectively annealed just like done before . Once again, luminance and luminance efficiency improvements were observed for the OLEDs with all the layers annealed. Moreover, improved half-life from 14 to 52 h was noted with all the layers annealed. These results show that annealing the cathode leads to enhanced electron injection. AFM scan of the Al cathode with all the layers annealed showed that annealing led to leveling out of the undulation of the Al film , which could possibly be the reason behind better operational stability.
Before addressing the various salient features of OS materials presently covered in this review, a quick look into the status of amorphous and poly-Si devices for large-area display and flexible electronics applications will be useful as a benchmark for frequent comparisons.
α-Si, though lacking in long-range order, contains abundant atomic species with unattached bonds, which give rise to localized trap energy states situated below the conduction band edge . Although a large density of such defects lowers the conductivity, it can be mitigated to some extent by hydrogenation  resulting in a hydrogenated amorphous state termed as α-Si:H. A low-temperature PECVD is available to deposit films with controlled hydrogen out-diffusion. Carrier transport in α-Si:H involves thermally assisted hopping between trap states , leading to a mobility approx. 1 cm2/Vs . Even this low mobility is acceptable having a number of other advantageous features such as low defect density resulting in good turn-on, subthreshold slope under 1 V/decade and threshold voltage VT<1 V . High-quality PECVD tools are available for depositing extremely uniform films on large-area glass substrate with very uniform device characteristics .
Material reliability is a major issue  with α-Si:H, where large density weak Si-Si bonds, present in a randomly distributed Si network, are prone to disruption under the influence of applied electric field. Besides, there is sufficient charge trapping at the poor-quality interface between α-Si:H and Si3N4. Both these effects lead to temporal shifts in device threshold voltage . Although n-type metal oxide semiconductor (NMOS) α-Si:H devices are good enough for TFTs in displays, they are not all appropriate for logic applications.
In polycrystalline Si (poly-Si), the next alternative, there are randomly distributed crystalline grains ranging from tens of nanometer to several microns in size having crystalline quality  surrounded by grain boundaries rich in defects. Here, the larger the grains, the better it is for device applications. For instance, if the grains are larger than the transistor channel length, device performance approaches that of single-crystal Si. Poly-Si films are deposited  either as a polycrystalline film as such or as an amorphous film followed by recrystallization, where larger grains are possible to have. Si deposition, in this context, uses any one of the methods involving low-pressure CVD (LPCVD), PECVD, or sputter deposition. In LPCVD, a deposition above 580°C produces poly-Si films and below it produces α-Si films. PECVD processes are typically run between 250°C and 400°C, whereas sputter deposition is possible between room temperature to as high as permitted by the tool. The recrystallization of as-deposited film  is performed either by thermal annealing at >600°C or laser annealing. Laser annealing produces large-grain poly-Si and is compatible with low-temperature substrates, despite raising the temperature above Si melting point. For instance, 30 ns excimer laser pulse take <1 ms  for heating/cooling cycle for each pulse. Thus, having an appropriate barrier layer beneath the Si film, the molten Si cools back to room temperature before enough heat diffuses into the substrate causing damage.
Mobility in poly-Si depends on the grain size. For instance, LPCVD poly-Si may have electron mobility approximately 10 cm2/Vs compared to the thermally annealed one having 30 cm2/Vs . Excimer laser crystallized poly-Si may have electron mobility as high as 100 to 200 cm2/Vs . Using special annealing methods such as sequential lateral solidification (SLS) or grain engineering, electron mobility has gone up to 300 to 500 cm2/Vs [745, 746] approaching the single-crystal Si value.
Comparing OS with the inorganic ones, few points emerge clearly. Despite the best efforts put in to grow organic crystals and deposit thin films, the mobilities are still an order of magnitude or two lower than in inorganic crystals, respectively. However, the organic thin-film devices have outperformed the α-Si devices because of nonexistent grain boundaries with excess defects.
Compared to polymers, small molecules give better crystals with improved mobility but produce poor thin films. Polymers form more ordered thin films but with reduced mobility due to a larger number of disorders. Thus, polymer thin films are better for OE applications where even lower mobilities may be useable. MBE has further promises for smaller molecules but needs more studies following a process possibly similar to gas source MBE , where appropriate precursors can be developed for their use in high-quality epitaxial films. Of course, ALD  is highly attractive for depositing high-quality metal and dielectric thin films for device fabrications, where large-scale fabrication tools are already available.
In polymers, inserting appropriate functional groups leads to better mobility via improved π-conjugation besides creating mechanical torsion in the backbone leading to higher band-gap resulting in better stability. Pentacene was investigated in detail by examining two basic parameters that characterize the hopping conduction, namely, reorganization energy of carrier exchange and intermolecular orbital coupling transfer integral . Having five benzene rings and a delocalized π-system with a rigid structure, pentacene molecules need smaller reorganization energy with larger orbit coupling transfer integral between neighboring molecules. It packs in a herringbone structure in solid state that provides 2D isotropic electronic structure on the substrates, which is also advantageous in the application of TFTs . Although pentacene appears as an ideal OS for OFETs, its chemical instability and higher oxidation potential make it susceptible to air oxidation that is harmful for device applications. With the observation of similar instabilities in higher oligoacenes such as naphthacene and hexacene, it was suggested that perhaps further improvements in mobility were not possible using acene structures. However, in this context, the enhancement of intermolecular interactions using heteroaromatic compounds of S and Se was worth exploring , as it was noted to provide conduction paths in organic charge-transfer salts and organic superconductors. However, using this approach in oligothiophene and oligoselenophene-based OFETs, the mobilities still remained very low . Further study of such unexpected results not only explained the observed behavior but also indicated the importance of the geometry of the frontier molecular orbitals in improving the mobility. Based on this experience of designing better molecules, it is worth noting the following results in this context. Vapor-deposited DPh-BTBT and DPh-BSBS thin films on Si/SiO2 exhibited mobility of 2.0 and 0.3 cm2/Vs in the respective devices [750, 751] accompanied by extremely good air-stability and more than 6 months of shelf-life under ambient conditions. Cn-BXBXs OFETs with a top contact configuration exhibited mobility better than 0.5 cm2/Vs; in particular, C13-BTBT showed the best performances with the maximum mobility of 2.8 cm2/Vs . In contrast, the Cn-BSBS-based devices showed two orders of magnitude lower mobility than that of C13-BTBT . Air-stable DNTT and DNSS-based OFETs showed mobility of 3.0 and 2.0 cm2/Vs, respectively [752, 753], with negligible degradation and low hysteresis in the device characteristics over long duration. When compared to pentacene, DNTT may be taken as standard OS in the next generation of OFET developments [677, 754].
The role of SAMs in modifying the interfaces between organic-organic, organic-inorganic, and metal-organic combinations is possibly more profound than those observed in inorganic semiconductors possibly due to very strong covalent bonds present there compared to van der Waals forces in OS. The influence of OS NPs with immense scope of changing their functionalities using core-shell configuration is another area worth examining with reference to their use as SAMs in this context.
Combination of metal NPs with CNTs is another area of great interest for their use in contact electrodes and interconnects. One example of rubber-like interconnect developed recently [755, 756] is a useful example in this context. Bundles of CNTs with BMITFSI  ionic liquid and rubber-like fluorinated copolymer were stirred together by sonication to prepare stretchable interconnect films after drying in air. These flexible interconnects could be stretched by 40% due to CNTs embedded in the rubber-like fluorinated copolymer. A new printing technology  was reported in this context involving microporous base coated with polyethylene, PVA, boracic acid, or other materials, which absorbed the solvent of the Ag ink. The ink included Ag NPs, alcohol, surface-active agent, and dispersant, which reacted with the base on the paper at room temperature without causing any damage to the OFETs. When the printed interconnect was stored, the lifetime was estimated approximately 6 months, with 10% increase in interconnect resistance after 1 month .
The extensive use of impurity dopings and ion implantation  is well known in microelectronics, while mass-producing devices and circuits with numerous modifications were introduced in the tool designs to handle larger-sized Si wafers in sufficiently large-sized batches aided by process automations and robotics. In the case of OS, the status of diffusion doping and ion implantation is at the beginning of understanding the basic processes involved before exploring their applications further. Nevertheless, organic devices are already there in a number of mobile device displays, in which the doping has enhanced the efficiency of optoelectronic devices and improved process reproducibility for commercial productions . Despite having succeeded in improving the performance of optoelectronic devices even with less than the complete knowledge of the precise mechanism of doping, it is yet to be put to use in organic solar cells and organic transistors, opening newer opportunities for research in times to come. Similarly, ion implantation has entered the OS device processing area in the form of surface passivation technique  for improving the device reliability, a welcome start. Like impurity doping, the process of ion implantation may also take some more time to find even better applications in the future.
Based on the recent findings of various research groups involved in studying OS for their uses in device applications, considerable improvements in basic understanding of the processes involved accompanied with more profound technological applications are anticipated as illustrated here with few specific examples.
Understanding the basic mechanism of charge carrier transport in OS is the key to improve the performance of the explored device structures after realizing the optimum conditions necessary for material growth and processes involved therein. In this context, it is necessary to have the right kind of diagnostic tool to verify the theoretical predictions based on the proposed models proposed by several groups. In this context, the recent development of using transmission electron microscopy (TEM) to map the material structure very precisely at atomic level is going to boost the level of understanding of charge carrier transport in OS in the near future. In this context, an SAM of pentathiophene derivative deposited on an electron beam transparent substrate was successfully characterized using TEM, where high-precision crystallographic structural maps were produced . These maps were used to decipher size, symmetry, and orientation of unit cell as well as the orientation, structure, and the degree of crystallinity inside the domains, which are essential for understanding the transport properties of the charge carriers in such semiconductor films . This kind of characterization is expected to experimentally determine the influence of various molecular functionalization schemes but also help in optimizing the same in a convenient way.
OS devices such as OLEDs, OPVSCs, memory elements, and OFETs are predicted to reduce fabrication costs and enable new functions. Besides optimizing the performances of these individual devices where the phenomena of electronic and optoelectronic material properties are explored for best possible performance, the feasibility of combining features of two different devices in one may be explored for some unique applications. In this context, combining the FET action with light generation and detection has been already explored in recent past. The development of newer compounds designed at molecular levels is worth considering in this direction as shown in the examples mentioned here.
The concept of organic light-emitting transistor (OLET), having the FET structure with the capability of light generation, is seriously being explored for bright multicolor electroluminescent displays with simpler driving circuits besides possessing technological potential for realizing intense nanoscale light sources and highly integrated optoelectronic systems, including electrically pumped organic laser . Although OLEDs are the most developed OS devices as the active matrix OLED displays are already in the market, the major associated problems of the exciton-charge interactions and the photon losses at the electrodes need further consideration for improving their reliability and life. High-density electrons and holes, present in the light-emitting layer, cause significant exciton-charge carrier quenching in the close proximity of the excitons . For enhancing efficiency, brightness, and stability, the exciton-charge carrier quenching is necessary to be minimized. With this background, the development of OLET is primarily motivated by the possibility of having an alternate display/light source by adapting a geometry to suppress the photon losses and exciton quenching inherent in the OLED architecture. In this context, the phenomenon of exciton-metal interaction has been, so far, minimized in ambipolar single-layer OLETs  by separating the light-emitting area from the metal contact. These single-layer devices still have the charge carrier accumulation and the exciton formation zones coinciding together, leading to severe exciton-charge quenching resulting in poor external quantum efficiency (EQE) of 0.2% but demonstrated remarkable results in terms of brightness . An alternate structure of horizontal p-n heterojunction OLET was also fabricated having light emission area far from the contacts; however, exciton-charge quenching could not be avoided . Alternatively, two types of bilayer configurations involving a highly efficient luminescent layer superimposed over a unipolar conducting layer or putting p- and n-type transport layers directly in contact with each other were attempted, which did not provide any control over exciton quenching or photon losses as such. Yet, in another work , a trilayer heterostructure OLET enabled simultaneous control of electrode-induced photon losses and exciton-metal and exciton-charge interactions. OLET devices with EQEs of 5% were demonstrated, which exceeded the best OLEDs based on the same emitting layer and optimized transport layers .
In a recent attempt, a basic OFET device was modified to produce light emission by employing self-assembled oligophenylene and phenylene-thiophene nanofibers  on FET platform appropriate for producing EL and photoluminescence (PL) simultaneously. The phenomenon of localized, polarized, and wave-guided EL was realized in aligned nanofibers on OFETs by applying AC bias on the gate electrodes , which caused sequential injection of holes and electrons resulting in charge carrier recombination leading to light emission from a small area near the metal-nanofiber interface. In this configuration, the nanofibers acted as optical waveguides and part of the generated light was therefore guided along the nanofibers, which finally radiated from the fiber end. Besides hexaphenylene nanofibers, this scheme was found equally applicable  in the case of nanofibers made from a variety of different molecules or material with altered spectral characteristics. The realization of an electrically biased organic nanoscale light emitter demonstrates the ability to fabricate on-chip light sources with a tunable emission spectrum via synthesis of appropriate molecular building blocks. Extending this concept further, PL and EL of two devices made from naphthyl end-capped oligothiophenes were investigated based on an organic light-emitting FET (OLEFET) configuration. Observing the similarity between EL and PL spectra from both materials, it indicates that the light emission is caused by the same electronic transitions. It was further observed that the intensity of the EL emission increased with increasing frequency of the AC gate voltage, and it was nonlinearly dependent on the amplitude of the AC gate voltage.
In another study , a single-crystal light-emitting transistor (SCLET) was realized by employing single-crystal optical waveguide, coupler, and resonator prepared from as-grown organic single crystals. Using parallel edges of the crystal as Fabry-Perot cavity, effective optical coupling between single-crystal waveguides and optical feedback resonator was realized by simple crystal lamination. Practical realization of the concept demonstrated spectral narrowing behavior as a clear evidence of the cavity effect. These results open a route to the development of color tunable and highly efficient SCLETs as well as electro-optical interconnecting devices .
Successful fabrication of air-stable ambipolar heterojunction-based organic light-emitting FETs (OLEFETs) was demonstrated using multidigitated top-contact device geometry . Active layers of p-type pentacene and n-type N,N′-ditrydecylperylene-3,4,9,10-tetracarbocyclic diimide (P13) with a protecting layer of 2,5-bis(4-biphenyl thiophene) (BP1T) were deposited using cluster beam deposition technique for well-ordered thin-film deposition. These results open up the newer possibilities of exploring the light generation capability of OLEFETs in the near future, where many problems of conventional device structures would be taken care of appropriately.
A real hybrid combination of LC-on-OFET was reported in another recent publication  to sense ultra-low-level gas flows. These devices were realized by mounting 4-cyano-49-pentylbiphenyl-5CB LC-on-OFET channel. The LC molecules on the channel layer enhanced the source-drain current due to dipoles of LC molecules. When low-intensity nitrogen gas flow was employed, the drain current increased depending on the intensity and time of the gas flows. These LC-on-OFET devices detected extremely low level of nitrogen flows (0.7 sccm–11 ml/s), which could not be felt by human skins. The successful demonstration of similar behavior of the LC-on-OFET devices onto a polymer film skin suggests a viable application in tactile sensing .
With the emergence of nanoscience and technology disciplines, the concept of “material by design” (S. Ahmad. Nanocrystals, superlattices and nanomaterials, communicated for publication) appears to be a reality in the near future. This concept has been primarily considered as an extension of the well-known principle of materials being the manifestations of the constituent atomic and molecular species arranged in specific orders. In case atoms and molecules of conventional materials are replaced by nanosize building blocks of numerous material species already known to date, still a bigger family of mesomaterials is possible to design and realize. As physical, chemical, and biological properties of nanosized material species are strong functions of size and shapes, it gives enormous latitude in designing materials by choosing from large varieties of material species, which is possible to further extend to another level of sophistication. In this context, even materials would be possible to design with preassigned features by choosing the right kind of building blocks resulting in the desired combination of physical, chemical, and biological features. Such synthetic materials are also named as mesomaterials or metamaterials belonging to the category of materials by design. In this context, the role of organic molecules are certainly going to play very significant as can be seen from the discussions presented here in this review. Combining the strength of organic molecules with more complex biomolecules, the possibilities of realizing newer families of synthetic materials are immense as can be verified by the marvels of nature where such complex modifications are playing their roles in evolving species over millions of years in the past.
Besides intensive efforts made by material scientists and chemists to design and synthesize newer molecules with immense possibilities of tailoring the properties appropriate for their use in device fabrication, it is equally important for the device scientists and process engineers to examine the gap areas for improving the device behavior and manufacturability at low cost especially required in niche areas as mentioned earlier. Various observations made in this context in the present review may be used accordingly hopefully.
The author gratefully acknowledges the support and encouragement extended by the Confederation of Indian Industry (CII) and the Center of Excellence in Green Nanotechnology, CII Western Region, Ahmedabad, Gujarat, India. The challenging R&D environment, created and supported by Mr. Anjan Das, Mr. G.K. Moinudeen, and colleagues at the Center of Excellence in Green Nanotechnology, was especially enjoyed by the author while conducting this study and the positive encouragement received is sincerely acknowledged. In connection with preparing the present manuscript, various concepts, developed over past few decades in attempting to improve the material characteristics for device applications, which have been elaborately discussed in various excellent review articles and especially the present status of the device-related developments reported in recent publications, have been used with due acknowledgements. The references provided here may not be that exhaustive as in most review papers mentioned above, but the contributions made by numerous researchers in this area are all duly acknowledged, directly or indirectly, through the references currently included in the text as well as those mentioned in the referred review articles.
Friend RH, Gymer RW, Holmes AB, Burroughes JH, Marks RN, Taliani C, Bradley DDC, Dos Santos DA, Brédas JL, Lögdlund M, Salaneck WR. Nature 1999, 397, 121–128.Google Scholar
Sirringhaus H, Brown PJ, Friend RH, Nielsen MM, Bechgaard K, Langeveld-Voss BMW, Spiering AJH, Janssen RAJ, Meijer EW, Herwig P, de Leeuw DM. Nature 1999, 401, 685–688.Google Scholar
Malliaras G, Friend R. Phys. Today 2005, 58, 53–58.Google Scholar
Jurchescu OD. Molecular organic semiconductors for electronic devices, Ph.D. thesis, Materials Science Centre, University of Groningen: The Netherlands, 2006.Google Scholar
Naber RCG, Tanase C, Blom PWM, Gelinck GH, Marsman AW, Touwslager FJ, Setayesh S, de Leeuw DM. Nat. Mater. 2005, 4, 243–248.Google Scholar
Tashiro K. In Ferroelectric Polymers Chemistry, Physics, and Applications, Nalwa HS, Ed., Marcel Dekker, Inc.: New York, 1955, pp. 63–181.Google Scholar
Bonfiglio A, De Rossi D, Kirstein T, Locher I, Mameli F, Paradiso R, Vozzi G. IEEE Trans. Inf. Tech. Biomed. 2005, 9, 319–324.Google Scholar
Rollable display. Available at http://phys.org/news6180.html.
Gundlach DJ, Jackson TN, Schlom DG, Nelson SF. Appl. Phys. Lett. 1999, 74, 3302–3304.Google Scholar
Meijer EJ, Gelinck GH, van Veenendaal E, Huisman B-H, de Leeuw DM, Klapwijk TM. Appl. Phys. Lett. 2003, 82, 4576–4578.Google Scholar
Kline RJ, McGehee MD, Kadnikova EN, Liu J, Fréchet JMJ. Adv. Mater. 2003, 15, 1519–1522.Google Scholar
Zhang Y, Petta JR, Ambily S, Shen Y, Ralph DC, Malliaras GG. Adv. Mater. 2003, 15, 1632–1635.Google Scholar
Podzorov V, Pudalov VM, Gershenson ME. Appl. Phys. Lett. 2003, 82, 1739–1741.Google Scholar
de Boer RWI, Klapwijk TM, Morpurgo AF. Appl. Phys. Lett. 2003, 83, 4345–4347.Google Scholar
Gelinck GH, Huitema HEA, van Veenendaal E, Cantatore E, Schrijnemakers L, van der Putten JBPH, Geuns TCT, Beenhakkers M, Giesbers JB, Huisman B-H, Meijer EJ, Benito EM, Touwslager FJ, Marsman AW, van Rens BJE, de Leeuw DM. Nat. Mater. 2004, 3, 106–110.CrossrefGoogle Scholar
Dimitrakopoulos CD, Malenfant PRL. Adv. Mater. 2002, 14, 99–117.Google Scholar
Bao Z, Rogers JA, Katz HE. J. Mater. Chem. 1999, 9, 1895–1904.Google Scholar
Horowitz G. J. Mater. Chem. 1999, 9, 2021–2026.Google Scholar
Mas-Torrent M, Rovira C. Chem. Soc. Rev. 2008, 37, 827–838.Google Scholar
Bao Z, Feng Y, Dodabalapur A, Raju VR, Lovinger AJ. Chem. Mater. 1997, 9, 1299–1301.Google Scholar
Bharathan J, Yang Y. Appl. Phys. Lett. 1998, 72, 2660–2662.Google Scholar
Chang S-C, Bharathan J, Yang Y, Helgeson R, Wudl F, Ramey MB, Reynolds JR. Appl. Phys. Lett. 1998, 73, 2561–2563.Google Scholar
Hebner TR, Sturm JC. Appl. Phys. Lett. 1998, 73, 1775–1777.Google Scholar
Hebner TR, Wu CC, Marcy D, Lu MH, Sturm JC. Appl. Phys. Lett. 1998, 72, 519–521.Google Scholar
Rogers JA, Bao Z, Raju RV. Appl. Phys. Lett. 1998, 72, 2716–2718.Google Scholar
Rogers JA, Bao Z, Makhija. Adv. Mater. 1999, 11, 741–745.Google Scholar
Rogers JA, Huang Y. Proc. Natl. Acad. Sci. 2009, 106, 10875–10876.Google Scholar
Sekitani T, Noguchi Y, Zschieschang U, Klauk H, Someya T. Proc. Natl. Acad. Sci. 2008, 105, 4976–4980.Google Scholar
Service RF. Science 1997, 383–384.Google Scholar
de Boer RWI, Gershenson ME, Morpurgo AF, Podzorov V. Phys. Stat. Solid. A 2004, 201, 1302–1331.Google Scholar
Reese C, Bao Z. Mater. Today 2007, 10, 20–27.Google Scholar
Hasegawa T, Takeya J. Sci. Technol. Adv. Mater. 2009, 10, 024314(1–16).Google Scholar
Jiang L, Dong H, Hu W. J. Mater. Chem. 2010, 20, 4994–5007.Google Scholar
Dimitrakopoulos CD, Mascaro DJ. IBM J. Res. Dev. 2001, 45, 11–28.Google Scholar
Warta W, Karl N. Phys. Rev. B 1985, 32, 1172–1182.Google Scholar
Sundar VC, Eisler H-J, Deng T, Chan Y, Thomas EL, Bawendi MG. Adv. Mater. 2004, 16, 2137–2141.Google Scholar
Burroughes JH, Bradley DDC, Brown AR, Marks RN, Mackay K, Friend RH, Burns PL, Holmes AB. Nature 1990, 347, 539–541.Google Scholar
Brabec CJ. Sol. Energy Mater. Sol. Cells 2004, 83, 273–292.Google Scholar
Brabec CJ, Sariciftci NS. In Semiconducting Polymers, Hadziiannou G, van Hutten PF, Eds., Wiley: New York, 2000, Ch. 15.Google Scholar
Gowrishankar V, Hardin BE, McGehee MD. Organic Photovoltaics: Materials, Device Physics and Manufacturing Technologies. Wiley-VCH Verlag GmbH & Co.: Weinheim, 2008, p 27.Google Scholar
Koster LJA, Mihailetchi VD, Lenes M, Blom PWM. In Organic Photovoltaics: Materials, Device Physics, and Manufacturing Technologies, Brabec CC, Dyakonov V, Scherf U, Eds., WILEY-VCH Verlag GmbH & Co.: KGaA, Weinheim, Copyright – 2008, ISBN: 978-3-527-31675-5.Google Scholar
Neuteboom EE, van Hal PA, Janssen RAJ. J. Chem. Eur. J. 2004, 10, 3907–3918.Google Scholar
Scerf U, Gutacker A, Koenen N. Acc. Chem. Res. 2008, 41, 1086–1097.Google Scholar
Schueppel R, Schmidt K, Uhrich C, Schulze K, Wynands D, Brédas JL, Baeuerle P, Maennig B, Pfeiffer M, Leo K. Phys. Rev. 2008, B77, 085311-24.Google Scholar
Sun SS. Sol. Energy Mater. Sol. Cells 2003, 79, 257–264.Google Scholar
Bao Z, Lovinger AJ, Dodabalapur A. Appl. Phys. Lett. 1996, 69, 3066–3068.Google Scholar
Horowitz G. Adv. Mater. 1998, 10, 365–377.Google Scholar
Chang J, Sun B, Breiby DW, Nielsen MM, Sölling TI, Giles M, McCulloch I, Sirringhaus H. Chem. Mater. 2004, 16, 4772–4776.Google Scholar
Sun Y, Liu Y, Zhu D. J. Mater. Chem. 2005, 15, 53–65.Google Scholar
Roncali J, Leriche P, Cravino A. Adv. Mater. 2007, 19, 2045–2060.Google Scholar
Cornil J, Beljonne D, Calbert J-P, Brédas J-L. Adv. Mater. 2001, 13, 1053–1067.Google Scholar
Brédas JL, Calbert JP, da Silva Filho DA, Cornil J. Proc. Natl. Acad. Sci. 2002, 99, 5804–5809.Google Scholar
Kazukauskas V, Tzeng H, Chen SA. Appl. Phys. Lett. 2002, 80, 2017–2019.Google Scholar
Smits ECP, Anthopoulos TD, Setayesh S, van Veenendaal E, Coehoorn R, Blom PWM, de Boer B, de Leeuw DM. Phys. Rev. B 2006, 73, 205316(1–9).Google Scholar
Inokuchi H. Org. Electron. 2006, 7, 62–76.Google Scholar
Akamatu H, Inokuchi H, Matsunaga Y. Natus 1954, 173, 168–169.Google Scholar
Tour JM, Wu R, Schumm JS. In Molecular Electronics – Science and Technology, Aviram A, Ed., American Institute of Physics: New York, 1992, pp. 77–84.Google Scholar
Yassar A, Delabouglise D, Hmyene M, Nessak B, Horowitz G, Garnier F. Adv. Mater. 1992, 4, 490–494.Google Scholar
Guay J, Kasai P, Diaz A, Wu R, Tour JM, Dao LH. Chem. Mater. 1992, 4, 1097–1105.Google Scholar
Bäuerle P, Fischer T, Bidlingmeier B, Rabe JP, Stabel A. Angew. Chem. Int. Ed. 1995, 34, 303–307.Google Scholar
Nakanishi H, Sumi N, Aso Y, Otsubo T. J. Org. Chem. 1998, 63, 8632–8633.Google Scholar
Noda T, Shirota Y. J. Am. Chem. Soc. 1998, 120, 9714–9715.Google Scholar
Noda T, Ogawa H, Shirota Y. Adv. Mater. 1999, 11, 283–285.Google Scholar
Kunugi Y, Takimiya K, Negishi N, Otsubo T, Aso Y. J. Mater. Chem. 2004, 14, 2840–2841.Google Scholar
Hotta S, Waragai K. J. Mater. Chem. 1991, 1, 835–842.Google Scholar
Bäuerle P, Götz G, Hiller M, Scheib S, Fischer T. Synth. Metals 1993, 61, 71–79.Google Scholar
Graf DD, Campbell JP, Miller LL, Mann KR. J. Am. Chem. Soc. 1996, 118, 5480–5481.Google Scholar
Hughes G, Bryce MR. J. Mater. Chem. 2005, 15, 94–107.Google Scholar
Shirota Y. J. Mater. Chem. 2005, 15, 75–93.Google Scholar
Okumoto K, Shirota Y. Chem. Mater. 2003, 15, 699–707.Google Scholar
Lupton JM, Samuel IDW, Beavington R, Frampton MJ, Burn PL, Bassler H. Phys. Rev. B 2001, 63, 155206(1–8).Google Scholar
Barron JA, Bernhard S, Houston PL, Abruña HD. Phys. Chem. A 2003, 107, 8130–8133.Google Scholar
Gambino S, Stevenson SG, Knights KA, Burn PL, Samuel IDW. Adv. Funct. Mater. 2009, 19, 317–323.Google Scholar
Dhanabalan A, van Duren JKJ, van Hal PA, van Dongen JLJ, Janssen RAJ. Adv. Funct. Mater. 2001, 11, 255–262.Google Scholar
Coakley KM, McGehee MD. Chem. Mater. 2004, 16, 4533–4542.Google Scholar
Colladet K, Fourier S, Cleij TJ, Lutsen L, Gelan J, Vanderzande D. Macromolecules 2007, 40, 65–72.Google Scholar
Demadrille R, Delbosc N, Kervella Y, Firon M, De Bettignies R, Billon M, Rannou P, Pron A. J. Mater. Chem. 2007, 17, 4661–4669.Google Scholar
Roncali J. Macromol. Rapid Commun. 2007, 28, 1761–1775.Google Scholar
Soci C, Hwang I, Moses D, Zhu Z, Waller D, Gaudiana R, Brabec CJ, Heeger AJ. Adv. Funct. Mater. 2007, 17, 632–636.Google Scholar
Zhu Z, Waller D, Gaudiana R, Morana M, Mühlbacher D, Scharber M, Brabec C. Macromolecules 2007, 40, 1981–1986.Google Scholar
Zhan X, Zhu D. Polym. Chem. 2010, 1, 409–419.Google Scholar
Shaheen SE, Brabec CJ, Sariciftci NS, Padinger F, Fromherz T, Hummelen JC. Appl. Phys. Lett. 2001, 78, 841–843.Google Scholar
Kobashi M, Takeuchi H. Macromolecules 1998, 31, 7273–7278.Google Scholar
Mozer AJ, Sariciftci NS, Pivrikas A, Österbacka R, Juška G, Brassat L, Bässler H. Phys. Rev. B 2005, 71, 035214(1–9).Google Scholar
Karl N. In Organic Electronic Materials, Farchioni R, Grosso G, Eds., Springer-Verlag: Berlin, 2001, pp. 283–326.Google Scholar
Hannewald K, Bobbert PA. Appl. Phys. Lett. 2004, 85, 1535–1537.Google Scholar
Hannewald K, Bobbert PA. AIP Conf. Proc. 2005, 772, 1101–1102.Google Scholar
Kiguchi M, Nakayama M, Shimada T, Saiki K. Phys. Rev. B 2005, 71, 035332(1–5).Google Scholar
de Boer RWI, Iosad NN, Stassen AF, Klapwijk TM, Morpurgo AF. Appl. Phys. Lett. 2005, 86, 032103.Google Scholar
Panzer MJ, Frisbie CD. Appl. Phys. Lett. 2006, 88, 203504.Google Scholar
Houili H, Picon JD, Zuppiroli L. Cond. Mater. 2005, 0510751. Available at http://arxiv.org/pdf/cond-mat/0510751v1.pdf.
Riordan M, Hoddeson L. Crystal Fire: The Birth of Information Age, W.W. Norton and Co.: New York, 1997.Google Scholar
Wu Y, Liu P, Li Y, Ong BS. In Lasers and Electro-Optics Society, 2006. LEOS 2006. Proc. 19th Annual Meeting of the IEEE, pp. 434–435.Google Scholar
Lim JA, Lee WH, Lee HS, Lee JH, Park YD, Cho K. Adv. Funct. Mater. 2008, 18, 229–234.Google Scholar
Wang Y-W, Cheng H-L. Solid-State Electron. 2009, 53, 1107–1111.Google Scholar
Huang J-H, Yang C-Y, Hsu C-Y, Chen C-L, Lin L-Y, Wang R-R, Ho K-C, Chu C-W. ACS Appl. Mater. Interf. 2009, 1, 2821–2828.Google Scholar
Choi KN, Kim KS, Chung KS, Hosun L. IEEE Trans. Dev. Mater. Reliab. 2009, 9, 489–493.Google Scholar
Klauk H, Halik M, Zschieschang U, Schmid G, Radlik W, Weber W. J. Appl. Phys. 2002, 92, 5259–5263.Google Scholar
Kang KS, Chen Y, Lim HK, Cho KY, Han KJ, Kim J. Thin Solid Films 2009, 517, 6096–6099.Google Scholar
Wu Y, Liu P, Ong BS, Srikumar T, Zao N, Botton G, Zhu S. Appl. Phys. Lett. 2005, 86, 142102.Google Scholar
Pingel P, Zen A, Neher D, Lieberwirth I, Wegner G, Allard S, Scherf U. Appl. Phys. A Mater. Sci. Process. 2009, 95, 67–72.Google Scholar
Ong BS, Wu Y, Liu P, Gardner S. J. Am. Chem. Soc. 2004, 126, 3378–3379.Google Scholar
Ong BS, Wu Y, Liu P, Gardner S. Adv. Mater. 2005, 17, 1141–1144.Google Scholar
Cahyadi T, Kasim J, Tan HS, Kulkarni SR, Ong BS, Wu Y, Chen Z-K, Ng CM, Shen Z-X, Mhaisalkar SG. Adv. Funct. Mater. 2009, 19, 378–385.Google Scholar
Liu P, Wu Y, Li Y, Ong BS, Zhu S. J. Am. Chem. Soc. 2006, 128, 4554–4555.Google Scholar
Wu Y, Li Y, Ong BS. J. Am. Chem. Soc. 2006, 128, 4202–4203.Google Scholar
Cosseddu P, Mattana G, Orgiu E, Bonfiglio A. Appl. Phys. A 2009, 95, 49–54.Google Scholar
Kato Y, Iba S, Teramoto R, Sekitani T, Someya T, Kawaguchi H, Sakurai T. Appl. Phys. Lett. 2004, 84, 3789–3791.Google Scholar
Parashkov R, Becker E, Ginev G, Riedl T, Johannes H-H, Kowalsky W. J. Appl. Phys. 2004, 95, 1594–1596.Google Scholar
Oku S, Nagase T, Nagamatsu S, Takashima W, Kaneto K. Jpn. J. Appl. Phys. 2010, 49, 01AB14(1–4).Google Scholar
Fukuda K, Sekitani T, Someya T. Appl. Phys. Lett. 2009, 95, 023302(1–3).Google Scholar
Yun HJ, Ham YH, Shin H-S, Jeong K-S, Park J-G, Choi D-S, Lee G-W. J. Nanosci. Nanotechnol. 2011, 11, 5640–5644.Google Scholar
Yun H-J, Baek K-H, Do L-M, Jeong K-S, Kim Y-M, Yang S-D, Lee S-Y, Lee H-D, Lee G-W. J. Nanosci. Nanotechnol. 2013, 13, 3313–3316.Google Scholar
Wu Y, Li Y, Ong BS. J. Am. Chem. Soc. 2007, 129, 1862–1863.Google Scholar
Li Y, Wu Y, Ong BS. J. Am. Chem. Soc. 2005, 127, 3266–3267.Google Scholar
Crispin X, Jakobsson FLE, Crispin A, Grim PCM, Andersson P, Volodin A, van Haesendonck C, Van der Auweraer M, Salaneck WR, Berggren M. Chem. Mater. 2006, 18, 4354–4360.Google Scholar
Sonar P, Lim JPF, Chan KL. Energy Environ. Sci. 2011, 4, 1558–1574.Google Scholar
Li Y, Sonar P, Murphy L, Hong W. Energy. Environ. Sci. 2013, 6, 1684–1710.Google Scholar
Liu S-W, Wen J-M, Lee C-C, Su W-C, Wang W-L, Chen H-C, Lin C-F. Thin Solid Films 2013, 534, 640–644.Google Scholar
Pasquier AD, Unalan HE, Kanwal A, Miller S, Chhowalla M. Appl. Phys. Lett. 2005, 87, 203511(1–3).Google Scholar
Bo XZ, Lee CY, Strano MS, Goldfinger M, Nuckolls C, Blanchet GB. Appl. Phys. Lett. 2005, 86, 182102–182103.Google Scholar
Liu S, Mannsfeld SCB, LeMieux MCH, Lee W, Bao Z. Appl. Phys. Lett. 2008, 92, 053306(1–3).Google Scholar
Geng Q, Guo Q, Cao C, Wang L. Ind. Eng. Chem. Res. 2008, 47, 2561–2568.Google Scholar
Hsieh G-W, Li FM, Beecher P, Nathan A, Wu Y, Ong BS, Milne WI. J. Appl. Phys. 2009, 106, 123706–123707.Google Scholar
Yeong DP, Lim JA, Kwak D, Cho JH, Cho K. Electrochem. Solid State Lett. 2009, 12, H312–H314.Google Scholar
Bahr JL, Tour JM. J. Mater. Chem. 2002, 12, 1952–1958.Google Scholar
Zhao C, Ji L, Liu H, Hu G, Zhang S, Yang M, Yang Z. J. Solid State Chem. 2004, 177, 4394–4398.Google Scholar
Oliveira MM, Zarbin AJG. Activity Report 2008. Available at http://lnls.cnpem.br/PDF/1084.pdf.
Stefopoulos AA, Chochos CL, Prato M, Pistolis G, Papagelis K, Petraki F, Kennou S, Kallitsis JK. Eur. J. 2008, 14, 8715–8724.Google Scholar
Villar-Rodil S, Paredes JI, Martínez-Alonso A, Tascón JMD. J. Mater. Chem. 2009, 19, 3591–3593.Google Scholar
Ansari S, Giannelis EP. J. Polym. Sci. B 2009, 47, 888–897.Google Scholar
Tkachenko LI, Efimov ON, Anoshkin IV, Kulova TL, Roshchupkina OS, Shul’ga YM, Petrova NK. Russ. J. Electrochem. 2009, 45, 296–303.Google Scholar
Song YJ, Lee JU, Jo WH. Carbon 2010, 48, 389–395.Google Scholar
Mickelson ET, Huffman CB, Rinzler AG, Smalley RE, Hauge RH, Margrave JL. Chem. Phys. Lett. 1998, 296, 188–194.Google Scholar
Kelly KF, Chiang IW, Mickelson ET, Hauge RH, Margrave JL, Wang X, Scuseria GE, Radloff C, Halas NJ. Chem. Phys. Lett. 1999, 313, 445–450.Google Scholar
Touhara H, Inahara J, Mizuno T, Yokoyama Y, Okanao S, Yanagiuch K, Mukopadhyay I, Kawasaki S, Okino F, Shirai H, Xu WH, Kyotani T, Tomita A. Fluorine Chem. 2002, 114, 181–188.Google Scholar
Stevens JL, Huang AAY, Peng H, Chiang IW, Khabashesku VN, Margrave JL. Nano. Lett. 2003, 3, 331–336.Google Scholar
Zhang Q, Chen D. J. Mater. Sci. 2004, 39, 1751–1757.Google Scholar
Hu H, Zhao B, Hamon MA, Kamaras K, Itkis ME, Haddon RC. J. Am. Chem. Soc. 2003, 125, 14893–14900.Google Scholar
Holzinger M, Steinmetz J, Samaille D, Glerup M, Paillet M, Bernier P, Ley L, Graupner R. Carbon 2004, 42, 941–947.Google Scholar
Unger E, Graham A, Kreupl F, Liebau M, Hoenlein W. Curr. Appl. Phys. 2002, 2, 107–111.Google Scholar
Kim KS, Bae DJ, Kim JR, Park KA, Lim SC, Kim J-J, Choi WB, Park CY, Lee YH. Adv. Mater. 2002, 14, 1818–1821.Google Scholar
Tagmatarchis N, Prato M. J. Mater. Chem. 2004, 14, 437–439.Google Scholar
Esumi K, Ishigami M, Nakajima A, Sawada K, Honda H. Carbon 1996, 34, 279–281.Google Scholar
Yu RQ, Chen L, Liu Q, Lin J, Tan K-L, Ng SC, Chan HSO, Xu G-Q, Hor TSA. Chem. Mater. 1998, 10, 718–722.Google Scholar
Sham ML, Kim JK. Carbon 2006, 44, 768–777.Google Scholar
Ma PC, Kim JK, Tang BZ. Carbon 2006, 44, 3232–3238.Google Scholar
Wang SC, Chang KS, Yuan CJ. Electrochim. Acta 2009, 54, 4937–4943.Google Scholar
Ávila-Orta CA, Cruz-Delgado VJ, Neira-Velázquez MG, Hernández-Hernández E, Méndez-Padilla MG, Medellín-Rodríguez FJ. Carbon 2009, 47, 1916–1921.Google Scholar
Liu P. Eur. Polym. J. 2005, 41, 2693–2703.Google Scholar
Hamon MA, Hui H, Bhowmik P, Itkis HME, Haddon RC. Appl. Phys. A 2002, 74, 333–338.Google Scholar
Liu J, Rinzler AG, Dai H, Hafner JH, Bradley RK, Boul PJ, Lu A, Iverson T, Shelimov K, Huffman CB, Rodriguez-Macias F, Shon Y-S, Lee TR, Colbert DT, Smalley RE. Science 1998, 280, 1253–1256.PubMedGoogle Scholar
Stephenson JJ, Sadana AK, Higginbotham AL, Tour JM. Chem. Mater. 2006, 18, 4658–4661.Google Scholar
McCarthy B, Coleman JN, Czerw R, Dalton AB, Carroll DL, Blau WJ. Synth. Metals 2001, 121, 1225–1226.Google Scholar
Hill DE, Lin Y, Rao AM, Allard LF, Sun Y-P. Macromolecules 2002, 35, 9466–9471.Google Scholar
Gong X, Liu J, Baskaran S, Voise RD, Young JS. Chem. Mater. 2000, 12, 1049–1052.Google Scholar
Cui S, Canet R, Derre A, Couzi M, Delhaes P. Carbon 2003, 41, 797–809.Google Scholar
Strano MS, Moore VC, Miller MK, Allen MJ, Haroz EH, Kittrell C, Hauge RH, Smalley RE. J. Nanosci. Nanotechnol. 2003, 3, 81–86.Google Scholar
Islam MF, Rojas E, Bergey DM, Johnson AT, Yodh AG. Nano. Lett. 2003, 3, 269–273.Google Scholar
Whitsitt EA, Barron AR. Nano. Lett. 2003, 3, 775–778.Google Scholar
Vaisman L, Wagner HD, Marom G. Adv. Colloid Interf. Sci. 2006, 128–130, 37–46.Google Scholar
Grossiord N, Loos J, Regev O, Koning CE. Chem. Mater. 2006, 18, 1089–1099.Google Scholar
Yu J, Grossiord N, Koning CE, Loos J. Carbon 2007, 45, 618–623.Google Scholar
Kim ST, Lim JY, Park BJ, Choi HJ. Macromol. Chem. Phys. 2007, 208, 514–519.Google Scholar
Georgakilas V, Gournis D, Tzitzios V, Pasquato L, Guldi DM, Prato M. J. Mater. Chem. 2007, 17, 2679–2694.Google Scholar
Tsang SC, Guo Z, Chen YK, Green MLH, Hill HAO, Hambley TW, Sadler PJ. Angew. Chem. Int. Ed. 1997, 36, 2198–2200.Google Scholar
Hirsch A. Angew. Chem. Int. Ed. 2002, 41, 1853–1859.Google Scholar
Li J, Ma PC, Chow WS, To CK, Tang BZ, Kim J-K. Adv. Funct. Mater. 2007, 17, 3207–3215.Google Scholar
Summary at http://www.scopus.com.
Du JH, Bai J, Cheng HM. Exp. Polym. Lett. 2007, 1, 253–273.Google Scholar
Moniruzzaman M, Winey KI. Macromolecules 2006, 39, 5194–5205.Google Scholar
Kim JY, Kim SH. J. Polym. Sci. B 2006, 44, 1062–1071.Google Scholar
Aneli JN, Khananasvili LM, Zaikov GE. Structuring and Conductivity of Polymer Composites. Nova Science Publishers: New York, 1998, pp. 1–100.Google Scholar
Sham ML, Li J, Ma PC, Kim JK. J. Compos. Mater. 2009, 43, 1537–1564.Google Scholar
Zhang W, Dehghani-Sanij AA, Blackburn RS. J. Mater. Sci. 2007, 42, 3408–3418.Google Scholar
Gul’ VE. Structure and Properties of Conducting Polymer Composites, VSP: Netherlands, 1996, pp. 27–123.Google Scholar
Sandler JKW, Kirk JE, Kinloch IA, Shaffer MSP, Windle AH. Polymer 2003, 44, 5893–5899.Google Scholar
Martin CA, Sandler JKW, Windle AH, Schwarz M-K, Bauhofer W, Schulte K, Shaffer MSP. Polymer 2005, 46, 877–886.Google Scholar
Moisala A, Li Q, Kinloch IA, Windle AH. Compos. Sci. Technol. 2006, 66, 1285–1288.Google Scholar
Ma PC, Kim J-K, Tang BZ. Compos. Sci. Technol. 2007, 67, 2965–2972.Google Scholar
Bauhofer W, Kovacs JZ. Compos. Sci. Technol. 2009, 69, 1486–1498.Google Scholar
Ma PC, Tang BZ, Kim J-K. Carbon 2008, 46, 1497–1505.Google Scholar
Buldum A, Lu JP. Phys. Rev. B 2001, 63, 161403-1-4.Google Scholar
Stadermann M, Papadakis SJ, Falvo MR, Novak J, Snow E, Fu Q, Liu J, Fridman Y. Phys. Rev. B 2004, 69, 201402-1-2-3.Google Scholar
Bokobza L, Rahmani M, Belin C, Bruneel J-L, El Bounia N-E. J. Polym. Sci. B 2008, 46, 1939–1951.Google Scholar
Li J, Wong PS, Kim JK. Mater. Sci. Eng. A 2008, 483–484, 660–663.Google Scholar
Kim J-K, Ma P-C. Carbon Nanotubes for Polymer Reinforcement, Boca Raton: CRC Press, 2011.Google Scholar
Sumfleth J, Adroher XC, Schulte K. J. Mater. Sci. 2009, 44, 3241–3247.Google Scholar
Sun Y, Bao H-D, Guo Z-X, Yu J. Macromolecules 2009, 42, 459–463.Google Scholar
Ma PC, Liu M-Y, Zhang H, Wang S-Q, Wang R, Wang K, Wong Y-K, Tang B-Z, Hong S-H, Paik K-W, Kim J-K. ACS Appl. Mater. Interf. 2009, 1, 1090–1096.Google Scholar
Jin Z, Lomeda JR, Price BK, Lu W, Zhu Y, Tour JM. Chem. Mater. 2009, 21, 3045–3047.Google Scholar
Lomeda JR, Doyle CD, Kosynkin DV, Hwang W-F, Tour JM. J. Am. Chem. Soc. 2008, 130, 16201–16206.Google Scholar
Bekyarova E, Itkis ME, Ramesh P, Berger C, Sprinkle M, de Heer WA, Haddon RC. J. Am. Chem. Soc. 2009, 131, 1336–1337.Google Scholar
Hossain MZ, Walsh MA, Hersam MC. J. Am. Chem. Soc. 2010, 132, 15399–15403.Google Scholar
Liu H, Ryu S, Chen Z, Steigerwald ML, Nuckolls C, Brus LE. J. Am. Chem. Soc. 2009, 131, 17099–17101.Google Scholar
Georgakilas V, Bourlinos A, Gournis D, Tsoufis T, Trapalis C, Mateo-Alonso A, Prato M. J. Am. Chem. Soc. 2008, 130, 8733–8740.Google Scholar
Georgakilas V, Guldi DM, Signorini R, Bozio R, Prato M. J. Am. Chem. Soc. 2003, 125, 14268–14269.Google Scholar
Kordatos K, Bosi S, Da Ros T, Bosi S, Vázquez E, Bergamin M, Cusan C, Pellarini F, Tomberli V, Baiti B, Pantarotto D, Georgakilas V, Spalluto G, Prato M. J. Org. Chem. 2001, 66, 4915–4920.Google Scholar
Cioffi C, Campidelli S, Brunetti FG, Meneghetti M, Prato M. Chem. Commun. 2006, 2129–2131.Google Scholar
Singh R, Pantarotto D, Lacerda L, Pastorin G, Klumpp C, Prato M, Bianco A, Kostarelos K. Proc. Natl. Acad. Sci. USA 2006, 103, 3357–3362.Google Scholar
Georgakilas V, Bourlinos AB, Zboril R, Steriotis TA, Dallas P, Stubos AK, Trapalis C. Chem. Commun. 2010, 46, 1766–1768.Google Scholar
Zhang X, Hou L, Cnossen A, Coleman AC, Ivashenko O, Rudolf P, van Wees BJ, Browne WR, Feringa BL. Chem. Eur. J. 2011, 17, 8957–8964.Google Scholar
He H, Gao C. Chem. Mater. 2010, 22, 5054–5064.Google Scholar
Zhong X, Jin J, Li S, Niu Z, Hu W, Li R, Ma J. Chem. Commun. 2010, 46, 7340–7342.Google Scholar
Chen T, Pan G-B, Yan H-J, Wan L-J, Matsuo Y, Nakamura E. J. Phys. Chem. C 2010, 114, 3170–3174.Google Scholar
Hummers WS, Offeman RE. J. Am. Chem. Soc. 1958, 80, 1339.Google Scholar
Huh JW, Moon J, Lee JW, Lee J, Cho D-H, Shin J-W, Han J-H, Hwang J, Joo CW, Lee J-I, Chu HY. Org. Electron. 2013, 14, 2039–2045.Google Scholar
Stankovich S, Dikin DA, Piner RD, Kohlhaas KA, Kleinhammes A, Jia Y, Wu Y, Nguyen SBT, Ruoff RS. Carbon 2007, 4, 1558–1565.Google Scholar
Stankovich S, Piner RD, Chen X, Wu N, Nguyen SBT, Ruoff RS. J. Mater. Chem. 2006, 16, 155–158.Google Scholar
Bourlinos AB, Gournis D, Petridis D, Szabó T, Szeri A, Dékány I. Langmuir 2003, 19, 6050–6055.Google Scholar
McAllister MJ, Li J-L, Adamson DH, Schniepp HC, Abdala AA, Liu J, Herrera-Alonso M, Milius DL, Car R, Prud’homme RK, Aksay IA. Chem. Mater. 2007, 19, 4396–4404.Google Scholar
Schniepp HC, Li J-L, McAllister MJ, Sai H, Herrera-Alonso M, Adamson DH, Prud’homme RK, Car R, Saville DA, Aksay IA. J. Phys. Chem. B 2006, 110, 8535–8539.Google Scholar
Niyogi S, Bekyarova E, Itkis ME, McWilliams JL, Hamon MA, Haddon RC. J. Am. Chem. Soc. 2006, 128, 7720–7721.Google Scholar
Su Q, Pang S, Alijani V, Li C, Feng X, Müllen K. Adv. Mater. 2009, 21, 3191–3195.Google Scholar
Liang Y, Wu Y, Feng D, Tsai S-T, Son H-J, Li G, Yu L. J. Am. Chem. Soc. 2009, 131, 56–57.Google Scholar
Chen GH, Weng WG, Wu D, Wu C. Eur. Polym. J. 2003, 39, 2329–2335.Google Scholar
Verdejo R, Barroso-Bujans F, Rodriguez-Perez MA, de Saja JA, Lopez-Manchado MA. J. Mater. Chem. 2008, 18, 2221–2226.Google Scholar
Patil AJ, Vickery JL, Scott TB, Mann S. Adv. Mater. 2009, 21, 3159–3164.Google Scholar
Moon IK, Lee J, Ruoff RS, Lee H. Nature Comm. 2010, 1, Art No: 73.Google Scholar
Xu Y, Liu Z, Zhang X, Wang Y, Tian J, Huang Y, Ma Y, Zhang X, Chen Y. Adv. Mater. 2009, 21, 1275–1279.Google Scholar
Karousis N, Sandanayaka ASD, Hasobe T, Economopoulos SP, Sarantopoulou E, Tagmatarchis N. J. Mater. Chem. 2011, 21, 109–117.Google Scholar
Zhang X, Feng Y, Huang D, Li Y, Feng W. Carbon 2010, 48, 3236–3241.Google Scholar
Karl N. Landolt-Börnstein, Group III, Semiconductors, Organic Semiconductors, Vol. 17, Madelung O, Ed., Springer, Berlin, 1985, pp. 106–218.Google Scholar
Silinsh EA. Organic Molecular Crystals – Their Electronic States, Springer Series in Solid State Sciences, Vol 16, Springer-Verlag: Berlin/Heidelberg /New York, 1980.Google Scholar
Pope S. Electronic Processes in Organic Crystals and Polymers, 2nd ed., Oxford University Press, 1999, pp. 337–340.Google Scholar
Skotheim TA, Elsembaumer RL, Reynolds JR, Eds. Handbook of Conducting Polymers, Marcel Dekker: New York, 1998.Google Scholar
Borsenberger PM, Weiss DS. Organic Photoreceptors for Imaging Systems, Marcel Dekker: New York, 1993.Google Scholar
Tang CW. Appl. Phys. Lett. 1986, 48, 183–185.Google Scholar
Koezuka H, Tsumura A, Ando T. Synth. Metals 1987, 270, 18, 699–704.Google Scholar
Burroughes JH, Jones CA, Friend RH. Nature 1988, 335, 137–141.Google Scholar
Horowitz G, Fichou D, Peng X, Xu Z, Garnier F. Solid State Commun. 1989, 72, 381.Google Scholar
Tang CW, VanSlyke SA. Appl. Phys. Lett. 1987, 51, 913–915.Google Scholar
Burrows PE, Forrest SR, Sapochak LS, Schwartz J, Fenter P, Buma T, Ban VS, Forrest JL. J. Cryst. Growth 1995, 156, 91–98.Google Scholar
Kloc Ch, Simpkins PG, Siegrist T, Laudise RA. J. Cryst. Growth 1997, 182, 416–427.Google Scholar
Baldo M, Deutsch M, Burrows P, Gossenberger H, Gerstenberg M, Ban V, Forrest S. Adv. Mater. 1998, 10, 1505–1514.Google Scholar
Laudise RA, Kloc C, Simpkins PG, Siegrist T. J. Cryst. Growth 1998, 187, 449–454.Google Scholar
Ostoja P, Guerri S, Rossini S, Servidori M, Taliani C, Zamboni R. Synth. Metals 1993, 54, 447–452.Google Scholar
Nelson SF, Lin Y-Y, Grundlach DJ, Jackson TN. Appl. Phys. Lett. 1998, 72, 1854.Google Scholar
Lin YY, Gundlach DI, Nelson SF, Jackson TN. IEEE Trans. Electron Devices 1997, 44, 1325–1331.Google Scholar
Karl N, Marktanner J, Stehle R, Warta W. Synth. Met. 1991, 41–43, 2473–2481.Google Scholar
Schön JH, Batlogg B. J. Appl. Phys. 2001, 89, 336–342.Google Scholar
Burland DM. Phys. Rev. Lett. 1974, 33, 833–835.Google Scholar
Silinsh EA, Capek V. Organic Molecular Crystals: Interaction, Localization, and Transport Phenomena, AIP: New York, 1994.Google Scholar
Martel R, Schmidt T, Shea HR, Hertel T. Appl. Phys. Lett. 1998, 73, 2447–2449.Google Scholar
Dimitrakopoulos CD, Furman BK, Graham T, Hegde S, Purushothaman S. Synth. Metals 1998, 92, 47–52.Google Scholar
Malenfant PRL, Dimitrakopoulos CD, Gelorme JD, Kosbar LL, Graham TO, Curioni A, Andreoni W. Appl. Phys. Lett. 2002, 80, 2517–2519.Google Scholar
Dimitrakopoulos CD, Brown AR, Pomp A. J. Appl. Phys. 1996, 80, 2501–2508.Google Scholar
Jentzsch T, Juepner HJ, Brzezinka K-W, Lau A. Thin Solid Films 1998, 315, 273–280.Google Scholar
Minakata T, Ozaki M, Imai H. J. Appl. Phys. 1992, 72, 4178–4182.Google Scholar
Minakata T, Ozaki M, Imai H. J. Appl. Phys. 1993, 74, 1079–1082.Google Scholar
Horowitz G, Peng X, Fichou D, Garnier F. Synth. Metals 1992, 51, 419–424.Google Scholar
Garnier F, Assar A, Hajiloui R, Horowitz G, Deloffre F, Servet B, Ries S, Alnot P. J. Am. Chem. Soc. 1993, 115, 8716–8721.Google Scholar
Lin Y-Y, Gundlach DJ, Jackson TN. 54th Annual Device Research Conference Digest, 1996, 80–81.Google Scholar
Campbell RB, Robertson JM, Trotter J. Acta Cryst. 1961, 14, 705–711.Google Scholar
Kosbar LL, Dimitrakopoulos CD, Mascaro DJ. Mater. Res. Soc. Symp. Proc. 2001, 665, C10.6.1.Google Scholar
Heringdorf F-J Mz, Reuter MC, Tromp RM. Nature 2001, 412, 517–520.Google Scholar
Horowitz G. J. Mater. Res. 2004, 19, 1946–1962.Google Scholar
Takeya J, Goldmann C, Haas S, Pernstich KP, Ketterer B, Batlogg B. J. Appl. Phys. 2003, 94, 5800–5804.Google Scholar
Butko VY, Chi X, Lang DV, Ramirez AP. Appl. Phys. Lett. 2003, 83, 4773–4775.Google Scholar
Goldmann C, Haas S, Krellner C, Pernstich KP, Gundlach DJ, Batlogg B. J. Appl. Phys. 2004, 96, 2080–2086.Google Scholar
Aleshin AN, Lee JY, Chu SW, Kim JS, Park YW. Appl. Phys. Lett. 2004, 84, 5383–5385.Google Scholar
Podzorov V, Menard E, Borissov A, Kiryukhin V, Rogers JA, Gershenson ME. Phys. Rev. Lett. 2 004, 93, 086602(1–4).Google Scholar
Menard E, Podzorov V, Hur S-H, Gaur A, Gershenson ME, Rogers JA. Adv. Mater. 2004, 16, 2097–2101.Google Scholar
Ahmad S. Microwave and Millimeter Wave Semiconductor Materials Technology. Tata-McGraw Hill Publishing Co., Ltd.: New Delhi, 1998.Google Scholar
Jiang H, Kloc C. MRS Bull. 2012, 38.Google Scholar
Ishiguro T, Yamaji K, Saito G. Organic Superconductors, 2nd ed., Springer-Verlag: Berlin, 1998.Google Scholar
Kulkarni AP, Tonzola CJ, Babel A, Jenekhe SA. Chem. Mater. 2004, 16, 4556–4573.Google Scholar
Hoppe H, Sariciftci NS. J. Mater. Res. 2004, 19, 1924–1945.Google Scholar
Gershenson ME, Podzorov V, Morpurgo AF. Rev. Mod. Phys. 2006, 78, 973.Google Scholar
Yang X, Wang L, Wang C, Long W, Shuai Z. Chem. Mater. 2008, 20, 3205–3211.Google Scholar
Jiang H, Yang X, Cui Z, et al. Appl. Phys. Lett. 2009, 94, 123308(1–3).Google Scholar
Ortmann F, Bechstedt F, Hannewald K. Phys. Stat. Solid. B 2011, 248, 511–525.Google Scholar
Mas-Torrent M, Durkut M, Hadley P, Ribas X, Rovira C. J. Am. Chem. Soc. 2004, 126, 984–985.Google Scholar
Briseno AL, Tseng RJ, Ling M-M, Falcao EHL, Yang Y, Wudl F, Bao Z. Adv. Mater. 2006, 18, 2320–2324.Google Scholar
Jiang H, Yang X, Cui Z, Liu Y, Li H, Hu W, Liu Y, Zhu D. Appl. Phys. Lett. 2007, 91, 123505.Google Scholar
Yamao T, Miki T, Akagami H, Nishimoto Y, Ota S, Hotta S. Chem. Mater. 2007, 19, 3748–3753.Google Scholar
Jiang L, Fu Y, Li H, Hu W. J. Am. Chem. Soc. 2008, 130, 3937–3941.Google Scholar
Pfattner R, Mas-Torrent M, Bilotti I, Brillante A, Milita S, Liscio F, Biscarini F, Marszalek T, Ulanski J, Nosal A, Gazicki-Lipman M, Leufgen M, Schmidt G, Molenkamp LW, Laukhin V, Veciana J, Rovira C. Adv. Mater. 2010, 22, 4198–4203.PubMedGoogle Scholar
Jiang H, Tan KJ, Zhang KK, Chen X, Klo C. J. Mater. Chem. 2011, 21, 4771–4773.Google Scholar
Selvakumar S, Sivajia K, Arulchakkaravarthi A, Balamurugan N, Sankar S, Ramasamy P. J. Cryst. Growth 2005, 282, 370–375.Google Scholar
Probst KH, Karl N. Phys. Stat. Solid. A 1975, 27, 499–508.Google Scholar
McArdle BJ, Sherwood JN, Damask AC. J. Cryst. Growth 1974, 22, 193–200.Google Scholar
Inokuchi H. Bull. Chem. Soc. Jpn. 1956, 29, 131–133.Google Scholar
Niemax J, Pflaum. J. Appl. Phys. Lett. 2005, 87, 241921.Google Scholar
Arulchakkaravarthi A, Santhanaraghavan P, Ramasamy P. J. Cryst. Growth 2001, 224, 89–94.Google Scholar
Bleay J, Hooper RM, Narang RS, Sherwood JN. J. Cryst. Growth 1978, 43, 589–596.Google Scholar
Arivanandhan M, Natarajan V, Sankaranarayanan K, Hayakawa Y. Direction Controlled Growth of Organic Single Crystals by Novel Growth Methods, Material Science, Advanced Topics on Crystal Growth, Olavo Ferreira, S, Ed., ISBN 978-953-51-1010-1.Google Scholar
Bridgman PW. Proc. Am. Acad. Arts Sci. 1925, 60, 305–383.Google Scholar
Brissaud M, Dolin C, Le Duigou J. J. Cryst. Growth 1977, 38, 134–138.Google Scholar
Tripathi AK, Heinrich M, Siegrist T, Pflaum J. Adv. Mater. 2007, 19, 2097–101.Google Scholar
Karl N. Crystals Growth, Properties, and Applications, 1st ed., Springer-Verlag: Berlin/Heidelberg.Google Scholar
Hong IH, Tan KJ, Toh M, Jiang H, Zhang K, Kloc C. J. Cryst. Growth 2013, 363, 61–68.Google Scholar
Kim Y, Cook S, Tuladhar SM, Choulis SA, Nelson J, Durrant JR, Bradley DDC, Giles M, McCulloch I, Ha C-S, Ree M. Nat. Mater. 2006, 5, 197–203.Google Scholar
Matsukawa T, Yoshimura M, Sasai K, Uchiyama M, Yamagishi M, Tominari Y, Takahashi Y, Takeya J, Kitaoka Y, Mori Y, Sasaki T. J. Cryst. Growth 2010, 312, 310–313.Google Scholar
Johannsen I, Groth-Andersen L, Nielsen KF. J. Cryst. Growth 1981, 51, 627–628.Google Scholar
Piper WW, Polich SJ. J. Appl. Phys. 1961, 32, 1278–1279.Google Scholar
Park S-W, Jeong SH, Choi J-M, Hwang JM, Kim JH, Im S. Appl. Phys. Lett. 2007, 91, 033506(1–3).Google Scholar
Dhanaraj G, Byrappa K, Prasad V, Dudley M, Eds., Springer Handbook of Crystal Growth, 1st ed., Springer-Verlag: Berlin, Germany, 2010.Google Scholar
Feigelson RS, Route RK, Kao T-M. J. Cryst. Growth 1985, 72, 585–594.Google Scholar
Tickle IJ, Prout CK. J. Chem. Soc., Perkin Trans. 2, 1973, 727–731.Google Scholar
Truong B. Chem. Phys. Lett. 1976, 44, 232–235.Google Scholar
Tang Q, Li H, He M, Hu W, Liu C, Chen K, Wang C, Liu Y, Zhu D. Adv. Mater. 2006, 18, 65–68.Google Scholar
Moon H, Zeis R, Borkent E-J, Besnard C, Lovinger AJ, Siegrist T, Kloc C, Bao Z. J. Am. Chem. Soc. 2004, 126, 15322–15323.Google Scholar
Reese C, Chung W-J, Ling M-M, Roberts M, Bao Z. Appl. Phys. Lett. 2006, 89, 202108.Google Scholar
Mannsfeld SCB, Locklin J, Reese C, Roberts ME, Lovinger AJ, Bao Z. Adv. Funct. Mater. 2007, 17, 1617, 3545–3553.Google Scholar
Briseno AL, Mannsfeld SCB, Formo E, Xiong Y, Lu X, Bao Z, Jenekhe SA, Xia Y. J. Mater. Chem. 2008, 18, 5395–5398.Google Scholar
Choi HY, Kim SH, Jang J. Adv. Mater. 2004, 16, 732–736.Google Scholar
Steudel S, Janssen D, Verlaak S, Genoe J, Heremans P. Appl. Phys. Lett. 2004, 85, 5550–5552.Google Scholar
Kelley TW, Baude PF, Gerlach C, Ender DE, Muyres D, Haase MA, Vogel DE, Theiss SD. Chem. Mater. 2004, 16, 4413–4422.Google Scholar
Klauk H, Gundlach DJ, Nichols JA, Jackson TN. IEEE Trans. 1999, ED-46, 1258–1263.Google Scholar
Liu S, Mannsfeld SCB, Briseno AL, You W, Locklin J, Lee HW, Xia Y, Bao Z. Adv. Funct. Mater. 2007, 17, 2891–2896.Google Scholar
Xiao K, Tao J, Puretzky AA, Ivanov IN, Liu Z, Geohegan DB. Adv. Mater. 2006, 18, 2184–2188.Google Scholar
Xiao K, Tao J, Pan Z, Puretzky AA, Ivanov IN, Pennycook SJ, Geohegan DB. Angew. Chem. Int. Ed. 2007, 46, 2650–2654.Google Scholar
Aizenberg J, Black AJ, Whitesides GM. J. Am. Chem. Soc. 1999, 121, 4500–4509.Google Scholar
Chabinyc ML, Wong WS, Salleo A, Paul KE, Street RA. Appl. Phys. Lett. 2002, 81, 4260–4262.Google Scholar
Lee KS, Blanchet GB, Gao F, Loo Y-L. Appl. Phys. Lett. 2005, 86, 074102.Google Scholar
Chandekar A, Whitten JE. Appl. Phys. Lett. 2007, 91, 113103.Google Scholar
Briseno AL, Aizenberg J, Han Y-J, Penkala RA, Moon H, Lovinger AJ, Kloc C, Bao Z. J. Am. Chem. Soc. 2005, 127, 12164–12165.Google Scholar
Mannsfeld SCB, Sharei A, Liu S, Roberts ME, McCulloch I, Heeney M, Bao Z. Adv. Mater. 2008, 20, 4044–4048.Google Scholar
Herman MA, Sitter H. Molecular Beam Epitaxy Fundamentals and Current Status, 1st ed., Springer-Verlag: Berlin, 1988.Google Scholar
O. Wada, Ed., Optoelectronic Integration -Physics, Technology and Applications, Kluwer Academic Publishers, Boston, 1994.Google Scholar
Fetterman HR, Ni DC. Microwave Opt. Technol. Lett. 1988, 1, 34–39.Google Scholar
Chao CP, Shiau G-J, Forrest SR. IEEE Photon Tech. Lett. 1994, 6, 1406–1408.Google Scholar
Sasaki T, Yamaguchi M, Kitamura M. J. Cryst. Growth 1994, 145, 846–851.Google Scholar
Shiau G-J, Chao C-P, Burrows PE, Forrest SR. Appl. Phys. Lett. 1994, 65, 892–894.Google Scholar
Coldren LA, Corzine SW. Diode Lasers and Photonic Integrated Circuits. John Wiley: New York, 1995.Google Scholar
Chemla DS, Zyss J. Nonlinear Optical Properties of Organic Molecules and Crystals. Academic Press: Orlando/New York, 1987, Vol. 1–2.Google Scholar
Hann RA, Bloor D. Organic Materials for Non-linear Optics II, Royal Society of Chemistry: Cambridge, 1991.Google Scholar
Mukamel S. Principles of Nonlinear Optical Spectroscopy, Oxford University Press: New York, 1995.Google Scholar
Maruyama Y, Hoshi H, Fang SL, Kohama K. Synth. Metals 1995, 71, 1653–1656.Google Scholar
Seto J, Asai N, Fujiwara I, Ishibashi T, Kamei T, Tamura S. Thin Solid Films 1996, 273, 97–104.Google Scholar
Simon J, André J-J. Molecular Semiconductors: Photoelectrical Properties and Solar Cells. Springer-Verlag: Berlin, 1985.Google Scholar
Leznoff CC, Lever ABP. Phthalocyanines: Properties and Applications, VCH Publishers, Inc.: New York, 1989.Google Scholar
Wohrle D, Meissne D. Adv. Mater. 1991, 3, 129–138.Google Scholar
Taylor RB, Burrows PE, Forrest SR. IEEE Photon. Tech. Lett. 1997, 9, 365–367.Google Scholar
So FF, Forrest SR, Shi YQ, Steier WH. Appl. Phys. Lett. 1990, 56, 674–676.Google Scholar
Agranovich VM. Phys. Scr. 1993, 149, 699.Google Scholar
Agranovich VM, La Rocca GC, Bassani F. Chem. Phys. Lett. 1995, 247, 355–358.Google Scholar
Wang N, Jenkins JK, Chernyak VV, Mukamel S. Phys. Rev. B 1994, 49, 17079–17091.Google Scholar
Zakhidov AA, Akashi T, Yoshino K. Synth. Metals 1994, 64, 155–165.Google Scholar
Nakada H, Tohma T. In Proc. 8th International Workshop on Electroluminescence. Berlin, 1996, p 385.Google Scholar
Burrows PE, Forrest SR. Appl. Phys. Lett. 1994, 64, 2285–2287.Google Scholar
Fang S, Kohama K, Hoshi H, Maruyama Y. Synth. Metals 1994, 64, 167–170.Google Scholar
Arbour C, Armstrong NR, Brina R, Collins G, Danziger J-P, Lee P, Nebesney KW, Pankow J, Waite S. Molecular Crystals and Liquid Crystals Incorporating Nonlinear Optics 1990, 183, 307–320.Google Scholar
Forrest SR, Kaplan ML, Schmidt PH, Feldmann WL, Yanowski E. Appl. Phys. Lett. 1982, 41, 90–92.Google Scholar
Lovinger AJ, Forrest SR, Kaplan ML, Schmidt PH, Venkatesan T. J. Appl. Phys. 1984, 55, 476–482.Google Scholar
Forrest SR, Kaplan ML, Schmidt PH. J. Appl. Phys. 1984, 55, 1492–1507.Google Scholar
Akers K, Aroca R, Hor AM, Loutfy RO. J. Phys. Chem. 1987, 91, 2954–2959.Google Scholar
Zimmermann U, Schnitzler G, Kar N, Umbach E. Thin Solid Films 1989, 175, 85–88.Google Scholar
Danziger J, Dodeiet J-P, Lee P, Nebesny KW, Armstrong NR. Chem. Mater. 1991, 3, 821–829.Google Scholar
Ludwig C, Gompf B, Glatz W, Petersen J, Eisenmenger W, Möbus M, Zimmermann U, Karl N. Z. Phys. B 1992, 86, 397–404.Google Scholar
Hoshino A, Isoda S, Kurata H, Kobayashi T. J. Appl. Phys. 1994, 76, 4113–4120.Google Scholar
Vincett PS, Popovic ZD, McIntyre L. Thin Solid Films 1981, 82, 357–376.Google Scholar
Debe MK, Kam KK, Liu JC, Poirier RJ. J. Vac. Sci. Technol. A Vac. Surf. Films 1988, 6, 1907–1911.Google Scholar
Mobus M, Karl N, Kobayashi T. J. Cryst. Growth 1992, 116, 495–504.Google Scholar
Tanigaki K, Kuroshima S, Ebbesen TW, Ichihashi T. J. Cryst. Growth 1991, 114, 3–6.Google Scholar
Yoshimura T, Tatsuura S, Sotoyama W. Appl. Phys. Lett. 1991, 59, 482–484.Google Scholar
Haskal EI, So FF, Burrows PE, Forrest SR. Appl. Phys. Lett. 1992, 60, 3223–3225.Google Scholar
Tada H, Kawaguchi T, Koma A. Appl. Phys. Lett. 1992, 61, 2021–2023.Google Scholar
Ludwig C, Gompf B, Petersen J, Strohmaier R, Eisenmenger W. Z Phys. B 1994, 93, 365–373.Google Scholar
Kendrick C, Kahn A, Forrest SR, Appl. Surf. Sci. 1996, 104/105, 586–594.Google Scholar
Yanagi H, Araki Y, Ohara T, Hotta S, Ichikawa M, Taniguchi Y. Adv. Funct. Mater. 2003, 13, 767–773.Google Scholar
Andreev A, Resel R, Smilgies D-M, Hoppe H, Matt G, Sitter H, Sariciftci NS, Meissner D, Plank H, Zrzavecka O. Synth. Metals 2003, 138, 59–63.Google Scholar
Kambayashi T, Ohta H, Hoshi H, Hirano M, Hosono H, Takezoe H, Ishikawa K. Cryst. Growth Des. 2005, 5, 143–146.Google Scholar
Koller G, Berkebile S, Krenn J, Tzvetkov G, Hlawacek G, Lengye O, Netzer FP, Teichert C, Rese R, Ramsey MG. Adv. Mater. 2004, 16, 2159–2162.Google Scholar
Ivanco J, Haber T, Krenn JR, Netzer FP, Resel R, Ramsey MG. Surf. Sci. 2007, 601, 178–187.Google Scholar
Sellam F, Schmitz-Hubsch T, Toerker M, Mannsfeld S, Proehl H, Fritz T, Leo K, Simpson C, Müllen K. Surf. Sci. 2001, 478, 113–121.Google Scholar
Haber T, Oehzelt M, Resel R, Andreev A, Thierry A, Sitter H, Smilgies D-M, Schaffer B, Grogger W, J. Nanosci. Nanotech. 2006, 6, 698–703.Google Scholar
Dienel T, Loppacher C, Mannsfeld SCB, Forker R, Fritz T. Adv. Mater. 2008, 20, 959–963.Google Scholar
Toda Y. Appl. Phys. Lett. 1996, 69, 2315–2317.Google Scholar
Yanagi H, Ohara T, Morikawa T. Adv. Mater. 2001, 13, 1452–1455.Google Scholar
Ichikawa M, Yanagi H, Shimizu Y, Hotta S, Suganuma N, Koyama T, Taniguchi Y. Adv. Mater. 2002, 14, 1272–1275.Google Scholar
Noh YY, Kim JJ, Yoshida Y, Yase K. Adv. Mater. 2003, 15, 699–702.Google Scholar
Noh YY, Kim J-J, Nagamatsu S. Appl. Phys. Lett. 2003, 83, 1243–1245.Google Scholar
Wang H, Zhu F, Yang J, Geng Y, Yan D. Adv. Mater. 2007, 19, 2168–2171.Google Scholar
Schreiber F. Phys. Status Solid. A 2004, 201, 1037–1054.Google Scholar
Möbus M, Karl N, Kobayashi T. J. Cryst. Growth 1992, 116, 495–504.Google Scholar
Umbach E, Fink R, Sokolowski M. Appl. Phys. A 1996, 63, 565–576.Google Scholar
Schmitz-Hübsch T, Fritz T, Sellam F, Staub R, Leo K. Phys Rev B 1997, 55, 7972–7976.Google Scholar
Fenter P, Schreiber F, Zhou P, Eisenberger, Forrest SR. Phys. Rev. B 1997, 56, 3046–3053.Google Scholar
Seidel C, Awater C, Liu XD, Ellerbrake R, Fuchs H. Surf. Sci. 1997, 371, 123–130.Google Scholar
Glöckler K, Seidel C, Soukopp A, Sokolowski M, Umbach E, Böhringer M, Berndt R, Schneider WD. Surf. Sci. 1998, 405, 1–20.Google Scholar
Krause B, Dürr AC, Ritely KA, Schreiber F, Dosch H, Smilgies D. Appl. Surf. Sci. 2001, 175, 332–336.Google Scholar
Krause B, Dürr AC, Ritley R, Schreiber F, Dosch H, Smilgies D. Phys. Rev. B, 2002, 66, 235404.Google Scholar
Krause B, Dürr AC, Schreiber F, Dosch H. Chem. Phys. 2003, 119, 3429–3435.Google Scholar
Möbus M, Karl N. Thin Solid Films 1992, 215, 213–217.Google Scholar
Chizhov I, Kahn A, Scoles G. J. Cryst. Growth 2000, 208, 449–458.Google Scholar
Dürr AC, Schreiber F, Monch M, Karl N, Krause B, Kruppa V, Dosch H. Appl. Phys. Lett. 2002, 81, 2276–2278.Google Scholar
Strohmaier R. Ph.D. thesis, Stuttgart, 1997.Google Scholar
Rogers JA, Bao Z, Dodabalapur A, Makhija A. IEEE EDL-21 2000, 100–103.Google Scholar
Schlettwein D. Supramolecular Photosensitive and Electroactive Materials, Nalwa HS, Ed., Academic Press: San Diego, 2001.Google Scholar
Yim S, Jones TS. Surf. Sci. 2002, 521, 151–159.Google Scholar
Alonso MI, Garriga M, Ossó JO, Schreiber F, Barrena E, Dosch H. J. Chem. Phys. 2003, 119, 6335–6340.Google Scholar
Bao Z, Lovinger A, Brown J. J. Am. Chem. Soc. 1998, 120, 207–208.Google Scholar
Meyer zu Heringdorf FJ, Reuter MC, Tromp RM. Nature 2001, 412, 517–520.Google Scholar
Shtein M, Mapel J, Benziger JB, Forrest SR. Appl. Phys. Lett. 2002, 81, 268–270.Google Scholar
Liu C, Minari T, Li Y, Kumatani A, Lee MV, Pan SHA, Takimiya K, Tsukagoshi K. J. Mater. Chem. 2012, 22, 8462–8469.Google Scholar
Ebata H, Izawa T, Miyazaki E, Takimiya K, Ikeda M, Kuwabara H, Yui T. J. Am. Chem. Soc. 2007, 129, 15732–15733.Google Scholar
Minari T, Kano M, Miyadera T, Wang S-D, Aoyagi Y, Tsukagoshi K. Appl. Phys. Lett. 2009, 94, 093307(1–3).Google Scholar
Kano M, Minari T, Tsukagoshi K. Appl. Phys. Lett. 2009, 94, 143304.Google Scholar
Liu B, Zeng HC. J. Am. Chem. Soc. 2003, 125, 4430–4431.Google Scholar
Wang X, Li Y. Angew. Chem. Int. Ed. 2002, 41, 4790–4793.Google Scholar
Kearns DR, Tollin G, Calvin M. J. Chem. Phys. 1960, 32, 1020–1025.Google Scholar
Lous EJ, Blom PWM, Molenkamp LW. Phys. Rev. B 1985, 51, 17251–17254.Google Scholar
Maitrot M, Guillaud G, Boudjema B, André JJ, Simon J. J. Appl. Phys. 1986, 60, 2396–400.Google Scholar
Andre JJ, Simon J, Even R, Boudjema B, Guillaud G, Maitrot M. Synth. Metals 1987, 18, 683–8.Google Scholar
Blochwitz J, Pfeiffer M, Fritz T, Leo K. Appl. Phys. Lett. 1998, 73, 729–731.Google Scholar
Huang J, Pfeiffer M, Werner A, Blochwitz J, Leo K, Shiyong L. Appl. Phys. Lett. 2002, 80, 139–141.Google Scholar
Zhou X, Pfeiffer M, Huang JS, Blochwitz-Nimoth J, Qin DS, Werner A, Drechsel J, Maennig B, Leo K. Appl. Phys. Lett. 2002, 81, 922–924.Google Scholar
Blochwitz J, Pfeiffer M, Hofmann M, Leo K. Synth. Metals 2002, 127, 169–173.Google Scholar
He G, Pfeiffer M, Leo K, Hofmann M, Birnstock J, Pudzich R, Salbeck J. Appl. Phys. Lett. 2004, 85, 3911–3913.Google Scholar
Huang Q, Walzer K, Pfeiffer M, Lyssenko V, He GF, Leo K. Appl. Phys. Lett. 2006, 88, 113515.Google Scholar
Meerheim R, Walzer K, Pfeiffer M, Leo K. Appl. Phys. Lett. 2006, 89, 061111.Google Scholar
Pfeiffer M, Beyer A, Plönnigs B, Nollau A, Fritz T, Leo K, Schlettwein D, Hiller S, Wöhrle D. Sol. Energy Mater. Sol. Cells 2000, 63, 83–99.Google Scholar
Gebeyehu D, Maennig B, Drechsel J, Leo K, Pfeiffer M. Sol. Energy Mater. Sol. Cells 2003, 79, 81–92.Google Scholar
Drechsel J, Männig B, Kozlowski F, Pfeiffer M, Leo K, Hoppe H. Appl. Phys. Lett. 2005, 86, 244102.Google Scholar
Schulze K, Uhrich C, Schüppel R, Leo K, Pfeiffer M, Brier E, Reinold E, Bäuerle P. Adv. Mater. 2006, 18, 2872–2875.Google Scholar
Uhrich C, Wynands D, Olthof S, Riede MK, Leo K, Sonntag S, Maennig B, Pfeiffer M. J. Appl. Phys. 2008, 104, 043107.Google Scholar
Wynands D, Levichkova M, Leo K, Uhrich C, Schwartz G, Hildebrandt D, Pfeiffer M, Riede M. Appl. Phys. Lett. 2010, 97, 073503.Google Scholar
Lüssem B, Riede M, Leo K. Phys. Stat. Solid. A 2013, 210, 9–43.Google Scholar
Kido J, Matsumoto T. Appl. Phys. Lett. 1998, 73, 2866–2868.Google Scholar
Brabec CJ, Dyakonov V. Photo Induced Charge Transfer in Bulk Heterojunctions Composites, Springer-Verlag: Berlin/ Heidelberg, 2003, pp 1–51.Google Scholar
Nollau A, Pfeiffer M, Fritz T, Leo K. J. Appl. Phys. 2000, 87, 4340–4343.Google Scholar
Tanaka S, Kanai K, Kawabe E, Iwahashi T, Nishi T, Ouchi Y, Seki K. Jpn. J. Appl. Phys. 2005, 44, 3760–3763.Google Scholar
Chan CK, Amy F, Zhang Q, Barlow S, Marder S, Kahn A. Chem. Phys. Lett. 2006, 431, 67–71.Google Scholar
Bloom CJ, Elliott CM, Schroeder PG, France CB, Parkinson BA. J. Phys. Chem. 2003, 107, 2933–2938.Google Scholar
Werner AG, Li F, Harada K, Pfeiffer M, Fritz T, Leo K. Appl. Phys. Lett. 2003, 82, 4495–4497.Google Scholar
Werner A, Li F, Harada K, Pfeiffer M, Fritz T, Leo K, Machill S. Adv. Funct. Mater. 2004, 14, 255–260.Google Scholar
Olthof S, Tress W, Meerheim R, Lüssem B, Leo K. J. Appl. Phys. 2009, 106, 103711.Google Scholar
Sze SM. Physics of Semiconductor Devices, 2nd ed., Wiley Interscience: New York, 1999.Google Scholar
Koch N, Vollmer A. Appl. Phys. Lett. 2006, 89, 162107.Google Scholar
Li T, Ruden PP, Campbell IH, Smith DL. J. Appl. Phys. 2003, 93, 4017.Google Scholar
Tessler N, Roichman Y. App Phys Lett 2001, 79, 2987–2989.Google Scholar
Watkins NJ, Yan L, Gao Y. Appl. Phys. Lett. 2002, 80, 4384–4386.Google Scholar
Brown TM, Kim JS, Friend RH, Cacialli F, Daik R, Feast WJ. Appl. Phys. Lett. 1999, 75, 1679.Google Scholar
Peisert H, Knupfer M, Fink J. Appl. Phys. Lett. 2002, 81, 2400.Google Scholar
Qiu CF, Xie Z, Chen H, Wong M, Kwok HS. J. Appl. Phys. 2003, 93, 3253–3258.Google Scholar
Puntambekar KP, Pesavento PV, Frisbie CD. Appl. Phys. Lett. 2003, 83, 5539.Google Scholar
Bürgi L, Richards TJ, Friend RH, Sirringhaus H. J. Appl. Phys. 2003, 94, 6129.Google Scholar
Pesevanto PV, Chesterfield RJ, Newman CR, Frisbie CD. J. Appl. Phys. 2004, 96, 7312.Google Scholar
Chen TH, Liou Y, Wu TJ, Chen JY. Appl. Phys. Lett. 2004, 85, 2092.Google Scholar
Hill IG. Appl. Phys. Lett. 2005, 87, 163505.Google Scholar
Park J, Choi JS. Synth. Metals 2005, 155, 657–661.Google Scholar
Kalakodimi RP, Nowak AM, McCreery RL. Chem. Mater. 2005, 17, 4939–4948.Google Scholar
Lim JA, Cho JH, Park YD, Kim DH, Hwang M, Cho K. Appl. Phys. Lett. 2006, 88, 082102.Google Scholar
Pesavento PV, Puntambekar KP, Frisbie CD, McKeen JC, Ruden PP. J. Appl. Phys. 2006, 99, 094504.Google Scholar
Vanoni C, Tsujino S, Jung TA. Appl. Phys. Lett. 2007, 90, 193119.Google Scholar
Ihm K, Heo H-E, Chunga S, Ahn J-R, Kim JH, Kang T-H. Appl. Phys. Lett. 2007, 90, 242111.Google Scholar
Sun Y, Liu Z, Pianetta P. J. Vac. Sci. Technol. 2007, A 25, 1351.Google Scholar
Wang G, Moses D, Heeger AJ, Zhang H-M, Narasimhan M, Demaray RE. J. Appl. Phys. 2004, 95, 316–322.Google Scholar
Pyo KS, Song CK. Thin Solid Films 2005, 485, 230.Google Scholar
Yang SY, Shin K, Park CE. Adv. Func. Mater. 2005, 15, 1806–1814.Google Scholar
Jang Y, Kim DH, Park YD, Cho JH, Hwang M, Cho K. Appl. Phys. Lett. 2005, 87, 152105.Google Scholar
Shin K, Yang C, Yang SY, Jeon H, Park CE. Appl. Phys. Lett. 2006, 88, 072109.Google Scholar
Lee S, Koo B, Shin J, Lee JE, Park H, Kim H. Appl. Phys. Lett. 2006, 88, 162109.Google Scholar
Manaka T, Lim E, Tamura R, I wamoto M. Thin Solid Films 2006, 499, 386–391.Google Scholar
Ng TS, Chabinyc ML, Marohn JA. J. Appl. Phys. 2006, 100, 084505.Google Scholar
Gu G, Kane MG. Appl. Phys. Lett. 2008, 92, 053305.Google Scholar
Schön JH. Synth. Metals 2001, 122, 157–160.Google Scholar
Karl N. Synth. Metals 2002, 649, 133–134.Google Scholar
Tsukagoshi K, Yagi I, Aoyagi Y. Sci. Technol. Adv. Mater. 2006, 7, 231.Google Scholar
Cui J, Wang A, Edleman NL, Ni J, Lee P, Armstrong NR, Marks TJ. Adv. Mater. 2001, 13, 1476–1480.Google Scholar
Morgado J, Charas A, Barbagallo N. Appl. Phys. Lett. 2002, 81, 933.Google Scholar
Klauk H, Schmid G, Radlik W, Weber W, Zhou L, Sheraw CD, Nichols JA, Jackson TN. Solid-State Electron. 2003, 47, 297–301.Google Scholar
Yoneya N, Noda M, Hirai N, Nomoto K, Wada M, Kasahara J. J. Appl. Phys. Lett. 2004, 85, 4663.Google Scholar
Kim SY, Lee J-L. Metal. Mater. Int. 2005, 11, 411–414.Google Scholar
Jin SH, Jung KD, Shin H, Park B-G, Lee JD. Synth. Metals 2006, 156, 196–201.Google Scholar
Bock C, Pham DV, Kunze U, Käfer D, Witte G, Terfort A. Appl. Phys. Lett. 2007, 91, 052110.Google Scholar
Diaom L, Frisbie CD, Schroepfer DD, Ruden PP. J. Appl. Phys. 2007, 101, 014510.Google Scholar
Watanabe Y, Iechi H, Kudo K. Jpn. J. Appl. Phys. 2007, 46, 2717–2721.Google Scholar
Wang SD, Minari T, Miyadera T, Aoyagi Y, Tsukagoshi K. Appl. Phys. Lett. 2008, 92, 063305.Google Scholar
Hong K, Lee JW, Yang SY, Shin K, Jeon H, Kim SH, Yang C, Park CE. Org. Electron. 2008, 9, 21–29.Google Scholar
Michaelson HB. J. Appl. Phys. 1997, 48, 4729.Google Scholar
Shen C, Kahn A. Org. Electron. 2001, 2, 89–95.Google Scholar
Song W, So SK, Kao L. Appl. Phys. A 2001, 72, 361–365.Google Scholar
Cahen D, Kahn A. Adv. Mater. 2003, 15, 271–277.Google Scholar
Pantisano L, Afanas’ev V, Pourtois G, Chen PJ. J. Appl. Phys. 2005, 98, 053712.Google Scholar
Tapajna M, Hušeková K, Espinos JP, Harmatha L, Fröhlich K. Mater. Sci. Semicond. Proc. 2006, 9, 969–974.Google Scholar
Hamadani BH, Natelson D. Appl. Phys. Lett. 2004, 84, 443.Google Scholar
von Hauff E, Dyakonov V, Parisi J. Sol. Energy Mater. Sol. Cells 2005, 87, 149–156.Google Scholar
Narioka S, Ishii H, Yoshimura D, Sei M, Ouchi Y, Seki K, Hasegawa S, Miyazaki T, Harima Y, Yamashita K. Appl. Phys. Lett. 1995, 67, 1899.Google Scholar
Shimada T, Hamaguchi K, Koma A, Ohuchi FS. Appl. Phys. Lett. 1998, 72, 1869.Google Scholar
Bharathan JM, Yang Y. J. Appl. Phys. 1998, 84, 3207.Google Scholar
Katuka G, Roe KJ, Kolodzey J, Swann CP, Desalvo G, Clarke RC, Eldridge G, Messham R. J. Electron. Mater. 2002, 31, 346–350.Google Scholar
Obata M, Sakuda T, Abe K, Hayashibe R, Kamimura K. Surf. Coat. Technol. 2004, 180, 136–139.Google Scholar
Moselund KE, Freiermuth JE, Dainesi P, Ionescu AM. IEEE Trans. ED-53 2006, 712–718.Google Scholar
Kim W-K, Lee J-L. Appl. Phys. Lett. 2006, 88, 262102.Google Scholar
Yun DJ, Lee D-K, Jeon H-K, Rhee SW. Org. Electron. 2007, 8, 690–694.Google Scholar
von Hauff E, Spethmann N, Parisi J. Z. Naturforsch. 2008, 63, 591–595.Google Scholar
Folkers JP, Gorman CB, Laibinis PE, Buchholz S, Whitesides GM, Nuzzo RG. Langmuir 1995, 11, 813–824.Google Scholar
Bürgi L, Sirringhaus H, Friend RH. Appl. Phys. Lett. 2002, 80, 2913.Google Scholar
Bürgi L, Richards T, Chiesa M, Friend RH, Sirringhaus H. Synth. Metals 2004, 146, 297–309.Google Scholar
Nichols JA, Gundlach DJ, Jackson TN. Appl. Phys. Lett. 2003, 83, 2366.Google Scholar
Miyazaki T, Kobayashi K, Ishida K, Hotta S, Horiuchi T, Yamada H, Matsushige K. Jpn. J. Appl. Phys. 2003, 42, 4852–4855.Google Scholar
Angst D, Simmons GW. Langmuir 1991, 7, 2236–2242.Google Scholar
Xia Y, Mrksich M, Kim E, Whitesides GM. J. Am. Chem. Soc. 1995, 117, 9576–9577.Google Scholar
Vallant T, Kattner J, Brunner H, Mayer U, Hoffmann H. Langmuir 1999, 15, 5339–5346.Google Scholar
Zuppiroli L, Si-Ahmed L, Kamaras K, Nüesch F, Bussac MN, Ades D, Siove A, Moons E, Grätzel M. Eur. Phys. J. B 1999, 11, 505–512.Google Scholar
Masuda Y, Seo WS, Koumoto K. Langmuir 2001, 17, 4876–4880.Google Scholar
Ren SL, Yang SR, Zhao YP, Zhou J, Xu T, Liu W. Tribol. Lett. 2002, 13, 233–239.Google Scholar
Wang Y, Lieberman M. Langmuir 2003, 19, 1159–1167.Google Scholar
Yan D, Jordan JL, Burapatana V, Jennings GK. Langmuir 2003, 19, 3357–3364.Google Scholar
Tremont RJ, Blasini DR, Cabrera CR. J. Electroanal. Chem. 2003, 556, 147–158.Google Scholar
Engelkes VB, Beebe JM, Frisbie CD. J. Am. Chem. Soc. 2004, 126, 14287–14296.Google Scholar
de Boer B, Hadipore A, Mandoc MM, van Woudenbergh T, Blom PWM. Adv. Mater. 2005, 17, 621–625.Google Scholar
Di CA, Yu G, Liu YQ, Xu X, Wei D, Song Y, Sun Y, Wang Y, Zhu D, Liu J, Liu X, Wu D. J. Am. Chem. Soc. 2006, 128, 16418–16419.Google Scholar
Khodabakhsh S, Poplavskyy D, Heutz S, Nelson J, Bradley DDC, Murata H, Jones TS. Adv. Funct. Mater. 2004, 14, 1205–1210.Google Scholar
Khodabakhsh S, Sanderson BM, Nelson J, Jones TS. Adv. Funct. Mater. 2006, 16, 95–100.Google Scholar
Kymissis I, Dimitrakopoulos CD, Purushothaman S. IEEE Trans. 2001, ED-48, 1060–1064.Google Scholar
Pernstich KP, Haas S, Oberhoff D, Goldmann C, Gundlach DJ, Batlogg B, Rashid AN, Schitter G. J. Appl. Phys. 2004, 96, 6431.Google Scholar
Abthagir PS, Ha Y-G, You E-A, Jeong S-H, Seo H-S, Choi J-H. J. Phys. Chem. B 2005, 109, 23918–23924.Google Scholar
Lin R, Qiang L, Ranade P, King T-J, Hu C. IEEE EDL 2002, 49–51.Google Scholar
Yun D-J, Rhee S-W. J. Electrochem. Soc. 2008, 155, H357–H362.Google Scholar
Rusu PC, Brocks G. Phys. Rev. B 2006, 74, 073414.Google Scholar
Jee SH, Kim SH, Park H, Kim D-J, Yoon Y-S. J. Korean Phys. Soc. 2006, 49, 2034.Google Scholar
Campbell IH, Kress JD, Martin RL, Smith DL, Barashkov NN, Ferraris JP. Appl. Phys. Lett. 1997, 71, 3528.Google Scholar
Facchetti A, Yoon M-H, Marks TJ. Adv. Mater. 2005, 17, 1705–1725.Google Scholar
Kelley TW, Frisbie CD. J. Phys. Chem. B 2001, 105, 4538–4540.Google Scholar
Ling MM, Bao Z. Chem. Mater. 2004, 16, 4824–4840.Google Scholar
Addadi L, Weiner S. Proc. Natl. Acad. Sci. USA 1985, 82, 4110–4114.Google Scholar
Mann S. Nature 1993, 365, 499–505.Google Scholar
Lee S-W, Lee SK, Belcher AM. Adv. Mater. 2003, 15, 689–692.Google Scholar
Han Y-J, Aizenberg J. Angew. Chem. Int. Ed. 2003, 42, 3668–3670.Google Scholar
Travaille AM, Kaptijn L, Verwer P, Hulsken B, Elemans JAAW, Nolte RJM, van Kempen H. J. Am. Chem. Soc. 2003, 125, 11571–11577.Google Scholar
Aizenberg J. Adv. Mater. 2004, 16, 1295–1302.Google Scholar
Noh D-Y, Willing GA, Han CY, Shin K-S, Geiser U, Wang HH. Chem. Mater. 2004, 16, 4777–4782.Google Scholar
Kasai H, Nalwa HS, Oikawa H, Okada S, Matsuda H, Minami N, Kakuta A, Ono K, Mukoh A, Nakanishi H. Jpn. J. Appl. Phys. 1992, 31, L1132–L1134.Google Scholar
Landfester K, Montenegro R, Scherf U, Güntner R, Asawapirom U, Patil S, Neher D, Kietzke T. Adv. Mater. 2002, 14, 651–655.Google Scholar
Köstler S, Rudorfer A, Haase A, et al. Adv. Mater. 2009, 21, 2505–2510.Google Scholar
Yasukuni R, Asahi T, Sugiyama T, Masuhara H, Sliwa M, Hofkens J, De Schryver FC, Van der Auweraer M, Herrmann A, Müllen K. Appl. Phys. A 2008, 93, 5–9.Google Scholar
Baier MC, Huber J, Mecking S. J. Am. Chem. Soc. 2009, 131, 14267–14273.Google Scholar
Hittinger E, Kokil A, Weder C. Angew. Chem. Int. Ed. Eng. 2004, 43, 1808–1811.Google Scholar
Henderson AMJ, Saunders JM, Mrkic J, Kent P, Gore J, Saunders BR. J. Mater. Chem. 2001, 11, 3037–3042.Google Scholar
Wosnick JH, Liao JH, Swager TM. Macromolecules 2005, 38, 9287–9290.Google Scholar
Reverchon E. J. Supercrit. Fluids 1999, 15, 1–21.Google Scholar
Jung J, Perrut M. J. Supercrit. Fluids 2001, 20, 179–219.Google Scholar
Weber, Thies. Understanding the RESS process. In Supercritical Fluid Technology in Materials Science and Engineering: Synthesis, Properties, and Applications, Sun Y-P, Ed., CRC Press: New York, 2002, pp 387–437.Google Scholar
Blasig A, Shi C, Enick RM, Thies MC. Ind. Eng. Chem. Res. 2002, 41, 4976–4983.Google Scholar
Sane A, Thies MC. J. Supercrit. Fluids 2007, 40, 134–143.Google Scholar
Sun YP, Rolling HW, Bandara J, Meziani JM, Bunker CE. Preparation and processing of nanoscale materials by supercritical fluid technology. In Supercritical Fluid Technology in Materials Science and Engineering:Synthesis, Properties, and Applications, Sun YP, Ed., New York: Marcel Dekker, 2002, pp. 491–576.Google Scholar
Meziani MJ, Pathak P, Hurezeanu R, Thies MC, Enick RM, Sun Y-P. Angew. Chem. Int. Ed. 2004, 43, 704–707.Google Scholar
Meziani MJ, Pathak P, Wang W, Desai T, Patil A, Sun Y-P. Ind. Eng. Chem. Res. 2005, 44, 4594–4598.Google Scholar
Schuck P, Dolivet A, Méjean S, Zhu P, Blanchard E, Jeantet R. J. Food Eng. 2009, 94, 199–204.Google Scholar
Arpagaus C, Schafroth N. Respir. Drug Deliv. Eur. 2009, 2, 269–274.Google Scholar
Chu CT, Dunn B. Appl. Phys. Lett. 1993, 55, 492–494.Google Scholar
Gao B, Yue GZ, Qiu Q, Cheng Y, Shimoda H, Fleming L, Zhou O. Adv. Mater. 2001, 13, 1770–1773.Google Scholar
Tada, Onoda. Adv. Funct. Mater. 2002, 12, 420–424.Google Scholar
Irvin G, Blanton T, Jagannathan S. Adv. Funct. Mater. 2006, 16, 747–753.Google Scholar
Deegan RD, Bakajin O, Dupont TF, Huber G, Nagel SR, Witten TA. Nature 1997, 389, 827–829.Google Scholar
An BK, Kwon SK, Park SY. Angew. Chem. Int. Ed. 2007, 46, 1978–1982.Google Scholar
Fisslthaler E, Blümel A, Landfester K, Scherf U, List EJW. Soft Matter 2008, 4, 2448–2453.Google Scholar
Kietzke T, Neher D, Kumke M, Montenegro R, Landfester K, Scherf U. Macromolecules 2004, 37, 4882–4890.Google Scholar
Heo J-S, Park N-H, Ryu J-H, Suh K-D. Adv. Mater. 2005, 17, 822–826.Google Scholar
Huebner CF, Carroll JB, Evanoff DD, Ying Y, Stevenson BJ, Lawrence JR, Houchins JM, Foguth AL, Sperry J, Foulger SH. J. Mater. Chem. 2008, 18, 4942–4948.Google Scholar
Huebner CF, Roeder RD, Foulger SH. Adv. Funct. Mater. 19, 3604–3609.Google Scholar
Kroto HW, Heath JR, O’brien SC, Curl RF, Smalley RE. Nature 1985, 318, 162–163.Google Scholar
Mirkin CA, Caldwell WB. Tetrahedron 1996, 52, 5113–5130.Google Scholar
Konisi T, Ikeda A, Shinkai S. Tetrahedron 2005, 61, 4881–4899.Google Scholar
Yu G, Gao J, Hummelen JC, Wudl F, Heeger AJ. Science 1995, 270, 1789–1791.Google Scholar
Morris RC, Palstra TTM, Hebard AF, Fleming RM. Appl. Phys. Lett. 1995, 67, 121–123.Google Scholar
Priebe G, Pietzak B, Konenkamp R. Appl. Phys. Lett. 1997, 71, 2160.Google Scholar
Anthopoulos TN, Singh B, Marjanovic N, Sariciftci NS, Ramil AM, Sitter H, Cölle M, de Leeuw DM. Appl. Phys. Lett. 2006, 89, 213504/1–3.Google Scholar
Ma H, Yip H-L, Huang F, Jen AK-Y. Adv. Funct. Mater. 2010, 20, 1371–1388.Google Scholar
Shi X, Caldwell WB, Chen K, Mirkin CA. J. Am. Chem. Soc. 1994, 116, 11598–11599.Google Scholar
Imahori H, Norieda H, Yamada H, Nishimura Y, Yamazaki I, Sakata Y, Fukuzumi S. J. Am. Chem. Soc. 2001, 123, 100–110.Google Scholar
Yamada H, Imahori H, Nishimura Y, Yamazaki I, Ahn TK, Kim SK, Kim D, Fukuzumi S. J. Am. Chem. Soc. 2003, 125, 9129–9139.Google Scholar
Cho Y-J, Ahn TK, Song H, Kim KS, Lee CY, Seo WS, Lee K, Kim SK, Kim D, Park JT. J. Am. Chem. Soc. 2005, 127, 2380–2381.Google Scholar
Shirai Y, Cheng L, Chen B, Tour JM. J. Am. Chem. Soc. 2006, 128, 13479–13489.Google Scholar
Matsuo Y, Kanaizuka K, Matsuo K, Zhong Y-W, Nakae T, Nakamura E. J. Am. Chem. Soc. 2008, 130, 5016–5017.Google Scholar
Matsuo Y, Ichiki T, Radhakrishnan SG, Guldi DM, Nakamura E. J. Am. Chem. Soc. 2010, 132, 6342–6348.Google Scholar
Tsukruk. Langmuir 1994, 10, 996–999.Google Scholar
Tsukruk. Langmuir 1996, 12, 3905–3911.Google Scholar
Lee H, Jeon IC. Synth. Metals 1997, 86, 2297–2298.Google Scholar
Wei T-X, Zhai J, Ge J-H, Gan L-B, Huang C-H, Luo G-B, Ying L-M, Liu T-T, Zhao X-S. Appl. Surf. Sci. 1999, 151, 153–158.Google Scholar
Wei T-X, Zhai J, Ge J, Gan L-B, Huang C-H, Luo G-B, Ying L-M, Liu T-T, Zhao X-S. J. Colloid Interf. Sci. 2000, 222, 262–264.Google Scholar
Gulino A, Bazzano S, Condorelli GG, Giuffrida S, Mineo P, Satriano C, Scamporrino E, Ventimiglia G, Vitalini D, Fragalà I. Chem. Mater. 2005, 17, 1079–1084.Google Scholar
Guérin D, Lenfant S, Godey S, Vuillaume D. J. Mater. Chem. 2010, 20, 2680–2690.Google Scholar
Takeya J, Nishikawa T, Takenobu T, Kobayashi S, Iwasa Y, Mitani T, Goldmann C, Krellner C, Batlogg B. Appl. Phys. Lett. 2004, 85, 5078–5080.Google Scholar
Tremblay NJ, Gorodetsky AA, Cox MP, Schiros T, Kim B, Steiner R, Bullard Z, Sattler A, So W-Y, Itoh Y, Toney MF, Ogasawara H, Ramirez AP, Kymissis I, Steigerwald ML, Nuckolls C. Chem. Phys. Chem. 2010, 11, 799–803.Google Scholar
Kim B. Single crystal growth of organic semiconductors and organic electronic applications, Ph.D. thesis, Columbia University, 2011.Google Scholar
Yamamoto Y, Zhang G, Jin W, Fukushima T, Ishii N, Saeki A, Seki S, Tagawa S, Minari T, Tsukagoshi K, Aida T. Proc. Natl. Acad. Sci. USA 2009, 106, 21051–21056.Google Scholar
Scida A. Ion implantation of organic thin films and electronic devices, Ph.D. thesis, University of Bologna: Italy, 2013.Google Scholar
Baglin JEE. Nucl. Instrum. Methods Phys. Res. B 1989, 39, 764–768.Google Scholar
Fraboni B, Scidà A, Cavallini A, Milita S, Cosseddu P, Bonfiglio A, Wang Y, Nastasi M. Synth. Metals 2012, 161, 2585–2588.Google Scholar
Basiricò L, Inkjet printing of organic transistor devices, Ph.D. thesis, Electronic and Computer Engineering Department of Electrical and Electronic Engineering, University of Cagliari, 2012.Google Scholar
Karwa A. Printing studies with conductive inks and exploration of new conducting polymer compositions, Master’s thesis, Center for Materials Science and Engineering, Rochester Institute of Technology: New York, 2006.Google Scholar
Coatanéa E, Kantola V, Kulovesi J, Lahti L, Lin R, Zavodchikova M. In Printed Electronics, Now and Future, Bit Bang – Rays to the Future, Neuvo Y, Ylönen S, Eds., Helsinki University of Technology, 2009.Google Scholar
Caglar U. Studies of inkjet printing technologies with focus on electronic materials, Ph.D. thesis, Tampere University of Technology, 2009.Google Scholar
Yu JS, Jo J, Kim DS, Kim DJ. J. Nanosci. Nanotechnol. 2008, 8, 4940–4944.Google Scholar
Jo J, Yu J-S, Lee T-M, Kim D-S, Kim K-Y. J. Nanosci. Nanotechnol. 2010, 10, 3595–3599.Google Scholar
Hon KKB, Li L, Hutchings IM. CIRP Ann. Manuf. Technol. 2008, 57, 601–620.Google Scholar
Tech Note, Micro Fab Technote, 1999, 99–101.Google Scholar
de Gans B-J, Duineveld PC, Schubert US. Adv. Mater. 2004, 16, 203–213.Google Scholar
Ballarin B, Fraleoni-Morgera A, Frascaro D, Marazzita S, Piana C, Setti L. Synth. Metals 2004, 146, 201–205.Google Scholar
Kawase T, Shimoda T, Newsome C, Sirringhaus H, Friend RH. Thin Solid Films 2003, 438–439, 279–287.Google Scholar
Al-Chami H. Inkjet printing of transducers, Master’s thesis, University of British Columbia, 2010.Google Scholar
Griffith OL, Anthony JE, Jones AG, Lichtenberger DL. J. Am. Chem. Soc. 2010, 132, 580–586.Google Scholar
Madec M-B, Smith PJ, Malandraki A, Wang N, Korvink JG, Yeates SG. J. Mater. Chem. 2010, 20, 9155–9160.Google Scholar
Lee SH, Choi MH, Han SH, Choo DJ, Jang J, Kwon SK. Org. Electron. 2008, 9, 721–726.Google Scholar
Lim JA, Kim J-H, Qiu L, Lee WH, Lee HS, Kwak D, Cho K. Adv. Funct. Mater. 2010, 20, 3292–3297.Google Scholar
Lee MW, Ryu GS, Lee YU, Pearson C, Petty MC, Song CK. Microel. Eng. 2012, 95, 1–4.Google Scholar
Xia Y, Whitesides GM. Angew. Chem. Int. Ed. 1998, 37, 550–575.Google Scholar
Gamota D, Paul B, Kalyanasundaram K, Zhang J, Eds., Printed Organics and Molecular Electronics, Kluwer Academic Publishers, 2004. ISBN 1402077076/1-4020-7707-6.Google Scholar
Michel B, Bernard A, Bietsch A, Delamarche E, Geissler M, Juncker D, Kind H, Renault J-P, Rothuizen H, Schmid H, Schmidt-Winkel P, Stutz R, Wolf H. IBM J. Res. Dev. 2001, 45, 697–719.Google Scholar
Shinamura S, Osaka I, Miyazaki E, Takimiya K. Heterocycles 2011, 83, 1187–1204.Google Scholar
Doi I, Kang MJ, Takimiya K. Curr. Appl. Phys. 2012, 12, e2–e5.Google Scholar
Cho YJ, Lee JY, Chin BD, Forrest SR. Org. Electron. 2013, 14, 1081–1085.Google Scholar
McCall KL, Ogier SD, Brown BA, Rutter SR, Palumbo M, Lee YU, Evans LA, Pease TJ. Low Voltage High Mobility Organic Semiconductors for Flexible Display Applications, Centre for Process Innovation Limited, Thomas Wright Way, NET Park: Sedgefield, TS21 3FG, UK.Google Scholar
Li FM, Hsieh G-W, Dalal S, Newton MC, Stott JE, Hiralal P, Nathan A, Warburton PA, Unalan HE, Beecher P, Flewitt AJ, Robinson I, Amaratunga G, Milne WI. IEEE Trans. 2008, ED-55, 3001–3011.Google Scholar
Shin H-J, Kim KK, Benayad A, Yoon S-M, Park HK, Jung I-S, Jin MH, Jeong H-K, Kim JM, Choi J-Y, Lee YH. Adv. Func. Mater. 2009, 19, 1987–1992.Google Scholar
Patent, 2011, Transistors and Methods of Making Them, PCT/ GB2012/051170.Google Scholar
Kim N. Fabrication and Characterization of Thin-film Encapsulation for Organic Electronics, Ph.D. thesis, Georgia Institute of Technology: USA, 2009.Google Scholar
McElvain J, Antoniadis H, Hueschen MR, Miller JN, Roitman DM, Sheats JR, Moon RL. J. Appl. Phys. 1996, 80, 6002–6007.Google Scholar
Cumpston BH, Parker ID, Jensen KF. J. Appl. Phys. 1997, 81, 3716–3720.Google Scholar
Bliznyuk VN, Carter SA. Macromolecules 1999, 32, 361–369.Google Scholar
Burrows PE, Bulovic V, Forrest SR, Sapochak LS, McCarty DM, Thompson ME. Appl. Phys. Lett. 1994, 65, 2922–2924.Google Scholar
Mandlik P, Gartside J, Han L, Cheng I-C, Wagner S, Silvernail JA, Ma R-Q, Hack M, Brown JJ. Appl. Phys. Lett. 2008, 92, 103309.Google Scholar
Kwon SH, Paik SY, Kwon OJ, Yoo JS. Appl. Phys. Lett. 2001, 79, 4450.Google Scholar
Ghosh AP, Gerenser LJ, Jarman CM, Fornalik JE. Appl. Phys. Lett. 2005, 86, 223, 503.Google Scholar
Yamashita K, Mori T, Mizutani T. J. Phys. D Appl. Phys. 2001, 34, 740–743.Google Scholar
Chwang AB, Rothman MA, Mao SY, Hewitt RH, Weaver MS, Silvernail JA, Rajan K, Hack M, Brown JJ, Chu X, Moro L, Krajewski T, Rutherford N. Appl. Phys. Lett. 2003, 83, 413.Google Scholar
Kim GH, Oh J, Yang YS, Do L-M, Suh KS. Thin Solid Films 2004, 467, 1–3.Google Scholar
da Silva Sobrinho AS, Latrèche M, Czeremuszkin G, Klemberg-Sapieha JE, Wertheimer MR. J. Vac. Sci. Technol. 1998, A 16, 3190.Google Scholar
Leterrier Y. Prog. Mater. Sci. 2003, 48, 1–55.Google Scholar
Huang W, Wang X, Sheng M, Xu L, Stubhan F, Luo L, Feng T, Wang X, Zhang F, Zou S. Mater. Sci. Eng. B 2003, 98, 248–254.Google Scholar
Potscavage WJ, Yoo S, Domercq B, Kippelen B. Appl. Phys. Lett. 2007, 90, 253511.Google Scholar
Dameron AA, Davidson SD, Burton BB, Carcia PF, McLean RS, George SM. J. Phys. Chem. C 2008, 112, 4573–4580.Google Scholar
Meyer J, Görrn P, Bertram F, Hamwi S, Winkler T, Johannes H-H, Weimann T, Hinze P, Riedl T, Kowalsky W. Adv. Mater. 2009, 21, 1845–1849.Google Scholar
Phatak R. Dependence of Dark Spot Growth on Cathode/ Organic Interfacial Adhesion in Organic Light Emitting Devices, Thesis, University of Waterloo: Canada, 2011.Google Scholar
Aziz H, Popovic ZD. Chem. Mater. 2004, 16, 4522–4532.Google Scholar
Subbarao SP, Bahlke ME, Kymissis I. IEEE Trans. ED 2010, 57, 153–156.Google Scholar
Lim SF, Ke L, Wang W, Chua SJ. Appl. Phys. Lett. 2001, 78, 2116.Google Scholar
Savvateev VN, Yakimov AV, Davidov D, Pogreb RM, Neumann R, Avny Y. Appl. Phys. Lett. 1997, 71, 3344.Google Scholar
Do L-M, Kim K, Zyung T, Shim H-K, Kim J-J. Appl. Phys. Lett. 1997, 70, 3470–3472.Google Scholar
Kawaharada M, Ooishi M, Saito T, Hasegawa E. Synth. Metals 1997, 91, 113–116.Google Scholar
Aziz H, Popovic Z, Tripp CP, Hu N-X, Hor A-M, Xu G. Appl. Phys. Lett. 1998, 72, 2642–2644; 756–758.Google Scholar
Schaer M, Nüesch F, Berner D, Leo W, Zuppiroli L. Adv. Funct. Mater. 2001, 11, 116–121.Google Scholar
Gao ZQ, Lai WY, Wong TC, Lee CS, Bello I, Lee ST. Appl. Phys. Lett. 1999, 74, 3269.Google Scholar
Gao ZQ, Lai WY, Wong TC, et al. Appl. Phys. Lett. 2000, 77, 3113.Google Scholar
Xu MS, Xu JB. Thin Solid Films 2005, 491, 317–322.Google Scholar
Kwong CY, Djurišic AB, Roy VAL, Lai PT, Chan WK. Thin Solid Films 2004, 458, 281–286.Google Scholar
Vitex Systems. Barix Thin Film Encapsulation. Available at: http://www.vitexsys.com/index.html.
Aziz HM, Liew Y-F, Popovic Z. Proc. SPIE 2004, 5214, 277.Google Scholar
Chan MY, Lai SL, Wong FL, Lengyel O, Lee CS, Lee ST. Chem. Phys. Lett. 2003, 371, 700–706.Google Scholar
Le Comber PG, Spear WE. Phys. Rev. Lett. 1970, 25, 509–511.Google Scholar
Good DA. Tech Report No. UCB/EECS-2007-62, University of California: Berkeley, 2007.Google Scholar
Tiedje T, Abele B, Morel DL, Moustakas TD, Wronski CR. Appl. Phys. Lett. 1980, 36, 695–697.Google Scholar
Sobbia R, Sansonnens L, Bondkowski J. J. Vac. Sci. Technol. 2005, A 23, 927.Google Scholar
Kaneko T, Hosokawa Y, Tadauchi M, Kita Y, Andoh H. Electron Devices, IEEE Transactions 1991, ED-38, 1086–1093.Google Scholar
Smith PM, Carey PG, Sigmon TW. Appl. Phys. Lett. 1997, 70, 342–344.Google Scholar
Subramanian V, Dankoski P, Degertekin L, Khuri-Yakub BT, Saraswat KC. IEEE 1997, EDL-18, 378–381.Google Scholar
Sameshima T, Usui S, Sekiya M. IEEE 1986, EDL-7, 276–278.Google Scholar
Crowder MA, Carey PG, Smith PM, Sposili RS, Cho HS, Im JS. IEEE 1998, EDL-19, 306–308.Google Scholar
Giust GK, Sigmon TW, Boyce JB, Ho J. IEEE 1999, EDL-20, 77–79.Google Scholar
Kunugi Y, Takimiya K, Yamane K, Yamashita K, Aso Y, Otsubo T. Chem. Mater. 2003, 15, 6–7.Google Scholar
Yoshida H, Sato N. Phys. Rev. B 2008, 77, 235205.Google Scholar
Takimiya K, Kunugi Y, Ebata H, Otsubo T. Chem. Lett. 2006, 35, 1200–1201.Google Scholar
Izawa T, Miyazaki E, Takimiya K. Chem. Mater. 2009, 21, 903–912.Google Scholar
Yamamoto T, Takimiya K. J. Photopolym. Sci. Tech. 2007, 20, 47–51.Google Scholar
Yamamoto T, Shinamura S, Miyazaki E, Takimiya K. Bull. Chem. Soc. Jpn. 2010, 83, 120–130.Google Scholar
Haas S, Takahashi Y, Takimiya K, Hasegawa T. Appl. Phys. Lett. 2009, 95, 022111.Google Scholar
Ishida K, Huang T-C. 2011. Available at http://icdesign.iis.u-tokyo.ac.jp/2011_38.pdf.
Shigeki Shino. Conducting Patterns Making Technology with Metal Nano-Particle Inks, 2008. Available at http://www.jpcashow.com/show2012/English/event_features/electronics-convention.html.
Jakob K-H, Joseph EN, da Silva Filho DA, Brédas JL, Rubahn H-G. Org. Electron. 2009, 10, 1228–1234.Google Scholar
Liu X, Wallmann I, Boudinov H, Kjelstrup-Hansen J, Schiek M, Lützen A, Rubahn H-G. Org. Electron. 2010, 11, 1096–1102.Google Scholar
Seo H-S, Kim D-K, Oh J-D, Shin E-S, Choi J-H. J. Phys. Chem. C 2013, 117, 4764–4770.Google Scholar
About the article
Published Online: 2014-03-13
Published in Print: 2014-06-01