- freely available
Electronics 2014, 3(1), 76-86; doi:10.3390/electronics3010076
Abstract: In this work, the electrical response of n-type organic field-effect transistors, achieved by evaporating PDIF-CN2 films on both bare and Hexamethyldisilazane (HMDS) treated SiO2 substrates, was investigated by standard electrical characterization and potentiometry. Morphological and charge transport characterizations demonstrated that the hydrophobic degree of the substrate surface has a huge impact on the final response of the devices. The PDIF-CN2 transistors on HMDS-treated substrates show a maximum mobility of 0.7 cm2/Volt·s, three orders of magnitude greater than in the case of the device without surface functionalization. The scanning Kelvin probe microscopy technique was used to perform surface potentiometry to image the local surface potential inside the channel during the transistor operation and has allowed us to identify the film morphological disorder as the primary factor that could compromise the effectiveness of the charge injection process from gold contacts to PDIF-CN2 films. For optimized devices on HMDS-treated substrates, SKPM was also used to analyze, over time, the evolution of the potential profile when negative VGS voltages were applied. The findings of these measurements are discussed taking into account the role of VGS-induced proton migration towards SiO2 bulk, in the operational stability of the device.
Organic field-effect transistors (OFET) are the key devices for the development of complex analog and digital electronic circuits which, based on low-temperature processed organic semiconductors, can be fabricated through cost-effective techniques . Nowadays, owing to the results of a wide number of accurate experiments, we are fully aware that the quality of the electrical response of an OFET basically relies on the robustness of the charge transport processes taking place across and along the interfaces separating the different device component parts. In particular, severe effects of contact resistance can strongly compromise the device behavior in the presence of a non-optimized charge injection condition. With regard to this last phenomenon, it can be largely dependent on the matching of the Fermi level of the injecting electrode and the LUMO (Low Unoccupied Molecular Orbital) or HOMO (High Occupied Molecular Orbital) level of the organic semiconductor, as well as on the morphological order of the active channel in the proximity of the metal contact .
At the same time, a crucial role is also played by the interface between the dielectric barrier and the organic channel, where the charge motion occurs in a very thin region involving few molecular layers . Starting from this consideration, it becomes clear that the chemical and physical nature of the dielectric/organic interface impacts extraordinarily on the final device performances, with the possible occurrence of trapping processes, limiting the carrier mobility and contemporarily giving rise to hysteresis and/or bias-stress effects (namely, the change over time of the drain-source IDS current when the device is driven in the accumulation regime) [4,5].
In the recent past, time and spatial resolved scanning Kelvin probe microscopy (SKPM) allowed us to gain detailed information about the basic mechanisms ruling the charge transport in the OFET . In this context, specific attention was devoted to processes like trapping energy , dynamics of trapping and detrapping , charge injection at the electrodes [9,10] and charge recombination mechanisms . The near totality of these reports was focused on p-type (hole-transporting) organic transistors, while very little experimental data concerning SKPM experiments performed on n-type (electron-transporting) OFET are still available today . This occurrence is basically due to the slowness with which organic semiconductors, displaying reliable electron-accumulation effects, have been developed with respect to p-type compounds. Indeed, considerable research has been necessary to synthesize conjugated compounds with contemporarily good self-assembling properties and large electron affinity, thus making the formation of radical anions possible, with sufficient insensitivity to oxidative processes by the ambient gases. In particular, within the last 10 years, Perylene diimide molecules functionalized with cyano groups in the bay regions (PDI_CY) have emerged as a new class of n-type organic semiconductors, with highly stable electrical performances under ambient conditions and effective charge injection from gold electrodes [13,14]. All these features are basically triggered by the low-lying LUMO levels (down to −4.5 eV) of these compounds, which come from the presence of strong electron-withdrawing moieties in their molecular structure.
Among the PDI_CY molecules, N,N0-1H,1H-perfluorobutyl-cyanoperylenediimide (PDIF-CN2) is the compound displaying the highest electron mobility (µ) both in form of thin film [15,16] and single crystal [17,18]. In particular, PDIF-CN2 single-crystal devices have demonstrated µ values up to 6 cm2/Volt·s, band-like transport features (namely, µ increases in a range of temperatures below room temperature)  and negligible bias stress effects . On the other hand, very recently, PDIF-CN2 thin-film transistors have also been shown to be able to operate steadily in aqueous environments, opening a new perspective for the use of this compound in the development of bio-sensing devices .
In this paper, we analyze the morphological and electrical properties of PDIF-CN2 films deposited by Joule evaporation from Knudsen cells on SiO2 substrates with different surface properties. Our results clearly highlight that the self-assembling properties of PDIF-CN2 molecules and the related electrical performances are strongly improved when hydrophobic surfaces are used for the film growth. SKPM was then applied to investigate basic aspects concerning the charge injection process in the PDIF-CN2 films and the poor operational stability of these devices when operated for a prolonged time in the depletion regime.
2. Experimental Section
PDIF-CN2 (Polyera ActivInk N1100) powder was purchased from POLYERA CORPORATION and used without any further purification. Bottom-contact bottom-gate transistors were fabricated by Joule evaporating PDIF-CN2 films on multilayered substrates (Figure 1c) composed of a 500 μm thick layer of highly doped Silicon (Si++) acting both as gate and substrate, a 200 nm thin SiO2 dielectric barrier and, finally, interdigitated source/drain gold electrodes (about 130 nm high) (W/L ratio was fixed to 550 where W is the channel width while L is the channel length). These substrates had been recently used to successfully investigate the electrical response of both p- and n-type molecules [22,23]. Bare substrates were inserted into the evaporation chamber after a basic cleaning by ultrasonic baths in acetone and ethanol. Hexamethyldisilazane (HMDS) treated substrates were obtained with a process lasting 7 days and described in detail in . Water-contact angle (θC) was about 60° for the bare substrates, while θC was about 110° after the HMDS treatment, index of the formation of a more hydrophobic surface. During the evaporation, both bare and HMDS-treated substrates were kept at about Tsub = 90 °C by warming the entire deposition chamber. The deposition rate was 0.5 nm/min and thickness was fixed to about 30 nm for all films analyzed in this work.
The morphological properties of the film surface were investigated by a XE100 Park AFM operating in air with amplitude regulation and oscillating near the cantilever resonance frequency. Images were acquired using silicon-doped cantilevers (resonance frequency around 300 KHz) provided by Nanosensor™. The same microscope was used to carry out the SKPM experiments in air and at room temperature, employing conducting cantilevers (NSC14 Cr/Au, MikroMasch™, resonance frequency around 170 KHz). These measurements were performed in a true non contact mode and dual frequency mode, acquiring contemporarily topography and surface potential profiles. We used an AC sinusoidal signal applied to the conductive tip at a frequency of 17 KHz as an electrostatic probe. First harmonic amplitude of the electrostatic force generated by the potential difference between tip and sample, was considered to perform local surface potential measurement .
The basic transistor characteristic curves were measured in vacuum (10−4 mbar) and darkness using a Janis probe station, connected to a Keithley 2612A Dual-Channel system source-meter instrument. Mobility values were extracted by using the standard MOSFET equations .
3. Results and Discussion
AFM images in Figure 1 report the surface morphology of PDIF-CN2 films evaporated on bare and HMDS-treated substrates, setting the substrate temperature (Tsub) at 90 °C. As shown, the film microstructure is strongly affected by the SiO2 functionalization. In particular, on the bare substrates (Figure 1b), the films show a poor morphological order, being composed of small and rounded grains with a lateral size lower than 100 nm. On the other hand, on HMDS substrates (Figure 1a), the long-range order appears to be considerably improved and the films are characterized by the presence of larger circular islands, with diameters approaching 1 µm in the best cases. Moreover, the film surface displays a well-defined terraced structure with molecular steps close to 2 nm, in agreement with previous reports , which may indicate an alignment of the long axis of the molecules close to the direction perpendicular to the substrate.
The electrical responses measured for the PDIF-CN2 transistors demonstrate a clear correlation between the film morphological properties and the related charge transport performance. First of all, the output curves reported in Figure 2a for a PDIF-CN2 transistor fabricated on bare SiO2 substrate, evidence the occurrence of peculiar and not ideal electrical features. Indeed, although the current behavior in the low VDS (<10 V) region seems apparently linear, all the IDS current curves overlap for VGS higher than 10 V and no further current modulation is observable.
The transfer-curves reported in Figure 2b,c in the linear and saturation regimes, respectively, confirm the poor electrical performances of this class of devices. In particular, besides the presence of a large hysteresis, the transfer-curve in the linear regime again makes clear that, under the application of small VDS voltages and for VGS exceeding 10 V, the device is not able to work properly, since the IDS current reaches a constant value that VGS is no longer able to modulate.
The inability to further control the channel conductance could be ascribed to the detrimental action of very large contact resistances (RC) at the source and drain electrodes, which, because of their weak dependence on VGS, are able to completely dominate the device electrical response in the full accumulation region [26,27,28].
To find an experimental confirmation regarding the RC role in the electrical behavior of these PDIF-CN2 transistors, we carried out SKPM measurements acquiring the potential profile along a line across the channel of the device in operation with VGS grounded and VDS < 10 V (in the specific, 4.5 Volt for not treated device and 7 Volt for the HMDS-treated one). In these driving conditions, the device works in the linear regime (see the output curve at VGS = 0 V in Figure 2a). Figure 3 depicts the measured potential profile in the transistor channel. As shown, while the potential follows the predicted linear behavior in the central part of the channel, a large drop (ΔVS ~ 1.4 V) appears in the film region (about 2 µm long) close the source electrode. In contrast, only a very small drop of about ΔVD ~ 0.1 V can be detected near the drain electrode. This experimental finding reveals that the contact resistance effect is related exclusively to the charge injection process taking place at the source contact and, hence, the RC value can be simply evaluated by dividing (ΔVS) for the current flowing in the channel (IDS = 0.44 μA), achieving about 3 MΩ. This value seems to confirm the idea that the device resistance (usually in the range between 3 and 5 MΩ) measured in the VGS-independent current region (VGS > 10 V in the transfer-curves), as reported in Figure 2, is completely determined by the contact resistance contribution.
In the saturation regime, the role of contact resistance is still clearly visible and the related transfer-curve shows that the IDS slope strongly reduces at increasing VGS. Here, the maximum trans-conductance (gm = ∂IDS/∂VGS) does not exceed 0.8 μS, while a rough estimation of the charge carrier mobility using the MOSFET equations gives a value of 0.002 cm2/Volt·s. Significantly (see the semi-log plots in the inset of Figure 2b,c), the IDS starts flowing in the channel for highly negative VGS, usually comprised between −30 V and −40 V. These VGS values are usually defined as the onset voltages (Von) of the transistor and, for PDIF-CN2 films on bare SiO2 substrates, they are even more negative than the corresponding values measured for the parent compound PDI8-CN2 on the same type of SiO2 surface . This occurrence supports the hypothesis that the negative Von in the PDI_CY-based transistors is related to unintentional charge doping effects given by the interaction between the perylene molecules and water molecules absorbed on the substrate surface [29,30]. On the basis of this concept, the larger the electron affinity of the n-type compounds the more negative will be the Von values.
Due to the improved morphological order, PDIF-CN2 transistors fabricated on HMDS-treated substrates exhibit noticeably improved electrical performances (Figure 2d,e,f), and no potential drop related to the RC effects is appreciable in the SKPM profile reported in Figure 3. In general, the electrical behavior of these devices follows the predictions of the ideal MOSFET model very closely, with the absence of significant hysteresis phenomena.
Mobility values extracted from the curves in the saturation regime are distributed on a homogeneous set of 23 samples, around a mean value of (0.37 ± 0.19) cm2/Volt·s. For one device, we measured a mobility of 0.71 cm2/Volt·s which exceeds the highest value ever reported for an evaporated PDIF-CN2 thin-film transistor . In several cases, the maximum IDS current measured in the saturation regime was higher than 4 mA, while the trans-conductance (gm) came close to the value of 0.2 mS in the best case. It is significant to outline that, according to the experimental data so far reported in literature, mobility values higher than 0.3 cm2/Volt·s have been measured only for films evaporated with Tsub ~ 130 °C . Since this temperature is considerably higher than that (Tsub = 90 °C) used in this work, we consider that the mobility enhancement obtained in our growth conditions is basically due to the quality of the adopted HMDS treatment, making the SiO2 surface highly hydrophobic with a water contact angle θc close to 110°. The impact of the HMDS surface coverage level on the response of the final transistor was already stressed in . Considering the effect of the SiO2 functionalization on the electrical behavior of the analyzed PDIF-CN2 transistors, it can be also concluded that the contact resistance effect observed for the devices fabricated on bare substrates must be mainly related to the poor structural order of the related films. However, a degradation effect of the injection process, associated to the water electrolysis phenomenon occurring on the metallic surface of the gold source electrode and the consequent protonization of the silanol groups on silicon surface closed to the source electrode, cannot be excluded .
Besides the strong influence on mobility and RC effects, HMDS-treatment also provided a Von shift toward the ideal 0 V, similar to what had recently been observed for PDI8-CN2 devices . This evidence is in agreement with the aforementioned discussion about the role of water molecules absorbed on the SiO2 surface in the Von determination. For most of the pristine PDIF-CN2 devices on HMDS, Von was found to range between −10 and −5 V. However, we also observed that these values can be largely modified during the device operation and, in particular, when the transfer-curves are recorded continuously in air for long periods. More specifically, it was found that a prolonged application of negative VGS voltages is able to induce a considerable shift of the Von values towards more negative values. This phenomenon had already been reported for PDI8-CN2 inkjet-printed OFET and is probably common to n-type devices based on organic semiconductors which have a large electron affinity .
While the effect of a continuous application of VGS (positive for n-type OFET) driving the devices into the accumulation regime (i.e., the so-called bias stress effect) can be analyzed in a direct manner through the observation of the IDS(t) behavior, it becomes much more difficult to investigate the time evolution of the device response in the depletion regime, where the IDS current is very low. To approach this interesting task, we have performed SKPM with VGS = −20 V and VDS = 7.5 V along a line across the channel for a time of 10 min. We observe (Figure 4) the presence of a vale in the potential profile inhibiting electron injection for this n-type device. As shown, this potential well gradually tends to disappear over the time scale of ten minutes, causing the device to move from the depletion regime to that of accumulation and the consequent Von shift. We observe that this time is much shorter than the typical time of recovery as already observed for PDI8-CN2 . This phenomenon agrees very well with the predictions of the so-called proton migration model, which was introduced originally for p-type devices  based on SiO2 gate dielectric and then applied, in a modified version, also to n-type transistors . According to this model, under the application of a negative gate voltage, H+ protons, which are present on the SiO2 surface because of acidification produced by water, can rapidly migrate toward the SiO2 bulk. In this way, as observed in our SKPM measurements, the potential well is gradually screened by the increase of the positive charges localized inside the dielectric.
In conclusion, the experimental work here reported demonstrates that the growth of PDIF-CN2 molecules is extremely sensitive on the hydrophobicity degree of the substrate surface. Indeed, while PDIF-CN2 films on HMDS-treated substrates are characterized by a poly-crystalline structure with a long-range morphological order, very small rounded grains with random spatial orientation characterize the microstructure of PDIF-CN2 films deposited on bare SiO2 substrates. Consequently, the electrical response of these latter devices results to be very poor with µ not exceeding 10−3 cm2/Volt·s and huge contact resistance effects. On the contrary, charge carrier mobility extracted for PDIF-CN2 transistors grown on HMDS-treated SiO2 surfaces has a mean value of about (0.37 ± 0.19) cm2/Volt·s. The highest mobility value measured in this work for a PDIF-CN2 transistor was higher than 0.7 cm2/Volt·s, which is among the best results ever reported in literature for n-type thin-film transistors. HMDS treatment was found also to shift toward positive values the Von voltages by more than 20 V, approaching in some cases the ideal Von = 0 V condition. In any case, the time stability of the Von values during the device operation remains an open issue also for the transistors fabricated on hydrophobic SiO2 surfaces. In particular, SKPM measurements reveal that the prolonged (of the order of minutes) application of negative VGS is able to change dramatically the channel conducting properties moving rapidly the device from depletion working regime to saturation regime. Further investigation will be needed to further clarify the mechanisms ruling this operational instability in high-mobility PDIF-CN2 transistors.
The authors would like to acknowledge the work of Mary Longrigg in the English editing and of A. Carella (Department of Chemistry, University of Naples) in functionalizing the SiO2 substrates by HMDS treatment. Financial support from EU FP7 project MAMA Grant Agreement No. 264098 is gratefully acknowledged.
Conflicts of Interest
The authors declare no conflict of interest.
- Bao, Z.; Locklin, J. Organic Field-Effect Transistors; Taylor & Francis: Boca Raton, FL, USA, 2007. [Google Scholar]
- Burgi, L.; Richards, T.J.; Friend, R.H.; Sirringhaus, H. Close look at charge carrier injection in polymer field-effect transistors. J. Appl. Phys. 2003, 94, 6129–6137. [Google Scholar]
- Shehu, A.; Quiroga, S.D.; DʼAngelo, P.; Albonetti, C.; Borgatti, F.; Scorzoni, A.; Stoliar, P.; Biscarini, F. Layered distribution of charge carriers in organic thin film transitors. Phys. Rev. Lett. 2010, 104, 246602. [Google Scholar] [CrossRef]
- Sirringhaus, H. Reliability of organic field-effect transistors. Adv. Mater. 2009, 21, 3859–3873. [Google Scholar] [CrossRef]
- Bobbert, P.A.; Sharma, A.; Mathijssen, S.G.J.; Kemerink, M.; de Leuuw, D.M. Operational stability of orgaic field-effect transistors. Adv. Mater. 2012, 24, 1146–1158. [Google Scholar] [CrossRef]
- Palermo, V.; Palma, M.; Samori, P. Electronic characterization of organic thin films by Kelvin probe force microscopy. Adv. Mater. 2005, 18, 145–164. [Google Scholar] [CrossRef]
- Hallam, T.; Lee, M.J.; Zhao, N.; Nandhakumar, I.; Kemerink, M.; Heeney, M.; McCulloch, I.; Sirringhaus, H. Local charge trapping in conjugated polymers resolved by scanning Kelvin probe microscopy. Phys. Rev. Lett. 2009, 103, 256803. [Google Scholar] [CrossRef]
- Mathijssen, S.G.J.; Kemerink, M.; Sharma, A.; Colle, M.; Bobbert, P.; Janssen, R.A.J.; de Leeuw, D.M. Charge trapping at the dielectric of organic transistors visualized in real time and space. Adv. Mater. 2008, 20, 975–979. [Google Scholar] [CrossRef]
- Ng, T.N.; Silveira, W.R.; Marohn, J.A. Dependence of charge injection on temperature, electric field, and energetic disorder in an organic semiconductor. Phys. Rev. Lett. 2007, 98, 066101. [Google Scholar] [CrossRef]
- Afsharimani, N.; Nysten, B. Electronic properties of dioctylterthiophene-based organic thin-film transistors: A Kelvin probe force microscopy study. Thin Solid Films 2013, 536, 295–301. [Google Scholar] [CrossRef]
- Matyba, P.; Maturova, K.; Kemerink, M.; Robinson, N.D.; Edman, L. The dynamic organic p–n junction. Nat. Mater. 2009, 8, 672. [Google Scholar] [CrossRef]
- Lüttich, F.; Lehmann, D.; Graaf, H.; Zahn, D.R.T.; von Borczyskowski, C. Kelvin-probe studies of n-conductive organic field-effect transistors during operation. Phys. Status Solidi C 2010, 7, 452–455. [Google Scholar] [CrossRef]
- Jones, B.A.; Facchetti, A.; Wasielewski, M.R.; Marks, T.J. Tuning orbital energetics in Arylene Diimide semiconductors: Materials design for ambient stability of n-Type charge transport. J. Am. Chem. Soc. 2007, 129, 15259. [Google Scholar] [CrossRef]
- Jones, B.A.; Facchetti, A.; Wasielewski, M.R.; Marks, T.J. Effects of Arylene Diimide thin film growth conditions on n-channel OFET performance. Adv. Funct. Mater. 2008, 18, 1329–1339. [Google Scholar] [CrossRef]
- Jones, B.A.; Ahrens, M.J.; Yoon, M.H.; Marks, A.F.T.J.; Wasielewski, M.R. High-mobility air-stable n-Type semiconductors with processing versatility: Dicyanoperylene-3,4:9,10-bis(dicarboximides). Angw. Chem. Int. Ed. 2004, 43, 6363–6366. [Google Scholar] [CrossRef]
- Soeda, J.; Uemura, T.; Mizuno, Y.; Nakao, A.; Nakazawa, Y.; Facchetti, A.; Takeya, J. High electron mobility in air for N,N'-1H,1H-perfluorobutyldicyanoperylene carboxydi-imide solution-crystallized thin-film transistors on hydrophobic surfaces. Adv. Mater. 2011, 23, 3681–3685. [Google Scholar] [CrossRef]
- Molinari, A.S.; Alves, H.; Chen, Z.; Facchetti, A.; Morpurgo, A.F. High electron mobility in vacuum and ambient for PDIF-CN2 single-crystal transistors. J. Am. Chem. Soc. 2009, 131, 2462–2463. [Google Scholar] [CrossRef]
- Willa, K.; Hausermann, R.; Mathis, T.; Facchetti, A.; Chen, Z.; Batlogg, B. Form organic single crystals to solution processed thin-films: Charge transport and trapping with varying degree of order. J. Appl. Phys. 2013, 113, 133707. [Google Scholar] [CrossRef]
- Minder, N.A.; Ono, S.; Chen, Z.; Facchetti, A.; Morpurgo, A.F. Band-like electron transport in organic transistors and implication of the molecular structure for performance optimization. Adv. Mater. 2012, 24, 503–508. [Google Scholar] [CrossRef]
- Barra, M.; di Girolamo, F.V.; Minder, N.A.; Lezama, I.G.; Chen, Z.; Facchetti, A.; Morpurgo, A.F.; Cassinese, A. Very low bias stress in n-type organic single-crystal transistors. Appl. Phys. Lett. 2012, 100, 133301. [Google Scholar] [CrossRef]
- Barra, M.; Viggiano, D.; Ambrosino, P.; Bloisi, F.; Girolamo, F.V.D.; Soldovieri, M.V.; Taglialatela, M.; Cassinese, A. Addressing the use of PDIF-CN2 molecules in the development of n-type organic field-effect transistors for biosensing applications. Biochimica et Biophysica Acta 2013, 1830, 4365–4373. [Google Scholar]
- Riccio, M.; Irace, A.; Breglio, G.; Rossi, L.; Barra, M.; di Girolamo, F.V.; Cassinese, A. Current distribution effects in organic sexithiophene field effect transistors investigated by lock-in thermography: Mobility evaluation issues. Appl. Phys. Lett. 2008, 93, 243504–245407. [Google Scholar] [CrossRef]
- Barra, M.; di Girolamo, F.V.; Chiarella, F.; Salluzzo, M.; Chen, Z.; Facchetti, A.; Anderson, L.; Cassinese, A. Transport property and charge trap comparison for N-channel perylene diimide transistors with different air-stability. J. Phys. Chem. C 2010, 114, 20387. [Google Scholar] [CrossRef]
- Sze, S. Physics of Semiconductor Devices; Wiley: New York, NY, USA, 1981. [Google Scholar]
- Weitz, R.T.; Amsharov, K.; Zschieschang, U.; Villas, E.B.; Goswami, D.K.; Burghard, M.; Dosch, H.; Jansen, M.; Kern, K.; Klauk, K. Organic n-channel transistors based on core-cyanated perylene carboxylic diimide derivatives. J. Am. Chem. Soc. 2008, 130, 4637–4645. [Google Scholar] [CrossRef]
- Necliudov, P.V.; Shur, M.S.; Gundlach, D.J.; Jackson, T.N. Modeling of organic thin film transistors of different designs. J. Appl. Phys. 2000, 88, 6594–6597. [Google Scholar] [CrossRef]
- Necliudov, P.V.; Shur, M.S.; Gundlach, D.J.; Jackson, T.N. Contact resistance extraction in pentacene thin film transistors. Solid State Electron. 2003, 47, 259–262. [Google Scholar] [CrossRef]
- Dimitrakopoulos, C.D.; Malenfant, P.R.L. Organic thin film transistors for large area electronics. Adv. Mater. 2002, 14, 99–117. [Google Scholar] [CrossRef]
- Di Girolamo, F.V.; Ciccullo, F.; Barra, M.; Carella, A.; Cassinese, A. Investigation on bias stress effects in n-type PDI8-CN2 thin-film transistors. Org. Electron. 2012, 13, 2281–2289. [Google Scholar]
- Grimaldi, I.A.; Barra, M.; Carella, A.; di Girolamo, F.V.; Loffredo, F.; Minarini, C.; Villani, F.; Cassinese, A. Bias stress effects investigated in charge depletion and accumulation regimes for inkjet-printed perylene diimide organic transistors. Synthetic Met. 2013, 176, 121–127. [Google Scholar]
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).