Studies of Photogenerated Charge Carriers from Donor−Acceptor

J. Phys. Chem. C 2010, 114, 20609–20613

20609

Studies of Photogenerated Charge Carriers from Donor-Acceptor Interfaces in Organic Field Effect Transistors. Implications for Organic Solar Cells† Manohar Rao,‡ Rocio Ponce Ortiz,§ Antonio Facchetti,*,§ Tobin J. Marks,*,§ and K. S. Narayan*,‡ Jawaharlal Nehru Centre for AdVanced Scientific Research, Jakkur P. O., Bangalore 560 064, India, and Department of Chemistry and the Materials Research Center, Northwestern UniVersity, 2145 Sheridan Road, EVanston, Illinois, 60208 ReceiVed: June 3, 2010; ReVised Manuscript ReceiVed: July 30, 2010

Bilayer organic field effect transistor (OFET) structures consisting of an optically active electron donor (D) and an electrically active electron acceptor (A) system offer a quantitative device tool for characterizing photoinduced charge transport processes. Here, we report an investigation of the photoinduced response of a bilayer OFET fabricated from a naphthalene-bis(dicarboximide)-based polymer (N2200) as the n-channel A transport layer and a p-channel regioregular poly-3-hexylthiophene (P3HT) top D layer. This FET exhibits characteristic steady-state spectral response as well as transient profiles as a function of the gate voltage (Vg), yielding valuable information on bulk and interfacial charge transport properties. Thus, the derived N2200 electron mobility is shown to be in good agreement with bulk measurements (significantly greater than that of PCBM), and the N2200/P3HT interface is shown to be a highly efficient structure for charge transfer and free carrier generation. Introduction Polymer field-effect transistors (PFETs) have been extensively studied over the past decade, and impressive progress has been achieved in enhancing device performance.1-4 For example, solution-processable p-channel5 and n-channel6 small-molecule and polymeric materials with high mobility and operational stability under ambient conditions have now been achieved. Ambipolar charge transport processes in PFETs are of great interest from both fundamental scientific and application points of view and have been investigated in bilayer,7 blend,8 and single-component polymer device systems.9 In this contribution, we report that stable n-channel FETs can be used as electrical templates to observe charge transfer and free carrier generation processes introduced by laminating the surface with an electron donor (D) polymer layer. The D polymer can be perturbed by photoexcitation and the ensuing electrical changes, closely monitored. Understanding processes occurring at such D-A interfaces should be of great value in understanding and enhancing analogous processes in organic heterostructure solar cells.10 The charge transport in such organic photovoltaic (OPV) cells is largely governed by interfacial energetic barriers in bilayer devices or by the greater spatial disorder in bulkheterojunction cells. In this regard, D-A interfaces play an important role in determining the free charge carrier yield generated by excitonic processes initiated in the donor medium.11 In bilayer structures, the D-A interface is spatially well-defined, and since the photogenerated charges are confined to either side of the junction, the bimolecular recombination probability is minimized, resulting in large densities of free carriers.12 Information on such systems is particularly relevant to organic solar cells, since there has been recent interest in solution†

Part of the “Mark A. Ratner Festschrift”. * Corresponding authors. E-mails: [email protected], a-facchetti@ northwestern.edu, [email protected]. ‡ Jawaharlal Nehru Centre for Advanced Scientific Research. § Northwestern University.

processed bilayer OPVs, which offer the possibility of greater exciton diffusion lengths in cleaner systems having lower defect concentrations and having long-range interactions to promote charge transfer over large length scales.13 In addition to the above considerations, there is considerable interest in examining interfacial photophysical processes in such systems under differing electrostatic conditions without invasive factors such as injection currents14 in the standard bilayer geometry. Here, exciton dissociation processes at the interface compete with back recombination mechanisms in which chargeseparated states decay back to the ground state, decreasing the free carrier yield.15 Spectroscopic and electrical characterization of such interfacial charge transfer (CT) complexes are valuable in understanding the efficiency of exciton dissociation into free charge carriers. Methods and approaches to simultaneously evaluate FET characteristics and intrinsic solar cell parameters in D-A structures should be interesting and informative. We report here the fabrication of a novel device structure consisting of a D-A bilayer in which the A layer is an underlying n-channel, bottom-gate, top-contact OFET device (Figure 1). Photogenerated charge carriers can be introduced in these devices by photoexciting the D layer, which leads to charge transfer across the D-A heterojunction, resulting in diffusion and drift of the electrons away from the interface and toward the bottom channel region. Consequently, charge carrier generation and charge transport can occur in spatially wellseparated zones. We previously described similar D-A bilayer n-transporting OFETs consisting of [6,6]-phenyl C61-butyric acid methyl ester (PCBM) as the A layer.16 However, in this case, µFET was significantly lower, and PCBM does not have a characteristic optical absorption window. In the present studies, we utilize N2200, a high-stability, high-mobility, A material with a characteristic optical absorption window that complements the D optical response region.6 This complementarity in the D/A optical absorption characteristics enables determination of the

10.1021/jp1051062  2010 American Chemical Society Published on Web 08/19/2010

20610

J. Phys. Chem. C, Vol. 114, No. 48, 2010

Rao et al.

Figure 1. Highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels of P3HT (D layer) and the N2200 (A layer) semiconductors. Structure of the D-A bilayer device fabricated on a BCB gate dielectric with an Al gate and S/D electrodes (L ) 40 µm, W ) 2.0 mm).

contribution of each carrier to the Vg-dependent spectral response of the drain-source current (Ids). The Vg dependence of the photophysical processes occurring at the D-A interface and semiconductor-dielectric interface are characterized using a combination of steady-state and transient pulse excitation measurements.

Figure 2. n-Channel FET response characteristics of a single-layer N2200 device fabricated on a BCB dielectric with Al gate and S/D electrodes (L ) 40 µm, W ) 2 mm) at different values of Vg. Inset shows the transfer characteristics (Ids - Vg) measured at Vds ) 60 V.

Results and Discussion Regioregular poly(3-hexylthiophene), RR-P3HT (American Dye Source Inc.), and naphthalene-bis(dicarboximide) NDI-based polymer poly{[N,N′-bis(2-octyldodecyl)-napthalene-1,4,5,8-bis(dicarboximide)-2,6-diyl-alt-5,5′-(2,2′-bithiophene)}, P(NDI20D-T2(Polyera ActivInk N2200), were used as the D and A species, respectively, with both components exhibiting good film-forming properties. N2200 is an air-stable, optically absorbant n-type polymer exhibiting excellent OTFT characteristics under ambient in combination with vapor-deposited Au contacts and various polymeric gate dielectrics.6 High mobility and high on-off ratios are observed with topcontact N2200 TFTs on Si/SiO2 substrates with gold source/ drain (S/D) contacts (See Figure S1, Supporting Information).6 The N2200 layers were deposited by spin-coating from 10 mg/ mL solutions of N2200 in dry CHCl3 (spin-coating conditions: 1500 rpm for 30 s) on hexamethyldisilazane (HMDS)-treated p-doped Si (001) wafers having a 300 nm thermally grown SiO2 dielectric layer. Trimethylsilation of the plasma-cleaned Si/SiO2 surfaces was carried out by exposing the silicon wafers to HMDS vapor at 25 °C in a closed, air-free container under nitrogen for one week. After spin-coating the N2200 semiconductor layer, films were annealed at 110 °C for 2 h in vacuum. The capacitance of the 300 nm SiO2 gate insulator was found to be 10-8 F cm-2. Characterization of the present devices was performed under vacuum in a custom high-vacuum probe station (∼10-6 Torr) using a Keithley semiconductor parameter analyzer. The saturation regime µFET and VTH parameters were extracted using the standard expression, Ids ) WµCi/L (Vg - Vth)2 based on the gradual channel approximation valid for OFETs, where µ is the mobility, Ci is the gate dielectric areal capacitance, Vg is the gate voltage, and Vth is the threshold voltage. The N2200 thin films on HMDS-treated substrates and using vapordeposited gold S/D contacts exhibit an average µFET of 0.033 cm2 V-1 s-1 and Vth of 20.0 V (Figure S1). Top contact FETs fabricated with N2200 films (thickness ∼ 200 nm) on a cross-linked benzocyclobutene polymer dielectric (BCB) with vapor-deposited Al S/D and gate electrodes exhibit n-type transport, as indicated from the output and transfer characteristics shown in Figure 2. Characterizing several devices and fitting the data to the above response model yields µFET

Figure 3. n-Channel FET response characteristics of a bilayer P3HT/ N2200 device fabricated on a BCB dielectric layer with Al gate and S/D electrodes (L ) 40 µm, W ) 2 mm) at different values of Vg. Inset: Transconductance characteristics of the bilayer FET.

∼10-3 cm2 V-1 s-1 and Vth ) 0.0-20 V. The large Vth range can reasonably be attributed to imperfections in the soft dielectric medium generating traps at the dielectric-semiconductor interface. P3HT/N2200-based bilayer structures (Figure 1) were fabricated using the nonsolvent casting methodology reported previously.16 Thus, P3HT films are suspended in water from chloroform solutions and then lifted off using an elastomeric polydimethylsiloxane substrate and subsequently laminated onto Al/BCB substrates coated with N2200 films. The resulting laminated films (thicknesses around 30-50 nm) were then annealed at 150 °C for 30 min in an N2-filled glovebox. For bilayer device structures, vapor-deposited Al was used as the S/D electrodes to ensure acceptable injecting contacts with the N2200 layer, while forming blocking contacts with the P3HT layer. It was noticed that in a coplanar configuration, the Al electrode exhibits blocking characteristics with respect to P3HT in both dark and light conditions, in agreement with previously reported observations, which also demonstrated that injection from Al could be observed only upon introduction of electron transporting acceptor species in the polymer matrix.17 FETs based on N2200 with Al as the S/D electrodes exhibit n-channel transfer response, and introduction of an additional (D) polymer layer on the N2200 introduces only a marginal shift in response characteristics (Figure 3). Thus, the presence of the P3HT D layer does not adversely affect the n-transport characteristics of the N2200 (A)-based FET.18 Importantly, the magnitude of

Charge Carriers in Field Effect Transistors

J. Phys. Chem. C, Vol. 114, No. 48, 2010 20611

Figure 4. Light intensity modulated photocurrent, Iph(λ), measured as a function of wavelength using lock-in techniques for a bilayer N2200/ P3HT OFET at the indicated Vg values (Vds ) 60 V; ω ) 17 Hz). The inset shows the Iph(λ) response for an N2200-only FET device.

Ids remains unchanged in all of the present donor-laminated acceptor FETs and corresponds to the single-layer N2200 FET value of µe. Moreover, the TFT on/off ratios of the bilayer devices in the dark are similar to those of single-layer N2200 transistors. The electrical characteristics of the pristine single-layer N2200 FETs change upon photoexcitation in a manner analogous to changes in p-channel FETs and can also be analyzed within the framework of the gradual channel approximation.19 Thus, the changes can be summarized as: (i) The spectral response of the intensity-modulated drain-source Iph(λ) current largely tracks the absorption profile of N2200, as indicated in Figure 4 (inset), which consists of two distinct maxima (also see the Supporting Information, Figures S3 and S4). (ii) The Ids changes with respect to intensity in the ON and OFF states can be described as photoconductive (linear responsivity ∼ 60 µA/W) and photovoltaic types (logarithmic response), respectively.In the case of the bilayer FET, the N2200 and P3HT layers have significant and complementary optical absorption features, which are reflected in the spectral response of the intensity-modulated Ids (Figure 4). Note that the photogenerated negative charge carriers can originate either in N2200 layer bulk or at the P3HTN2200 interface, which ultimately contributes to the total channel current as observed in Figure 4. The spectral region corresponding to the N2200 absorption, 300 < λ < 450 nm, is observed to have a stronger Vg dependence than the carriers originating from P3HT absorption region, 450 < λ < 600 nm, as expected due to the proximity of the dielectric to the N2200 layer. Since the measured signal results primarily from Ids confined to the channel at the N2200/dielectric interface, the negative free carriers constitute the dominant species. The electrical transport process arising from the photogenerated positive carriers is a slower trap-mediated process that finally is involved in a slow recombination process.20 The differences in the carrier dynamics can also be assayed by measuring an open circuit voltage developed at the gate electrode with respect to shorted S/D electrodes.21 A signal in the range +70 mV is observed upon illuminating the device (λ 532 nm, 100 µW/ cm2), indicative of increased n-carrier density in the channel. The Ids(t) response measured upon single pulse excitation is very informative. The advantage of transient techniques over stationary I-V measurements is that one can study the dynamics of the processes that otherwise are integrated over time. Transient current measurements introduced by pulsing Vds have been used in studying FETs to evaluate device parameters, such as contact resistance and carrier mobility in polymers.22 The

Figure 5. Transient response of a uniformly illuminated bilayer N2200/ P3HT OFET to pulse photoexcitation (10-8 s, 532 nm, 140 µW/cm2) averaged over 100 pulses, repetition rate )100 Hz, at different Vg values (Vds ) 60 V), measured using a 1 GHz Le-Croy digital oscilloscope across a 1 MΩ resistor. Insets: (a) Schematic of the different time regimes on a semilog plot. (b) Reciprocal of rise time versus Vg, indicating linear dependence of 1/τtr on Vg in the enhancement regime.

present approach is based on studying transient Ids induced by excess charge-carriers introduced by single pulse photoexcitation at the D-A interface. It is important to distinguish the present technique from conventional time-of-flight (TOF) measurements employed for extracting mobility in thick polymer films sandwiched between electrodes. In the traditional TOF method, the charge carrier motion appears in the form of photocurrent, where the Iph(t) represents the average charge carrier concentration and the mean velocity at a given time, t, and the measured current density is in the direction of light propagation. In contrast, in the present FET geometry, the excess carriers do not contribute to the current (Ids) significantly until they are in the vicinity of the channel. We rely on the dominant signatures in Ids(t) to describe the underlying processes. The inflection point in the Ids(t) profile observed is taken as the carrier transit time, and assuming a diffusion model or a drift model with uniform electric field (E), the mobility can be estimated. A typical Ids(t) profile upon a single-pulse, uniform photoexcitation of a N2200/P3HT bilayer device as observed in Figure 5 exhibits the following features: (i) A distinctive delay period prior to an increase in Ids following the 10 ns pulse, denoted by a time scale, τ. The delay period is observed to be independent of Vg and can be attributed to the initial diffusion process after the interfacial charge transfer process. This appearance of delay is unlike that of conventional TOF profiles in which the current rise corresponds to the excitation pulse. (ii) Ids(t) gradually rises beyond the τonset, indicative of a current pathway for the photogenerated carriers. The electrical transport of the carriers in these systems can be described in the Gaussian disorder formalism. The initial narrow charge carrier pulse broadens as it propagates through the N2200 bulk. An inflection point in the increasing Ids(t) at τtr indicates the culmination of the carrier pulse in the channel. We attribute this regime in Ids(t) as a measure of the transit time τtr. The observation of the independence of τtr with respect to Vds (as compared with the linear dependence of tpeak on Vds) and the linear response of 1/τtr with

20612

J. Phys. Chem. C, Vol. 114, No. 48, 2010

respect to Vg suggests τtr as a more appropriate representation of a transit time. Note that 1/τtr scales linearly with the field E, indicating the independence of µ on E in the limited low-E regime, as observed in the inset (b) of Figure 5. Note that τtr is also independent of the intensity of the photoexciation pulse. The variation of the field within the bulk semiconductor region is limited to a considerable extent. (iii) The photocarriers equilibrate as they approach the zone of Vg influence (dielectricsemiconductor interface), which further modifies the dispersive transient profile. The rate of increase in Ids(t) progressively slows, and Ids(t) goes through a broad maximum before it gradually decays to the original state. The transit time (τtr) can be used to estimate the bulk mobility of N2200. Following the carrier generation process, the observed photocurrent response pulse, which is delayed from the optical pulse, can be initially modeled as a pure diffusion process. The photocurrent essentially then is the electron concentration gradient times the diffusion coefficient, D, for the electrons and the elementary charge, q. If τtr is taken to be the representative time constant signifying a net diffusion period for carrier traversal from the D-A interface to the channel, then τTR ∼ L2/D, indicating a value of D ∼ 8 × 10-5 cm2/s (for L ) 200 nm, and τtr ) 5 × 10-6 s) and assuming the Einstein relation, µ ∼ 3 × 10-3 cm2/(V s). If an exclusive drift model is used to describe the Ids(t), then in the absence of information of the field profile within the device, we can expect the vertical field to be in the range of the built-in-field. Upon assuming the value of E to be similar to the magnitude derived from (∆V)G,S/D ∼ 70 mV from short-circuit voltage measurements (i.e., E ∼ 4 × 104 V/cm), then µ is estimated to be ∼2.5 × 10-3 cm2/(V s). It is interesting to observe a similar value from both of these different models. A complete drift-diffusion model is more appropriate to describe the photogenerated carrier motion away from the interface and could possibly simulate the entire Ids(t) without segmentation into different time zones based on different dominant processes. Recent measurements of N2200 bulk electron transport in N2200 using both time-of- flight and electron-only current measurements yield µe ∼ 2.5-5.0 × 10-3 cm2/(V s).23 Thus, the present data are nicely in agreement with these independent measurements but permit mobility estimates for far thinner samples. The Ids(t) values are inherently low field measurements, since the voltages utilized for drawing the carriers to the channel are significantly lower than Vg. This value of transverse field across the N2200 layer can be justified by noting that the Vg drop is largely across the dielectric, and the potential difference across the photocarrier transport region constitutes a fraction of the actual bias. The transit time decreases with the increasing bias voltage, as expected. It may be possible to confirm the magnitude of the vertical electric field with numerical simulation procedures and scanning potentiometry experiments24 using a localized region for the incident light and scanning it across the channel. Note that it may not be appropriate to compare the absolute value of µFET (∼3 × 10-2 cm2/(V s)) to the present estimate using a simple diffusion model or a drift model with an assumed value of E. However, this range of µbulk points to three-dimensional order in the N2200 thin films. The Ids(t) obtained at different Vg values can be utilized to reconstruct the dynamic transconductance (Ids - Vg)(t). Figure 6 depicts the Ids(Vg) at different t values, with t ) 0 corresponding to the dark state. The Ids(Vg) at different t’s is also a measure of the trajectory of the photocarriers. The characteristic square law Ids(Vg) dependence evolves to a more exponential dependence beyond t > 10-6 s. The effective ON-OFF ratio of the

Rao et al.

Figure 6. Dynamic transconductance Ids - Vg responses for bilayer N2200/P3HT OFETs obtained from transient measurements at increasing time intervals. Data were derived from Ids(t) at different Vg values, as shown in Figure 5.

device in this interim period (10-6 s < t < 10-4 s) increases along with a shift in Vth. These trends can be qualitatively explained on the basis of the carrier density increase. It is instructive to compare the results on the present bilayer N2200/P3HT OFETs with those on bilayer PCBM/P3HT OFETs. The photoinduced Ids is orders of magnitude larger in the present bilayer N2200/P3HT OFETs than in bilayer PCBM/ P3HT OFETs.16 The net excess charge upon a single-pulse photoexcitation for N2200/P3HT devices can be estimated by the ∫Ids(t) dt. For a typical response at Vg ) 60 V, the integral corresponds to 2.6 × 1013/cm3, which corresponds to charge collection efficiency of at least 0.01 charge/photon, as compared with 10-3 charge/photon in the PCBM/P3HT devices for similar device and experimental parameters. The transit time in N2200 layer is also observed to be consistently smaller than in the PCBM layers, with all other device parameters being equivalent. This feature points to a bulk mobility that is significantly higher in N2200 than in PCBM-C60. Other factors contributing to the significant photoinduced Ids in N2200/P3HT bilayer OFETs could arise from more efficient primary steps in the C-T and charge separation processes. The origin of this trend could arise from molecular/macromolecular architectural compatibility of the electronic structures of the D and A systems, affording an optimum interface or from mechanical aspects of an optimally structured bilayer interface in the case of P3HT/N2200 versus P3HT/PCBM. These results suggest the possibility that such optimized bilayer structures may be more suitable than bulk heterostructures for P3HT/ N2200 OPVs. Conclusions In summary, we have studied the significant effects arising from steady and pulsed photoexcitation of N2200-based OFETs having an additional P3HT donor top layer. The photoexcitation was tuned to excite either the donor-P3HT layer or the acceptorN2200 transport layer, and the ensuing changes in the OFET characteristics were characterized. Importantly, the results from photoexciting the P3HT layer then can be used to estimate, simultaneously, both the bulk and field effect mobility of the N2200 layer. This novel approach of studying the light-induced response characteristics of N2200/P3HT heterostructures in the OFET geometry reveals promising bilayer strategies for polymer solar cell design. Acknowledgment. Dedicated to Mark Ratner, stimulating collaborator, colleague, and kind friend. We acknowledge the

Charge Carriers in Field Effect Transistors support of DST, the Government of India and Northwestern University through the bilateral JNCASR-NU program for Research Excellence in Advanced Materials, AFOSR (FA955008-01-0331), and the NSF-MRSEC program through the Northwestern Materials Research Center (DMR-0520513). We also thank the Indo-US Science & Technology Forum for supporting the collaborative work through the joint center award to JNCASR and Northwestern University. K.S.N. acknowledges DAE and the Government of India for partial funding, and R.P.O. acknowledges funding from the European Community’s Seventh Framework Programme under Grant Agreement 234808. Supporting Information Available: Electrical characterization of single-layer N2200 devices before and upon photoexcitation. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) (a) Baklar, M.; Woebkenberg, P. H.; Sparrowe, D.; Goncalves, M.; McCulloch, I.; Heeney, M.; Anthopoulos, T.; Stingelin, N. J. Mater. Chem. 2010, 20, 1927. (b) Xia, Y.; Zhang, W.; Ha, M.; Cho, J. H.; Renn, M. J.; Kim, C. H.; Frisbie, C. D. AdV. Funct. Mater. 2010, 20, 587. (c) Voigt, M. M.; Guite, A.; Chung, D.-Y.; Khan, R. U. A.; Campbell, A. J.; Bradley, D. D. C.; Meng, F.; Steinke, J. H. G.; Tierney, S.; McCulloch, I.; Penxten, H.; Lutsen, L.; Douheret, O.; Manca, J.; Brokmann, U.; Soennichsen, K.; Huelsenberg, D.; Bock, W.; Barron, C.; Blanckaert, N.; Springer, S.; Grupp, J.; Mosley, A. AdV. Funct. Mater. 2010, 20, 239. (d) Ortiz, R. P.; Facchetti, A.; Marks, T. J. Chem. ReV. 2010, 110, 205. (2) (a) Liu, C.; Sirringhaus, H. J. Appl. Phys. 2010, 107, 014516/1. (b) Rieger, R.; Beckmann, D.; Pisula, W.; Steffen, W.; Kastler, M.; Muellen, K. AdV. Mater. 2010, 22, 83. (c) Zhan, X.; Tan, Z.; Zhou, E.; Li, Y.; Misra, R.; Grant, A.; Domercq, B.; Zhang, X.-H.; An, Z.; Zhang, X.; Barlow, S.; Kippelen, B.; Marder, S. R. J. Mater. Chem. 2009, 19, 5794. (d) Souharce, B.; Kudla, C. J.; Forster, M.; Steiger, J.; Anselmann, R.; Thiem, H.; Scherf, U. Macromol. Rapid Commun. 2009, 30, 1258. (e) Zhang, W.; Smith, J.; Hamilton, R.; Heeney, M.; Kirkpatrick, J.; Song, K.; Watkins, S. E.; Anthopoulos, T.; McCulloch, I. J. Am. Chem. Soc. 2009, 131, 10814. (f) Kong, H.; Chung, D. S.; Kang, I.-N.; Park, J.-H.; Park, M.-J.; Jung, I. H.; Park, C. E.; Shim, H.-K. J. Mater. Chem. 2009, 19, 3490. (g) Guo, X.; Kim, F. S.; Jenekhe, S. A.; Watson, M. D. J. Am. Chem. Soc. 2009, 131, 7206. (h) Zou, Y.; Gendron, D.; Badrou-Aich, R.; Najari, A.; Tao, Y.; Leclerc, M. Macromolecules 2009, 42, 2891. (i) Yang, C.; Cho, S.; Chiechi, R. C.; Walker, W.; Coates, N. E.; Moses, D.; Heeger, A. J.; Wudl, F. J. Am. Chem. Soc. 2008, 130, 16524. (3) (a) Zaumseil, J.; Sirringhaus, H. Chem. ReV. 2007, 107, 1296. (b) Sundar, V. C.; Zaumseil, J.; Podzorov, V.; Menard, E.; Willett, R. L.; Someya, T.; Gershenson, M. E.; Rogers, J. A. Science 2005, 303, 1644. (c) Stutzmann, N. S. S.; Smits, E. C. P.; Wondergem, H.; Tanase, C.; Blom, P. W. M.; Smith, P.; De Leeuw, D. M. Nat. Mater. 2005, 4, 601. (4) (a) Hamilton, R.; Heeney, M.; Anthopoulos, T.; McCulloch, I. Org. Electron. 2010, 393. (b) Daniele, B.; Horowitz, G. AdV. Mater. 2009, 21, 1. (c) Hutchinson, G. R.; Ratner, M. A.; Marks, T. J. J. Am. Chem. Soc. 2005, 127, 16866. (d) Hutchinson, G. R.; Ratner, M. A.; Marks, T. J. J. Am. Chem. Soc. 2005, 127, 2339. (5) (a) Smith, J.; Hamilton, R.; McCulloch, I.; Heeney, M.; Anthony, J. E.; Bradley, D. D. C.; Anthopoulos, T. Synth. Met. 2009, 159, 2365. (b) Kang-Jun, B.; Dongyoon, K.; Dong-Yun, K.; Jae Bon, K.; In-Kyu, Y.; Won San, C.; Yong-Young, N. Thin Solid Films 2010, 518, 4024. (c) Lee, S. S.;

J. Phys. Chem. C, Vol. 114, No. 48, 2010 20613 Kim, C. S.; Gomez, E. D.; Purushothaman, B.; Toney, M. F.; Wang, C.; Hexemer, A.; Anthony, J. E.; Loo, Y.-L. AdV. Mater. 2009, 21, 3605. (6) (a) Zhihua, C.; Yan, Z.; He, Y.; Facchetti, A. J. Am. Chem. Soc. 2009, 131, 8. (b) Yan, H.; Chen, Z.; Zheng, Y.; Newman, C.; Quinn, J. R.; Do¨tz, F.; Kastler, M.; Facchetti, A. Nature 2009, 457, 679–U1. (7) (a) Shi, J. W.; Wang, H. B.; Song, D.; Tian, H. K.; Geng, Y. H.; Yan, D. H. AdV. Funct. Mater. 2007, 17, 397. (b) Shinuk, C.; Jonathan, Y.; Jin Young, K.; Kwanghee, L.; Heeger, A. J.; Sangyun, L. Appl. Phys. Lett. 2008, 92, 063505. (c) Liu, C.; Sirringhaus, H. Org. Electron. 2010, 11, 558. (d) Capelli, R.; Toffanin, S.; Generali, G.; Usta, H.; Facchetti, A.; Muccini, M. Nat. Mater. 2010, 9, 496. (e) Capelli, R.; Dinelli, F.; Loi, M. A.; Murgia, M.; Zamboni, R.; Muccini, M. J. Phys.: Condens. Matter 2006, 18, S2127. (f) Loi, M. A.; Rost-Bietsch, C.; Murgia, M.; Karg, S.; Riess, W.; Muccini, M. AdV. Funct. Mater. 2006, 16, 41. (g) Dinelli, F.; Capelli, R.; Loi, M. A.; Murgia, M.; Muccini, M.; Facchetti, A.; Marks, T. J. AdV. Mater. 2006, 18, 1416–1420. (8) (a) Szendrei, K.; Jarzab, D.; Chen, Z.; Facchetti, A.; Loi, M. A. J. Mater. Chem. 2010, 20, 1317. (b) Babel, A.; Wind, J. D.; Jenekhe, S. A. AdV. Funct. Mater. 2004, 14, 891. (9) (a) Zaumseil, J.; Donley, C. L.; Kim, J. S.; Friend, R. H.; Sirringhaus, H. AdV. Mater. 2006, 18, 2708. (b) Gwinner, M. C.; Khodabakhsh, S.; Giessen, H.; Sirringhaus, H. Chem. Mater. 2009, 21, 4425. (c) Usta, H.; Risko, C.; Wang, Z.; Huang, H.; Deliomeroglu, M. K.; Zhukhovitskiy, A.; Facchetti, A.; Marks, T. J. J. Am. Chem. Soc. 2009, 131, 5586. (d) Burgi, L.; Turbiez, M.; Pfeiffer, R.; Bienewald, F.; Kirner, H.-J.; Winnewisser, C. AdV. Mater. 2008, 20, 2217. (e) Kim, F. S.; Guo, X.; Watson, M. D.; Jenekhe, S. A. AdV. Mater. 2010, 22, 478. (10) (a) Tracey, M. C.; Durrant, J. R. Chem. ReV. 2010In press. (b) Serap, G.; Neugebauer, H.; Sariciftci, N. S. Chem. ReV. 2007, 107, 1324. (c) Yoon, W.-J.; Jung, K.-Y.; Liu, J.; Duraisamy, T.; Revur, R.; Teixeira, F. L.; Sengupta, S.; Berger, P. R. Sol. Energy Mater. Sol. Cells 2010, 94, 128. (d) Potscavage, W. J., Jr.; Sharma, A.; Kippelen, B. Acc. Chem. Res. 2009, 42, 1758. (e) Kippelen, B.; Bre´das, J.-L. Energy EnViron. Sci. 2009, 2, 251. (f) Yoon, W. J.; Berger, P. R. Appl. Phys. Lett. 2008, 92, 013306. (11) (a) De Lucia, F. C.; Gustafson, T. L.; Wang, D.; Epstein, A. J. Phys. ReV. B 2002, 65, 235204. (b) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science 1995, 270, 1789. (12) Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E. AdV. Mater. 2007, 19, 1551. (13) Dong, H. W.; Hang, K. L.; Dae-Geun, C.; Jong, H. P.; Park, O. O. Appl. Phys. Lett. 2009, 95, 043505. (14) Barth, S.; Ba¨ssler, H.; Rost, H.; Horhold, H. H. Phys. ReV. B 1997, 56, 3844. (15) Lemaur, V.; Steel, M.; Beljonne, D.; Bredas, J. L.; Cornil, J. J. Am. Chem. Soc. 2005, 127, 6077. (16) Rao, M.; Narayan, K. S. Appl. Phys. Lett. 2009, 95, 183306. (17) Cho, S.; Yuen, J.; Kim, J. Y.; Lee, K.; Heeger, A. J. Appl. Phys. Lett. 2006, 89, 153505. (18) To verify the active channel-transport region in the bilayer FET, the output characteristics for drain current in positive and negative drainsource voltage were measured, (Figure S2, Supporting Information). The results show that Ids exhibits saturation current in positive drain voltage and diverging Ids in negative voltage, indicative of unipolar n-channel FET. (19) Narayan, K. S.; Kumar, N. Appl. Phys. Lett. 2001, 79, 1891. (20) Dutta, S.; Narayan, K. S. Phys. ReV. B 2003, 68, 125208. (21) Dutta, S.; Narayan, K. S. AdV. Mater. 2004, 16, 2151. (22) Weis, M.; Lin, J.; Taguchi, D.; Manaka, T.; Iwamoto, M. J. Phys. Chem. C 2009, 113, 18459. (23) Steyrleuthner, R.; Schubert, M.; Jaiser, F.; Blakesley, J. C.; Chen, Z.; Facchetti, A.; Neher, D. AdV. Mater. 2010, 22, 2799. (24) Seshadri, K.; Frisbie, C. D. Appl. Phys. Lett. 2001, 78, 993.

JP1051062