Letter pubs.acs.org/NanoLett
Semiconducting Polymers with Nanocrystallites Interconnected via Boron-Doped Carbon Nanotubes Kilho Yu,†,‡ Ju Min Lee,∥ Junghwan Kim,†,‡ Geunjin Kim,†,‡ Hongkyu Kang,†,‡ Byoungwook Park,†,‡ Yung Ho Kahng,§ Sooncheol Kwon,†,‡ Sangchul Lee,† Byoung Hun Lee,† Jehan Kim,⊥ Hyung Il Park,∥ Sang Ouk Kim,*,∥ and Kwanghee Lee*,†,‡,§ †
Department of Nanobio Materials and Electronics, School of Materials Science and Engineering, ‡Heeger Center for Advanced Materials (HCAM), and §Research Institute for Solar and Sustainable Energies (RISE), Gwangju Institute of Science and Technology (GIST), Gwangju 500-712, Republic of Korea ∥ Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Department of Materials Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Republic of Korea ⊥ Pohang Accelerator Laboratory (PAL), Pohang University of Science and Technology (POSTECH), Pohang 790-784, Republic of Korea S Supporting Information *
ABSTRACT: Organic semiconductors are key building blocks for future electronic devices that require unprecedented properties of low-weight, flexibility, and portability. However, the low charge-carrier mobility and undesirable processing conditions limit their compatibility with low-cost, flexible, and printable electronics. Here, we present significantly enhanced field-effect mobility (μFET) in semiconducting polymers mixed with boron-doped carbon nanotubes (B-CNTs). In contrast to undoped CNTs, which tend to form undesired aggregates, the B-CNTs exhibit an excellent dispersion in conjugated polymer matrices and improve the charge transport between polymer chains. Consequently, the B-CNT-mixed semiconducting polymers enable the fabrication of high-performance FETs on plastic substrates via a solution process; the μFET of the resulting FETs reaches 7.2 cm2 V−1 s−1, which is the highest value reported for a flexible FET based on a semiconducting polymer. Our approach is applicable to various semiconducting polymers without any additional undesirable processing treatments, indicating its versatility, universality, and potential for high-performance printable electronics. KEYWORDS: Semiconducting polymer, carbon nanotube, polymer nanocrystallite, nanocomposite, field-effect transistor, room-temperature process rganic field-effect transistors (OFETs) based on semiconducting polymers have been intensively investigated for future ubiquitous devices in which unprecedented device features such as low-weight, flexibility, and portability are required.1−4 Moreover, the solution processability of the semiconducting polymers enables the cost-effective, large-scale fabrication of OFETs on plastic substrates via high-throughput printing technologies.5−7 Nevertheless, their poor charge-carrier mobility (μ) has hindered the realization of their potential advantages.8 Because of remarkable advances in the development of processing techniques, the state of the art of solutionprocessed semiconducting polymers recently reached a remarkable stage9 of μ ∼ 36 cm2 V−1 s−1. However, because these techniques still require undesirable processing treatments, such as thermal annealing at high temperature (T)10,11 (typically, T ≥ 200 °C) and/or macroscopic alignment processes9,12−15 using mechanically grooved inorganic substrates, they are not compatible with the concept of “flexible and printable electronics”
O
© XXXX American Chemical Society
using plastic substrates. Fabricating high-μ semiconducting polymer layers on plastic substrates from a solution process at roomtemperature (RT) has been a formidable challenge. One promising approach involves introducing carbon nanotubes (CNTs) into semiconducting polymers.16 CNTs possess all the outstanding properties as a μ-enhancer in OFETs, including a high-μ, good mechanical flexibility, and a high aspect ratio.17 CNTs in semiconducting polymers act as ballistic conduction bridges18 between the polymer chains and/or induce a local alignment of the polymer chains located in the vicinity of CNTs,19,20 thereby significantly improving the μ value of the polymers. However, because CNTs tend to easily aggregate via van der Waals interaction,21,22 most CNT-based OFETs have suffered from poor device performance and electrical breakdown.23 Although many different solutions for overcoming this Received: September 17, 2014
A
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 1. Comparison of the morphology of the P3HT/CNT composite films. (a) Chemical structure of P3HT and schematic illustration of the formation of the P3HT/U-CNT or B-CNT composite films. (b) TEM image of a P3HT/U-CNT composite film. (c) TEM image of a P3HT/B-CNT composite film. Note that in b and c, the concentration of CNTs in the composite films is 10 wt % with respect to P3HT. Scale bars, 5 μm (lowmagnification images) and 500 nm (high-magnification images).
aggregation problem have been explored including surface functionalization24−26 and in situ polymerization,27−29 these efforts have thus far failed to satisfy the requirements for highperformance flexible FETs. In this work, we report for the first time significantly enhanced field-effect μ in conjugated polymers mixed with a small amount of boron-doped CNTs (B-CNTs). In contrast to undoped CNTs (U-CNTs), which tend to form undesired aggregates, the B-CNTs exhibit excellent dispersion in a conjugated polymer matrix. The substitutional doping of B atoms also effectively increases the work function of CNTs and facilitates holetransport between the semiconducting polymers and CNTs. More importantly, the well-dispersed CNTs on the molecular scale induce a local crystallization of the polymer chains even at RT, enabling the fabrication of high performance flexible OFETs with a top-gate geometry. The field-effect μ of our devices, which were fabricated on plastic substrates via a solution process at RT, reaches 7.2 cm2 V−1 s−1, which has never been previously achieved in a flexible FET based on a semiconducting polymer. The B-CNTs were processed using commercially available multiwalled CNTs with an average length of ∼1 μm and a diameter of 10−15 nm. After the CNTs were purified, B-doping was performed via thermal treatment under an Ar/NH3 gas stream with vaporized B2O3 powder,30 as explained in detail in the Supporting Information. We subsequently dispersed B-CNTs in o-dichlorobenzene (o-DCB) at various concentrations using ultrasonication. Poly(3-hexylthiophene) (P3HT) was used as a representative conjugated polymer material to explore the effect of the CNTs in this study.31 P3HT was dissolved in chloroform (CF) and then simply mixed with the CNT solutions, as described in detail in the Supporting Information. Both U-CNT and B-CNT were mixed well with P3HT in a solution state. However, during thin-film formation via spin-casting of the composite solutions, the U-CNTs began to aggregate (Figure 1a), as clearly observed in the transmission electron microscopy (TEM) images in Figure 1b. Even with a small concentration of U-CNTs (1 wt % with respect to P3HT), the U-CNTs formed aggregates with an average size of ∼2 μm in the P3HT/U-CNT composite films (see Supporting Information, Figure S1). When the concentration of U-CNTs was
increased to 5 and 10 wt %, the aggregates became larger (2−50 μm) over the entire films. These large aggregates are known to hinder the effective use of CNTs as conduction bridges in conjugated polymers and eventually cause device breakdown. In contrast, the B-CNTs exhibit a superb dispersion in the polymer matrix even at high concentration as evident in Figure 1c. This difference originates from the strong electrostatic repulsion with neighboring B-CNTs induced by the local positive charges of B atoms, which significantly reduce the van der Waals interactions among B-CNTs during film formation. The uniformly dispersed B-CNTs act more effectively as conducting bridges between the polymer chains than the aggregated U-CNTs on the molecular scale. To investigate the charge transport properties of the P3HT/ CNT composite films, we fabricated typical polymer-based FETs with a structure of bottom-gate and bottom-contact (BGBC) FETs (Figure 2a). Figure 2b shows the transfer characteristics of P3HT with or without 1 wt % CNTs. Both FETs with U-CNTs or B-CNTs exhibit an improved on-current (Ids) of the devices. However, whereas the average saturation μ of the P3HT/ U-CNT device is ∼0.012 cm2 V−1 s−1, which is 6-fold greater than that of pristine P3HT (∼0.0019 cm2 V−1 s−1), the P3HT/B-CNT exhibits a much higher μ of ∼0.044 cm2 V−1 s−1, which is 23-fold greater than that of pristine P3HT. The performance enhancement of B-CNT over U-CNT was also observed in the μ and on/off ratio profiles of the FET devices as a function of the CNT concentration (Figure 2c,d and Supporting Information, Figure S2). The P3HT/B-CNT device exhibits rapidly increasing μ values even with a small amount of B-CNTs (0.1−1 wt %) and then saturates with a slow increase until 10 wt % (see Table S1 in Supporting Information). In contrast, the device with U-CNTs starts to decrease again with U-CNT concentrations greater than 5 wt %. This performance degradation is attributed to the large U-CNT aggregates in the P3HT/U-CNT devices. Because we used multiwalled CNTs, the U-CNT aggregates in the FET channel are dominated by a metallic nature and appear to obstruct the field-effect behavior induced by the gate-source voltage (Vgs). Therefore, the P3HT/ U-CNT FETs with large U-CNT aggregates exhibit very poor on/off ratios and output characteristics (Figure 2d and B
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 2. Charge transport characteristics of the P3HT/CNT composite films. (a) Cross-sectional diagram of the P3HT/CNT composite FET. The SiO2 with 200 nm thickness serves as the gate dielectric layer (capacitance Ci = 17.2 nF cm−2) and the Au with a Ni adhesion layer serves as the source and drain contacts with a channel length and width of 10 and 1000 μm, respectively. (b) The transfer characteristics of P3HT (with or without 1 wt % CNTs with respect to P3HT) FETs under a drain-source voltage (Vds) of −60 V. (c) Hole μ of the P3HT/CNT composite FETs at various CNT concentrations. μ was extracted from the slope of the square root of the drain current in the saturation regime. (d) On/off ratio of the P3HT/CNT composite FETs at various CNT concentrations.
did the U-CNTs (see Supporting Information, Figure S6 and Supplementary Note). To obtain further insight into the charge transport mechanisms and the effects of CNTs in the composite films, we measured the temperature-dependent μ values of the P3HT/ CNT composite FETs. The μ values were extracted from the linear region in the transfer curves under a low drain voltage and a sufficiently high gate voltage (Vgs ≫ Vds = −1 V) (Supporting Information, Figure S7). Arrhenius plots of the temperaturedependent μ are presented in Figure 3a. The decrease of μ with decreasing temperature (dμ/dT > 0) for all the devices indicates that the charge transport is limited by localized states and is thermally activated.34 We obtained activation energies for the CNT composite films using the Arrhenius relation, μ ∝ exp(Ea/kT), where Ea is the activation energy and k is the Boltzmann constant. The estimated activation energy of the pristine P3HT is 42 meV, which is comparable to other reported values for P3HT35,36 and for other semiconducting polymers.34,37 However, the addition of CNTs into P3HT results in a significant reduction in the activation energy of P3HT: 13 meV in the case of U-CNT and 7.4 meV in the case of B-CNT. The small activation energies of the CNT composites suggest that the degree of energetic disorder in the channel region considerably decreased. In particular, the smaller activation energy of P3HT/B-CNT agrees well with the excellent dispersion and well-matched energy level of the B-CNT-mixed polymers. Another possibility for the improved hole-transport in the B-CNT-mixed polymers is the enhanced local crystallization via B-CNTs, as observed by grazing incidence wide-angle X-ray
Supporting Information, Figure S3, S4). In contrast, the P3HT/ B-CNT devices exhibit higher on/off ratios than P3HT/U-CNT because of the higher on-current the lower off-current. This observation implies that the P3HT/B-CNT films maintain better semiconducting properties and that the well-dispersed B-CNT conducting bridges are more likely to facilitate interchain holetransport of the P3HT chain networks than the aggregated U-CNTs. However, a B-CNT concentration greater than 5 wt % causes the films to reaches the percolation threshold and also increases the off-current at which point the films show metallic characteristics rather than semiconducting properties. Therefore, we conclude that the optimum amount of B-CNTs is approximately 1 wt %, which yields the best field-effect characteristics in the composite FETs. Although the improved performance of the P3HT/B-CNT devices is mainly attributed to the better dispersion of B-CNTs, the modulated WF of the B-CNTs also contributes to the μ-enhancing effect of the B-CNTs. For efficient hole-transfer at the organic semiconductor/metallic CNT interface, the WF of the CNTs should be close to the highest occupied molecular orbital (HOMO) of the organic semiconducting molecules.32,33 Our WF measurements of the CNTs using ultraviolet photoelectron spectroscopy (UPS) analysis indicate that the electrondeficient B-doping significantly increases the WF of CNTs from 4.7 to 5.2 eV (See Supporting Information, Figure S5) and matches well with the HOMO of P3HT and with those of a wide range of other semiconducting polymers. Moreover, because of the modulated WF, the B-CNTs near the source contact improved the hole-injection properties to a greater extent than C
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 3. Temperature-dependence and structural characteristics of P3HT/CNT composite FETs. (a) Temperature-dependence of the field-effect μ of P3HT/CNT composite FETs. The μ values are estimated in the linear regime under a small Vds of −1 V. The straight lines represent the best Arrhenius fitting, and the slopes were used to extract the activation energies. The channel length and width of the devices are 50 and 1000 μm, respectively. (b,c) Normalized 2D GIWAXS patterns of (b) P3HT and (c) P3HT/CNT composite films with various CNT concentrations. (d) Normalized 1D profiles along the in-plane direction obtained from the 2D GIWAXS profiles. (e) Average crystallite size of P3HT with various CNT concentrations.
films. However, in the case of B-CNT concentrations greater than 5 wt %, the average crystallite size again decreases. Moreover, the evolution of the (100) peak in the in-plane profile of a film with a B-CNT concentration greater than 5 wt % indicates that the crystal orientation is more randomly distributed with a larger amount of B-CNTs, as also observed in the azimuthal profiles for the (100) peaks (Supporting Information, Figure S8). These results are consistent with the aforementioned field-effect characteristics. In contrast to the addition of B-CNTs, the addition of U-CNTs did not greatly affect the polymer morphology. This observation implies that the aggregated U-CNTs rarely interact with polymer chains on the molecular scale. We demonstrated that a simple mixing of B-CNTs induces crystallization in semiconducting polymers and improves charge transport. To take full advantage of B-CNTs, we fabricated flexible FETs based on RT-processed conjugated polymer/B-CNT composites directly on commercially available poly(ethylene-2, 6-naphthalate)
scattering (GIWAXS) measurements of the P3HT/CNT composite films. Figure 3b,c presents two-dimensional (2D) GIWAXS patterns of the P3HT and P3HT/CNT composite films, respectively. In the 2D patterns, two distinctive types of patterns are observed: one for (h00) peaks with a lamellar spacing ∼16.4 Å along the out-of-plane direction and the other for (010) peaks with a π−π spacing of ∼3.73 Å along the in-plane direction. To evaluate the relationship between the structural characteristics and the charge transport properties, we analyzed the in-plane (010) peaks using 1D profiles, as shown in Figure 3d. We subsequently estimated the average crystallite size on the basis of Scherrer’s equation using the full-width at half-maximum (fwhm) of the (010) peaks (Figure 3e). We observed that the average crystallite size substantially increased with an increase in the amount of B-CNTs (0.1−1 wt %). We believe that the increased crystallite size also contributes to the enhanced hole-μ, even when very few B-CNT bridges are present in the composite D
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 4. Conjugated polymer/B-CNT composite flexible TGBC-FETs. (a) A photograph and cross-sectional device structure of the flexible devices. Scale bar, 1 mm. (b) The transfer characteristics of P3HT (with or without 1 wt % B-CNT) FETs under a Vds of −60 V. The channel length and width are 50 and 1000 μm, respectively. PMMA (t ∼ 500 nm) was used for the gate insulator. (c,d) The output characteristics of (c) pristine P3HT and (d) P3HT/B-CNT devices. (e) Cycle stability test of the P3HT/B-CNT device with alternating gate pulses of −60 V (on-state) and 10 V (off-state) (1 Hz) and a constant Vds of −60 V before and after 500 bending cycles (R ∼ 5 mm). The photographs show the flexible device during the tests. (f−h) The transfer characteristics of (f) PBTTT, (g) DT-PDPP2T-TT, and (h) PDVT-10 (with or without B-CNT 1 wt %) FETs. The channel lengths are 50, 10, and 20 μm for the PBTTT, DT-PDPP2T-TT, and PDVT-10 devices, respectively. The channel widths are 1000 μm for all the devices. For the gate insulating layer, PMMA (t ∼ 500 nm) was used for the PBTTT devices, and CYTOP (t ∼ 500 nm) was used for the DT-PDPP2T-TT and PDVT-10 devices.
The transfer and output characteristics of the P3HT flexible FETs with or without B-CNT (1 wt %) are shown in Figure 4b−d. The measured saturation μ of the pristine P3HT device is centered at ∼0.095 cm2 V−1 s−1 (with a maximum value of ∼0.15 cm2 V−1 s−1). However, the P3HT/B-CNT exhibits a dramatically improved μ of ∼0.88 cm2 V−1 s−1 (with an average value of ∼0.7 cm2 V−1 s−1), the highest value reported for an FET based on P3HT. To utilize B-CNTs in flexible applications, verification that the composite devices are robust under bias stress and mechanical bending is important. Figure 4e presents the results of a cycle test on the P3HT/B-CNT device under repeated gate voltage pulse (1 Hz) before and after 500 bending cycles. The on- and off-state currents (Vgs = −60 and 10 V, respectively) remain constant even after 1000 switching cycles. In addition, the corresponding transfer characteristics show no threshold voltage shift or hysteresis between the forward and reverse Vgs sweep after the electrical switching (Supporting Information, Figure S11). Furthermore, repeated bending for up to 500 cycles with a bending radius
(PEN) films (thickness t = 125 μm) with a top-gate and bottomcontact (TGBC) configuration. For the fabrication of the TGBCFETs, the top surface of the semiconducting layer should be smooth. Otherwise, a rough surface of polymer layers results in an unfavorable semiconductor/dielectric interface and causes low performance or device breakdown. Indeed, P3HT/U-CNT composite FETs in the TGBC configuration exhibited very poor performance or electrical breakdown because of the rough surface due to the large U-CNT aggregates with a height of ∼800 nm (Supporting Information, Figure S9). In contrast to the U-CNT composites, the B-CNT-doped polymers ensure a successful fabrication of TGBC-FETs with very smooth surface of conjugated polymer/B-CNT composite films (Supporting Information, Figure S10). Figure 4a presents a photograph and illustrates the device structure of our flexible devices. Notably, the composite semiconducting layers were fabricated via simple spin-coating without any additional processes such as high-temperature annealing or alignment techniques. E
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
■
R ∼ 5 mm did not result in any change in the electrical properties of the device. To verify the universal applicability of our B-CNT technique, we also investigated flexible FETs using other conjugated polymers: poly(2,5-bis(3-dodecylthiophen-2-yl)thieno[3,2-b]thiophene) (PBTTT),38 diketopyrrolopyrrole-based polymer (DT-PDPP2T-TT),11 and poly[2,5-bis(2-decyltetradecyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione-alt-5,5′-di(thiophen2-yl)-2,2′-(E)-2-(2-(thiophen-2-yl)vinyl)thiophene] (PDVT-10)10 (their molecular structures are shown in Supporting Information, Figure S12). Figure 4f−h shows the transfer characteristics of the flexible devices based on PBTTT, DT-PDPP2T-TT, and PDVT-10 with or without 1 wt % B-CNTs, respectively; the output characteristics of the corresponding devices are shown in Supporting Information, Figure S13. The liquid crystalline polymer PBTTT (without heat treatment) exhibited a rather low μ of ∼0.015 cm2 V−1 s−1 (with a maximum value of ∼0.022 cm2 V−1 s−1). However, after the addition of B-CNTs (1 wt %), the devices exhibit a substantially improved μ of ∼0.090 cm2 V−1 s−1 (with a maximum value of ∼0.12 cm2 V−1 s−1). A similar effect was also observed in the DT-PDPP2T-TT devices; with B-CNTs, a high μ of ∼1.7 cm2 V−1 s−1 (with a maximum value of ∼2.3 cm2 V−1 s−1) was achieved, whereas the pristine devices exhibited an average μ of ∼0.40 cm2 V−1 s−1 (with a maximum value of ∼0.60 cm2 V−1 s−1). Notably, the highest hole-μ reported for this polymer is ∼0.80 cm2 V−1 s−1 in the case of a high-temperature annealing process conducted at 200 °C. Finally, we fabricated PVDT-10 based devices. PDVT-10 has also been reported to require high-temperature processing (∼180 °C) to induce structural ordering and a high chargecarrier μ. Our pristine PDVT-10 devices processed at RT exhibited an average hole-μ of 0.24 cm2 V−1 s−1 (with a maximum value of ∼0.31 cm2 V−1 s−1). However, the B-CNT-mixed PDVT-10 exhibited a significantly increased μ of ∼5.0 cm2 V−1 s−1 (with a maximum value of ∼7.2 cm2 V−1 s−1), which is more than 20 times greater than that of the pristine devices. To the best of our knowledge, this field-effect mobility value is the highest such value reported for a flexible polymer FET fabricated using a plastic substrate and is even comparable to that of indiumgallium-zinc oxide (IGZO) FETs fabricated on flexible substrates.39−41 In summary, B-CNTs exhibit superior dispersion in polymer matrices even at high concentrations, whereas U-CNTs tend to form undesired aggregates. Therefore, using the conjugated polymers mixed with 1 wt % B-CNTs, we achieved a hole-μ of 7.2 cm2 V−1 s−1 in FET devices fabricated on plastic substrates at RT. This value is the highest value reported for a polymer-based flexible FET to date. Our results also indicate that the B-CNT technique is universal and works well with various semiconducting polymers without any additional undesirable processing treatments for future low-cost and printable electronics applications.
■
Letter
AUTHOR INFORMATION
Corresponding Authors
*E-mail: (K.L.) [email protected] . *E-mail: (S.O.K.) [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank the Heeger Center for Advanced Materials (HCAM) and the Research Institute of Solar & Sustainable Energies (RISE) at the Gwangju Institute of Science and Technology (GIST) of Korea for assistance with the device fabrication and characterization. K.L. also acknowledges support from a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (NRF-2014R1A2A1A09006137 and No. 2008-0062606, CELA-NCRC). S.O.K. acknowledges support from IBS-R004-G1-2014-a00.
■
REFERENCES
(1) Forrest, S. R. Nature 2004, 428, 911−918. (2) Rogers, J. A.; Someya, T.; Huang, Y. Science 2010, 327, 1603−1607. (3) Kaltenbrunner, M.; Sekitani, T.; Reeder, J.; Yokota, T.; Kuribara, K.; Tokuhara, T.; Drack, M.; Schwodiauer, R.; Graz, I.; Bauer-Gogonea, S.; Bauer, S.; Someya, T. Nature 2013, 499, 458−463. (4) Di, C.-a.; Zhang, F.; Zhu, D. Adv. Mater. 2013, 25, 313−330. (5) Zhao, Y.; Di, C.-a.; Gao, X.; Hu, Y.; Guo, Y.; Zhang, L.; Liu, Y.; Wang, J.; Hu, W.; Zhu, D. Adv. Mater. 2011, 23, 2448−2453. (6) Søndergaard, R. R.; Hösel, M.; Krebs, F. C. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 16−34. (7) Jung, M.; Kim, J.; Noh, J.; Lim, N.; Lim, C.; Lee, G.; Kim, J.; Kang, H.; Jung, K.; Leonard, A. D.; Tour, J. M.; Cho, G. IEEE Trans. Electron Devices 2010, 57, 571−580. (8) Noriega, R.; Rivnay, J.; Vandewal, K.; Koch, F. P. V.; Stingelin, N.; Smith, P.; Toney, M. F.; Salleo, A. Nat. Mater. 2013, 12, 1038−1044. (9) Luo, C.; Kyaw, A. K. K.; Perez, L. A.; Patel, S.; Wang, M.; Grimm, B.; Bazan, G. C.; Kramer, E. J.; Heeger, A. J. Nano Lett. 2014, 14, 2764− 2771. (10) Chen, H.; Guo, Y.; Yu, G.; Zhao, Y.; Zhang, J.; Gao, D.; Liu, H.; Liu, Y. Adv. Mater. 2012, 24, 4618−4622. (11) Li, W.; Hendriks, K. H.; Roelofs, W. S. C.; Kim, Y.; Wienk, M. M.; Janssen, R. A. J. Adv. Mater. 2013, 25, 3182−3186. (12) van de Craats, A. M.; Stutzmann, N.; Bunk, O.; Nielsen, M. M.; Watson, M.; Müllen, K.; Chanzy, H. D.; Sirringhaus, H.; Friend, R. H. Adv. Mater. 2003, 15, 495−499. (13) Tseng, H. R.; Ying, L.; Hsu, B. B. Y.; Perez, L. A.; Takacs, C. J.; Bazan, G. C.; Heeger, A. J. Nano Lett. 2012, 12, 6353−6357. (14) Hsu, B. B. Y.; Seifter, J.; Takacs, C. J.; Zhong, C.; Tseng, H.-R.; Samuel, I. D. W.; Namdas, E. B.; Bazan, G. C.; Huang, F.; Cao, Y.; Heeger, A. J. ACS Nano 2013, 7, 2344−2351. (15) Tseng, H.-R.; Phan, H.; Luo, C.; Wang, M.; Perez, L. A.; Patel, S. N.; Ying, L.; Kramer, E. J.; Nguyen, T.-Q.; Bazan, G. C.; Heeger, A. J. Adv. Mater. 2014, 26, 2993−2998. (16) Spitalsky, Z.; Tasis, D.; Papagelis, K.; Galiotis, C. Prog. Polym. Sci. 2010, 35, 357−401. (17) Maiti, U. N.; Lee, W. J.; Lee, J. M.; Oh, Y.; Kim, J. Y.; Kim, J. E.; Shim, J.; Han, T. H.; Kim, S. O. Adv. Mater. 2014, 26, 40−67. (18) Li, H. J.; Lu, W. G.; Li, J. J.; Bai, X. D.; Gu, C. Z. Phys. Rev. Lett. 2005, 95, 086601. (19) Alig, I.; Lellinger, D.; Dudkin, S. M.; Pötschke, P. Polymer 2007, 48, 1020−1029. (20) Lu, K.; Grossiord, N.; Koning, C. E.; Miltner, H. E.; Mele, B. v.; Loos, J. Macromolecules 2008, 41, 8081−8085. (21) Wijewardane, S. Sol. Energy 2009, 83, 1379−1389. (22) Ma, P.-C.; Siddiqui, N. A.; Marom, G.; Kim, J.-K. Composites, Part A 2010, 41, 1345−1367.
ASSOCIATED CONTENT
* Supporting Information S
Detailed experimental procedures; additional results from FET, TEM, and GIWAXS characterizations; UPS spectra of CNTs; optical microscopy images of P3HT/CNT composite films; molecular structures of semiconducting polymers used for the flexible devices; and detailed discussion about charge-injection at source contacts and transport properties of P3HT/CNT composites in FETs. This material is available free of charge via the Internet at http://pubs.acs.org. F
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
(23) Park, Y. D.; Lim, J. A.; Jang, Y.; Hwang, M.; Lee, H. S.; Lee, D. H.; Lee, H.-J.; Baek, J.-B.; Cho, K. Org. Electron. 2008, 9, 317−322. (24) Chang, C. H.; Chien, C. H. IEEE Electron Device Lett. 2011, 32, 1457−1459. (25) Park, S.; Jin, S.; Jun, G.; Jeon, S.; Hong, S. Nano Res. 2011, 4, 1129−1135. (26) Jun, G. H.; Jin, S. H.; Park, S. H.; Jeon, S.; Hong, S. H. Carbon 2012, 50, 40−46. (27) Park, S. J.; Cho, M. S.; Lim, S. T.; Choi, H. J.; Jhon, M. S. Macromol. Rapid Commun. 2003, 24, 1070−1073. (28) Song, Y. J.; Lee, J. U.; Jo, W. H. Carbon 2010, 48, 389−395. (29) Bai, X.; Hu, X.; Zhou, S.; Yan, J.; Sun, C.; Chen, P.; Li, L. Electrochim. Acta 2013, 87, 394−400. (30) Lee, J. M.; Park, J. S.; Lee, S. H.; Kim, H.; Yoo, S.; Kim, S. O. Adv. Mater. 2011, 23, 629−633. (31) Bao, Z.; Dodabalapur, A.; Lovinger, A. J. Appl. Phys. Lett. 1996, 69, 4108−4110. (32) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. Adv. Mater. 1999, 11, 605− 625. (33) Braun, S.; Salaneck, W. R.; Fahlman, M. Adv. Mater. 2009, 21, 1450−1472. (34) Salleo, A.; Chen, T. W.; Völkel, A. R.; Wu, Y.; Liu, P.; Ong, B. S.; Street, R. A. Phys. Rev. B 2004, 70, 115311. (35) Kline, R. J.; McGehee, M. D.; Kadnikova, E. N.; Liu, J.; Fréchet, J. M. J.; Toney, M. F. Macromolecules 2005, 38, 3312−3319. (36) Hamadani, B. H.; Natelson, D. Appl. Phys. Lett. 2004, 84, 443− 445. (37) Ha, T.-J.; Sonar, P.; Dodabalapur, A. Phys. Chem. Chem. Phys. 2013, 15, 9735−9741. (38) McCulloch, I.; Heeney, M.; Bailey, C.; Genevicius, K.; MacDonald, I.; Shkunov, M.; Sparrowe, D.; Tierney, S.; Wagner, R.; Zhang, W.; Chabinyc, M. L.; Kline, R. J.; McGehee, M. D.; Toney, M. F. Nat. Mater. 2006, 5, 328−333. (39) Kim, Y.-H.; Heo, J.-S.; Kim, T.-H.; Park, S.; Yoon, M.-H.; Kim, J.; Oh, M. S.; Yi, G.-R.; Noh, Y.-Y.; Park, S. K. Nature 2012, 489, 128−132. (40) Hwang, B.-U.; Kim, D.-I.; Cho, S.-W.; Yun, M.-G.; Kim, H. J.; Kim, Y. J.; Cho, H.-K.; Lee, N.-E. Org. Electron. 2014, 15, 1458−1464. (41) Chen, H.; Cao, Y.; Zhang, J.; Zhou, C. Nat. Commun. 2014, 5, 4097.
G
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Semiconducting Polymers with Nanocrystallites Interconnected via Boron-Doped Carbon Nanotubes Kilho Yu,†,‡ Ju Min Lee,∥ Junghwan Kim,†,‡ Geunjin Kim,†,‡ Hongkyu Kang,†,‡ Byoungwook Park,†,‡ Yung Ho Kahng,§ Sooncheol Kwon,†,‡ Sangchul Lee,† Byoung Hun Lee,† Jehan Kim,⊥ Hyung Il Park,∥ Sang Ouk Kim,*,∥ and Kwanghee Lee*,†,‡,§ †
Department of Nanobio Materials and Electronics, School of Materials Science and Engineering, ‡Heeger Center for Advanced Materials (HCAM), and §Research Institute for Solar and Sustainable Energies (RISE), Gwangju Institute of Science and Technology (GIST), Gwangju 500-712, Republic of Korea ∥ Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Department of Materials Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Republic of Korea ⊥ Pohang Accelerator Laboratory (PAL), Pohang University of Science and Technology (POSTECH), Pohang 790-784, Republic of Korea S Supporting Information *
ABSTRACT: Organic semiconductors are key building blocks for future electronic devices that require unprecedented properties of low-weight, flexibility, and portability. However, the low charge-carrier mobility and undesirable processing conditions limit their compatibility with low-cost, flexible, and printable electronics. Here, we present significantly enhanced field-effect mobility (μFET) in semiconducting polymers mixed with boron-doped carbon nanotubes (B-CNTs). In contrast to undoped CNTs, which tend to form undesired aggregates, the B-CNTs exhibit an excellent dispersion in conjugated polymer matrices and improve the charge transport between polymer chains. Consequently, the B-CNT-mixed semiconducting polymers enable the fabrication of high-performance FETs on plastic substrates via a solution process; the μFET of the resulting FETs reaches 7.2 cm2 V−1 s−1, which is the highest value reported for a flexible FET based on a semiconducting polymer. Our approach is applicable to various semiconducting polymers without any additional undesirable processing treatments, indicating its versatility, universality, and potential for high-performance printable electronics. KEYWORDS: Semiconducting polymer, carbon nanotube, polymer nanocrystallite, nanocomposite, field-effect transistor, room-temperature process rganic field-effect transistors (OFETs) based on semiconducting polymers have been intensively investigated for future ubiquitous devices in which unprecedented device features such as low-weight, flexibility, and portability are required.1−4 Moreover, the solution processability of the semiconducting polymers enables the cost-effective, large-scale fabrication of OFETs on plastic substrates via high-throughput printing technologies.5−7 Nevertheless, their poor charge-carrier mobility (μ) has hindered the realization of their potential advantages.8 Because of remarkable advances in the development of processing techniques, the state of the art of solutionprocessed semiconducting polymers recently reached a remarkable stage9 of μ ∼ 36 cm2 V−1 s−1. However, because these techniques still require undesirable processing treatments, such as thermal annealing at high temperature (T)10,11 (typically, T ≥ 200 °C) and/or macroscopic alignment processes9,12−15 using mechanically grooved inorganic substrates, they are not compatible with the concept of “flexible and printable electronics”
O
© XXXX American Chemical Society
using plastic substrates. Fabricating high-μ semiconducting polymer layers on plastic substrates from a solution process at roomtemperature (RT) has been a formidable challenge. One promising approach involves introducing carbon nanotubes (CNTs) into semiconducting polymers.16 CNTs possess all the outstanding properties as a μ-enhancer in OFETs, including a high-μ, good mechanical flexibility, and a high aspect ratio.17 CNTs in semiconducting polymers act as ballistic conduction bridges18 between the polymer chains and/or induce a local alignment of the polymer chains located in the vicinity of CNTs,19,20 thereby significantly improving the μ value of the polymers. However, because CNTs tend to easily aggregate via van der Waals interaction,21,22 most CNT-based OFETs have suffered from poor device performance and electrical breakdown.23 Although many different solutions for overcoming this Received: September 17, 2014
A
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 1. Comparison of the morphology of the P3HT/CNT composite films. (a) Chemical structure of P3HT and schematic illustration of the formation of the P3HT/U-CNT or B-CNT composite films. (b) TEM image of a P3HT/U-CNT composite film. (c) TEM image of a P3HT/B-CNT composite film. Note that in b and c, the concentration of CNTs in the composite films is 10 wt % with respect to P3HT. Scale bars, 5 μm (lowmagnification images) and 500 nm (high-magnification images).
aggregation problem have been explored including surface functionalization24−26 and in situ polymerization,27−29 these efforts have thus far failed to satisfy the requirements for highperformance flexible FETs. In this work, we report for the first time significantly enhanced field-effect μ in conjugated polymers mixed with a small amount of boron-doped CNTs (B-CNTs). In contrast to undoped CNTs (U-CNTs), which tend to form undesired aggregates, the B-CNTs exhibit excellent dispersion in a conjugated polymer matrix. The substitutional doping of B atoms also effectively increases the work function of CNTs and facilitates holetransport between the semiconducting polymers and CNTs. More importantly, the well-dispersed CNTs on the molecular scale induce a local crystallization of the polymer chains even at RT, enabling the fabrication of high performance flexible OFETs with a top-gate geometry. The field-effect μ of our devices, which were fabricated on plastic substrates via a solution process at RT, reaches 7.2 cm2 V−1 s−1, which has never been previously achieved in a flexible FET based on a semiconducting polymer. The B-CNTs were processed using commercially available multiwalled CNTs with an average length of ∼1 μm and a diameter of 10−15 nm. After the CNTs were purified, B-doping was performed via thermal treatment under an Ar/NH3 gas stream with vaporized B2O3 powder,30 as explained in detail in the Supporting Information. We subsequently dispersed B-CNTs in o-dichlorobenzene (o-DCB) at various concentrations using ultrasonication. Poly(3-hexylthiophene) (P3HT) was used as a representative conjugated polymer material to explore the effect of the CNTs in this study.31 P3HT was dissolved in chloroform (CF) and then simply mixed with the CNT solutions, as described in detail in the Supporting Information. Both U-CNT and B-CNT were mixed well with P3HT in a solution state. However, during thin-film formation via spin-casting of the composite solutions, the U-CNTs began to aggregate (Figure 1a), as clearly observed in the transmission electron microscopy (TEM) images in Figure 1b. Even with a small concentration of U-CNTs (1 wt % with respect to P3HT), the U-CNTs formed aggregates with an average size of ∼2 μm in the P3HT/U-CNT composite films (see Supporting Information, Figure S1). When the concentration of U-CNTs was
increased to 5 and 10 wt %, the aggregates became larger (2−50 μm) over the entire films. These large aggregates are known to hinder the effective use of CNTs as conduction bridges in conjugated polymers and eventually cause device breakdown. In contrast, the B-CNTs exhibit a superb dispersion in the polymer matrix even at high concentration as evident in Figure 1c. This difference originates from the strong electrostatic repulsion with neighboring B-CNTs induced by the local positive charges of B atoms, which significantly reduce the van der Waals interactions among B-CNTs during film formation. The uniformly dispersed B-CNTs act more effectively as conducting bridges between the polymer chains than the aggregated U-CNTs on the molecular scale. To investigate the charge transport properties of the P3HT/ CNT composite films, we fabricated typical polymer-based FETs with a structure of bottom-gate and bottom-contact (BGBC) FETs (Figure 2a). Figure 2b shows the transfer characteristics of P3HT with or without 1 wt % CNTs. Both FETs with U-CNTs or B-CNTs exhibit an improved on-current (Ids) of the devices. However, whereas the average saturation μ of the P3HT/ U-CNT device is ∼0.012 cm2 V−1 s−1, which is 6-fold greater than that of pristine P3HT (∼0.0019 cm2 V−1 s−1), the P3HT/B-CNT exhibits a much higher μ of ∼0.044 cm2 V−1 s−1, which is 23-fold greater than that of pristine P3HT. The performance enhancement of B-CNT over U-CNT was also observed in the μ and on/off ratio profiles of the FET devices as a function of the CNT concentration (Figure 2c,d and Supporting Information, Figure S2). The P3HT/B-CNT device exhibits rapidly increasing μ values even with a small amount of B-CNTs (0.1−1 wt %) and then saturates with a slow increase until 10 wt % (see Table S1 in Supporting Information). In contrast, the device with U-CNTs starts to decrease again with U-CNT concentrations greater than 5 wt %. This performance degradation is attributed to the large U-CNT aggregates in the P3HT/U-CNT devices. Because we used multiwalled CNTs, the U-CNT aggregates in the FET channel are dominated by a metallic nature and appear to obstruct the field-effect behavior induced by the gate-source voltage (Vgs). Therefore, the P3HT/ U-CNT FETs with large U-CNT aggregates exhibit very poor on/off ratios and output characteristics (Figure 2d and B
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 2. Charge transport characteristics of the P3HT/CNT composite films. (a) Cross-sectional diagram of the P3HT/CNT composite FET. The SiO2 with 200 nm thickness serves as the gate dielectric layer (capacitance Ci = 17.2 nF cm−2) and the Au with a Ni adhesion layer serves as the source and drain contacts with a channel length and width of 10 and 1000 μm, respectively. (b) The transfer characteristics of P3HT (with or without 1 wt % CNTs with respect to P3HT) FETs under a drain-source voltage (Vds) of −60 V. (c) Hole μ of the P3HT/CNT composite FETs at various CNT concentrations. μ was extracted from the slope of the square root of the drain current in the saturation regime. (d) On/off ratio of the P3HT/CNT composite FETs at various CNT concentrations.
did the U-CNTs (see Supporting Information, Figure S6 and Supplementary Note). To obtain further insight into the charge transport mechanisms and the effects of CNTs in the composite films, we measured the temperature-dependent μ values of the P3HT/ CNT composite FETs. The μ values were extracted from the linear region in the transfer curves under a low drain voltage and a sufficiently high gate voltage (Vgs ≫ Vds = −1 V) (Supporting Information, Figure S7). Arrhenius plots of the temperaturedependent μ are presented in Figure 3a. The decrease of μ with decreasing temperature (dμ/dT > 0) for all the devices indicates that the charge transport is limited by localized states and is thermally activated.34 We obtained activation energies for the CNT composite films using the Arrhenius relation, μ ∝ exp(Ea/kT), where Ea is the activation energy and k is the Boltzmann constant. The estimated activation energy of the pristine P3HT is 42 meV, which is comparable to other reported values for P3HT35,36 and for other semiconducting polymers.34,37 However, the addition of CNTs into P3HT results in a significant reduction in the activation energy of P3HT: 13 meV in the case of U-CNT and 7.4 meV in the case of B-CNT. The small activation energies of the CNT composites suggest that the degree of energetic disorder in the channel region considerably decreased. In particular, the smaller activation energy of P3HT/B-CNT agrees well with the excellent dispersion and well-matched energy level of the B-CNT-mixed polymers. Another possibility for the improved hole-transport in the B-CNT-mixed polymers is the enhanced local crystallization via B-CNTs, as observed by grazing incidence wide-angle X-ray
Supporting Information, Figure S3, S4). In contrast, the P3HT/ B-CNT devices exhibit higher on/off ratios than P3HT/U-CNT because of the higher on-current the lower off-current. This observation implies that the P3HT/B-CNT films maintain better semiconducting properties and that the well-dispersed B-CNT conducting bridges are more likely to facilitate interchain holetransport of the P3HT chain networks than the aggregated U-CNTs. However, a B-CNT concentration greater than 5 wt % causes the films to reaches the percolation threshold and also increases the off-current at which point the films show metallic characteristics rather than semiconducting properties. Therefore, we conclude that the optimum amount of B-CNTs is approximately 1 wt %, which yields the best field-effect characteristics in the composite FETs. Although the improved performance of the P3HT/B-CNT devices is mainly attributed to the better dispersion of B-CNTs, the modulated WF of the B-CNTs also contributes to the μ-enhancing effect of the B-CNTs. For efficient hole-transfer at the organic semiconductor/metallic CNT interface, the WF of the CNTs should be close to the highest occupied molecular orbital (HOMO) of the organic semiconducting molecules.32,33 Our WF measurements of the CNTs using ultraviolet photoelectron spectroscopy (UPS) analysis indicate that the electrondeficient B-doping significantly increases the WF of CNTs from 4.7 to 5.2 eV (See Supporting Information, Figure S5) and matches well with the HOMO of P3HT and with those of a wide range of other semiconducting polymers. Moreover, because of the modulated WF, the B-CNTs near the source contact improved the hole-injection properties to a greater extent than C
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 3. Temperature-dependence and structural characteristics of P3HT/CNT composite FETs. (a) Temperature-dependence of the field-effect μ of P3HT/CNT composite FETs. The μ values are estimated in the linear regime under a small Vds of −1 V. The straight lines represent the best Arrhenius fitting, and the slopes were used to extract the activation energies. The channel length and width of the devices are 50 and 1000 μm, respectively. (b,c) Normalized 2D GIWAXS patterns of (b) P3HT and (c) P3HT/CNT composite films with various CNT concentrations. (d) Normalized 1D profiles along the in-plane direction obtained from the 2D GIWAXS profiles. (e) Average crystallite size of P3HT with various CNT concentrations.
films. However, in the case of B-CNT concentrations greater than 5 wt %, the average crystallite size again decreases. Moreover, the evolution of the (100) peak in the in-plane profile of a film with a B-CNT concentration greater than 5 wt % indicates that the crystal orientation is more randomly distributed with a larger amount of B-CNTs, as also observed in the azimuthal profiles for the (100) peaks (Supporting Information, Figure S8). These results are consistent with the aforementioned field-effect characteristics. In contrast to the addition of B-CNTs, the addition of U-CNTs did not greatly affect the polymer morphology. This observation implies that the aggregated U-CNTs rarely interact with polymer chains on the molecular scale. We demonstrated that a simple mixing of B-CNTs induces crystallization in semiconducting polymers and improves charge transport. To take full advantage of B-CNTs, we fabricated flexible FETs based on RT-processed conjugated polymer/B-CNT composites directly on commercially available poly(ethylene-2, 6-naphthalate)
scattering (GIWAXS) measurements of the P3HT/CNT composite films. Figure 3b,c presents two-dimensional (2D) GIWAXS patterns of the P3HT and P3HT/CNT composite films, respectively. In the 2D patterns, two distinctive types of patterns are observed: one for (h00) peaks with a lamellar spacing ∼16.4 Å along the out-of-plane direction and the other for (010) peaks with a π−π spacing of ∼3.73 Å along the in-plane direction. To evaluate the relationship between the structural characteristics and the charge transport properties, we analyzed the in-plane (010) peaks using 1D profiles, as shown in Figure 3d. We subsequently estimated the average crystallite size on the basis of Scherrer’s equation using the full-width at half-maximum (fwhm) of the (010) peaks (Figure 3e). We observed that the average crystallite size substantially increased with an increase in the amount of B-CNTs (0.1−1 wt %). We believe that the increased crystallite size also contributes to the enhanced hole-μ, even when very few B-CNT bridges are present in the composite D
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 4. Conjugated polymer/B-CNT composite flexible TGBC-FETs. (a) A photograph and cross-sectional device structure of the flexible devices. Scale bar, 1 mm. (b) The transfer characteristics of P3HT (with or without 1 wt % B-CNT) FETs under a Vds of −60 V. The channel length and width are 50 and 1000 μm, respectively. PMMA (t ∼ 500 nm) was used for the gate insulator. (c,d) The output characteristics of (c) pristine P3HT and (d) P3HT/B-CNT devices. (e) Cycle stability test of the P3HT/B-CNT device with alternating gate pulses of −60 V (on-state) and 10 V (off-state) (1 Hz) and a constant Vds of −60 V before and after 500 bending cycles (R ∼ 5 mm). The photographs show the flexible device during the tests. (f−h) The transfer characteristics of (f) PBTTT, (g) DT-PDPP2T-TT, and (h) PDVT-10 (with or without B-CNT 1 wt %) FETs. The channel lengths are 50, 10, and 20 μm for the PBTTT, DT-PDPP2T-TT, and PDVT-10 devices, respectively. The channel widths are 1000 μm for all the devices. For the gate insulating layer, PMMA (t ∼ 500 nm) was used for the PBTTT devices, and CYTOP (t ∼ 500 nm) was used for the DT-PDPP2T-TT and PDVT-10 devices.
The transfer and output characteristics of the P3HT flexible FETs with or without B-CNT (1 wt %) are shown in Figure 4b−d. The measured saturation μ of the pristine P3HT device is centered at ∼0.095 cm2 V−1 s−1 (with a maximum value of ∼0.15 cm2 V−1 s−1). However, the P3HT/B-CNT exhibits a dramatically improved μ of ∼0.88 cm2 V−1 s−1 (with an average value of ∼0.7 cm2 V−1 s−1), the highest value reported for an FET based on P3HT. To utilize B-CNTs in flexible applications, verification that the composite devices are robust under bias stress and mechanical bending is important. Figure 4e presents the results of a cycle test on the P3HT/B-CNT device under repeated gate voltage pulse (1 Hz) before and after 500 bending cycles. The on- and off-state currents (Vgs = −60 and 10 V, respectively) remain constant even after 1000 switching cycles. In addition, the corresponding transfer characteristics show no threshold voltage shift or hysteresis between the forward and reverse Vgs sweep after the electrical switching (Supporting Information, Figure S11). Furthermore, repeated bending for up to 500 cycles with a bending radius
(PEN) films (thickness t = 125 μm) with a top-gate and bottomcontact (TGBC) configuration. For the fabrication of the TGBCFETs, the top surface of the semiconducting layer should be smooth. Otherwise, a rough surface of polymer layers results in an unfavorable semiconductor/dielectric interface and causes low performance or device breakdown. Indeed, P3HT/U-CNT composite FETs in the TGBC configuration exhibited very poor performance or electrical breakdown because of the rough surface due to the large U-CNT aggregates with a height of ∼800 nm (Supporting Information, Figure S9). In contrast to the U-CNT composites, the B-CNT-doped polymers ensure a successful fabrication of TGBC-FETs with very smooth surface of conjugated polymer/B-CNT composite films (Supporting Information, Figure S10). Figure 4a presents a photograph and illustrates the device structure of our flexible devices. Notably, the composite semiconducting layers were fabricated via simple spin-coating without any additional processes such as high-temperature annealing or alignment techniques. E
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
■
R ∼ 5 mm did not result in any change in the electrical properties of the device. To verify the universal applicability of our B-CNT technique, we also investigated flexible FETs using other conjugated polymers: poly(2,5-bis(3-dodecylthiophen-2-yl)thieno[3,2-b]thiophene) (PBTTT),38 diketopyrrolopyrrole-based polymer (DT-PDPP2T-TT),11 and poly[2,5-bis(2-decyltetradecyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione-alt-5,5′-di(thiophen2-yl)-2,2′-(E)-2-(2-(thiophen-2-yl)vinyl)thiophene] (PDVT-10)10 (their molecular structures are shown in Supporting Information, Figure S12). Figure 4f−h shows the transfer characteristics of the flexible devices based on PBTTT, DT-PDPP2T-TT, and PDVT-10 with or without 1 wt % B-CNTs, respectively; the output characteristics of the corresponding devices are shown in Supporting Information, Figure S13. The liquid crystalline polymer PBTTT (without heat treatment) exhibited a rather low μ of ∼0.015 cm2 V−1 s−1 (with a maximum value of ∼0.022 cm2 V−1 s−1). However, after the addition of B-CNTs (1 wt %), the devices exhibit a substantially improved μ of ∼0.090 cm2 V−1 s−1 (with a maximum value of ∼0.12 cm2 V−1 s−1). A similar effect was also observed in the DT-PDPP2T-TT devices; with B-CNTs, a high μ of ∼1.7 cm2 V−1 s−1 (with a maximum value of ∼2.3 cm2 V−1 s−1) was achieved, whereas the pristine devices exhibited an average μ of ∼0.40 cm2 V−1 s−1 (with a maximum value of ∼0.60 cm2 V−1 s−1). Notably, the highest hole-μ reported for this polymer is ∼0.80 cm2 V−1 s−1 in the case of a high-temperature annealing process conducted at 200 °C. Finally, we fabricated PVDT-10 based devices. PDVT-10 has also been reported to require high-temperature processing (∼180 °C) to induce structural ordering and a high chargecarrier μ. Our pristine PDVT-10 devices processed at RT exhibited an average hole-μ of 0.24 cm2 V−1 s−1 (with a maximum value of ∼0.31 cm2 V−1 s−1). However, the B-CNT-mixed PDVT-10 exhibited a significantly increased μ of ∼5.0 cm2 V−1 s−1 (with a maximum value of ∼7.2 cm2 V−1 s−1), which is more than 20 times greater than that of the pristine devices. To the best of our knowledge, this field-effect mobility value is the highest such value reported for a flexible polymer FET fabricated using a plastic substrate and is even comparable to that of indiumgallium-zinc oxide (IGZO) FETs fabricated on flexible substrates.39−41 In summary, B-CNTs exhibit superior dispersion in polymer matrices even at high concentrations, whereas U-CNTs tend to form undesired aggregates. Therefore, using the conjugated polymers mixed with 1 wt % B-CNTs, we achieved a hole-μ of 7.2 cm2 V−1 s−1 in FET devices fabricated on plastic substrates at RT. This value is the highest value reported for a polymer-based flexible FET to date. Our results also indicate that the B-CNT technique is universal and works well with various semiconducting polymers without any additional undesirable processing treatments for future low-cost and printable electronics applications.
■
Letter
AUTHOR INFORMATION
Corresponding Authors
*E-mail: (K.L.) [email protected] . *E-mail: (S.O.K.) [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank the Heeger Center for Advanced Materials (HCAM) and the Research Institute of Solar & Sustainable Energies (RISE) at the Gwangju Institute of Science and Technology (GIST) of Korea for assistance with the device fabrication and characterization. K.L. also acknowledges support from a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (NRF-2014R1A2A1A09006137 and No. 2008-0062606, CELA-NCRC). S.O.K. acknowledges support from IBS-R004-G1-2014-a00.
■
REFERENCES
(1) Forrest, S. R. Nature 2004, 428, 911−918. (2) Rogers, J. A.; Someya, T.; Huang, Y. Science 2010, 327, 1603−1607. (3) Kaltenbrunner, M.; Sekitani, T.; Reeder, J.; Yokota, T.; Kuribara, K.; Tokuhara, T.; Drack, M.; Schwodiauer, R.; Graz, I.; Bauer-Gogonea, S.; Bauer, S.; Someya, T. Nature 2013, 499, 458−463. (4) Di, C.-a.; Zhang, F.; Zhu, D. Adv. Mater. 2013, 25, 313−330. (5) Zhao, Y.; Di, C.-a.; Gao, X.; Hu, Y.; Guo, Y.; Zhang, L.; Liu, Y.; Wang, J.; Hu, W.; Zhu, D. Adv. Mater. 2011, 23, 2448−2453. (6) Søndergaard, R. R.; Hösel, M.; Krebs, F. C. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 16−34. (7) Jung, M.; Kim, J.; Noh, J.; Lim, N.; Lim, C.; Lee, G.; Kim, J.; Kang, H.; Jung, K.; Leonard, A. D.; Tour, J. M.; Cho, G. IEEE Trans. Electron Devices 2010, 57, 571−580. (8) Noriega, R.; Rivnay, J.; Vandewal, K.; Koch, F. P. V.; Stingelin, N.; Smith, P.; Toney, M. F.; Salleo, A. Nat. Mater. 2013, 12, 1038−1044. (9) Luo, C.; Kyaw, A. K. K.; Perez, L. A.; Patel, S.; Wang, M.; Grimm, B.; Bazan, G. C.; Kramer, E. J.; Heeger, A. J. Nano Lett. 2014, 14, 2764− 2771. (10) Chen, H.; Guo, Y.; Yu, G.; Zhao, Y.; Zhang, J.; Gao, D.; Liu, H.; Liu, Y. Adv. Mater. 2012, 24, 4618−4622. (11) Li, W.; Hendriks, K. H.; Roelofs, W. S. C.; Kim, Y.; Wienk, M. M.; Janssen, R. A. J. Adv. Mater. 2013, 25, 3182−3186. (12) van de Craats, A. M.; Stutzmann, N.; Bunk, O.; Nielsen, M. M.; Watson, M.; Müllen, K.; Chanzy, H. D.; Sirringhaus, H.; Friend, R. H. Adv. Mater. 2003, 15, 495−499. (13) Tseng, H. R.; Ying, L.; Hsu, B. B. Y.; Perez, L. A.; Takacs, C. J.; Bazan, G. C.; Heeger, A. J. Nano Lett. 2012, 12, 6353−6357. (14) Hsu, B. B. Y.; Seifter, J.; Takacs, C. J.; Zhong, C.; Tseng, H.-R.; Samuel, I. D. W.; Namdas, E. B.; Bazan, G. C.; Huang, F.; Cao, Y.; Heeger, A. J. ACS Nano 2013, 7, 2344−2351. (15) Tseng, H.-R.; Phan, H.; Luo, C.; Wang, M.; Perez, L. A.; Patel, S. N.; Ying, L.; Kramer, E. J.; Nguyen, T.-Q.; Bazan, G. C.; Heeger, A. J. Adv. Mater. 2014, 26, 2993−2998. (16) Spitalsky, Z.; Tasis, D.; Papagelis, K.; Galiotis, C. Prog. Polym. Sci. 2010, 35, 357−401. (17) Maiti, U. N.; Lee, W. J.; Lee, J. M.; Oh, Y.; Kim, J. Y.; Kim, J. E.; Shim, J.; Han, T. H.; Kim, S. O. Adv. Mater. 2014, 26, 40−67. (18) Li, H. J.; Lu, W. G.; Li, J. J.; Bai, X. D.; Gu, C. Z. Phys. Rev. Lett. 2005, 95, 086601. (19) Alig, I.; Lellinger, D.; Dudkin, S. M.; Pötschke, P. Polymer 2007, 48, 1020−1029. (20) Lu, K.; Grossiord, N.; Koning, C. E.; Miltner, H. E.; Mele, B. v.; Loos, J. Macromolecules 2008, 41, 8081−8085. (21) Wijewardane, S. Sol. Energy 2009, 83, 1379−1389. (22) Ma, P.-C.; Siddiqui, N. A.; Marom, G.; Kim, J.-K. Composites, Part A 2010, 41, 1345−1367.
ASSOCIATED CONTENT
* Supporting Information S
Detailed experimental procedures; additional results from FET, TEM, and GIWAXS characterizations; UPS spectra of CNTs; optical microscopy images of P3HT/CNT composite films; molecular structures of semiconducting polymers used for the flexible devices; and detailed discussion about charge-injection at source contacts and transport properties of P3HT/CNT composites in FETs. This material is available free of charge via the Internet at http://pubs.acs.org. F
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
(23) Park, Y. D.; Lim, J. A.; Jang, Y.; Hwang, M.; Lee, H. S.; Lee, D. H.; Lee, H.-J.; Baek, J.-B.; Cho, K. Org. Electron. 2008, 9, 317−322. (24) Chang, C. H.; Chien, C. H. IEEE Electron Device Lett. 2011, 32, 1457−1459. (25) Park, S.; Jin, S.; Jun, G.; Jeon, S.; Hong, S. Nano Res. 2011, 4, 1129−1135. (26) Jun, G. H.; Jin, S. H.; Park, S. H.; Jeon, S.; Hong, S. H. Carbon 2012, 50, 40−46. (27) Park, S. J.; Cho, M. S.; Lim, S. T.; Choi, H. J.; Jhon, M. S. Macromol. Rapid Commun. 2003, 24, 1070−1073. (28) Song, Y. J.; Lee, J. U.; Jo, W. H. Carbon 2010, 48, 389−395. (29) Bai, X.; Hu, X.; Zhou, S.; Yan, J.; Sun, C.; Chen, P.; Li, L. Electrochim. Acta 2013, 87, 394−400. (30) Lee, J. M.; Park, J. S.; Lee, S. H.; Kim, H.; Yoo, S.; Kim, S. O. Adv. Mater. 2011, 23, 629−633. (31) Bao, Z.; Dodabalapur, A.; Lovinger, A. J. Appl. Phys. Lett. 1996, 69, 4108−4110. (32) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. Adv. Mater. 1999, 11, 605− 625. (33) Braun, S.; Salaneck, W. R.; Fahlman, M. Adv. Mater. 2009, 21, 1450−1472. (34) Salleo, A.; Chen, T. W.; Völkel, A. R.; Wu, Y.; Liu, P.; Ong, B. S.; Street, R. A. Phys. Rev. B 2004, 70, 115311. (35) Kline, R. J.; McGehee, M. D.; Kadnikova, E. N.; Liu, J.; Fréchet, J. M. J.; Toney, M. F. Macromolecules 2005, 38, 3312−3319. (36) Hamadani, B. H.; Natelson, D. Appl. Phys. Lett. 2004, 84, 443− 445. (37) Ha, T.-J.; Sonar, P.; Dodabalapur, A. Phys. Chem. Chem. Phys. 2013, 15, 9735−9741. (38) McCulloch, I.; Heeney, M.; Bailey, C.; Genevicius, K.; MacDonald, I.; Shkunov, M.; Sparrowe, D.; Tierney, S.; Wagner, R.; Zhang, W.; Chabinyc, M. L.; Kline, R. J.; McGehee, M. D.; Toney, M. F. Nat. Mater. 2006, 5, 328−333. (39) Kim, Y.-H.; Heo, J.-S.; Kim, T.-H.; Park, S.; Yoon, M.-H.; Kim, J.; Oh, M. S.; Yi, G.-R.; Noh, Y.-Y.; Park, S. K. Nature 2012, 489, 128−132. (40) Hwang, B.-U.; Kim, D.-I.; Cho, S.-W.; Yun, M.-G.; Kim, H. J.; Kim, Y. J.; Cho, H.-K.; Lee, N.-E. Org. Electron. 2014, 15, 1458−1464. (41) Chen, H.; Cao, Y.; Zhang, J.; Zhou, C. Nat. Commun. 2014, 5, 4097.
G
dx.doi.org/10.1021/nl503574h | Nano Lett. XXXX, XXX, XXX−XXX
