Flexible organic field-effect transistors based on electrospun conjugated polymer nanofibers with high bending stability

Organic Electronics 15 (2014) 1056–1061

Contents lists available at ScienceDirect

Organic Electronics journal homepage: www.elsevier.com/locate/orgel

Flexible organic field-effect transistors based on electrospun conjugated polymer nanofibers with high bending stability Alessandro Manuelli a, Luana Persano a,b, Dario Pisignano a,b,c,⇑ a

Center for Biomolecular Nanotechnologies @UNILE, Istituto Italiano di Tecnologia, via Barsanti, I-73010 Arnesano, LE, Italy National Nanotechnology Laboratory of Istituto Nanoscienze-CNR, Università del Salento, via Arnesano, I-73100 Lecce, Italy c Dipartimento di Matematica e Fisica ‘‘Ennio De Giorgi’’, Università del Salento, via Arnesano, I-73100 Lecce, Italy b

a r t i c l e

i n f o

Article history: Received 20 November 2013 Received in revised form 13 February 2014 Accepted 20 February 2014 Available online 5 March 2014 Keywords: Polymer nanofibers Electrospinning Plastic electronics Flexible electronics Poly(alkylthiophene)s

a b s t r a c t We report on flexible, single electrospun nanofiber field-effect transistors made by a blend of poly(3-decylthiophene) and poly(3-hexylthiophene), assessing for the first time the performances of this class of devices in terms of stability upon repeated tensile bending. Charge-carrier mobilities in the nanofiber-based device are estimated of the order of 103 cm2/(V s). Repeated cycles of bending and relaxing are performed, and the evolution of the device current–voltage characteristics is monitored up to 1000 cycles. We find that during bending the mobility is higher than that measured in planar conditions, and that after about 100 bending cycles it rapidly stabilizes. The here observed bending stability suggests a high compatibility of electrospun nanofibers with devices fabricated by rollto-roll processes, and with bendable or wearable electronics. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction The availability of solution-based methods for nanopatterning, together with low cost and flexibility, are universally desired features which make organic electronics of very wide interest in both the scientific community and industry [1–8]. In the last decade, the development of methods fully exploiting bendable, rugged and extremely cheap organic and plastic materials has provided electronics with functionalities not achievable with conventional, waferbased devices, leading, for instance, to paperlike electronic displays [9], novel pressure-sensing architectures and electronic skins [8,10–13], flexible vertical light-emitting diodes [14] and batteries [15,16], and bendable memories [17] and

⇑ Corresponding author at: Dipartimento di Matematica e Fisica ‘‘Ennio De Giorgi’’, Università del Salento, via Arnesano, I-73100 Lecce, Italy. Tel.: +39 0832298104. E-mail addresses: [email protected] (A. Manuelli), luana.persano@ nano.cnr.it (L. Persano), [email protected] (D. Pisignano). http://dx.doi.org/10.1016/j.orgel.2014.02.016 1566-1199/Ó 2014 Elsevier B.V. All rights reserved.

circuits [18], all of which could be further integrated with solution-processable electronic materials such as semiconducting polymers. In particular, the well-assessed electronic properties of p-conjugated polymers, and their capability to be processed by large-scale and cost-effective printing and spinning techniques, have been applied to diverse applications such as radio frequency identification tags, integrated logic circuits, displays and sensors [1,5,6,18]. Among other conjugated compounds, poly(alkylthiophene)s (PATs), with their peculiar self-assembling properties and environmental stability, are likely the most used family of polymers in organic thin-film field-effect transistors (OFETs) [18–24]. Allowing these materials to be patterned or nanostructures composed of them to be produced, nanofabrication technologies have the potential both to increase the component density in single chips, through channel size reduction and realization of addressable matrices, and to provide devices with enhanced functionalities. Electrospinning [25,26], which is based on an electric field-induced, uniaxial elongation of a polymer solution with sufficient molecular entanglement, is particularly

A. Manuelli et al. / Organic Electronics 15 (2014) 1056–1061

relevant in this respect, allowing active fibers with potentially kilometric length to be realized. Nanofibers made of semiconducting polymers can exhibit remarkable performances [20,27–32] and generally improved supramolecular order compared to thin films. Unfortunately, electrospinning of conjugated polymers is often made difficult by the poor viscoelastic behavior of those solutions, which motivates using blends incorporating electrically inert compounds such as poly(ethylene oxide) [18,30,32] or poly(e-caprolactone) [30,32]. In addition, the process can easily determine the incorporation of oxygen and moisture if performed in air, which can lead to electronic defects [28]. Furthermore, despite of many claims supporting the ease of integration of electrospinning technologies with flexible electronics, the stability of nanofiber-based device conduction during and after bending is still not assessed. In this work, we propose active nanofibers, due to their geometry and conformability, as building blocks for reducing bending-induced damage in organic electronics, thus allowing devices with very high bending stability to be realized. The stability of bendable transistors and inverters has been investigated by pentacene films used in combination with C60 (found to be stable over few bending cycles) [33], with hexadecafluorocopperphtalocyanine in devices sandwiched between the substrate and an encapsulation layer to induced compressive and tensile strains to cancel each other [34], and with gold nano-particles in memories (over several hundreds of cycles) [35]. Transistors with a printed triarylamine copolymer and a perfluoropolymer dielectric layer show a saturation current increased by 60% after 1000 bending cycles [36]. Here, nanofiber-based OFETs are, for the first time, measured under bending conditions with repeated tensile stressing. During initial bending the charge-carrier mobility (l) is higher than that measured in planar conditions, and after 100 bending cycles it stabilizes, up to 1000 performed cycles.

2. Results and discussion Our flexible OFETs are based on a blend [24] of regioregular poly(3-decylthiophene) (P3DT) and regioregular poly(3-hexylthiophene) (P3HT) electrospun fibers as active elements, and fully plastic dielectric and gate electrode, both processed from solution with no further process step such as UV curing used in previous reports on nanofiberbased transistors. Notwithstanding its relatively lower conduction properties, P3DT is used for blending due to its better processability, which makes possible electrospinning pure PAT-based nanofibers. The fabrication steps are shown in Fig. 1. P3DT/P3HT blends (89%/11% w/w referred to solid content) are dissolved in chloroform at 54 °C with polymer/solvent concentration of 0.3–6% w/w. Electrospinning is performed in a glove box system with [O2] < 0.1 ppm at 21 °C by injecting 0.4 mL of highly-concentrated solution in a syringe tipped with a 27-gauge stainless steel needle. The used flow rate is 7.5  103 lL/ min, and the total bias is 11 kV, applied over a spinneretcollector distance of 20 cm. Poly(ethylene terephthalate)

1057

(PET) substrates with gold interdigitated electrodes are positioned on the collector for fiber deposition. Afterwards, a 1.3-lm thick film of poly(methyl methacrylate) (PMMA) is deposited by a doctor blade system and a poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) gate electrode is finally realized by drop-casting. Having a very smooth organic-dielectric interface is critical for device operation and stability. To this aim a single solvent is conveniently used in electrospinning, avoiding eventual fiber porosity which may arise from phaseseparation phenomena favored by using immiscible components and solvent mixtures [37]. The atomic force microscopy (AFM) micrographs of doctored P3DT/P3HT films and electrospun fibers (Fig. 2a and b) evidence homogeneous and smooth surfaces, with an overall peakto-peak roughness of 1 nm or lower (inset of Fig. 2b), as needed for thin-film OFETs with state-of-art bending stability [34]. Each spun fiber reveals a neat rounded structure which is well retained upon deposition across gaps between metal electrodes (Fig. 2c). The resulting flexible chip consists of 28 transistors in top-gate configuration, arranged in 4 rows (Fig. 3a), and the device response is measured before, during, and after bending (Fig. 3b–d, details in the Materials and methods Section). The substrate is bent along an axis passing exactly through the measured transistors. Bending radii, Rb, are of about 5 mm, corresponding to a tensile strain, e ffi D/2Rb, where D is the substrate thickness, of about 1.7%, applied parallel to the current flowing between source and drain electrodes (i.e. parallel to the nanofiber longitudinal axis). Fig. 3b and c presents the current–voltage characteristics of nanofiber transistors, for gate voltages (VGS) decreasing from 0 to 50 V, before and during bending, respectively. The drain current (IDS) dependence on the drain voltage (VDS) highlights the typical p-type behavior of PAT-based OFET, working in accumulation mode. The device regularly operates in its bent state. IDS variations from before to during bending are within 3% (Fig. 3d), and by the slope of |IDS|1/ 2 one cannot appreciate significant variations of the threshold voltage (VTH). VTH has positive values, indicating accumulated holes which are intrinsically present in the conduction channel, which can be related to residual impurities from the fabrication process as frequently observed in nanofiber-based transistors [28]. In the supposed saturation regime one has IDS = W Cl(VGS  VTH)2, where C is 2L gate dielectric capacitance per unit area, W is the width of the semiconducting element (here roughly given by the diameters of fibers bridging source and drain), and L is the channel length, respectively, and the calculated charge carriers mobility is about 5.0  103 cm2/(V s) and 3.8  103 cm2/(V s) before and during bending, respectively. These values are more than one order of magnitude higher than those measured in films with the same active blend, for which l = 6–7  105 cm2/(V s). One should bear in mind that, while the focus of the present study is not on achieving high OFET mobility, at least two strategies can be followed in order to improve conduction performances up to l values of 1 cm2/(V s) of more with our device geometry, namely either directly doping nanofibers (for instance by NOPF6 solutions) [38], or using ion-gel, polyelectrolyte gate dielectrics [32,39]. Both these methods allow an

1058

A. Manuelli et al. / Organic Electronics 15 (2014) 1056–1061

Fig. 1. Scheme of the process for realizing OFETs based on electrospun, conjugated polymer nanofibers on flexible substrates. Upon the definition of metal electrodes, the fiber is directly spun on the gap in order to bridge the Au regions, and the PMMA dielectric layer is deposited by doctor blading. The fiber diameters range between about 100 nm and 3 lm depending on processing. The polymeric gate electrode (PEDOT:PSS) is finally cast on top of PMMA. L: conduction channel length, W: channel width.

Fig. 2. (a) and (b) AFM planar view of the surface of a deposited P3DT/P3HT film and of a corresponding electrospun fiber. (a) Scale bar = 2 lm, vertical scale = 18 nm. (b) Scale bar = 800 nm, vertical scale = 190 nm. The inset in (b) shows the profile of the top of the fiber, measured along the dotted line in the main figure. Arrows and lines in the inset highlight the observable surface roughness of the fiber, which is 6 1 nm. (c) SEM micrograph, in false colors, of a fiber bridging Au electrodes in a FET device. Scale bar = 5 lm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

increase of mobilities of three orders of magnitude to be observed compared to pristine PATs. In order to assess the device stability performances following the mechanical stress, we perform repeated cycles of bending and relaxing and measure the evolution of the IDS(VDS) characteristics (Fig. 4). After the second bending, the mobility decreases by about 17% (Fig. 4) when measured in planar, released configuration, whereas it increases by about 25% when measured during bending (inset of Fig. 4). In the subsequent cycles, variations rapidly stabilize, especially under bending, and the values of l keep almost constant. Tests of up to 1000 cycles of bending and relaxing indicate no significant changes of the IDS(VDS) curves. In all the experiments, IDS is of (70 ± 5) nA at VDS = VGS = 50 V when the devices is in planar configuration, and the average value of l is about 4.4  103 cm2/(V s). The behavior observed under bending in the first cycles can be explained by considering a likely improved

adhesion of the spun fiber to electrode underneath, as well as an improved interfacing of the dielectric layer with the fiber during the initial tensile stress. During initial cycles, in fact, the conjugated polymer may undergo a rearrangement due to interfacial forces exerted by the dielectrics on top of it. In particular, the underlying mechanism improving the organics-dielectrics adhesion can be related to the penetration of the plastic PMMA layer underneath the lateral edges of the deposited fiber, due to the tensile stress on the device. Immediately after deposition of the dielectric layer, voids can generate in correspondence of the substrate region shadowed by the fiber rounded edges, as previously observed in nanowire and nanofiber-based lithographies [18,40]. These voids can be partially filled during device bending due to the consequent polymer rearrangements, leading to an overall extended and improved interface between the involved materials, as schematized in Fig. 4b. This suggests a high compatibility of electrospun nanofiber media with devices which are

A. Manuelli et al. / Organic Electronics 15 (2014) 1056–1061

1059

Fig. 3. (a) Photograph of the P3DT/P3HT fiber-based FET arrays. (b and c) Output characteristics [IDS(VDS)] of planar (b) and bent (c) devices (L = 30 lm), for VGS values from 0 to 50 V. (d) Corresponding transfer characteristics [IDS(VGS), left vertical scale], together with the curve, |IDS|1/2(VGS) (right scale), for devices in the planar (dots and diamonds) and bent (continuous lines) state, at a drain voltage of 50 V. The dotted lines are linear fits of |IDS|1/2 for |VGS| above 30 V.

Fig. 4. (a) Electrospun fiber-based device stability, measured by repeated tensile bending tests. The output characteristic [IDS(VDS)] is shown for an applied gate voltage, VGS = 50 V, for the OFET in its pristine conditions and after 10, 100 and 1000 cycles of bending with Rb = 5 mm. Inset: corresponding chargecarrier mobility (circles). The mobility measured with the device in the bent state is also shown for the first ten cycles (squares). (b) Schematics of the device cross-section showing a possible rearrangements of voids nearby the fibers, improving the quality of the conjugated-polymer/dielectrics interface upon bending (features not in scale).

directly fabricated under conditions of mechanical stress, such as displays produced by continuous roll-to-roll processes, but also with bendable, rollable or even wearable devices which need to resist a lot of stress cycles. In per-

spective, the fiber geometry may even help in keeping the active element of the transistor, and the interface between the channel and the dielectric material, under neutral strain conditions, especially for bending deformations

1060

A. Manuelli et al. / Organic Electronics 15 (2014) 1056–1061

applied perpendicularly to the fiber longitudinal axis, as immediate consequence of the reduced lateral extension of the active material, namely of the sub-lm fiber diameter achievable by electrospinning (W  Rb).

3. Conclusion The performances of single electrospun nanofiber OFETs have been characterized under bending conditions following repeated tensile stress experiments. During initial bending cycles, the charge-carrier mobility is found to be higher than that measured in planar conditions, and after about 100 bending cycles it stabilizes, devices being fully functional up to 1000 cycles. IDS of (70 ± 5) nA are supported by the devices at VDS = VGS = 50 V. This behavior is likely related to improved contact mechanics at the involved material interfaces following bending, and opens interesting perspectives for exploiting active electrospun nanofibers in wearable and roll-to-roll electronics.

4.3. Electrical characterization A Keithley 4200 semiconductor analyzer is used for the electrical characterization of the devices, which is carried out in air at room temperature, and without encapsulation. Au-coated needle probes are used to establish electrical contacts. Output and transfer characteristics are initially recorded without applying any bending. Tensile bending tests are then performed while recording the electrical characteristics of transistors using the setup shown in Fig. S1 in the Supporting Information, where the sample is fixed on a circular holder of light metal alloy with a diameter of 10 mm by means of two clamps. Alternating characterizations are performed under and without stress, and measurements are repeated fixing one sample side the holder and bending it for 100 times, then recording characteristics, and carrying out this procedure for ten times reaching a total of 1000 cycles of bending and unbending on the same device. Acknowledgements

4. Materials and methods 4.1. Materials P3DT, P3HT, chloroform, butyl acetate, ethyl acetate and PMMA (molecular weight = 350,000) are purchased from Aldrich. PEDOT/PSS (Clevios PH 1000) and PET (optically clear, 175 lm) substrates are kindly supplied by Heraeus and by Coveme, respectively. All materials are used as received.

4.2. Device fabrication and morphological characterization Gold interdigitated electrodes are realized by evaporation of 50 nm of metal, photolithography and wet etching by a KI/I2 solution. The doctor blade system for the deposition of the dielectric layer is driven by an Erichsen Coatmaster 510. A 25 lL volume of 8.5% (w/w) PMMA solution in a mixture of butyl acetate/ethyl acetate (70%/ 30% v/v) is deposited by a blade, shifting at a velocity of 3 mm/s at 300 lm from the device. For purpose of comparison, thin-film devices are produced by doctoring the P3DT/P3HT solution and then immediately depositing the dielectric layer with the same experimental conditions. During electrospinning, a + 8 kV voltage is applied to the needle by a power supply (Glassman High Voltage Inc. series EL), and the collector is negatively biased at 3 kV. All samples are annealed at 80 °C for 24 h under controlled atmosphere prior characterization, to remove solvent residues and improve adhesion to contacts. The morphological analysis of films and fibers is performed by tapping AFM in air, using a Nanoscope III controller with a Multimode head (Veeco) integrated with an E-scanner (P-doped Si tips are used with an 8 nm nominal curvature radius and a resonant frequency of 190 kHz) and by a scanning electron microscope (SEM) with a Nova NanoSEM 450 system (FEI), using an acceleration voltage around 5 kV and an aperture size of 30 lm.

The authors acknowledge the financial support from the Italian Minister of University and Research FIRB RBFR08DJZI ‘‘Futuro in Ricerca’’. Dr. S. Girardo and Dr. G. Potente are acknowledged for SEM measurements. Appendix A. Supplementary material Supplementary material associated with this article can be found, in the online version, at http://dx.doi.org/ 10.1016/j.orgel.2014.02.016. References [1] H. Sirringhaus, T. Kawase, R.H. Friend, T. Shimoda, M. Inbasekaran, W. Wu, E.P. Woo, Science 290 (2000) 2123–2126. [2] G.H. Gelinck, H.E.A. Huitema, E. van Veenendaal, E. Cantatore, L. Schrijnemakers, J.B.P.H. van der Putten, T.C.T. Geuns, M. Beenhakkers, J.B. Giesbers, B.-H. Huisman, E.J. Meijer, E.M. Benito, F.J. Touwslager, A.W. Marsman, B.J.E. van Rens, D.M. de Leeuw, Nat. Mater. 3 (2004) 106–110. [3] H. Klauk, Organic Electronics: Materials, Manufacturing, and Applications, Wiley-VCH, Weinheim, Germany, 2006. [4] G. Gelinck, P. Heremans, K. Nomoto, T.D. Anthopoulos, Adv. Mater. 22 (2010) 3778–3798. [5] G. De Luca, W. Pisula, D. Credgington, E. Treossi, O. Fenwick, G.M. Lazzerini, R. Dabirian, E. Orgiu, A. Liscio, V. Palermo, K. Müllen, F. Cacialli, P. Samorì, Adv. Funct. Mater. 21 (2011) 1279–1295. [6] C. Wang, H. Dong, W. Hu, Y. Liu, D. Zhu, Chem. Rev. 112 (2012) 2208– 2267. [7] S.C.B. Mannsfeld, B.C.-K. Tee, R.M. Stoltenberg, C.V.H.-H. Chen, S. Barman, B.V.O. Muir, A.N. Sokolov, C. Reese, Z. Bao, Nat. Mater. 9 (2010) 859–864. [8] L. Persano, C. Dagdeviren, Y. Su, Y. Zhang, S. Girardo, D. Pisignano, Y. Huang, J.A. Rogers, Nat. Commun. 4 (2013) 1633. [9] J.A. Rogers, Z. Bao, K. Baldwin, A. Dodabalapur, B. Crone, V.R. Raju, V. Kuck, H. Katz, K. Amundson, J. Ewing, P. Drzaic, Proc. Natl. Acad. Sci. U.S.A. 98 (2001) 4835–4840. [10] T. Someya, T. Sekitani, S. Iba, Y. Kato, H. Kawaguchi, T. Sakurai, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 9966–9970. [11] D. Pisignano, Polymer Nanofibers, Royal Society of Chemistry, Cambridge, UK, 2013. [12] C. Pang, G.-Y. Lee, T. Kim, S.M. Kim, H.N. Kim, S.-H. Ahn, K.-Y. Suh, Nat. Mater. 11 (2012) 795–801. [13] C. Wang, D. Hwang, Z. Yu, K. Takei, J. Park, T. Chen, B. Ma, A. Javey, Nat. Mater. (2013) 899–904.

A. Manuelli et al. / Organic Electronics 15 (2014) 1056–1061 [14] R.-H. Kim, S. Kim, Y.M. Song, H. Jeong, T. Kim, J. Lee, X. Li, K.D. Choquette, J.A. Rogers, Small 8 (2012) 3123–3128. [15] L. Hu, H. Wu, F. La Mantia, Y. Yang, Y. Cui, ACS Nano 4 (2010) 5843– 5848. [16] S. Xu, Y. Zhang, J. Cho, J. Lee, X. Huang, L. Jia, J.A. Fan, Y. Su, J. Su, H. Zhang, H. Cheng, B. Lu, C. Yu, C. Chuang, T. Kim, T. Song, K. Shigeta, S. Kang, C. Dagdeviren, I. Petrov, P.V. Braun, Y. Huang, U. Paik, J.A. Rogers, Nat. Commun. 4 (2013) 1543. [17] Y. Zhou, S.-T. Han, Z.-X. Xu, V.A.L. Roy, Nanotechnology 23 (2012) 344014. [18] S.-Y. Min, T.-S. Kim, B.J. Kim, H. Cho, Y.-Y. Noh, H. Yang, J. H Cho, T.W. Lee, Nat. Commun. 4 (2013) 1773. [19] H. Sirringhaus, P.J. Brown, R.H. Friend, M.M. Nielsen, K. Bechgaard, B.M.W. Lanveld-Voss, A.J.H. Spiering, R.A.J. Jannsen, E.W. Meijer, P. Herwig, D.M. de Leeuw, Nature 401 (1999) 685–688. [20] J.A. Merlo, C.D. Frisbie, J. Phys. Chem. B 108 (2004) 19169–19179. [21] P.-T. Wu, G. Ren, C. Li, R. Mezzenga, S.A. Jenekhe, Macromolecules 42 (2009) 2317–2320. [22] J. Ah Lim, F. Liu, S. Ferdous, M. Muthukumar, A.L. Briseno, Mater. Today 13 (2010) 14–24. [23] P.M. Beaujuge, J.M.J. Fréchet, J. Am. Chem. Soc. 133 (2011) 20009– 20029. [24] A. Babel, S.A. Jenekhe, J. Phys. Chem. B 107 (2003) 1749–1754. [25] D.H. Reneker, I. Chun, Nanotechnology 7 (1996) 216. [26] D. Li, Y. Xia, Adv. Mater. 16 (2004) 1151–1170. [27] R. Gonzalez, N.J. Pinto, Synth. Met. 151 (2005) 275–278.

1061

[28] H. Liu, C.H. Reccius, H.G. Craighead, Appl. Phys. Lett. 87 (2005) 253106. [29] A. Babel, D. Li, Y. Xia, S.A. Jenekhe, Macromolecules 38 (2005) 4705– 4711. [30] S. Lee, G.D. Moon, U. Jeong, J. Mater. Chem. 19 (2009) 743–748. [31] S. Samitsu, T. Shimomura, S. Heike, T. Hashizume, K. Ito, Macromolecules 43 (2010) 7891–7894. [32] S.W. Lee, H.J. Lee, J.H. Choi, W.G. Koh, J.M. Myoung, J.H. Hur, J.J. Park, J.H. Cho, U. Jeong, Nano Lett. 10 (2010) 347–351. [33] X.-H. Zhang, W.J. Potscavage, S. Choi, B. Kippelen, Appl. Phys. Lett. 94 (2009) 043312. [34] T. Sekitani, U. Zschieschang, H. Klauk, T. Someya, Nat. Mater. 9 (2010) 1015–1022. [35] S.-J. Kim, J.-S. Lee, Nano Lett. 10 (2010) 2884–2890. [36] X. Boddaert, B. Bensaid, P. Benaben, R. Gwoziecki, R. Coppard, Microelectron. Reliab. 50 (2010) 1884–1887. [37] P.K. Bhattacharjee, G.C. Rutledge, Electrospinning and polymer nanofibers: process fundamentals, in: P. Ducheyne, K.E. Healy, D.W. Hutmacher, D.W. Grainger, C.J. Kirkpatrick (Eds.), Comprehensive Biomaterials, Elsevier, Oxford, UK, 2011. [38] Y. Xuan, X. Liu, S. Desbief, P. Leclère, M. Fahlman, R. Lazzaroni, M. Berggren, J. Cornil, D. Emin, X. Crispin, Phys. Rev. B 82 (2010) 115454. [39] S.H. Kim, K. Hong, W. Xie, K.H. Lee, S. Zhang, T.P. Lodge, C.D. Frisbie, Adv. Mater. 25 (2013) 1822–1846. [40] D. Tu, S. Pagliara, A. Camposeo, G. Potente, E. Mele, R. Cingolani, D. Pisignano, Adv. Funct. Mater. 21 (2011) 1140–1145.