Impact of MoO3 interlayer on the energy level alignment of pentacene-C60 heterostructure , Ye Zou , Hongying Mao, Qing Meng, and Daoben Zhu
Citation: The Journal of Chemical Physics 144, 084706 (2016); doi: 10.1063/1.4942480 View online: http://dx.doi.org/10.1063/1.4942480 View Table of Contents: http://aip.scitation.org/toc/jcp/144/8 Published by the American Institute of Physics
THE JOURNAL OF CHEMICAL PHYSICS 144, 084706 (2016)
Impact of MoO3 interlayer on the energy level alignment of pentacene-C60 heterostructure Ye Zou,1,a) Hongying Mao,2,3 Qing Meng,1 and Daoben Zhu1 1
Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China 2 Department of Physics, Hangzhou Normal University, Hangzhou 310036, China 3 State Key Laboratory of Silicon Materials, School of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, China
(Received 3 December 2015; accepted 9 February 2016; published online 25 February 2016) Using in situ ultraviolet photoelectron spectroscopy, the electronic structure evolutions at the interface between pentacene and fullerene (C60), a classical organic donor-acceptor heterostructure in organic electronic devices, on indium-tin oxide (ITO) and MoO3 modified ITO substrates have been investigated. The insertion of a thin layer MoO3 has a significant impact on the interfacial energy level alignment of pentacene-C60 heterostructure. For the deposition of C60 on pentacene, the energy difference between the highest occupied molecular orbital of donor and the lowest unoccupied molecular orbital of acceptor (HOMOD-LUMOA) offset of C60/pentacene heterostructure increased from 0.86 eV to 1.54 eV after the insertion of a thin layer MoO3 on ITO. In the inverted heterostructrure where pentacene was deposited on C60, the HOMOD-LUMOA offset of pentacene/C60 heterostructure increased from 1.32 to 2.20 eV after MoO3 modification on ITO. The significant difference of HOMOD-LUMOA offset shows the feasibility to optimize organic electronic device performance through interfacial engineering approaches, such as the insertion of a thin layer high work function MoO3 films. C 2016 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4942480]
I. INTRODUCTION
Organic-organic heterojunction structures have drawn increasing research attention due to the importance in determining the performance of organic electronic devices including organic light-emitting diodes (OLEDs), organic solar cells (OSCs), and organic thin film transistors (OTFTs).1–12 For example, in a typical double-layer OLED, the energy level alignment at the heterostructure interface always depends on the excitons generation efficiency to emit light and hence reduces the onset voltage of OLEDs enormously.1,7 At the interface for exciton dissociation, the energy offset between the highest occupied molecular orbital of donor and the lowest unoccupied orbital of acceptor (HOMOD-LUMOA) limits the maximum open-circuit voltage (VOC) of heterojunction OSCs, and the power conversion efficiency of OSCs can be improved more than one order of magnitude by optimizing the interfacial energy level alignment at organic heterojunction interface.8–10 In OFETs, because of the strong charge transfer effect at the donor-acceptor heterojunction interface, organic heterostructure has been introduced to realize and/or enhance ambipolar OFETs to transport both holes and electrons within a single device.11 To better understand the function and to explore multifunctional role of organic-organic heterostructure in organic devices, it is thus of great importance to systematically study the heterostructure interfacial properties, among which the energy a)Author to whom correspondence should be addressed. Electronic mail:
[email protected] 0021-9606/2016/144(8)/084706/6/$30.00
level alignment at organic-organic heterostructure interface plays a crucial role in the interfacial property, and hence on the electronic process taking place in that region. For instance, using low work function (WF) Mg or CsF thin films, Lee’s group has successfully reduced the substrate WF and investigated the impact of low substrate WF on the energy level alignment at organic-organic heterostructure interfaces as well as the OSCs performance.13–15 Ratcliff et al. have also investigated the impact of a high electrode WF (NiOx) on the energy level alignment for a solution processed organic donor-acceptor heterostructure interface in OSCs.16 However, there is to date still short of systematical study about the effect of a high work function electrode on the electronic energy level structure of organic-organic heterostructure through the rarely reported results.16–18 In the present study, the interface energy level alignment between a classical donor-acceptor heterostructures, pentacene and fullerene (C60), was studied by introducing a high WF material MoO3 (WF higher than 6.0 eV18–20) as the substrate buffer layer to investigate substrate effect on the energy level alignment of the heterostructures. Specifically, the interfacial dipole formation and the energy level alignment at pentacene-C60 heterostructures interface with reverse deposition sequence either on ITO or on MoO3 modified ITO substrate were systematically studied by in situ ultraviolet photoelectron spectroscopy (UPS) measurements. We demonstrate that the high substrate work function has critical influences on the interfacial energy level alignment of pentacene-C60 heterostructures. In consideration of the fact that MoO3 is a commonly used p-type dopant or anode
144, 084706-1
© 2016 AIP Publishing LLC
084706-2
Zou et al.
buffer layer, our studies will facilitate better understanding of the interfacial energy level alignment of organic donoracceptor type heterostructure and provide novel design ideas of interface properties engineering and device optimization. II. EXPERIMENTAL
All UPS experiments were carried out in a Kratos AXIS Ultra DLD ultrahigh vacuum photoelectron spectroscopy connected to a custom-made high vacuum thermal evaporation system. The base pressure of the analysis chamber and the evaporation chamber was better than 5 × 10−10 and 5 × 10−9 Torr, respectively. ITO substrates were ultrasonically cleaned by de-ionized water, alcohol, and acetone in sequence before loading into the evaporation chamber. The pentacene (99.99%, Aldrich), C60 (99.995%, Aldrich), and MoO3 (99.9%, J&K Scientific) were separately evaporated at a pressure of about 2 × 10−8 Torr. The film deposition rate and thickness were in situ monitored using a calibrated quartz crystal microbalance (QCM) sensor (STM-2XM, Syncon Instrument). After each deposition, the samples were immediately transferred to the analysis chamber without breaking the vacuum for UPS analysis. UPS measurements were performed with He I (hν = 21.2 eV) as the excitation source at a pressure of 3 × 10−8 Torr, and the energy resolution is ∼100 meV. Vacuum level (VL) shifts were measured from the linear extrapolation of the low kinetic energy part of UPS spectra with a −9 V sample bias. The Fermi level (EF) was calibrated from a UPS spectrum of sputtered clean (Ar+ ion) Au substrate. All measurements were done at room temperature. III. RESULTS AND DISCUSSION
To study the electronic structure at C60/pentacene interface on ITO, a 7.5 nm pentacene and a 7.5 nm C60 are deposited onto the ITO substrate successively. Figs. 1(a) and
FIG. 1. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm C60 on 7.5 nm pentacene on ITO.
J. Chem. Phys. 144, 084706 (2016)
1(b) depict the evolution of the thickness-dependent secondary electron cutoff (SECO) and the valence band (HOMO) region of UPS spectra for 7.5 nm C60 on 7.5 nm pentacene on bare ITO. As depicted in Fig. 1(a), a gradually shift (∼0.16 eV) of the VL towards lower kinetic energy part with the increasing coverage of pentacene can be observed, or the work function decreases from 4.76 to 4.60 eV after the growth of 7.5 nm pentacene on ITO.4,21 Meanwhile, the HOMO leading edge of pentacene slightly increases towards higher binding energy part and finally locates at ∼0.34 eV below substrate EF, suggesting the HOMO or ionization potential (IP) of pentacene on ITO is ∼4.94 eV. After the deposition of C60 on pentacene/ITO, a progressive attenuation of the characteristic spectral features from pentacene is observed which are subsequently replaced by those of C60. With the increasing coverage of C60, no apparent VL shift can be observed at C60/pentacene interface, indicating a negligible or very weak charge transfer at the heterostructure interface. Meanwhile, there is no obvious band shift for C60 HOMO peak and the HOMO leading edge with increasing C60 thickness, which locate at the binding energy of ∼2.30 eV and ∼1.78 eV, respectively. The detailed energy level structure derived from Fig. 1 is depicted in Fig. 3(a). To investigate the impact of substrate WF on the energy level alignment of above-mentioned heterostructure, 5 nmthick MoO3 is first evaporated onto ITO. As a result, the substrate WF increases from ∼4.76 eV for bare ITO to ∼6.65 eV after MoO3 modification as shown in Fig. 2(a). The pentacene film is then deposited onto MoO3 modified ITO substrate. After the evaporation of 1.0 nm pentacene, a sharp downward shift of the VL by ∼1.25 eV is observed, and then it gradually decreases to ∼4.91 eV with further increasing the pentacene thickness to 7.5 nm. However, no shift of the HOMO leading edge of pentacene can be observed, which locates at ∼0.32 eV with the increasing coverage of pentacene, indicating a Fermi level pinning behavior of
FIG. 2. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm C60 on 7.5 nm pentacene on ITO/MoO3.
084706-3
Zou et al.
pentacene.22,23 For the growth of pentacene on MoO3 modified ITO, typical HOMO feature of pentacene can be observed after the deposition 4.5 nm pentacene. The IP of pentacene on MoO3 modified ITO is ∼5.23 eV, which is ∼0.29 eV higher than that of pentacene on bare ITO. It has been reported that the IP of ordered π-conjugated organic thin films strongly depends on the molecular orientation and packing mode.3,24 In the case of pentacene films, a standing-up configuration results in a lower IP, while lying-down configuration leads to a higher IP value.25–28 According to our UPS results, the standing-up configuration is more preferential for the growth of pentacene molecules on bare ITO substrate than on MoO3 modified ITO substrate. With the further deposition of C60 on pentacene on MoO3/ITO, upward VL shifts of ∼0.2 eV can be observed, implying ∼0.2 eV interfacial dipole exists at the interface. Meanwhile, a typical C60 valance feature emerges at the valence band region of the spectra. The HOMO peak and leading edge of C60 locate at ∼1.60 eV and ∼1.08 eV, respectively. The IP of C60 on pentacene on MoO3 modified ITO is ∼6.20 eV, which is ∼0.24 eV lower than that of C60 on pentacene on bare ITO (Fig. 1). Because of the spherically symmetric nature of C60 molecules, the IP difference of C60 should be unaffected by the orientation of pentacene molecules underneath. The IP difference is ascribed to the morphology change of C60 films on pentacene on MoO3 modified ITO compared to that on pentacene on bare ITO.29 The schematic energy level diagrams of the C60/pentacene heterostructures both on ITO and on MoO3 modified ITO substrates are shown in Figs. 3(a) and 3(b), respectively, where the HOMO positions are directly derived from the UPS measurements, and the LUMO edges are estimated by adding the charge transport gaps of 3.10, 2.30, and 2.30 eV for, respectively, MoO3, pentacene, and C60 to their corresponding HOMO energy level.18,30–32 For C60/pentacene heterostructure on bare ITO (Fig. 3(a)), since the substrate EF locates between the HOMO and LUMO of both pentacene and
J. Chem. Phys. 144, 084706 (2016)
C60, a negligible interface dipole is formed at the interface, leading to the flat energy level alignments at pentacene/ITO and C60/pentacene interface. However, for C60/pentacene heterostructure on MoO3 modified ITO substrate (Fig. 3(b)), the substrate WF increases to ∼6.65 eV, which is much higher than the IP of pentacene (∼5.23 eV). As a result, a spontaneous electron transfer from pentacene to MoO3/ITO substrate occurs upon contact, leading to a hole accumulation layer within pentacene films, and hence an abrupt VL shift at pentacene/MoO3 interface until the HOMO of pentacene is pining at substrate EF.22,23 As shown in Figs. 3(a) and 3(b), the HOMO leading edge of pentacene is about 0.33 ± 0.01 eV above the EF, no matter ITO or MoO3 modified ITO is used as substrate. On the other hand, the substrate WF has a great impact on the energy position of HOMO of C60. The HOMO leading edge of C60 locates at ∼1.78 and ∼1.08 eV for the heterostructure on bare and MoO3 modified ITO, respectively. As a result, the HOMOD-LUMOA offset for C60/pentacene heterostructure changes from ∼0.86 to ∼1.54 eV when the ITO substrate is replaced by MoO3 modified ITO. The substrate dependent energy level alignment effect at organic/organic heterojunction interface has been previously reported by using a low work function Mg substrate.15 However, the result in Ref. 15 shows that the EF is pinned at about 0.2-0.3 eV below the LUMO of acceptor (C60) in the second layer, and the HOMOD-LUMOA offset difference is tuned via shifting the EF position in donor (CuPc in Ref. 15) layer. In this work, the EF is pinned at about 0.3 eV above the HOMO of donor (pentacene) in the first layer, and the HOMOD-LUMOA offset difference can be tuned via shifting the EF position in acceptor (C60). Clearly, high work function substrate can also have a great impact on the energy level alignment of organic/organic heterostructure. Next, the interfacial electronic structure of pentacene on C60 (inverted heterostructure33) on bare and MoO3 modified ITO substrates have also been studied. As shown in the
FIG. 3. Schematic energy level diagrams for the C60/pentacene heterostructure (a) on ITO and (b) on MoO3 modified ITO.
084706-4
Zou et al.
J. Chem. Phys. 144, 084706 (2016)
FIG. 4. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm pentacene on 7.5 nm C60 on ITO.
FIG. 5. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm pentacene on 7.5 nm C60 on ITO/MoO3.
thickness-dependent UPS spectra (Fig. 4), an upward VL shift of ∼0.33 eV is observed after the deposition of 1.0 nm C60 on ITO, indicating the charge (electron) transfer from ITO to C60 upon contact. According to the energy level diagram shown in Fig. 6(a), the energy difference between the EF of ITO substrate and IP of C60 is ∼1.72 eV, which is much higher than that between the EF of ITO and electron affinity (EA) of C60 (∼0.58 eV). Such interfacial energy level alignment induces the electron transfer from ITO to C60. However, there is no recognizable VL shift with the further increasing coverage of C60. After the deposition of 7.5 nm C60, the WF is ∼5.15 eV and HOMO leading edge of C60 locates at ∼1.35 eV. With the growth of pentacene on C60, a progressive attenuation of the two pronounced characteristic photoemission peaks of C60 is observed. A gradual downward shift of VL, together with a slightly downward shift of the HOMO leading edge, is observed with the growth of 7.5 nm pentacene on C60/ITO. The schematic energy level diagram is depicted in Fig. 6(a). For comparison, when the C60 layer is first evaporated on MoO3 modified ITO, the substrate WF gradually decreases from ∼6.76 to ∼6.07 eV, and the C60 HOMO leading edge shifts from ∼0.26 to ∼0.42 eV as shown in Fig. 5, all of which make the HOMO of C60 seem to be pinned at the EF.34,35 With the increasing coverage of pentacene, the HOMO peak and the HOMO leading edge are kept almost constant, while the VL continually gradually decreases to ∼4.72 eV. That again means the HOMO of pentacene is pinning at the EF. The energy level structure derived from Fig. 5 is depicted in Fig. 6(b). For the inverted heterostructures, the HOMO leading edge of C60 on ITO locates at ∼1.35 eV below substrate EF, while the EF is pinned at ∼0.37 eV above the HOMO in pentacene layer, resulting in a HOMOD-LUMOA offset of ∼1.32 eV at the interface of pentacene/C60 on ITO. However, for C60 on MoO3 modified ITO substrate (Fig. 6(b)), the WF of MoO3/ITO (∼6.76 eV) is higher than the IP of C60 (∼6.49 eV),
a spontaneous electron transfer from C60 to MoO3/ITO substrate occurs until the HOMO of C60 is pining at substrate EF. After the deposition of pentacene films, since the WF of C60 on MoO3/ITO (∼6.07 eV) is still higher than the IP of pentacene (∼5.04 eV), a spontaneous electron transfer from pentacene to C60 occurs, and the pentacene layer also shows HOMO pinning effect, making pentacene/C60 heterostructure a flat HOMO and LUMO band alignment on MoO3/ITO substrate, with the HOMOD-LUMOA offset calculated to be ∼2.20 eV. Here again we show the substrate dependent energy level alignment effect to the organic/organic heterostructure. Recently, Oehzelt et al. developed a theoretical framework via a self-consistent numerical electrostatic model to simulate the energy level alignment at organic/electrode and organic/organic interface by taking into account a continuous density of states (DOS) both for the occupied and the unoccupied levels in the organic semiconductors.36,37 Here we qualitatively compared our results with their model predictions. In order to simplify the comparison, the molecular growth mode and the IP/EA difference as well as the other model-used parameters difference between this work and calculation of Oehzelt et al.37 are neglected. Pentacene-C60 belongs to the OSCs-relevant scenario of organic type II heterojunction as illustrated in Fig. 4 in Ref. 37. As for C60/pentacene on ITO (Fig. 3(a)), where the substrate EF lies within the fundamental gap of both donor (pentacene) and acceptor (C60), both the two organic layers stay in charge-neutral states and almost VL alignment happens among substrate, donor, and acceptor, which is consistent with central panel of Fig. 4(a) in Ref. 37. While for C60/pentacene on MoO3/ITO (Fig. 3(b)), where the substrate EF lies much deeper than the HOMO of donor (pentacene), holes thus accumulate in the pentacene HOMO level due to Fermi-Dirac occupation, leading to an overall downward shift in the VL at donor/substrate interface accordingly as indicated in the right panel of Fig. 4(a) in Ref. 37. After further deposition
084706-5
Zou et al.
J. Chem. Phys. 144, 084706 (2016)
FIG. 6. Schematic energy level diagrams for the pentacene/C60 heterostructure (a) on ITO and (b) on MoO3 modified ITO.
of acceptor (C60) layer, nearly VL alignment between donor (pentacene) and acceptor (C60) happens. Nevertheless, as for the reverse stacking sequence pentacene/C60 structure on ITO (Fig. 6(a)), although the substrate EF lies within the gap of both acceptor (C60) and donor (pentacene) again, VL shifts for both C60/ITO and pentacene/C60 interfaces are observed due to the model unpredicted interface charge transfer or interface dipole exists between the ITO substrate and C60. While for pentacene/C60 on MoO3/ITO (Fig. 6(b)), where the substrate EF lies deeper than the HOMO of both acceptor and donor layer, hole accumulation and associated VL shifts occur at both C60/substrate and pentacene/C60 interfaces, which is consistent with the right panel scenario of Fig. 4(b) in Ref. 37. IV. CONCLUSION
In summary, we have carried out systematic investigations on the impact of MoO3 interlayer on the energy level alignment of pentacene-C60 heterostructure using in situ UPS measurements. We show that the energy level alignment of both the direct and inverted architectures of pentacene-C60 heterostructure can be effectively tuned by the high work function substrate, leading to a tunable value of energy difference between HOMOD-LUMOA offset. This research provides critical evidence of substrate dependence of the energy level alignment for organic/organic heterostructures. And the result of this research can be used to potentially enhance the performance of OSCs and other organic electronic devices. ACKNOWLEDGMENTS
We are grateful to the financial support by National Natural Science Foundation of China (Grant Nos. 61401136 and 51203167), the 135 Project of ICCAS, the State Key
Lab of Silicon Materials (No. SKL2014-9) and the Chinese Academy of Sciences. 1H.
Ishii, K. Sugiyama, E. Ito, and K. Seki, Adv. Mater. 11, 605 (1999). Braun, W. R. Salaneck, and M. Fahlman, Adv. Mater. 21, 1450 (2009). 3W. Chen, D.-C. Qi, H. Huang, X. Gao, and A. T. S. Wee, Adv. Funct. Mater. 21, 410 (2011). 4S. Zhong, J. Q. Zhong, H. Y. Mao, J. L. Zhang, J. D. Lin, and W. Chen, Phys. Chem. Chem. Phys. 14, 14127 (2012). 5F. Bussolotti, J. Yang, A. Hinderhofer, Y. Huang, W. Chen, S. Kera, A. T. S. Wee, and N. Ueno, Phys. Rev. B 89, 115319 (2014). 6N. Koch, Phys. Status Solidi RRL 6, 277 (2012). 7C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett. 51, 913 (1987). 8H. Spanggaard and F. C. Krebs, Sol. Energy Mater. Sol. Cells 83, 125 (2004). 9B. P. Rand, D. P. Burk, and S. R. Forrest, Phys. Rev. B 75, 115327 (2007). 10C. W. Tang, Appl. Phys. Lett. 48, 183 (1986). 11H. Wang and D. Yan, NPG Asia Mater. 2, 69 (2010). 12H. Y. Mao, F. Bussolotti, D.-C. Qi, R. Wang, S. Kera, N. Ueno, A. T. S. Wee, and W. Chen, Org. Electron. 12, 534 (2011). 13J. X. Tang, K. M. Lau, C. S. Lee, and S. T. Lee, Appl. Phys. Lett. 88, 232103 (2006). 14T. W. Ng, M. F. Lo, Z. T. Liu, F. L. Wong, S. L. Lai, M. K. Fung, C. S. Lee, and S. T. Lee, J. Appl. Phys. 106, 114501 (2009). 15Y. Zhou, Z. T. Liu, J. X. Tang, C. S. Lee, and S. T. Lee, J. Electron Spectrosc. Relat. Phenom. 174, 35 (2009). 16E. L. Ratcliff, J. Meyer, K. X. Steirer, N. R. Armstrong, D. Olson, and A. Kahn, Org. Electron. 13, 744 (2012). 17S. Braun, M. P. de Jong, W. Osikowicz, and W. R. Salaneck, Appl. Phys. Lett. 91, 202108 (2007). 18X. Liu, S. Yi, C. Wang, C. Wang, and Y. Gao, J. Appl. Phys. 115, 163708 (2014). 19D. Y. Kim, J. Subbiah, G. Sarasqueta, F. So, H. Ding, Irfan, and Y. Gao, Appl. Phys. Lett. 95, 093304 (2009). 20Z. Chen, I. Santoso, R. Wang, L. F. Xie, H. Y. Mao, H. Huang, Y. Z. Wang, X. Y. Gao, Z. K. Chen, D. Ma, A. T. S. Wee, and W. Chen, Appl. Phys. Lett. 96, 213104 (2010). 21N. Koch, J. Phys.: Condens. Matter 20, 184008 (2008). 22N. Koch, A. Elschner, J. P. Rabe, and R. L. Johnson, Adv. Mater. 17, 330 (2005). 23N. Koch and A. Vollmer, Appl. Phys. Lett. 89, 162107 (2006). 24S. Duhm, G. Heimel, I. Salzmann, H. Glowatzki, R. L. Johnson, A. Vollmer, J. P. Rabe, and N. Koch, Nat. Mater. 7, 326 (2008). 25S. Verlaak, D. Beljonne, D. Cheyns, C. Rolin, M. Linares, F. Castet, J. Cornil, and P. Heremans, Adv. Funct. Mater. 19, 3809 (2009). 26P. G. Schroeder, C. B. France, J. B. Park, and B. A. Parkinson, J. Phys. Chem. B 107, 2253 (2003). 2S.
084706-6 27X.
Zou et al.
Liu, Y. Zhan, S. Braun, F. Li, and M. Fahlman, Phys. Rev. B 80, 115401 (2009). 28G. Heimel and N. Koch, in Interface Controlled Organic Thin Films, Springer Proceedings in Physics Vol. 129, edited by K. Al-Shamery, G. Horowitz, H. Sitter, and H.-G. Rubahn (Springer, Berlin, Heidelberg, 2009), pp. 141–145. 29F. J. Zhang, A. Vollmer, J. Zhang, Z. Xu, J. P. Rabe, and N. Koch, Org. Electron. 8, 606 (2007). 30C. Wang, A. J. Turinske, and Y. Gao, Appl. Phys. B 113, 361 (2013). 31J. Hwang, A. Wan, and A. Kahn, Mater. Sci. Eng., R 64, 1 (2009).
J. Chem. Phys. 144, 084706 (2016) 32W.
Han, H. Yoshida, N. Ueno, and S. Kera, Appl. Phys. Lett. 103, 123303 (2013). 33D. M. Nanditha, M. Dissanayake, R. A. Hatton, R. J. Curry, and S. R. P. Silva, Appl. Phys. Lett. 90, 113505 (2007). 34L. Chai, R. T. White, M. T. Greiner, and Z. H. Lu, Phys. Rev. B 89, 035202 (2014). 35Irfan, M. Zhang, H. Ding, C. W. Tang, and Y. Gao, Org. Electron. 12, 1588 (2011). 36M. Oehzelt, N. Koch, and G. Heimeil, Nat. Commun. 5, 4174 (2014). 37M. Oehzelt, K. Akaike, N. Koch, and G. Heimel, Sci. Adv. 1, e1501127 (2015).
Citation: The Journal of Chemical Physics 144, 084706 (2016); doi: 10.1063/1.4942480 View online: http://dx.doi.org/10.1063/1.4942480 View Table of Contents: http://aip.scitation.org/toc/jcp/144/8 Published by the American Institute of Physics
THE JOURNAL OF CHEMICAL PHYSICS 144, 084706 (2016)
Impact of MoO3 interlayer on the energy level alignment of pentacene-C60 heterostructure Ye Zou,1,a) Hongying Mao,2,3 Qing Meng,1 and Daoben Zhu1 1
Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China 2 Department of Physics, Hangzhou Normal University, Hangzhou 310036, China 3 State Key Laboratory of Silicon Materials, School of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, China
(Received 3 December 2015; accepted 9 February 2016; published online 25 February 2016) Using in situ ultraviolet photoelectron spectroscopy, the electronic structure evolutions at the interface between pentacene and fullerene (C60), a classical organic donor-acceptor heterostructure in organic electronic devices, on indium-tin oxide (ITO) and MoO3 modified ITO substrates have been investigated. The insertion of a thin layer MoO3 has a significant impact on the interfacial energy level alignment of pentacene-C60 heterostructure. For the deposition of C60 on pentacene, the energy difference between the highest occupied molecular orbital of donor and the lowest unoccupied molecular orbital of acceptor (HOMOD-LUMOA) offset of C60/pentacene heterostructure increased from 0.86 eV to 1.54 eV after the insertion of a thin layer MoO3 on ITO. In the inverted heterostructrure where pentacene was deposited on C60, the HOMOD-LUMOA offset of pentacene/C60 heterostructure increased from 1.32 to 2.20 eV after MoO3 modification on ITO. The significant difference of HOMOD-LUMOA offset shows the feasibility to optimize organic electronic device performance through interfacial engineering approaches, such as the insertion of a thin layer high work function MoO3 films. C 2016 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4942480]
I. INTRODUCTION
Organic-organic heterojunction structures have drawn increasing research attention due to the importance in determining the performance of organic electronic devices including organic light-emitting diodes (OLEDs), organic solar cells (OSCs), and organic thin film transistors (OTFTs).1–12 For example, in a typical double-layer OLED, the energy level alignment at the heterostructure interface always depends on the excitons generation efficiency to emit light and hence reduces the onset voltage of OLEDs enormously.1,7 At the interface for exciton dissociation, the energy offset between the highest occupied molecular orbital of donor and the lowest unoccupied orbital of acceptor (HOMOD-LUMOA) limits the maximum open-circuit voltage (VOC) of heterojunction OSCs, and the power conversion efficiency of OSCs can be improved more than one order of magnitude by optimizing the interfacial energy level alignment at organic heterojunction interface.8–10 In OFETs, because of the strong charge transfer effect at the donor-acceptor heterojunction interface, organic heterostructure has been introduced to realize and/or enhance ambipolar OFETs to transport both holes and electrons within a single device.11 To better understand the function and to explore multifunctional role of organic-organic heterostructure in organic devices, it is thus of great importance to systematically study the heterostructure interfacial properties, among which the energy a)Author to whom correspondence should be addressed. Electronic mail:
[email protected] 0021-9606/2016/144(8)/084706/6/$30.00
level alignment at organic-organic heterostructure interface plays a crucial role in the interfacial property, and hence on the electronic process taking place in that region. For instance, using low work function (WF) Mg or CsF thin films, Lee’s group has successfully reduced the substrate WF and investigated the impact of low substrate WF on the energy level alignment at organic-organic heterostructure interfaces as well as the OSCs performance.13–15 Ratcliff et al. have also investigated the impact of a high electrode WF (NiOx) on the energy level alignment for a solution processed organic donor-acceptor heterostructure interface in OSCs.16 However, there is to date still short of systematical study about the effect of a high work function electrode on the electronic energy level structure of organic-organic heterostructure through the rarely reported results.16–18 In the present study, the interface energy level alignment between a classical donor-acceptor heterostructures, pentacene and fullerene (C60), was studied by introducing a high WF material MoO3 (WF higher than 6.0 eV18–20) as the substrate buffer layer to investigate substrate effect on the energy level alignment of the heterostructures. Specifically, the interfacial dipole formation and the energy level alignment at pentacene-C60 heterostructures interface with reverse deposition sequence either on ITO or on MoO3 modified ITO substrate were systematically studied by in situ ultraviolet photoelectron spectroscopy (UPS) measurements. We demonstrate that the high substrate work function has critical influences on the interfacial energy level alignment of pentacene-C60 heterostructures. In consideration of the fact that MoO3 is a commonly used p-type dopant or anode
144, 084706-1
© 2016 AIP Publishing LLC
084706-2
Zou et al.
buffer layer, our studies will facilitate better understanding of the interfacial energy level alignment of organic donoracceptor type heterostructure and provide novel design ideas of interface properties engineering and device optimization. II. EXPERIMENTAL
All UPS experiments were carried out in a Kratos AXIS Ultra DLD ultrahigh vacuum photoelectron spectroscopy connected to a custom-made high vacuum thermal evaporation system. The base pressure of the analysis chamber and the evaporation chamber was better than 5 × 10−10 and 5 × 10−9 Torr, respectively. ITO substrates were ultrasonically cleaned by de-ionized water, alcohol, and acetone in sequence before loading into the evaporation chamber. The pentacene (99.99%, Aldrich), C60 (99.995%, Aldrich), and MoO3 (99.9%, J&K Scientific) were separately evaporated at a pressure of about 2 × 10−8 Torr. The film deposition rate and thickness were in situ monitored using a calibrated quartz crystal microbalance (QCM) sensor (STM-2XM, Syncon Instrument). After each deposition, the samples were immediately transferred to the analysis chamber without breaking the vacuum for UPS analysis. UPS measurements were performed with He I (hν = 21.2 eV) as the excitation source at a pressure of 3 × 10−8 Torr, and the energy resolution is ∼100 meV. Vacuum level (VL) shifts were measured from the linear extrapolation of the low kinetic energy part of UPS spectra with a −9 V sample bias. The Fermi level (EF) was calibrated from a UPS spectrum of sputtered clean (Ar+ ion) Au substrate. All measurements were done at room temperature. III. RESULTS AND DISCUSSION
To study the electronic structure at C60/pentacene interface on ITO, a 7.5 nm pentacene and a 7.5 nm C60 are deposited onto the ITO substrate successively. Figs. 1(a) and
FIG. 1. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm C60 on 7.5 nm pentacene on ITO.
J. Chem. Phys. 144, 084706 (2016)
1(b) depict the evolution of the thickness-dependent secondary electron cutoff (SECO) and the valence band (HOMO) region of UPS spectra for 7.5 nm C60 on 7.5 nm pentacene on bare ITO. As depicted in Fig. 1(a), a gradually shift (∼0.16 eV) of the VL towards lower kinetic energy part with the increasing coverage of pentacene can be observed, or the work function decreases from 4.76 to 4.60 eV after the growth of 7.5 nm pentacene on ITO.4,21 Meanwhile, the HOMO leading edge of pentacene slightly increases towards higher binding energy part and finally locates at ∼0.34 eV below substrate EF, suggesting the HOMO or ionization potential (IP) of pentacene on ITO is ∼4.94 eV. After the deposition of C60 on pentacene/ITO, a progressive attenuation of the characteristic spectral features from pentacene is observed which are subsequently replaced by those of C60. With the increasing coverage of C60, no apparent VL shift can be observed at C60/pentacene interface, indicating a negligible or very weak charge transfer at the heterostructure interface. Meanwhile, there is no obvious band shift for C60 HOMO peak and the HOMO leading edge with increasing C60 thickness, which locate at the binding energy of ∼2.30 eV and ∼1.78 eV, respectively. The detailed energy level structure derived from Fig. 1 is depicted in Fig. 3(a). To investigate the impact of substrate WF on the energy level alignment of above-mentioned heterostructure, 5 nmthick MoO3 is first evaporated onto ITO. As a result, the substrate WF increases from ∼4.76 eV for bare ITO to ∼6.65 eV after MoO3 modification as shown in Fig. 2(a). The pentacene film is then deposited onto MoO3 modified ITO substrate. After the evaporation of 1.0 nm pentacene, a sharp downward shift of the VL by ∼1.25 eV is observed, and then it gradually decreases to ∼4.91 eV with further increasing the pentacene thickness to 7.5 nm. However, no shift of the HOMO leading edge of pentacene can be observed, which locates at ∼0.32 eV with the increasing coverage of pentacene, indicating a Fermi level pinning behavior of
FIG. 2. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm C60 on 7.5 nm pentacene on ITO/MoO3.
084706-3
Zou et al.
pentacene.22,23 For the growth of pentacene on MoO3 modified ITO, typical HOMO feature of pentacene can be observed after the deposition 4.5 nm pentacene. The IP of pentacene on MoO3 modified ITO is ∼5.23 eV, which is ∼0.29 eV higher than that of pentacene on bare ITO. It has been reported that the IP of ordered π-conjugated organic thin films strongly depends on the molecular orientation and packing mode.3,24 In the case of pentacene films, a standing-up configuration results in a lower IP, while lying-down configuration leads to a higher IP value.25–28 According to our UPS results, the standing-up configuration is more preferential for the growth of pentacene molecules on bare ITO substrate than on MoO3 modified ITO substrate. With the further deposition of C60 on pentacene on MoO3/ITO, upward VL shifts of ∼0.2 eV can be observed, implying ∼0.2 eV interfacial dipole exists at the interface. Meanwhile, a typical C60 valance feature emerges at the valence band region of the spectra. The HOMO peak and leading edge of C60 locate at ∼1.60 eV and ∼1.08 eV, respectively. The IP of C60 on pentacene on MoO3 modified ITO is ∼6.20 eV, which is ∼0.24 eV lower than that of C60 on pentacene on bare ITO (Fig. 1). Because of the spherically symmetric nature of C60 molecules, the IP difference of C60 should be unaffected by the orientation of pentacene molecules underneath. The IP difference is ascribed to the morphology change of C60 films on pentacene on MoO3 modified ITO compared to that on pentacene on bare ITO.29 The schematic energy level diagrams of the C60/pentacene heterostructures both on ITO and on MoO3 modified ITO substrates are shown in Figs. 3(a) and 3(b), respectively, where the HOMO positions are directly derived from the UPS measurements, and the LUMO edges are estimated by adding the charge transport gaps of 3.10, 2.30, and 2.30 eV for, respectively, MoO3, pentacene, and C60 to their corresponding HOMO energy level.18,30–32 For C60/pentacene heterostructure on bare ITO (Fig. 3(a)), since the substrate EF locates between the HOMO and LUMO of both pentacene and
J. Chem. Phys. 144, 084706 (2016)
C60, a negligible interface dipole is formed at the interface, leading to the flat energy level alignments at pentacene/ITO and C60/pentacene interface. However, for C60/pentacene heterostructure on MoO3 modified ITO substrate (Fig. 3(b)), the substrate WF increases to ∼6.65 eV, which is much higher than the IP of pentacene (∼5.23 eV). As a result, a spontaneous electron transfer from pentacene to MoO3/ITO substrate occurs upon contact, leading to a hole accumulation layer within pentacene films, and hence an abrupt VL shift at pentacene/MoO3 interface until the HOMO of pentacene is pining at substrate EF.22,23 As shown in Figs. 3(a) and 3(b), the HOMO leading edge of pentacene is about 0.33 ± 0.01 eV above the EF, no matter ITO or MoO3 modified ITO is used as substrate. On the other hand, the substrate WF has a great impact on the energy position of HOMO of C60. The HOMO leading edge of C60 locates at ∼1.78 and ∼1.08 eV for the heterostructure on bare and MoO3 modified ITO, respectively. As a result, the HOMOD-LUMOA offset for C60/pentacene heterostructure changes from ∼0.86 to ∼1.54 eV when the ITO substrate is replaced by MoO3 modified ITO. The substrate dependent energy level alignment effect at organic/organic heterojunction interface has been previously reported by using a low work function Mg substrate.15 However, the result in Ref. 15 shows that the EF is pinned at about 0.2-0.3 eV below the LUMO of acceptor (C60) in the second layer, and the HOMOD-LUMOA offset difference is tuned via shifting the EF position in donor (CuPc in Ref. 15) layer. In this work, the EF is pinned at about 0.3 eV above the HOMO of donor (pentacene) in the first layer, and the HOMOD-LUMOA offset difference can be tuned via shifting the EF position in acceptor (C60). Clearly, high work function substrate can also have a great impact on the energy level alignment of organic/organic heterostructure. Next, the interfacial electronic structure of pentacene on C60 (inverted heterostructure33) on bare and MoO3 modified ITO substrates have also been studied. As shown in the
FIG. 3. Schematic energy level diagrams for the C60/pentacene heterostructure (a) on ITO and (b) on MoO3 modified ITO.
084706-4
Zou et al.
J. Chem. Phys. 144, 084706 (2016)
FIG. 4. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm pentacene on 7.5 nm C60 on ITO.
FIG. 5. Evolution of UPS spectra at (a) the low kinetic energy region (secondary electron cutoff) and (b) the low binding energy region (HOMO region) during the sequential deposition of 7.5 nm pentacene on 7.5 nm C60 on ITO/MoO3.
thickness-dependent UPS spectra (Fig. 4), an upward VL shift of ∼0.33 eV is observed after the deposition of 1.0 nm C60 on ITO, indicating the charge (electron) transfer from ITO to C60 upon contact. According to the energy level diagram shown in Fig. 6(a), the energy difference between the EF of ITO substrate and IP of C60 is ∼1.72 eV, which is much higher than that between the EF of ITO and electron affinity (EA) of C60 (∼0.58 eV). Such interfacial energy level alignment induces the electron transfer from ITO to C60. However, there is no recognizable VL shift with the further increasing coverage of C60. After the deposition of 7.5 nm C60, the WF is ∼5.15 eV and HOMO leading edge of C60 locates at ∼1.35 eV. With the growth of pentacene on C60, a progressive attenuation of the two pronounced characteristic photoemission peaks of C60 is observed. A gradual downward shift of VL, together with a slightly downward shift of the HOMO leading edge, is observed with the growth of 7.5 nm pentacene on C60/ITO. The schematic energy level diagram is depicted in Fig. 6(a). For comparison, when the C60 layer is first evaporated on MoO3 modified ITO, the substrate WF gradually decreases from ∼6.76 to ∼6.07 eV, and the C60 HOMO leading edge shifts from ∼0.26 to ∼0.42 eV as shown in Fig. 5, all of which make the HOMO of C60 seem to be pinned at the EF.34,35 With the increasing coverage of pentacene, the HOMO peak and the HOMO leading edge are kept almost constant, while the VL continually gradually decreases to ∼4.72 eV. That again means the HOMO of pentacene is pinning at the EF. The energy level structure derived from Fig. 5 is depicted in Fig. 6(b). For the inverted heterostructures, the HOMO leading edge of C60 on ITO locates at ∼1.35 eV below substrate EF, while the EF is pinned at ∼0.37 eV above the HOMO in pentacene layer, resulting in a HOMOD-LUMOA offset of ∼1.32 eV at the interface of pentacene/C60 on ITO. However, for C60 on MoO3 modified ITO substrate (Fig. 6(b)), the WF of MoO3/ITO (∼6.76 eV) is higher than the IP of C60 (∼6.49 eV),
a spontaneous electron transfer from C60 to MoO3/ITO substrate occurs until the HOMO of C60 is pining at substrate EF. After the deposition of pentacene films, since the WF of C60 on MoO3/ITO (∼6.07 eV) is still higher than the IP of pentacene (∼5.04 eV), a spontaneous electron transfer from pentacene to C60 occurs, and the pentacene layer also shows HOMO pinning effect, making pentacene/C60 heterostructure a flat HOMO and LUMO band alignment on MoO3/ITO substrate, with the HOMOD-LUMOA offset calculated to be ∼2.20 eV. Here again we show the substrate dependent energy level alignment effect to the organic/organic heterostructure. Recently, Oehzelt et al. developed a theoretical framework via a self-consistent numerical electrostatic model to simulate the energy level alignment at organic/electrode and organic/organic interface by taking into account a continuous density of states (DOS) both for the occupied and the unoccupied levels in the organic semiconductors.36,37 Here we qualitatively compared our results with their model predictions. In order to simplify the comparison, the molecular growth mode and the IP/EA difference as well as the other model-used parameters difference between this work and calculation of Oehzelt et al.37 are neglected. Pentacene-C60 belongs to the OSCs-relevant scenario of organic type II heterojunction as illustrated in Fig. 4 in Ref. 37. As for C60/pentacene on ITO (Fig. 3(a)), where the substrate EF lies within the fundamental gap of both donor (pentacene) and acceptor (C60), both the two organic layers stay in charge-neutral states and almost VL alignment happens among substrate, donor, and acceptor, which is consistent with central panel of Fig. 4(a) in Ref. 37. While for C60/pentacene on MoO3/ITO (Fig. 3(b)), where the substrate EF lies much deeper than the HOMO of donor (pentacene), holes thus accumulate in the pentacene HOMO level due to Fermi-Dirac occupation, leading to an overall downward shift in the VL at donor/substrate interface accordingly as indicated in the right panel of Fig. 4(a) in Ref. 37. After further deposition
084706-5
Zou et al.
J. Chem. Phys. 144, 084706 (2016)
FIG. 6. Schematic energy level diagrams for the pentacene/C60 heterostructure (a) on ITO and (b) on MoO3 modified ITO.
of acceptor (C60) layer, nearly VL alignment between donor (pentacene) and acceptor (C60) happens. Nevertheless, as for the reverse stacking sequence pentacene/C60 structure on ITO (Fig. 6(a)), although the substrate EF lies within the gap of both acceptor (C60) and donor (pentacene) again, VL shifts for both C60/ITO and pentacene/C60 interfaces are observed due to the model unpredicted interface charge transfer or interface dipole exists between the ITO substrate and C60. While for pentacene/C60 on MoO3/ITO (Fig. 6(b)), where the substrate EF lies deeper than the HOMO of both acceptor and donor layer, hole accumulation and associated VL shifts occur at both C60/substrate and pentacene/C60 interfaces, which is consistent with the right panel scenario of Fig. 4(b) in Ref. 37. IV. CONCLUSION
In summary, we have carried out systematic investigations on the impact of MoO3 interlayer on the energy level alignment of pentacene-C60 heterostructure using in situ UPS measurements. We show that the energy level alignment of both the direct and inverted architectures of pentacene-C60 heterostructure can be effectively tuned by the high work function substrate, leading to a tunable value of energy difference between HOMOD-LUMOA offset. This research provides critical evidence of substrate dependence of the energy level alignment for organic/organic heterostructures. And the result of this research can be used to potentially enhance the performance of OSCs and other organic electronic devices. ACKNOWLEDGMENTS
We are grateful to the financial support by National Natural Science Foundation of China (Grant Nos. 61401136 and 51203167), the 135 Project of ICCAS, the State Key
Lab of Silicon Materials (No. SKL2014-9) and the Chinese Academy of Sciences. 1H.
Ishii, K. Sugiyama, E. Ito, and K. Seki, Adv. Mater. 11, 605 (1999). Braun, W. R. Salaneck, and M. Fahlman, Adv. Mater. 21, 1450 (2009). 3W. Chen, D.-C. Qi, H. Huang, X. Gao, and A. T. S. Wee, Adv. Funct. Mater. 21, 410 (2011). 4S. Zhong, J. Q. Zhong, H. Y. Mao, J. L. Zhang, J. D. Lin, and W. Chen, Phys. Chem. Chem. Phys. 14, 14127 (2012). 5F. Bussolotti, J. Yang, A. Hinderhofer, Y. Huang, W. Chen, S. Kera, A. T. S. Wee, and N. Ueno, Phys. Rev. B 89, 115319 (2014). 6N. Koch, Phys. Status Solidi RRL 6, 277 (2012). 7C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett. 51, 913 (1987). 8H. Spanggaard and F. C. Krebs, Sol. Energy Mater. Sol. Cells 83, 125 (2004). 9B. P. Rand, D. P. Burk, and S. R. Forrest, Phys. Rev. B 75, 115327 (2007). 10C. W. Tang, Appl. Phys. Lett. 48, 183 (1986). 11H. Wang and D. Yan, NPG Asia Mater. 2, 69 (2010). 12H. Y. Mao, F. Bussolotti, D.-C. Qi, R. Wang, S. Kera, N. Ueno, A. T. S. Wee, and W. Chen, Org. Electron. 12, 534 (2011). 13J. X. Tang, K. M. Lau, C. S. Lee, and S. T. Lee, Appl. Phys. Lett. 88, 232103 (2006). 14T. W. Ng, M. F. Lo, Z. T. Liu, F. L. Wong, S. L. Lai, M. K. Fung, C. S. Lee, and S. T. Lee, J. Appl. Phys. 106, 114501 (2009). 15Y. Zhou, Z. T. Liu, J. X. Tang, C. S. Lee, and S. T. Lee, J. Electron Spectrosc. Relat. Phenom. 174, 35 (2009). 16E. L. Ratcliff, J. Meyer, K. X. Steirer, N. R. Armstrong, D. Olson, and A. Kahn, Org. Electron. 13, 744 (2012). 17S. Braun, M. P. de Jong, W. Osikowicz, and W. R. Salaneck, Appl. Phys. Lett. 91, 202108 (2007). 18X. Liu, S. Yi, C. Wang, C. Wang, and Y. Gao, J. Appl. Phys. 115, 163708 (2014). 19D. Y. Kim, J. Subbiah, G. Sarasqueta, F. So, H. Ding, Irfan, and Y. Gao, Appl. Phys. Lett. 95, 093304 (2009). 20Z. Chen, I. Santoso, R. Wang, L. F. Xie, H. Y. Mao, H. Huang, Y. Z. Wang, X. Y. Gao, Z. K. Chen, D. Ma, A. T. S. Wee, and W. Chen, Appl. Phys. Lett. 96, 213104 (2010). 21N. Koch, J. Phys.: Condens. Matter 20, 184008 (2008). 22N. Koch, A. Elschner, J. P. Rabe, and R. L. Johnson, Adv. Mater. 17, 330 (2005). 23N. Koch and A. Vollmer, Appl. Phys. Lett. 89, 162107 (2006). 24S. Duhm, G. Heimel, I. Salzmann, H. Glowatzki, R. L. Johnson, A. Vollmer, J. P. Rabe, and N. Koch, Nat. Mater. 7, 326 (2008). 25S. Verlaak, D. Beljonne, D. Cheyns, C. Rolin, M. Linares, F. Castet, J. Cornil, and P. Heremans, Adv. Funct. Mater. 19, 3809 (2009). 26P. G. Schroeder, C. B. France, J. B. Park, and B. A. Parkinson, J. Phys. Chem. B 107, 2253 (2003). 2S.
084706-6 27X.
Zou et al.
Liu, Y. Zhan, S. Braun, F. Li, and M. Fahlman, Phys. Rev. B 80, 115401 (2009). 28G. Heimel and N. Koch, in Interface Controlled Organic Thin Films, Springer Proceedings in Physics Vol. 129, edited by K. Al-Shamery, G. Horowitz, H. Sitter, and H.-G. Rubahn (Springer, Berlin, Heidelberg, 2009), pp. 141–145. 29F. J. Zhang, A. Vollmer, J. Zhang, Z. Xu, J. P. Rabe, and N. Koch, Org. Electron. 8, 606 (2007). 30C. Wang, A. J. Turinske, and Y. Gao, Appl. Phys. B 113, 361 (2013). 31J. Hwang, A. Wan, and A. Kahn, Mater. Sci. Eng., R 64, 1 (2009).
J. Chem. Phys. 144, 084706 (2016) 32W.
Han, H. Yoshida, N. Ueno, and S. Kera, Appl. Phys. Lett. 103, 123303 (2013). 33D. M. Nanditha, M. Dissanayake, R. A. Hatton, R. J. Curry, and S. R. P. Silva, Appl. Phys. Lett. 90, 113505 (2007). 34L. Chai, R. T. White, M. T. Greiner, and Z. H. Lu, Phys. Rev. B 89, 035202 (2014). 35Irfan, M. Zhang, H. Ding, C. W. Tang, and Y. Gao, Org. Electron. 12, 1588 (2011). 36M. Oehzelt, N. Koch, and G. Heimeil, Nat. Commun. 5, 4174 (2014). 37M. Oehzelt, K. Akaike, N. Koch, and G. Heimel, Sci. Adv. 1, e1501127 (2015).
