Journal of ELECTRONIC MATERIALS

DOI: 10.1007/s11664-017-5605-7 Ó 2017 The Minerals, Metals & Materials Society

Properties of Vanadium-Doped Indium Oxide Deposited at Room Temperature as Transparent Conductor for Inverted Polymer Solar Cells MIN-JUN CHOI,1 KEUN YONG LIM,2,3 HYUN-WOO PARK,1 HAN-KI KIM,2 DO KYUNG HWANG,3 SUNG-JIN LIM,4 JAE WON SHIM,4,5 and KWUN-BUM CHUNG 1,6 1.—Division of Physics and Semiconductor Science, Dongguk University, Seoul 100715, Republic of Korea. 2.—Department of Advanced Materials Engineering for Information and Electronics, Kyung Hee University, Yongin 446-701, Republic of Korea. 3.—Materials and Life Science Research Division, Korea Institute of Science and Technology (KIST), Seoul 02792, Republic of Korea. 4.—Department of Electronics and Electrical Engineering, Dongguk University, Seoul 100715, Republic of Korea. 5.—e-mail: [email protected]. 6.—e-mail: [email protected]

The properties of vanadium-doped indium oxide (IVO) deposited at room temperature as a transparent conductor for inverted polymer solar cells have been investigated as a function of the vanadium doping concentration. IVO film prepared with V doping concentration of 0.03% showed optimal properties for use as a transparent conductor with figure of merit of 4.35 9 10 3 Ohm 1, related to altered band alignment between the Fermi level and conductionband minimum. In the optimal optoelectrical conditions for the IVO film, performance optimization of PTB7:PC70BM inverted polymer solar cells resulted in maximum power conversion efficiency of 4.7 ± 0.4% under simulated air mass 1.5 global illumination at 100 mW/cm2. Key words: Indium vanadium oxide, inverted polymer solar cell, electronic structure

INTRODUCTION Polymer solar cells (PSCs) are considered to be promising renewable and ecofriendly energy sources due to their unique electrical and optical properties, as well as their potentially cost-effective processability.1–4 In general, a PSC comprises a donor–acceptor system-based bulk heterojunction photoactive layer sandwiched between two chargecollecting electrodes, a hole-collecting electrode (HCE) and an electron-collecting electrode (ECE). Conventionally, the HCE with large work function (WF) and high transparency sits on the bottom of the device, while the ECE with small WF and reflective property is located on top of the cell.3,4 The performance of a PSC depends strongly on the optoelectrical properties of the photoactive layer

(Received November 1, 2016; accepted May 19, 2017)

and the two charge-collecting electrodes, and much effort has been focused on improvement of these properties.5–8 Transparent conducting electrodes (TCEs) have often been adopted as the HCE.9,10 For efficient operation of the HCE, such a TCE should fulfill several requirements, including low electrical resistivity (10 3 Ohm cm), high optical transmittance (>80%) in the visible-light range (k = 380 nm to 800 nm), and sufficiently large WF value.11 Tindoped indium oxide (ITO) has been most widely used for HCEs due to its low resistivity and high transparency; however, the relatively large energylevel mismatch between the WF of ITO and the ionization energy level of the donor polymer requires an additional buffer layer such as conductive polymer poly(3,4-ethylene dioxylene thiophene):poly(styrene sulfonic acid) (PEDOT:PSS) or metal oxide NiO to enhance the hole collection efficiency.5,12 In addition, a thermal annealing

Choi, K.Y. Lim, Park, Kim, Hwang, S.-J. Lim, Shim, and Chung

process is required to achieve ITO with high conductivity, limiting the versatility of this material.13,14 Besides ITO, many other candidate TCE materials such as fluorine-doped tin oxide (FTO),15 aluminum oxide (Al2O3):zinc oxide (ZnO) nanolaminates,16 and highly conductive PEDOT:PSS (PH 1000)17 have been utilized for PSCs; however, they still suffer from several issues related to effects such as low conductivity, difficulties with scale-up, and instability. Vanadium-doped indium oxide (IVO) is an attractive alternative TCE material due to its sufficiently large WF value (>5.4 eV), excellent optical and electrical properties, and the variety of stoichiometric compositions of VOx oxides, such as VO2, V2O3, V2O5, and V6O11.18–21 Recently, Li et al. employed IVO films for the TCE in organic light-emitting diodes (OLEDs). Because of the appropriate WF value and enhanced hole injection properties of IVO films, OLEDs with IVO TCE exhibited improved electroluminescence (EL) performance compared with devices using an ITO TCE.21 Also, Kim et al. and Guo et al. reported use of buffer-free thermally annealed IVO and ultraviolet/ozone-treated IVO, respectively, as the HCE of PSCs.22,23 Even if the acidic buffer layer was thereby eliminated from the PSC, understanding of the effects of vanadium doping remains insufficient, and thermal processing was still required for the IVO. Moreover, the conventional geometry adopted in previous PSC studies is known to result in device instability due to use of air-sensitive low-WF metals, such as Ca, on the top of such devices. In this work, the properties of IVO films processed at room temperature as transparent conductors for PSCs were investigated as functions of the V doping concentration, along with the origin of the changes in their physical properties and electronic structure. In addition, PSCs with inverted geometry and using IVO with various V concentrations were fabricated and evaluated. The correlations between the obtained photovoltaic and optoelectrical properties of the IVO films were also examined. EXPERIMENTAL PROCEDURES IVO films were deposited on soda lime glass substrates at room temperature using a radiofrequency (RF) cosputtering system with V2O5 and In2O3 as dual targets. The base pressure was below 2.7 9 10 7 kPa, and the working pressure was 3.3 kPa with Ar flow rate of 10 sccm. The RF power for the V2O5 target was controlled at 4 W, 8 W, 19 W, and 34 W with constant RF power for the In2O3 target of 75 W. The actual V doping concentration was determined by time-of-flight secondaryion mass spectrometry (TOF–SIMS) using Cs ions. To fabricate inverted PSCs, ZnO film (40 nm) was formed on top of IVO or ITO substrates using a sol– gel method by spin-coating at 3000 rpm for 40 s with zinc acetate dihydrate [Zn(CH3COO)2Æ2H2O,

Aldrich, 99.9%, 1 g] as ZnO precursor and ethanolamine (NH2CH2CH2OH, Aldrich, 99.5%, 0.28 g) in Aldrich, 2-methoxyethanol (CH3OCH2CH2OH, 99.8%, 10 mL). The substrates were subsequently annealed on a hot plate for 1 h at 200°C in air. A solution of PTB7:PC70BM (1-materials) with weight ratio of 1:1.5 (PTB7:PC70BM) and total PTB7 + PC70BM concentration of 25 mg/mL was prepared in mixed solvent solution of chlorobenzene (Aldrich) and 1,8-diiodooctane (Aldrich; 97:3, v/v). Before use, the PTB7:PC70BM solution was stirred overnight in a nitrogen-filled glovebox. PTB7:PC70BM films (80 nm) were formed on all substrates by spin-coating at 1000 rpm for 40 s, followed by drying at room temperature in a glovebox. Device fabrication was completed by evaporation of MoOx (10 nm) and Ag (100 nm) onto the PTB7:PC70BM active layer through a shadow mask under high-vacuum condition (2.7 9 10 7 kPa) to form the hole-collecting electrode. The active area was approximately 0.09 cm2, as determined accurately under a microscope for individual devices. The thickness of each PSC layer was measured using either a crystal thickness monitor (MoOx and Ag), profilometer (PTB7:PC70BM), or spectroscopic ellipsometer (IVO and ZnO). Scanning electron microscopy (SEM; Daeil Systems) was carried out to investigate the cross-section of the PSC, prepared by focused ion beam (FIB) milling. Hall-effect measurements were carried out to determine the electrical properties (carrier concentration, resistivity, and mobility) of the IVO films. Optical transmittance was measured by ultraviolet– visible (UV–Vis) spectroscopy. The physical structure was investigated by x-ray diffraction (XRD) analysis, and the surface morphology by atomic force microscopy (AFM). The electronic structure was analyzed using spectroscopic ellipsometry (SE) and x-ray photoelectron spectroscopy (XPS) to investigate changes in band alignment. SE measurements were performed using a rotating-analyzer system with automatic retarder in the energy range from 0.74 eV to 5.5 eV at angle of incidence of 65°, 70°, and 75°. XPS data were obtained using a monochromatic Al Ka source with pass energy of 23.5 eV. The J–V characteristic was measured under illumination using a source meter (2400, Keithley Instruments) controlled by a LabVIEW program. An air mass 1.5 global (AM1.5G) solar simulator (Oriel 91160, Oriel Instruments) with irradiance of IL = 100 mW cm 2 was used as light source. A calibrated reference silicon solar cell certified by the National Renewable Energy Laboratory (NREL) was used to confirm the measurement conditions of the setup. RESULTS AND DISCUSSION Figure 1a shows the TOF–SIMS spectra of the IVO films obtained using RF power for the V2O5 target of 4 W, 8 W, 19 W, and 34 W, hereafter

Properties of Vanadium-Doped Indium Oxide Deposited at Room Temperature as Transparent Conductor for Inverted Polymer Solar Cells

Fig. 1. (a) TOF–SIMS depth profile as function of V2O5 target RF power, (b) sheet resistance of IVO films, (c) optical transmittance of IVO films, and (d) figure of merit of IVO films calculated from sheet resistance (Rsh) and average optical transmittance (T) at 380 nm to 800 nm as function of V doping concentration.

indexed as having doping concentration of 0.01%, 0.03%, 0.10%, and 0.33%, respectively, based on the TOF–SIMS analysis. Figure 1b shows the sheet resistance of the IVO films as a function of the V doping concentration. With increasing amount of V doping, the sheet resistance of the IVO films exhibited a parabolic tendency with lowest value of 74.83 Ohm/sq for V doping concentration of 0.03%. When the V doping concentration was increased above 0.03%, the sheet resistance increased, reaching 112.48 Ohm/sq at V doping concentration of 0.33%. As shown in Fig. 1c, the V doping of the indium oxide films slightly changed the optical transmittance in the visible region, showing a similar tendency regardless of the V doping concentration. To optimize the properties of the IVO film as a transparent conductor, an attempt was made to calculate the figure of merit (FOM) using the sheet resistance (Rsh) and average optical transmittance (T) between 380 nm and 800 nm. As reported by Haacke, the FOM value calculated as (T10/Rsh) is useful for determination of the optimal performance of transparent conducting films,

because a higher value of this FOM indicates higher optical transmittance and conductivity.24 As seen from Fig. 1d, similar to the variation of the sheet resistance, the IVO film with V doping concentration of 0.03% exhibited the highest FOM value of 4.35 9 10 3 Ohm 1. The increase of the FOM was caused by the decrease of the sheet resistance depending on the V doping concentration. Figure 2a and b show the diffraction pattern and surface morphology obtained by XRD and AFM measurements, respectively. Regardless of the V doping concentration, the structure of the IVO film was polycrystalline with bixbyite structure of indium oxide.25 As the V doping concentration was increased, the (222) peak shifted slightly to higher angle due to lattice shrinkage induced by substitu˚ ) for In ion (0.80 A ˚ ). Also, the tion of V ion (0.79 A surface morphology and roughness varied slightly, with root-mean-square (RMS) roughness between 0.9 nm and 1.07 nm. These changes indicate that the structural properties and surface morphology had little influence on the conductive properties. To interpret the enhanced properties of the IVO film as

Choi, K.Y. Lim, Park, Kim, Hwang, S.-J. Lim, Shim, and Chung

Fig. 2. (a) X-ray diffraction patterns and (b) AFM images of IVO films with various V doping concentrations.

a transparent conductor as functions of the V doping concentration, we discuss below the electronic structure, including the band alignment. Figure 3a, b, and c show the bandgap (Eg) spectrum, valence-band spectrum, and schematic energy diagram as functions of the V doping concentration. The Eg spectra were extracted using a four-phase model comprising a Si substrate, SiO2 layer, conducting IVO layer, and ambient layer. The binding energy scale of the XPS spectra was calibrated using the reference energy value of 531 eV for the stoichiometric oxygen–metal bond of O 1s. Based on the bandgap and the valenceband spectra, the schematic energy diagram for the valence-band offset (energy difference between the maximum of the valence band and the Fermi energy, DEVB) and the conduction-band offset (energy difference between the minimum of the conduction band and the Fermi energy, DECB) could be obtained and are shown in Fig. 3c and Table I. To reduce the error in extrapolated values, we performed linear fitting of the epsilon 2 spectra from 3.2 eV to 3.5 eV, where the value of the x-axis at the baseline of the y-axis yields the bandgap (Eg). In the same way, the energy difference from the valence-band maximum to the Fermi level (DEVB) was determined by performing linear fitting for the region of 2 eV to 3.25 eV of the valence-band spectra. The linear fitting energy region of the spectra was determined by the two minimum points of inflection through analysis of the second derivative of the raw spectra excluding the region near the band-edge states. The conduction-band offset (DECB) decreased together with the V doping concentration up to 0.03%, whereas it increased with the V doping concentration above 0.03%; this effect is related to the change of the sheet resistance due to enhanced probability of electron transfer from occupied states to unoccupied states in the conduction band.26

A schematic cross-sectional view of the inverted PSC with glass/IVO(170 nm)/ZnO/PTB7:PC70BM/ MoOx/Ag layered structure and corresponding cross-sectional scanning electron microscopy images are displayed in Fig. 4a. To study the dependence of the photovoltaic performance on the V concentration in the IVO film, IVO with the four different V doping concentrations of 0.01%, 0.03%, 0.10%, and 0.33% was employed. Figure 4b shows the J–V characteristics of these devices with different V doping concentrations. The PSC using the IVO with the highest V doping concentration of 0.33% exhibited the poorest device performance with power conversion efficiency (PCE) of 1.3 ± 0.3%. The photovoltaic parameters are presented in Table II as functions of the V doping concentration, where each value is averaged over three or four PSCs. As the V doping concentration was decreased, the performance of the PSC substantially improved with the highest PCE value of up to 4.7 ± 0.4% achieved by PSCs with V doping concentration of 0.03%. However, the PSCs using IVO with further reduced V doping concentration of 0.01% showed degraded performance with lower PCE of 2.6 ± 0.9%. The evolution of the performance as a function of the V doping concentration, especially the short-circuit current density (JSC), seemed to correlate strongly with the FOM value of the IVO films. As discussed above, the FOM value of the TCE is associated with Rsh and the film transmittance, with a higher FOM indicating a more conductive and transparent film; namely, IVO film with higher FOM value (V doping concentration of 0.03%) induces lower optical loss within the TCE, leading to higher photon absorption in the photoactive layer. Also, the lower Rsh of the IVO with higher FOM value facilitates charge carrier (electron) collection at the electrode, resulting in higher photocurrent, fill factor, and PCE. Meanwhile, no significant change in the opencircuit voltage (VOC) was observed among the PSCs,

Properties of Vanadium-Doped Indium Oxide Deposited at Room Temperature as Transparent Conductor for Inverted Polymer Solar Cells

Fig. 3. (a) Imaginary dielectric function (e2) spectra with optical bandgap (Eg) and (b) valence-band spectra for IVO films based on SE and XPS analyses. (c) Design of band alignment for IVO films through combined optical bandgap and valence-band offset.

Table I. Band alignment of IVO films as function of V doping concentration V Doping concentration IVO IVO IVO IVO

(V: (V: (V: (V:

0.01%) 0.03%) 0.10%) 0.33%)

Eg (eV) 3.07 3.08 3.08 3.06

± ± ± ±

0.005 0.005 0.004 0.004

indicating that a sufficiently low WF can be achieved by the ZnO layer, leading to proper energy level alignment between the organic materials and TCE. To compare the performance of the PSCs with IVO electrode with other commonly used TCEs, PSCs were also fabricated using ITO with ITO/ZnO/ PTB7:PC70BM/MoOx/Ag structure. Figure 4c compares the J-V characteristics under illumination measured for PSCs with either IVO or ITO electrode. Note that the ITO (100 nm, Rsh: 76.9 ± 0.1

DEVB (eV) 1.86 1.97 1.86 1.90

± ± ± ±

0.022 0.018 0.023 0.018

DECB (eV) 1.21 1.11 1.22 1.16

± ± ± ±

0.027 0.022 0.027 0.023

Ohm/sq) used in this study was fabricated using the same RF sputtering system that was utilized for IVO film deposition. The devices with ITO electrode exhibited comparable VOC values of 739 ± 12 mV, but their JSC and FF values of 14.9 ± 0.3 mA/cm2 and 33.1 ± 2.4, respectively, were slightly lower compared with those of the PSCs with IVO (with V doping concentration of 0.03%). The PSCs fabricated on ITO exhibited slightly lower PCE values of 3.7 ± 0.3%.

Choi, K.Y. Lim, Park, Kim, Hwang, S.-J. Lim, Shim, and Chung

Fig. 4. (a) Device structure of PTB7:PC70BM solar cells with IVO transparent conducting electrode (TCE) (left) and corresponding crosssectional scanning electron microscopy (SEM) image. Representative J–V characteristics under illumination for PTB7:PC70BM-based polymer solar cell with (b) IVO TCE with various V doping concentrations (0.01%, square; 0.03%, circle; 0.10%, triangle; 0.33%, diamond), and (c) IVO TCE (0.03%, circle) and ITO TCE (star).

Table II. Average device performance of PTB7:PC70BM-based polymer solar cells with IVO TCE as function of V doping concentration and ITO TCE (laboratory made using RF sputtering) (average calculated over four devices) Transparent conducting electrode (TCE)

VOC (mV)

ITO IVO IVO IVO IVO

739 564 739 743 709

(V: (V: (V: (V:

0.01%) 0.03%) 0.10%) 0.33%)

CONCLUSIONS The properties of V-doped indium oxide (IVO) deposited at room temperature as transparent conductor for inverted polymer solar cells (PSCs) were investigated as functions of the V doping concentration. IVO film with V doping concentration of 0.03% showed optimal transparent conducting properties with sheet resistance of 74.83 Ohm/sq

± ± ± ± ±

12 25 9 6 24

JSC (mA/cm2) 14.9 12.9 16.5 14.4 7.7

± ± ± ± ±

0.3 1.6 0.2 1.3 1.7

FF 33.1 35.9 38.6 26.4 23.7

± ± ± ± ±

PCE (%) 2.4 5.6 2.7 4.2 3.4

3.7 2.6 4.7 2.9 1.3

± ± ± ± ±

0.3 0.9 0.4 0.7 0.3

and average transmittance of 89.4%. The energy band offset between the Fermi level and conductionband maximum decreased to 1.02 eV without changes in the physical structure or surface properties. The photovoltaic performance was also strongly dependent on the optoelectrical properties of the IVO film. Inverted PSCs using IVO film with V doping concentration of 0.03% reached high PCE

Properties of Vanadium-Doped Indium Oxide Deposited at Room Temperature as Transparent Conductor for Inverted Polymer Solar Cells

of 4.7 ± 0.4% under AM1.5G illumination at 100 mW/cm2, above that of reference PSCs with ITO TCE. ACKNOWLEDGEMENTS This research was supported by the Dongguk University Research Fund of 2015, the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Science, ICT, and Future Planning (NRF-2015R1C1A1A01051841 and NRF2016R1A6A1A03012877), and NanoÆMaterial Technology Development Program through the NRF funded by the Ministry of Science, ICT, and Future Planning (2009-0082580). REFERENCES 1. F.C. Krebs, Sol. Energy Mater. Sol. Cells 93, 394 (2009). 2. F. Padinger, R.S. Rittberger, and N.S. Sariciftci, Adv. Funct. Mater. 13, 85 (2003). 3. B. Kippelen and J.L. Bredas, Energy Environ. Sci. 2, 251 (2009). 4. C.J. Brabec, S. Gowrisanker, J.J.M. Halls, D. Laird, S.J. Jia, and S.P. Williams, Adv. Mater. 22, 3839 (2010). 5. J.W. Shim, C. Fuentes-Hernandez, A. Dindar, Y.H. Zhou, T.M. Khan, and B. Kippelen, Org. Electron. 14, 2802 (2013). 6. Y.H. Zhou, C. Fuentes-Hernandez, J. Shim, J. Meyer, A.J. Giordano, H. Li, P. Winget, T. Papadopoulos, H. Cheun, J. Kim, M. Fenoll, A. Dindar, W. Haske, E. Najafabadi, T.M. Khan, H. Sojoudi, S. Barlow, S. Graham, J.L. Bredas, S.R. Marder, A. Kahn, and B. Kippelen, Science 336, 327 (2012). 7. M.D. Irwin, B. Buchholz, A.W. Hains, R.P.H. Chang, and T.J. Marks, Proc. Natl. Acad. Sci. 105, 2783 (2008). 8. J.B. You, L.T. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C.C. Chen, J. Gao, G. Li, and Y. Yang, Nat. Commun. 4, 1 (2013).

9. G. Li, V. Shrotyriya, J.S. Huang, Y. Yao, T. Moriarty, K. Emery, and Y. Yang, Nat. Mater. 4, 864 (2005). 10. M. Pagliaro, R. Ciriminna, and G. Palmisano, ChemSusChem 1, 880 (2008). 11. T. Minami, Semicond. Sci. Technol. 20, S35 (2005). 12. K. Schulze, B. Maennig, K. Leo, Y. Tomita, C. May, J. Hu¨pkes, E. Brier, E. Reinold, and P. Ba¨uerle, Appl. Phys. Lett. 91, 073521 (2007). 13. S.I. Na, S.S. Kim, J. Jo, and D.Y. Kim, Adv. Mater. 20, 4061 (2008). 14. M. Jorgensen, K. Norrman, and F.C. Krebs, Sol. Energy Mater. Sol. Cells 92, 686 (2008). 15. Y. Zhou, J.W. Shim, C. Fuentes-Hernandez, T.M. Khan, and B. Kippelen, Thin Solid Films 554, 54 (2014). 16. H. Cheun, C. Fuentes-Hernandez, J. Shim, Y. Fang, Y. Cai, H. Li, A.K. Sigdel, J. Meyer, J. Maibach, A. Dindar, Y. Zhou, J.J. Berry, J. Bredas, A. Kahn, K.H. Sandhage, and B. Kippelen, Adv. Funct. Mater. 22, 1531 (2012). 17. Y. Zhou, T.M. Khan, J.W. Shim, A. Dindar, C. FuentesHernandez, and B. Kippelen, J. Mater. Chem. A 2, 3492 (2014). 18. I. Hancox, L.A. Rochford, D. Clare, M. Walker, J.J. Mudd, P. Sullivan, S. Schumann, C.F. Conville, and T.S. Jones, J. Phys. Chem. C 117, 49 (2013). 19. K. Zrberg, S. Trost, J. Meyer, A. Kahn, A. Behrendt, D. Lu¨tzenkirchen-Hecht, R. Frahm, and T. Riedl, Adv. Funct. Mater. 21, 4776 (2011). 20. V.V. Atuchin, B.M. Ayupov, V.A. Kochubey, L.D. Pokrovsky, C.V. Ramana, and Yu.M. Rumiantsev, Opt. Mater. 30, 1145 (2008). 21. Huibin Li, Ning Wang, and Xingyuan Liu, Opt. Express 16, 194 (2008). 22. H.M. Lee, Y.J. Noh, S.I. Na, and H.K. Kim, ECS Solid State Lett. 3, P145 (2014). 23. X.-Y. Guo, J.-S. Luo, H. Chen, and X.-Y. Liu, Chin. Phys. Lett. 29, 088801 (2012). 24. G. Haacke, J. Appl. Phys. 47, 4086 (1975). 25. D.C. Paine, T. Whitson, D. Janiac, R. Beresford, C.O. Yang, and B. Lewis, J. Appl. Phys. 85, 8445 (1999). 26. B.K. Kim, J.S. Park, D.H. Kim, and K.B. Chung, Appl. Phys. Lett. 104, 182106 (2014).