Solar Energy Materials & Solar Cells 117 (2013) 203–208

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells

0927-02http://d

n CorrE-m

journal homepage: www.elsevier.com/locate/solmat

The annealing effects of tungsten oxide interlayer based on organicphotovoltaic cells

Sang Bin Lee, Jin Ho Beak, Byung hyun Kang, Ki-Young Dong, Youn-Yeol Yu, Yang Doo Lee,Byeong-Kwon Ju n

Display and Nanosystem Laboratory, College of Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-713, Republic of Korea

a r t i c l e i n f o

Article history:Received 18 February 2013Received in revised form23 April 2013Accepted 9 May 2013Available online 26 June 2013

Keywords:Tungsten oxideThermal annealingOrganic solar cellOrganic photovoltaic cell

48/$ - see front matter & 2013 Elsevier B.V. Ax.doi.org/10.1016/j.solmat.2013.05.020

esponding author. Tel.: +82 2 3290 3671.ail address: bkju@korea.ac.kr (B.-K. Ju).

a b s t r a c t

We investigated the effect of tungsten oxide (WO3) interlayer as a hole collection layer on theperformance of organic photovoltaic cells according to the thickness and temperature of the interlayer.The characteristics of organic photovoltaic cells such as fill factor, current density, and open circuitvoltage are continuously improved by increasing the temperature of the WO3 interlayer. The surface of atreated WO3 film promotes the crystallization of P3HT because a treated WO3 film is more hydrophobicthan a pristine WO3 film. Furthermore, the energy barrier between P3HT and the WO3 interlayer isminimized since the work function of the WO3 film after annealing progressively increases until a holecan be smoothly transferred. Therefore, organic photovoltaic cells using an interlayer of treated WO3 filmhave higher hole mobility and better efficiency. The efficiency of an organic photovoltaic cell with a40 nm-thick WO3 interlayer is significantly enhanced from 0.94 to 3.04% as the temperature changesfrom room temperature to 350 1C under AM 1.5G illumination.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Organic photovoltaic cells (OPVs) have been studied as a type ofnext generation energy source because they are lightweight andflexible, showing much potential for large scale fabricationthrough a low-cost process [1]. During the past 30 years sincethe development of OPVs, OPVs technologies have focused onincreasing power conversion efficiency (PCE), lifetime and relia-bility. Many methods have been proposed to improve the perfor-mance of OPVs, such as the use of a bulk heterojunction structure[2], the use of an optical spacer over the active layer using titaniumoxide (TiO2) [3], annealing to improve the crystallization of poly-3(hexylthiophene) (P3HT) polymer [4], and O2-plasma treatment onindium tin oxide (ITO) for work function modification [5].

Also, thin interlayers between the active layer and the electro-des were introduced to extract diffused holes or electrons, easily.A polymer complex of poly(3,4-ethylenedioxythiophene) and poly(styrene sulfonate) (PEDOT:PSS) has been preferred for OPVs [6,7]because it has flexibility, high transparency and good hole mobilityon an ITO layer. Furthermore, it has a high work function, whichcan be used to match the band structure between the donor layerand the anode. However, the PEDOT:PSS interlayer has severaldisadvantages that reduce the performance of OPVs, such as

ll rights reserved.

stability and life-time due to the fact that PSS, which is highlyacidic, corrodes the ITO layer chemically and that PEDOT:PSSdegrades under UV illumination [8–11]. For solving these pro-blems, several proposals such as alternative materials of PEDOT:PSS, ITO-free devices [1,12,13], and inverted architecture device[14] have been studied.

Recently, as replacement for PEDOT:PSS, metal oxides havebeen used to form attractive hole extracting layers because they donot have side effects such as degradation and chemical reactionwith ITO. Nickel oxide (NiO) [15], vanadium oxide (V2O5), molyb-denum oxide (MoO3) [16] have been researched as materials forthe interlayer, such as the hole-selective or electron-selectivitylayer. Especially, tungsten oxide (WO3) has been investigated as aneffective hole collection layer because it has a high work function,high ionization potential (IP) and p-type electronic property [17].Also, the surface morphology of a WO3 layer is much smoother thanthose of MoO3 and V2O5 layers. It can also reduce the leakage currentbecause the bias-dependent carrier recombination is preventedbecause the local high electric fields were not formed at a smooth-ened surface of an interlayer or electrode [18]. The WO3 interlayerdeposited by thermal evaporation as an anodic interlayer wasreported to have increased the performance of OPVs [19]. In invertedOPVs, the WO3 layer between the active layer and the top electrodeimproved the efficiency of the OPVs from 0.13 to 2.58% [20].

In this work, OPVs are fabricated with WO3 interlayers, insteadof the PEDOT:PSS layer, between the active layer and the anode tocollect holes. We investigated the effect of WO3 layers of various

S. Bin Lee et al. / Solar Energy Materials & Solar Cells 117 (2013) 203–208204

thickness and annealing temperatures which changes from roomtemperature to 350 1C on the performance of organic photovoltaiccells. We discussed the function of an annealed WO3 interlayer, whichsignificantly improved the performance of an OPVs device. Thephysical, optical and electronic properties of a thermal evaporatedWO3 thin film are measured by AFM, IPCE, XPS and XRD.

Table 1Short-circuit current density (Jsc), open-circuit voltage (Voc), FF, and power conver-sion efficiency (PCE) of OPVs according to the thickness of WO3.

2. Experimental

Fig. 1 shows the structures of OPVs with WO3 interlayers ofdifferent thicknesses and temperatures. The patterned ITO sub-strates of 25 Ω/square were ultra-sonically cleaned several timesin acetone, methanol and de-ionized water sequentially. Thecleaned substrates underwent UV-Ozone treatment for 15 minand treated with oxygen plasma at 90 W for 240 s. A WO3

(purchased from KOJNDO KOREA, 99.99%) thin film was depositedonto an ITO substrate by thermal evaporation in vacuum of about1�10−6 Torr at evaporation rate of 4.0 Å/s. The thicknesses of theWO3 films were detected in situ by a quartz monitor.

The deposited WO3 thin films were annealed in Ar atmospherein a glove box at different temperatures (room temperature (RT),200 1C, and 350 1C) for 30 min using a hot plate. The ITO glasssubstrate below WO3 was kept in the glove box of Ar environmentfor 10 h after thermal annealing to cool down.

In order to fabricate bulk hetero-junction organic solar cells,P3HT (purchased from Rieke metals, Inc.) and PCBM (purchasedfrom Nano-C, Inc.) were dissolved in mono-chlorobenzene(CB, 40 mg/ml) for 14 h and blended at a mixing ratio of 1:0.6

Fig. 1. Structure of OPVs with WO3 interlayer of different thicknesses for differenttemperatures: ITO/WO3/P3HT:PCBM/LiF/Al.

Fig. 2. J–V characteristics under 100 mW/cm2 white light illumination in air for deviceswith WO3 interlayers of different thicknesses between the active layer and ITO.

for 2 h in a glove box (Ar atmosphere). The blended active mixturewas spin coated at 900 rpm for 30 s to form a film, which was thenthermally annealed at 140 1C for 15 min in Ar atmosphere. The Alcathode (100 nm) and the interlayer (LiF 0.6 nm) were thermallyevaporated in vacuum greater than 5�10−7 Torr.

The current–voltage (I–V) characteristics of the fabricateddevices were subjected to simulated AM 1.5 global solar irradiationof 100 mW/cm2 incident power density using a 300 W Xe lamp(Mc science) and measured using a Keithley 2400 source meter.Moreover, Incident-Photon-to-electron Conversion Efficiency

Fig. 3. (a) The J–V characteristics under 100 mW/cm2 white light illumination in airfor devices according to different temperature. (b) The Incident-Photon-to-electronConversion Efficiency (IPCE) according to different temperatures.

S. Bin Lee et al. / Solar Energy Materials & Solar Cells 117 (2013) 203–208 205

(IPCE) was measured as a function of wavelength from 300 to800 nm (PV Measurement).

After the WO3 thin film was thermally evaporated on ITO, itssurface morphology and grain size were measured by atomic forcemicroscopy (XE-100, Park systems Inc). X-ray photoelectron spec-troscopy (XPS) measurements were performed with monochro-matized Al Ka X-ray photons (hv¼1486.6 eV for XPS) (K-Alpha,Thermo Electron, Inc.). The contact angle was measured in atmo-spheric air at room temperature using a commercial contact anglemeter. X-ray diffraction (XRD) measurements were obtained by anX-ray diffractometer (X'pert APD, PHILIPS). The work function of aWO3 film under a UV source was measured by a photoelectronspectrometer (AC-2, Riken Keiki Inc.) at atmospheric pressure.

3. Result and discussion

The optimal thickness of the WO3 layer for increasing theperformance of OPVs has been reported previously [19–21]. But,the properties of the WO3 layer differ according to how to prepare

Table 2Short-circuit current density (Jsc), open-circuit voltage (Voc), FF, and power conver-sion efficiency (PCE) of OPVs according to the annealing temperature.

Fig. 4. AFM images (a) bare-ITO, (b) WO3 on ITO (RT), (c

material or equipment and process. For this reason, we investi-gated several thicknesses in the range between 5 nm and 60 nm atthe same temperature to find the adequate efficiency of OPVs.Fig. 2 and Table 1 show the performances of OPVs for differentthicknesses of the WO3 layer under 100 mW/cm2 white lightillumination. The performance of OPVs improved with increasingWO3 thickness up to 40 nm. The efficiency of OPVs, however,decreased with the thickness of WO3 layer over 40 nm. A WO3

layer of 40 nm thickness was found to be the optimal thickness foryielding the best performance of OPVs.

Fig. 3(a) and Table 2 show the characteristics of the OPVsaccording to temperature with a 40 nm-thick WO3 layer. Wefocused on the thermal annealing effect because it significantlyincreased OPVs performance, as described. But the processingtemperature was limited to 350 1C because the crystalline natureof WO3 would be changed at temperatures over 400 1C. The devicewith the pristine WO3 layer exhibited power conversion efficiency(PCE) of 0.94% at a very low fill factor (FF) of 29.9%. The short-current density (Jsc) and open voltage (Voc) are 6.00 mA/cm2 and0.52 V, respectively. However, the device with the annealed WO3

layer showed improved PCE of 3.04% at FF¼64.47%, Jsc¼7.79 mA/cm2

and Voc¼0.60 V. The improved performance characteristics ofOPVs, such as FF, Jsc, and PCE, are closely related to the decreaseof the series resistance (Rs) of devices [22]. We demonstrated anIPCE spectrum of the fabricated devices in Fig. 3(b). The IPCE isdefined as the ratio of photogenerated charge carrier per incidentphoton. As shown, IPCE improved for every wavelength by thethermal annealing process. The thermal treated device at 350 1Chad a maximum IPCE of 71% at 444 nm; on the other hand, the

) WO3 on ITO (200 1C) and (d) WO3 on ITO (350 1C).

S. Bin Lee et al. / Solar Energy Materials & Solar Cells 117 (2013) 203–208206

pristine device had a maximum IPCE of 57% at the same wave-length. Since the absorption spectra of P3HT exist near visiblespectrum of about 300–700 nm and the maximum peak is about450 nm [23], this enhanced IPCE (annealed WO3 film than pristineWO3 film) at 444 nm region is supposed to be oriented formcharacteristic of P3HT layer using annealed WO3 interlayer. Toanalyze these phenomena, surface analysis was performed byAFM, XPS, XRD.

We measured the surface morphology of the WO3 layer byAtomic Force Microscopy (AFM) to investigate the effect ofannealing after it was evaporated from the ITO. AFM images ofthe surface morphologies of the bare-ITO and WO3 film annealedat different temperatures (RT, 200 1C and 350 1C) are shown inFig. 4. As shown in Fig. 4(b), the surface morphology of the WO3

film deposited on the ITO without the annealing process issmoother than that of the bare-ITO at Fig. 4(a), as reported byother researchers about WO3 films [24,25]. Also, the surfacemorphology of the WO3 film with increasing temperature inFig. 4(c) and (d) is more rough than that of the pristine WO3 film.Nevertheless, WO3 films annealed at 350 1C still have smoothsurface, amorphous state and nanosized grains because treatedtemperature is low for changing the nature of WO3 crystallization[24]. The Root Mean Square (RMS, Rq) values of roughness were3.94 nm, 1.46 nm, 1.55 nm and 1.71 nm. Therefore, the WO3 inter-layer has a smoother morphology than the bare-ITO. Thissmoother morphology could be the reason for the increasedperformance of OPVs because it is known to be very effective indecreasing dark current and contact resistance [26].

Chemical characterization of the WO3 film with increasingtemperature was carried out by XPS. Fig. 5 shows the W4f peakand the O1s peak of the WO3 films at different temperatures,which were obtained by XPS analysis. The W4f peak did not

Fig. 5. XPS: variations of the W(4f) peak and the O(1s

remarkably change the state of the tungsten compounds accordingto increasing temperature. O1s peak was deconvoluted into threecomponents such as oxygen atoms (O2−), which are involved in theformations of strong W¼O bonds at binding energy of about531 eV, hydroxyl group (OH−) at about 532.5 eV and watermolecule (H2O), which are absorbed at the surface of WO3 filmat about 533.5 eV [27]. The pristine WO3 film had a compound ofwater molecules at about 533.5 eV in Fig. 5(a). However, the H2Opeaks disappeared after the annealing process at over 200 1Cbecause the trapped water molecules on the surface of the WO3

film were diffused by thermal energy, as shown in Fig. 5(b) and (c).Also, oxygen atoms (O2−) peak moved to lower energies withincreasing annealing temperature. These changes mean thatamount of the state of W¼O bonding in the WO3 film wasincreased [28]. Therefore, a heat treated WO3 film has a morehydrophobic surface than a pristine WO3 film.

The wetting property of the interlayer under an active layer canaffect the growth of P3HT films. The contact angles of a WO3 filmaccording to different temperatures are described in Fig. 6 inset. Thecontact angle measurements were performed using a 2 μl DI-water. Adroplet was injected on the surface of the thermally evaporated WO3

films after the annealing process. WO3 surface after annealing processis more hydrophobic than the pristine WO3 surface. According toprevious studies, hydrophobic surfaces or self-assembled monolayer(SAM) treatment promotes the crystallization of P3HT films [29,30].Furthermore, Li Destri et al. reported that a hydrophobic surfacegreatly prevents the aggregation of PCBM and highly in-plane orientedP3HT lamellae [31]. As shown in Fig. 6, the P3HT films that weregrown on the thermal treated WO3 films demonstrated higher degreeof ordering than those that were grown on the pristineWO3 films. Thehigher crystallization of P3HT on the annealing WO3 film is expectedto improve the hole mobility of P3HT films.

) peak for different thermal annealing processes.

Fig. 8. Work function variation of WO3 film after annealing at different tempera-tures.

S. Bin Lee et al. / Solar Energy Materials & Solar Cells 117 (2013) 203–208 207

To study the effect of increased P3HT crystallization onhole mobility by using an annealed WO3 film, which has a morehydrophobic surface than a pristine WO3 film, we measured holemobilities using hole-only devices having the structure of ITO/WO3 (40 nm)/P3HT (140 nm)/Au with different WO3 films. Fig. 7shows the current density (J)–voltage (V) characteristics of thehole-only devices as log–log plots. The treated WO3 film baseddevices exhibited a better device-to-device reproducibility thanthe pristine WO3 film based devices. Also, each hole mobility wascalculated by the following space-charge-limited current (SCLC)equation [32]:

J ¼ 9ε0εrμV2

8L3

where ε0 is the vacuum permittivity, εr is the relative dielectricconstant (value of P3HT is 3.4 [33]), μ is the carrier mobility, and Lis the P3HT film thickness. The mobility of each hole extractedfrom this equation was 2.0�10―4 cm2/V s for the pristine WO3 filmbased device, 2.4�10―4 cm2/V s for the 200 1C annealed deviceand 3.6�10―4 cm2/V s, for the 350 1C annealed device. The effi-ciency of the OPVs based, annealed WO3 films was improved bythe increased carrier mobility, which decreased the possibility of

Fig. 7. The current density (J)–voltage (V) curves of hole mobility using a hole-onlydevice in a structure of ITO/WO3 (40 nm)/P3HT (140 nm)/Au for different annealingtemperatures.

Fig. 6. X-ray diffraction spectra of P3HT films on WO3 films for different annealingtemperatures. Inset: the contact angles of the WO3 films with increasing tempera-ture (a) RT, (b) 200 1C and (c) 350 1C.

space-charge build-up [34] or reduced bias-dependent recombi-nation [35].

We studied the work function of the WO3 film with increasingannealing temperature to determine its effect on the energydecrease of the device between the interlayer and the active layer.Fig. 8 shows the work function variations of the WO3 film treatedby thermal treatment. The work function of the WO3 filmincreased with increasing temperature. The obtained work func-tions were 5.03 eV, 5.12 eV and 5.17 eV for the pristine WO3 film,for the WO3 film annealed at 200 1C and for the WO3 film annealedat 350 1C, respectively. Generally, the hole collection at the inter-face of the anode and active layer increased when the workfunction of the interlayer well matched the work function of thedonor type material such as P3HT [36]. As the work function ofthe annealed WO3 film gradually approached the energy level ofthe HOMO of P3HT (5.2 eV), the energy barrier between P3HT andWO3 film was minimized. Thus, the hole which was created inthe active layer easily transferred from the active layer to theanode. Also, a higher work function of the treated WO3 filmimproved the apparent hole mobility and then increased the holemobility induced by FF [37]. Therefore, the treated WO3 film canbe expected to be an efficient hole collection interlayer material.

4. Conclusion

We demonstrated tungsten oxide as an interlayer between ITOand P3HT:PCBM for different thicknesses and temperatures for BHJstructure organic photovoltaic cells. The device fabricated with aWO3 film of 40 nm thickness and 350 1C showed the highest FFand PCE of OPVs. Increased performance of OPVs was studied bymeasuring the characteristics of WO3 films and devices that wereannealed. The morphology of the treated WO3 film was notsignificantly changed, but its surface became more hydrophobicto improve P3HT crystallization and increase hole mobility of thedevice. Furthermore, the increased work-function of the treatedWO3 film improved hole-collection efficiency because it reducedthe barrier between P3HT and ITO. The heat treated WO3 film maybe used as an efficient hole collection material for OPVs.

Acknowledgment

This work was supported by the RFID R&D program of MKE/KEIT. (10035225, Development of core technology for high perfor-mance AMOLED on plastic).

S. Bin Lee et al. / Solar Energy Materials & Solar Cells 117 (2013) 203–208208

References

[1] R.R. Søndergaard, M. Hösel, F.C. Krebs, Roll-to-roll fabrication of large areafunctional organic materials, Journal of Polymer Science Part B: PolymerPhysics 51 (2013) 16–34.

[2] G. Yu, J. Gao, J.C. Hummelen, Polymer photovoltaic cells: enhanced efficienciesvia a network of internal donor–acceptor heterojunctions, Science 270 (1995)1789–1791.

[3] J.Y. Kim, S.H. Kim, H.-H. Lee, K. Lee, W. Ma, X. Gong, et al., New architecture forhigh-efficiency polymer photovoltaic cells using solution-based titaniumoxide as an optical spacer, Advanced Materials 18 (2006) 572–576.

[4] X. Yang, J. Loos, S.C. Veenstra, W.J.H. Verhees, M.M. Wienk, J.M. K1roon, et al.,Nanoscale morphology of high-performance polymer solar cells, Nano Letters5 (2005) 579–583.

[5] S. Schäfer, A. Petersen, T.A. Wagner, R. Kniprath, D. Lingenfelser, A. Zen, et al.,Influence of the indium tin oxide/organic interface on open-circuit voltage,recombination, and cell degradation in organic small-molecule solar cells,Physical Review B 83 (2011) 165311-1–165311-13.

[6] S. Gunes, H. Neugebauer, N.S. Sariciftci, Conjugated polymer-based organicsolar cells, Chemical Reviews 107 (2007) 1324–1338.

[7] A.C. Arias, M. Granstrom, D.S. Thomas, K. Petritsch, R.H. Friend, Dopedconducting-polymer–semiconducting-polymer interfaces: their use in organicphotovoltaic devices, Physical Review B 60 (1999) 1854–1860.

[8] N. Grossiord, J.M. Kroon, R. Andriessen, P.W.M. Blom, Degradation mechanismsin organic photovoltaic devices, Organic Electronics 13 (2012) 432–456.

[9] M. Jørgensen, K. Norrman, S.A. Gevorgyan, T. Tromholt, B. Andreasen,F.C. Krebs, Stability of polymer solar cells, Advanced Materials 24 (2012)580–612.

[10] M. Jørgensen, K. Norrman, F.C. Krebs, Stability/degradation of polymer solarcells, Solar Energy Materials and Solar Cells 92 (2008) 686–714.

[11] K. Norrman, M.V. Madsen, S.A. Gevorgyan, F.C. Krebs, Degradation patterns inwater and oxygen of an inverted polymer solar cell, Journal of the AmericanChemical Society 132 (2010) 16883–16892.

[12] T. Stubhan, J. Krantz, N. Li, F. Guo, I. Litzov, M. Steidl, M. Richter, G.J. Matt,C.J. Brabec, High fill factor polymer solar cells comprising a transparent, lowtemperature solution processed doped metal oxide/metal nanowire compositeelectrode, Solar Energy Materials and Solar Cells 107 (2012) 248–251.

[13] D.S. Ghosh, R. Betancur, T.L. Chen, V. Pruneri, J. Martorell, Semi-transparentmetal electrode of Cu–Ni as a replacement of an ITO in organic photovoltaiccells, Solar Energy Materials and Solar Cells 95 (2011) 1228–1231.

[14] S.K. Hau, H.-L. Yip, A.K.-Y. Jen, A review on the development of the invertedpolymer solar cell architecture, Polymer Reviews 50 (2010) 474–510.

[15] M.D. Irwin, D.B. Buchholz, A.W. Hains, R.P.H. Chang, T.J. Marks, p-Typesemiconducting nickel oxide as an efficiency-enhancing anode interfaciallayer in polymer bulk-heterojunction solar cells, Proceedings of the NationalAcademy of Sciences 105 (2008) 2783–2787.

[16] V. Shrotriya, G. Li, Y. Yao, C.-W. Chu, Y. Yang, Transition metal oxides as thebuffer layer for polymer photovoltaic cells, Applied Physics Letters 88 (2006)073508-1–073508-3.

[17] M. Kröger, S. Hamwi, J. Meyer, T. Riedl, W. Kowalsky, A. Kahn, Role of the deep-lying electronic states of MoO3 in the enhancement of hole-injection inorganic thin films, Applied Physics Letters 95 (2009) 123301-1–123301-3.

[18] S. Han, W.S. Shin, M. Seo, D. Gupta, S.-J. Moon, S. Yoo, Improving performanceof organic solar cells using amorphous tungsten oxides as an interfacial bufferlayer on transparent anodes, Organic Electronics 10 (2009) 791–797.

[19] M. Ghasemi Varnamkhasti, H.R. Fallah, M. Mostajaboddavati, R. Ghasemi,A. Hassanzadeh, Comparison of metal oxides as anode buffer layer for small

molecule organic photovoltaic cells, Solar Energy Materials and Solar Cells 98(2012) 379–384.

[20] C. Tao, S. Ruan, G. Xie, X. Kong, L. Shen, F. Meng, et al., Role of tungsten oxide ininverted polymer solar cells, Applied Physics Letters 94 (2009) 043311-1–043311-3.

[21] T. Stubhan, N. Li, N.A. Luechinger, S.C. Halim, G.J. Matt, C.J. Brabec, High fillfactor polymer solar cells incorporating a low temperature solution processedWO3 hole extraction layer, Advanced Energy Materials 2 (2012) 1433–1438.

[22] J.D. Servaites, S. Yeganeh, T.J. Marks, M.A. Ratner, Efficiency enhancement inorganic photovoltaic cells: consequences of optimizing series resistance,Advanced Functional Materials 20 (2010) 97–104.

[23] E. Bundgaard, F.C. Krebs, Low band gap polymers for organic photovoltaics,Solar Energy Materials and Solar Cells 91 (2007) 954–985.

[24] G.R. Jafari, A.A. Saberi, R. Azimirad, A.Z. Moshfegh, S. Rouhani, The effectof annealing temperature on the statistical properties of WO3 surface,Journal of Statistical Mechanics Theory and Experiment 2006 (2006) 09017-1–09017-14.

[25] M. Usta, S. Kahraman, F. Bayansal, H.A. Çetinkara, Effects of annealing onmorphological, structural and electrical properties of thermally evaporatedWO3 thin films, Superlattices and Microstructures 52 (2012) 326–335.

[26] T. Hori, T. Shibata, V. Kittichungchit, H. Moritou, J. Sakai, H. Kubo, et al., MoO3

buffer layer effect on photovoltaic properties of interpenetrating heterojunc-tion type organic solar cells, Thin Solid Films 518 (2009) 522–525.

[27] G. Leftheriotis, S. Papaefthimiou, P. Yianoulis, A. Siokou, D. Kefalas, Structuraland electrochemical properties of opaque sol–gel deposited WO3 layers,Applied Surface Science 218 (2003) 276–281.

[28] R. Azimirad, N. Naseri, O. Akhavan, A.Z. Moshfegh, Hydrophilicity variation ofWO3 thin films with annealing temperature, Journal of Physics D: AppliedPhysics 40 (2007) 1134–1137.

[29] X. Bulliard, S.-G. Ihn, S. Yun, Y. Kim, D. Choi, J.-Y. Choi, et al., Enhancedperformance in polymer solar cells by surface energy control, AdvancedFunctional Materials 20 (2010) 4381–4387.

[30] S.K. Hau, H.-L. Yip, O. Acton, N.S. Baek, H. Ma, A.K.-Y. Jen, Interfacialmodification to improve inverted polymer solar cells, Journal of MaterialsChemistry 18 (2008) 5113–5119.

[31] G. Li Destri, T.F. Keller, M. Catellani, F. Punzo, K.D. Jandt, G. Marletta, Interfacialfree energy driven nanophase separation in poly(3-hexylthiophene)/[6,6]-phenyl-C61-butyric acid methyl ester thin films, Langmuir 28 (2012)5257–5266.

[32] A.M. Goodman, A. Rose, Double extraction of uniformly generated electron–hole pairs from insulators with noninjecting contacts, Journal of AppliedPhysics 42 (1971) 2823–2830.

[33] L.J.A. Koster, V.D. Mihailetchi, P.W.M. Blom, Ultimate efficiency of polymer/fullerene bulk heterojunction solar cells, Applied Physics Letters 88 (2006)093511-1–093511-3.

[34] V.D. Mihailetchi, H.X. Xie, B. de Boer, L.J.A. Koster, P.W.M. Blom, Chargetransport and photocurrent generation in poly(3-hexylthiophene): methano-fullerene bulk-heterojunction solar cells, Advanced Functional Materials 16(2006) 699–708.

[35] P. Schilinsky, C. Waldauf, J. Hauch, C.J. Brabec, Simulation of light intensitydependent current characteristics of polymer solar cells, Journal of AppliedPhysics 95 (2004) 2816–2819.

[36] E.L. Ratcliff, B. Zacher, N.R. Armstrong, Selective interlayers and contacts inorganic photovoltaic cells, Journal of Physical Chemistry Letters 2 (2011)1337–1350.

[37] D. Yang, L. Zhou, L. Chen, B. Zhao, J. Zhang, C. Li, Chemically modified grapheneoxides as a hole transport layer in organic solar cells, Chemical Communica-tions 48 (2012) 8078–8080.