Impact on electronic structure of donor/acceptor blend in organic photovoltaics by decontamination of molybdenum-oxide surface Yuta Ito, Kouki Akaike, Takeshi Fukuda, Daisuke Sato, Takuya Fuse, Takashi Iwahashi, Yukio Ouchi, and Kaname Kanai
Citation: Journal of Applied Physics 123, 205501 (2018); doi: 10.1063/1.5027574 View online: https://doi.org/10.1063/1.5027574 View Table of Contents: http://aip.scitation.org/toc/jap/123/20 Published by the American Institute of Physics
JOURNAL OF APPLIED PHYSICS 123, 205501 (2018)
Impact on electronic structure of donor/acceptor blend in organic photovoltaics by decontamination of molybdenum-oxide surface Yuta Ito,1 Kouki Akaike,1,a) Takeshi Fukuda,2,b) Daisuke Sato,1 Takuya Fuse,1 Takashi Iwahashi,3 Yukio Ouchi,3 and Kaname Kanai1
1 Department of Physics, Faculty of Science and Technology, Tokyo University of Science, Noda-City, Chiba 278-8510, Japan 2 Department of Functional Materials Science, Saitama University, Saitama 338-8570, Japan 3 Depatment of Materials Science and Engineering, Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan
(Received 3 March 2018; accepted 3 May 2018; published online 23 May 2018) Molybdenum oxide (MoOx) is widely used as the hole-transport layer in bulk-heterojunction organic photovoltaics (BHJ-OPVs). During the fabrication of solution-processed BHJ-OPVs on vacuum-deposited MoOx film, the film must be exposed to N2 atmosphere in a glove box, where the donor/acceptor blends are spin-coated from a mixed solution. Employing photoelectron spectroscopy, we reveal that the exposure of the MoOx film to such atmosphere contaminates the MoOx surface. Annealing the contaminated MoOx film at 160 C for 5 min, prior to spin-coating the blend film, can partially remove the carbon and oxygen adsorbed on the MoOx surface during the exposure of MoOx. However, the contamination layer on the MoOx surface does not affect the energylevel alignment at the interface between MoOx and the donor/acceptor blend. Hence, significant improvement in the performance of BHJ-OPVs by mildly annealing the MoOx layer, which was previously reported, can be explained by the reduction of undesired contamination. Published by AIP Publishing. https://doi.org/10.1063/1.5027574
I. INTRODUCTION
Transition metal oxides with high work function, up to approximately 7 eV, have been used as hole transport layers in bulk heterojunction organic photovoltaics (BHJ-OPVs) to facilitate hole extraction, thereby, optimizing the device performance.1–5 Among them, molybdenum oxide (MoOx) is a typical selection because the film preparation is simple, with both dry2,6,7 and wet processes.3,8,9 The insertion of MoOx between a donor/acceptor blended film and an anode, such as indium tin oxide (ITO), leads to an increase in the shortcircuit current density (Jsc),2,10 open-circuit voltage (Voc), and fill factor (FF),2 as well as the device lifetime, compared to OPVs that employ poly(3,4-ethylenedioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) as the hole transport layer.9,11 The greater performance is attributed to the improvement in the hole-collecting efficiency due to the effective suppression of the electron migration to ITO by an energy barrier10 and/or the formation of a favorable built-in potential at the MoOx/organic interface.12,13 Moreover, Shrotriya et al. indicated that the suppression of the unwanted chemical reaction between PEDOT:PSS and ITO leads to better performance compared to PEDOT:PSS-employed devices,2 resulting in longer shelf-lifetime, in addition. These reports highlight the significance of the electronic and chemical properties of MoOx, in the improvement of the OPV performance when MoOx interlayer is used as the hole-transport layer. The electronic property of the MoOx surface is sensitive to the atmosphere to which the MoOx layer is exposed,14,15 a)
E-mail:
[email protected] Present address: Sekisui Chemical Co., Ltd., Osaka 618-0021, Japan.
b)
0021-8979/2018/123(20)/205501/6/$30.00
the post-annealing conditions,16,17 the chemical properties of the metal substrate on which MoOx is deposited,18 and the surface treatment.19 They influence the ratio of Mo5þ to Mo6þ by the formation of oxygen vacancy, resulting in a work-function shift18 and lifetime of the excitons generated in the active layers by photoabsorption.19 Recently, a noteworthy finding regarding the posttreatment of MoOx was reported by Kobori et al.20 In their work, the low-temperature annealing of vacuum-deposited MoOx films, under N2 atmosphere, was found to be practical for drastically improving the Jsc and FF of BHJ-OPVs.20 The thermal treatment was relatively mild (160 C for 5 min), compared to the condition under which the concentration of the oxygen vacancy increases.16,17 The improved performance was attributed to the decrease in the carrier-transport resistance;21 however, its chemical and/or physical origins remain elusive. As thermal treatment in a glove box filled with an inert gas is common in OPV fabrication processes, it is crucial to elucidate why such low-temperature annealing leads to substantial improvement in the OPV performance. In this study, employing ultraviolet and X-ray photoelectron spectroscopy (UPS and XPS, respectively), we reveal that contamination layers form on the MoOx surface, when the vacuum-deposited MoOx is exposed to the atmosphere in a glove box, where the donor/acceptor blends are solution-processed on a regular basis. Mildly annealing the exposed MoOx in the glove box at 160 C reduces the contamination layer. The energy-level alignment at the interface between the blend of regioregular poly(3-hexylthiophene) (P3HT) and the [6,6]-phenyl-C71-butyric acid methyl ester (PC71BM), and MoOx layers is unaffected by the thermal
123, 205501-1
Published by AIP Publishing.
205501-2
Ito et al.
J. Appl. Phys. 123, 205501 (2018)
treatment, as determined using UPS. Therefore, the improved performance of a BHJ-OPV prepared on a preheated MoOx layer can be attributed to the enhancement of the tunneling probability of the holes from the photoactive layer to MoOx, due to the reduction in the thickness of the probably insulating contamination layers. II. EXPERIMENTAL METHODS
A 150-nm ITO-patterned glass substrate was sequentially cleaned with detergent, deionized water, acetone, and isopropyl alcohol, under ultrasonication, and then subjected to UV-ozone cleaning for 20 min. Further, 40 nm of MoOx was thermally evaporated onto the substrate in a vacuum deposition chamber. ex-MoOx denotes the film exposed to N2 atmosphere in a glove box used for fabricating OPV cells at the Saitama University.20 ex-MoOx was subsequently heated at 160 C for 5 min in the glove box and is referred to as h-MoOx. For comparison, a fresh MoOx film of 20 nm was prepared on polycrystalline gold or solvent-cleaned ITO substrate, under ultrahigh vacuum, and is referred to as f-MoOx, in this manuscript. f-MoOx was characterized in situ using photoemission spectroscopy at the Tokyo Institute of Technology (Tokyo Tech.). Regioregular P3HT (purity 99.995%) and PC71BM (purity 99%), and PEDOT:PSS (Clevios P VP A1 4083) were purchased from Aldrich and Heraeus, respectively, and used without further purification. ex-MoOx and h-MoOx were transferred to the Tokyo University of Science (TUS) for spincoating the P3HT:PC71BM blended films (weight ratio ¼ 2:3), without air exposure. The films were spin-coated on the exMoOx and h-MoOx substrates, respectively, with chlorobenzene mixed solution (5 mg/ml) at 3000 and 3500 rpm for 30 s in an N2-filled glove box at the TUS. The rotation speeds were optimized to obtain uniform blended films on the respective substrates. For comparison, a neat P3HT film was spin-casted from chlorobenzene solution (4.4 mg/ml) on PEDOT:PSScoated ITO substrates at 1500 rpm for 30 s, after an acceleration of 5 s. Prior to this process, PEDOT:PSS was spin-coated at 3000 rpm for 30 s, after an acceleration of 5 s, and then annealed at 200 C for 5 min, under ambient atmosphere. The spin-coated films were sealed, under N2 atmosphere, and transferred to a UPS/XPS equipment at Tokyo Tech. The MoOx and the P3HT:PC71BM blended films were loaded into the sample-introduction chamber of the UPS/XPS apparatus, under N2, using a glove bag. UPS and XPS measurements were performed, under ultrahigh vacuum (base pressure
