- Email: [email protected]
Simple fabrication of perovskite solar cells with enhanced efficiency, stability, and flexibility under ambient air
Journal of Power Sources 442 (2019) 227216
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Simple fabrication of perovskite solar cells with enhanced efficiency, stability, and flexibility under ambient air Wei-Hsiang Chen a, b, Linlin Qiu b, Zhishan Zhuang b, Lixin Song b, Pingfan Du b, *, Jie Xiong a, b, **, Frank Ko c a b c
Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education, Hangzhou, 310018, PR China Silk Institute, College of Materials and Textiles, Zhejiang Sci-Tech University, Hangzhou, 310018, PR China Department of Materials Engineering, University of British Columbia, Vancouver, Canada
H I G H L I G H T S
G R A P H I C A L A B S T R A C T
� Positive tBP effect on perovskite can be greatly enhanced via PEDOT:PSS interlayer. � Perovskite grain with 1 μm size and quasi all-in-one structure can be obtained. � Flexible PSCs with excellent efficiency and moisture stability can be fulfilled. � Fabrication procedure can be performed greatly in ambient air condition.
A R T I C L E I N F O
A B S T R A C T
Keywords: 4-tert-butylpyridine Ambient-air fabrication Flexible perovskite solar cells Morphology-modifying Moisture stability
In order to increase the applicability and commercial potential of perovskite solar cells, simple fabrication of high-photovoltaic-performance flexible perovskite solar cells with excellent moisture stabilities without the use of a glove-box and an antisolvent is required. In this paper, we present a simple fabrication strategy involving introduction of 4-tert-butylpyridine into CH3NH3PbI3 and significantly enhancing 4-tert-butylpyridine morphology-modifying effect via a reduction-active flexible poly (3,4-ethylenedioxythiophene):poly (styr enesulfonate) interlayer. Owing to the specific oxidation facilitation by the poly (3,4-ethylenedioxythiophene): poly (styrenesulfonate) polymer, a perovskite film with large (~1 μm) and quasi-all-in-one-structured grains can be obtained, which significantly enhance the efficiency and the stability of the perovskite solar cells. Further more, the high-efficiency flexible perovskite solar cells fabricated by the simple strategy exhibit excellent moisture resistances owing to the stronger coordination and the outside-covering effect of the hydrophobic 4-tertbutylpyridine. The fabrication can be carried out under ambient air (without glovebox, relative humidi ty > 40%), which paves the way for wearable device application and commercialization.
* Corresponding author. ** Corresponding author. Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education, Hangzhou, 310018, PR China. E-mail addresses: [email protected] (P. Du), [email protected] (J. Xiong). https://doi.org/10.1016/j.jpowsour.2019.227216 Received 8 February 2019; Received in revised form 16 September 2019; Accepted 24 September 2019 Available online 1 October 2019 0378-7753/© 2019 Elsevier B.V. All rights reserved.
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
1. Introduction
2. Experimental methods
Organometal halide perovskite solar cells (PSCs) have attracted significant attentions owing to their promising characteristics such as high photovoltaic performances, low-cost solution-processable fabrica tion, flexible device designs, and potentials for commercialization. Over the past few years, the power conversion efficiency (PCE) has been improved amazingly from 3.8% to 25.2% [1–6]. In spite of large po tentials of PSCs for photovoltaic applications, various fundamental and practical issues have to be addressed for commercialization such as the outdoor moisture stabilities and low-cost mass-manufacture of highly efficient flexible and wearable PSCs. In this regard, extensive studies have been carried out including mixed-halide engineering [7–9], cation substitution [10–13], additive incorporation [14–16], high-stability charge-transport material selections [17–19], interface engineering [20–23], encapsulation [24,25] and other techniques [26,27]. Among the above approaches, introducing additive into perovskite is considered the most effective and simple option. However, PSCs are intrinsically limited by the drawback of high concentration additives in perovskite layer, which would lead to untuneable band gaps and poor charge transmission. In addition, so far a glove box is still necessary to avoid contacting moisture for most of high performance PSCs fabrication, which may seriously hamper the mass-manufacture and real applica tions. Therefore, a new and simple strategy that can overcome the lim itation of high concentration additives and improve the stabilities of the PSCs fabricated in ambient air is desperately needed. To simply fabricate PSCs that simultaneously exhibit excellent PCEs and long-term stabilities in ambient air, 4-tert-butylpyridine (tBP) has been used as a morphology-modifying agent to improve the perfor mances of PSCs [28–31]. This additive has a hydrophobic end, which can enhance the moisture resistance of the perovskite [31]. However, the improvements in photovoltaic performance and stability were very limited owing to the adverse effects caused by high concentration tBP in perovskite. In order to overcome this issue, Wu et al. [32] introduced tBP into the antisolvent instead of CH3NH3PbI3 and obtained better perov skite grain size and efficiency. However, the commonly applied anti solvents such as toluene and chlorobenzene [33,34] are highly toxic. In addition, a very large amount of antisolvent (>150 μL cm 2) is required and usually not reclaimed [35] for the PSCs fabrication, which is neither practical nor environment-friendly. Moreover, the antisolvent method is not suitable for a large-scale substrate, leading to PSCs scale-up and commercialization issues. Therefore, an intrinsic improvement in the tBP effect on the perovskite with less amount of tBP and without anti solvent treatment is advantageous. In this study, we propose a simple approach to enhance both tBP coordination with the perovskite and the PSCs flexibility by introducing a reduction-active flexible PEDOT:PSS interlayer. Owing to the specific chemical oxidation, the coordination between tBP and the perovskite could be significantly enhanced. A perovskite with very large (~1 μm) and quasi-all-in-one-structured grains could be obtained without the use of an antisolvent. Moreover, the amount of tBP could be largely reduced avoiding the unfavourable electric resistances and leading to higher short-circuit current density and stability. For the simplest and unopti mised ITO/PEDOT:PSS/CH3NH3PbI3-tBP/PC61BM/Ag PSC, the PCE and the fill factor (FF) were enhanced to 16.16% and 0.79 on a glass sub strate and to 13.54% and 0.7 on a flexible PET substrate, respectively. Furthermore, the PSCs exhibited excellent moisture resistances owing to the stronger coordination with the perovskite and the outside-covering effect of the hydrophobic tBP. The PSCs fabrication could be carried out even at an average relative humidity (RH) above 40% with a small reduction in the PCEs. This indicates that the use of a glove-box, which significantly increases the fabrication difficulties and the costs, is not required.
2.1. Materials CH3NH3I (MAI) and PbI2 were purchased from Kunshan Sunlaite New Energy Technology Co., Ltd. phenyl-C61-butyric acid methyl ester (PC61BM), poly (3,4-ethy-lenedioxythiophene) poly (styrenesulfonate) (PEDOT:PSS) and 4-tert-butylpyridine (tBP) were purchased from Xi’an Polymer Light Technology Corporation. Glass/ITO and PET/ITO sub strates were supplied from Advanced Election Technology Co., Ltd. The other materials were purchased from Aladdin. 2.2. Device fabrication Inverted-planar rigid and flexible perovskite solar cells (PSCs) were fabricated on laser-patterned, indium tin oxide (ITO) coated glass and PET (10 Ω sq 1) substrates, respectively. Both of them were sequentially ultrasonically cleaned with deionised water, acetone, ethyl alcohol, and isopropyl alcohol (IPA) for 15 min followed by a high-temperature drying with clean dry nitrogen and a treatment in an ultraviolet (UV) ozone oven for 20 min. After the cleaning, a thin layer of compact PEDOT:PSS was spin-coated on the clean substrate at 2000 rpm for 10 s and at 5000 rpm for 30 s in succession, and then annealed at 120 � C for 1 h. The PbI2 precursor (1.2 M, dissolved in DMF and stirred for 1 h at 70 � C) without and with different concentrations of tBP were continu ously spin-coated on the PEDOT:PSS substrate at 4500 rpm for 30 s and dried at 70 � C for 15 min. After the formation of the dried PbI2 (or PbI2⋅xtBP) film, CH3NH3I (12 mg/mL, dissolved in the IPA solvent) was spin-coated at 4000 rpm for 30 s to form the perovskite and annealed at 100 � C for 1 h. For the electron-transport layer (ETL), PC61BM (20 mg/ mL in chlorobenzene) was spin-coated on top of the perovskite layer at 2000 rpm for 30 s and annealed at 80 � C for 1 h. The fabrication was completed by thermal evaporation of Ag as an electrode with a thickness of approximately 100 nm and an effective area of 0.06 cm2. The whole fabrication was performed under ambient air (RH > 40%), which in dicates that the use of a glovebox is not necessary. The fabrication is illustrated in Fig. S1. 2.3. Characterisation The photocurrent density–voltage (J–V) characteristics of the cells were measured using a Keithley 2400-SCS source meter under AM 1.5 illumination with an intensity of 100 mW/cm2. The crystal structures of PbI2 and MAPbI3 were analysed by X-ray diffraction (XRD, Thermo ARLX’TRA, America) with Cu Kα radiation (λ ¼ 1.5418 Å). Field-emission scanning electron microscopy (FESEM, Ultra55, ZEISS, Germany) was employed to analyse the morphologies. The incident photon–to–current conversion efficiencies (IPCEs) of the cells were measured with a quantum-efficiency (QE)/IPCE test system (Solar Cell Scan100/Zolix). Electrochemical impedance spectroscopy (EIS) measurements were carried out using an electrochemical workstation (Zennium Pro, Ger many). The contact properties were characterised by a contact-angle instrument (Krüss Optronic, Germany). X-Ray photoelectron spectros copy (XPS) was measured with a K-Alpha instrument. The UV–visible (UV–vis) absorption was measured using a UV–vis spectrophotometer (Lambda 900, America). 3. Results and discussion For achieving high photovoltaic performances, stabilities, low costs, and simple fabrication of flexible perovskite solar cells (PSCs), 4-tertbutylpyridine (tBP) was introduced into the PbI2 precursor solution as a morphology-modifying agent on the reduction-active poly (3,4-ethyl enedioxythiophene):poly (styrenesulfonate) (PEDOT:PSS) interlayer. Fig. 1a–d shows the morphologies of the pristine PbI2, PbI2⋅xtBP, and the formed perovskite films on ITO/PEDOT:PSS, respectively. The insets in 2
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Fig. 1. SEM top-view images of (a) pristine PbI2, (b) CH3NH3PbI3, (c) PbI2⋅xtBP, (d) CH3NH3PbI3⋅xtBP films formed on ITO/PEDOT:PSS substrate. Insets show the theoretical model of each film. (e) The XRD patterns of PbI2 film with and without tBP, insets show the water contact angle measurements for both PbI2 film, respectively.
Fig. 1 show the theoretical model of each film. Fig. 1a shows a top-view SEM image of the pristine PbI2 film. A compact layered dense PbI2 film was formed owing to the high volatility of DMF [36]. This PbI2 film would not be completely converted into perovskite after the following CH3NH3I spin coating, leading to small perovskite grains with large numbers of pinholes and defects, as shown in Fig. 1b. In contrast, a free-standing porous PbI2⋅xtBP film was obtained by introducing tBP into the PbI2 precursor solution, which was then spin-coated on the ITO/PEDOT:PSS substrate, as shown in Fig. 1c. The self-assembled porous structure of the film was obtained owing to the coordination of tBP groups with Pb2þ in PbI2 [30–32,37]. To certify this mechanism, XPS measurements were performed and shown in Fig. S2. The experi mental results demonstrate that the binding energy of Pb 4f7/2 shifted to a lower energy level from 138.98 eV to 138.37 eV for PbI2 treated with tBP, which indicates the coordination of tBP with the center Pb2þ in PbI2 and formed porous PbI2⋅xtBP film. The XRD and water-contact-angle experimental results in Fig. 1e again demonstrate the existence and higher hydrophobicity of the PbI2⋅xtBP film. The complete conversion to the perovskite could be easily obtained by spin-coating CH3NH3I on the porous PbI2⋅xtBP film, owing to the enlarged reaction area for CH3NH3I provided by the higher porosity of the film. Fig. 1d shows a top-view image of the perovskite film, whose formation was facilitated by the porous PbI2⋅xtBP film. A considerably compact and pinhole-free perovskite film with very large grain sizes, even larger than 1 μm, was obtained. This significant change to perovskite by the insertion of tBP could significantly enhance the efficiencies and the long-term stabilities of the PSCs. The lower-magnification SEM image in Fig. S3 demonstrates that CH3NH3PbI3 treated with tBP on the reduction-active PEDOT:PSS interlayer has large grains. In order to increase the efficiencies and the stabilities of the PSCs, large diffusion lengths of the charge carriers, smaller numbers of defects, and low charge recombination are required. For this purpose, perovskite films with high qualities and good morphologies are crucial. We demonstrate that the morphology-modified tBP-induced perovskite through the reduction-active flexible PEDOT:PSS interlayer has better performances than those on a typical FTO/TiO2 and other substrates reported in recent studies [28–31]. For this purpose, we fabricated a pinhole-free dense TiO2 film on glass/FTO and grew a tBP-induced perovskite on its surface under the same conditions, as shown in Fig. 2a and b. The tBP-induced perovskite grains on the FTO/TiO2 substrate had similar sizes (approximately 400 nm) to those in the pre vious studies [28–31]; i.e., a smaller improvement was obtained than that on the PEDOT:PSS layer (~1 μm). Owing to the chemical oxidation of the metal halide perovskite by the reduction-active PEDOT:PSS HTL, the work function and the ionization potential of the perovskite may be
largely changed [38]. This considerable change could sufficiently enhance the coordination between tBP and CH3NH3PbI3, thus improving the morphology-modifying effect on the perovskite. Fig. 2c shows the structure of tBP functional group and a schematic of the operation principle of PEDOT:PSS and CH3NH3PbI3-tBP. As the electrons migrate from CH3NH3PbI3-tBP into the oxidized PEDOT chains, the ionization potential of the perovskite may change enhancing the coordination of Pb2þ in CH3NH3PbI3 with a nitrogen atom (pyridine nitrogen, which works as an electron-donating unit) at the end of tBP. This enhancement could significantly improves the perovskite morphology, the photovol taic performance, and the moisture resistance. In addition, PEDOT:PSS has an excellent flexibility, which is beneficial for flexible and wearable device applications (Fig. 2d–f). In order to optimize the amount of tBP to prevent the untuneable band gaps, unfavourable electric resistances and enhance the photo voltaic performances, different concentrations of tBP were introduced into the PbI2 precursor solution for further investigation. Fig. 3a–d shows SEM images of the CH3NH3PbI3⋅xtBP films with different con centrations of the tBP additive in the range of 20–50 μL/mL. An obvious morphology phase transition of the perovskite was observed. For the tBP content of 20 μL/mL, a particle morphology of the film was observed, similar with that of the perovskite whose formation was facilitated by the pristine PbI2. With the increasing in the concentration of tBP to 30 μL/mL, the perovskite consisted of coexisting particles and big flat grains with some pinholes. The further increasing in the tBP content to 40 μL/mL led to a considerably compact and pinhole-free perovskite film with large grain sizes, even larger than 1 μm. This indicates the signifi cant morphology-modifying effect in the tBP-induced perovskite facili tated by the reduction-active PEDOT:PSS interlayer. It is worth noting that no further considerable morphology changes could be observed at tBP concentrations larger than 40 μL/mL, which indicates that 40 μL/mL is the threshold concentration for the morphology improvement. This threshold concentration of tBP is only approximately 30–40% of those of other systems [28–31], which can be attributed to the stronger coordi nation between the tBP and the perovskite through the reduction-active PEDOT:PSS interlayer. The reduced content of tBP is beneficial to avoid untuneable band gaps and unfavourable electric resistances, improving the electron diffusions in the perovskite. The perovskite compositions and the photovoltaic performances of the inverted-planar PSCs with and without tBP incorporation were investigated. Fig. 4a shows XRD patterns of CH3NH3PbI3 with and without tBP, respectively. For the perovskite with tBP, smaller PbI2 peaks were detected with the increase in the tBP concentration. This implies that PbI2 could be converted into CH3NH3PbI3 more efficiently after the introduction of tBP. In addition, higher peak intensities of 3
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Fig. 2. SEM top-view images of CH3NH3PbI3⋅xtBP film grew on (a) ITO/TiO2, (b) ITO/PEDOT:PSS substrate. (c) The structure of tBP functional group and schematic illustration of the working principle between PEDOT:PSS and CH3NH3PbI3-tBP. Schematic drawings of morphology-modifying effect of perovskite induced by tBP on (d) ITO/TiO2, (e,f) ITO/PEDOT:PSS substrate.
CH3NH3PbI3 with tBP were observed, which indicates the improved crystallization of the perovskite. Fig. 4b shows J–V curves of the different types of inverted-planar rigid PSCs. The PSC fabricated without the tBP additive exhibits a very low PCE of 3.02%. As expected, the devices with incorporated tBP exhibited higher PCEs, as shown in Table 1. The best device (with a tBP concentration of 40 μL/mL) exhibited significantly improved performances with Voc ¼ 0.99 V, Jsc ¼ 20.66 mA cm 2, FF ¼ 0.79, and PCE ¼ 16.16%. The higher shortcircuit current density was achieved mainly by the complete conver sion of PbI2 into CH3NH3PbI3, highly crystalline grains, and lower concentration of tBP. It is worth noting that the PSCs with higher con centrations of tBP (50, 60 μL/mL) exhibited lower short-circuit current densities owing to the unfavourable electric resistances, which usually occurs due to a high content of additive. Further, the UV–vis absorption and corresponding IPCE spectra in Fig. 4c and d demonstrate the in creases in external quantum efficiency (EQE) and IPCE, respectively. The statistical devices PCEs, Jsc, Voc and FF of 2 sets of 15 PSCs devices each with CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) film were
also investigated and shown in Fig. S4. These above results show that the proper incorporation of tBP could largely improve the PCEs and repro ducibility of the PEDOT:PSS-based PSCs. In order to reveal the origin of the large improvement in photovoltaic performances, a cross-sectional SEM analysis of the perovskite was carried out. Fig. 5a and b shows cross-section images of the CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) films on ITO/PEDOT:PSS, respectively. Fig. 5c and d shows theoretical models of the fabricated PSCs with the perovskite films presented in Fig. 5a and b, respectively. As shown in Fig. 5a, a large number of defects could be observed in the small CH3NH3PbI3 grains. These defects and pinholes could lead to penetration of the ETL to the HTL, which might lead to a significant electron–hole recombination at the interface. Therefore, the opencircuit voltage and the FF of the PSCs significantly reduced, which decreased the photovoltaic performances. However, for the CH3NH3PbI3⋅xtBP film (x ¼ 40 μL/mL), considerably compact and very large CH3NH3PbI3⋅xtBP grains were observed, which is consistent with our previous results (Fig. 1d). In addition, the large CH3NH3PbI3⋅xtBP 4
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
The highly crystalline grains provide various merits such as fewer de fects and a large diffusion length of the charge carriers [39,40]. In addition, the quasi-all-in-one structure of the perovskite is suitable as a blocking layer to prevent ETL infiltration, suppress the electron–hole recombination at the interface, and thus significantly improve the open-circuit voltage and the FF of the inverted-planar PSCs, as shown in Fig. 5d. The open-circuit voltage of the PSC was significantly increased from 0.76 to 0.99 V. In addition, the FF for the rigid device fabricated under ambient air increased from 0.43 to 0.79, demonstrating its large potential. Moreover, the defect-free structure of the perovskite could largely suppress the traps at the interfaces of HTL/perovskite and per ovskite/ETL. Fig. 5e shows J–V curves of the inverted-planar PSCs with (40 μL/mL) and without tBP additive; no obvious hysteresis of the Table 1 The detailed photovoltaic parameters for fabricated PSCs with different con centrations of tBP additive. Fig. 3. SEM top-view images of CH3NH3PbI3⋅xtBP film with different concen trations of tBP additive: (a) 20, (b) 30, (c) 40, (d) 50 μL/ml.
grains exhibited the quasi-all-in-one structure with a thickness of approximately 300 nm, as shown in Fig. 5b. This morphology indicates that almost no grain boundaries existed in the perpendicular direction.
μL/mL
J (mA/cm2)
V (V)
FF (%)
PCE (%)
0 20 30 40 50 60
9.26 13.81 16.76 20.66 19.44 18.07
0.76 0.88 0.96 0.99 0.99 0.99
43 58 67 79 79 78
3.02 7.03 10.80 16.16 15.20 13.95
Fig. 4. (a) The XRD patterns of CH3NH3PbI3⋅xtBP film with 0, 20, 30, 40 μL/ml tBP additive. (b) J–V curves; (c) UV–Vis absorbance spectra; (d) IPCEs of fabricated PSCs with different concentrations of tBP additive. 5
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Fig. 5. (a,b) SEM cross-section images of CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (with 40 μL/ml tBP) films on ITO/PEDOT:PSS, respectively. (c,d) Theoretical models of the fabricated PSCs with the perovskite film from (a,b). (e) Typical J–V characteristics of fabricated PSC with or without tBP incorporation. (f) PL spectra of the CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) and conventional CH3NH3PbI3 films grown on PEDOT:PSS hole transport layer. (g) Nyquist plots at V ¼ 0.9 V bias for fabricated PSC with different concentrations of tBP.
photocurrent could be observed for the PSC with a tBP concentration of 40 μL/mL upon the changes in the sweep direction or the rate. The detail information about short-circuit current density, open-circuit voltage and fill factor of each PSC are all provided in Table 2. The results confirm the (almost) absence of traps and defects in this device, which are the main origin of photocurrent hysteresis [41]. The PL quenching efficiency of CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) onto PEDOT:PSS films was also investigated in Fig. 5f, which shows higher quenching efficiency than the conventional CH3NH3PbI3-based device; this implies better interfa cial contact and hole transfer in the CH3NH3PbI3⋅xtBP -based devices. In order to again validate our hypothesis about the electron–hole recom bination, EIS was carried out in the range of 1 Hz to 1 MHz under dark conditions at a bias of 0.9 V. Fig. 5g shows the characteristic Nyquist patterns of the fabricated PSCs with the small-particle CH3NH3PbI3 and the large-grain CH3NH3PbI3⋅xtBP (Fig. 5c and d, respectively). The main arcs in the low-frequency region represent the electron–hole recombi nation resistance, which was significantly enhanced in the PSC with the large-grain CH3NH3PbI3⋅xtBP. This further confirms that the quasi-all-in-one morphology of the perovskite could effectively block the electron–hole recombination. The larger morphology-modifying effect in the CH3NH3PbI3⋅xtBP film facilitated by the reduction-active PEDOT:PSS interlayer not only significantly enhanced the PCEs of the PSCs, but also provided good
moisture stabilities. In order to evaluate the moisture resistance of the CH3NH3PbI3⋅xtBP film, a PSC without encapsulation was prepared to analyse the long-term ambient stability. The moisture stabilities of the devices based on CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (40 μL/mL) are presented in Fig. 6a. The PSC with CH3NH3PbI3 exhibited a rapid deterioration in performance; the PCE decreased to 10% after 5 days and became almost zero after 10 days under the ambient air (RH ~ 50%). In contrast, the PSC with CH3NH3PbI3⋅xtBP exhibited a lower deterioration rate; the PCE kept over 80% after 10 days moisture exposure. The experimental results demonstrate the enhanced moisture resistances of the CH3NH3PbI3⋅xtBP film. This enhancement could provide long-term ambient stabilities, improving the properties and the applicability of the PSCs. The two small insets in Fig. 6a show the water-contact-angle measurement results for the CH3NH3PbI3 and CH3NH3PbI3⋅xtBP films. The water contact angle increased from 50.6� to 66.6� upon the tBP incorporation (40 μL/mL). The increase in the contact angle indicates that CH3NH3PbI3⋅xtBP had a higher hydrophobicity than that of the pristine CH3NH3PbI3. This is attributed to the strong coordination be tween the nitrogen atom at the end of tBP and Pb2þ in CH3NH3PbI3 facilitated by the reduction-active PEDOT:PSS interlayer, leading to a surrounding hydrophobic tertiary butyl group of tBP, as shown in the inset in Fig. 6c. The crystalline of CH3NH3PbI3 film with or without tBP incorporation under the high moisture level (50–60%) for several days has also been investigated by XRD measurements as demonstrated in Fig. 6b. For pristine CH3NH3PbI3 film, the characteristic PbI2 peak dramatically increased with the decreasing CH3NH3PbI3 peaks after 7 days moisture exposure. This indicates rapid decomposition of the perovskite into PbI2 by moisture, which might seriously hinder the application of the PSCs. In contrast, strong diffraction peaks character istic of the CH3NH3PbI3 perovskite structure were observed even after 7 days moisture exposure for CH3NH3PbI3⋅xtBP film, demonstrating excellent moisture resistances and potentials for commercialization. In order to further evaluate the moisture resistance, we fabricated PSCs
Table 2 The detailed photovoltaic parameters for fabricated PSCs with or without tBP incorporation. Sample
Scanning directions
Jsc (mA/ cm2)
Voc (V)
FF (%)
PCE (%)
CH3NH3PbI3 w/ tBP CH3NH3PbI3 w/o tBP
Reverse Forward Reverse Forward
20.66 20.47 9.26 8.7
0.99 0.99 0.76 0.94
79 78 43 23
16.16 15.81 3.02 1.99
6
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Table 3 Detailed photovoltaic parameters for fabricated PSCs with 40 μL/ml tBP incor poration at different moisture condition. RH 40–50% RH 50–60% RH 60–70%
J (mA/cm2)
V (V)
FF (%)
PCE (%)
20.66 19.69 18.02
0.99 0.97 0.87
79 74 65
16.16 14.13 10.19
Table 4 Detailed photovoltaic parameters for fabricated rigid and flexible PSCs with 40 μL/ml tBP incorporation. Glass/ITO rigid substrate PET/ITO flexible substrate
J (mA/cm2)
V (V)
FF (%)
PCE (%)
20.66 19.74
0.99 0.98
79 70
16.16 13.54
demonstrating the potentials for a low-cost mass manufacture (see Table 4). As the reduction-active PEDOT:PSS layer had a high flexibility, we replaced the rigid glass/ITO substrate by a flexible PET/ITO substrate to increase the applicability. The photovoltaic performance of the fabri cated inverted-planar flexible PSC with tBP incorporation (40 μL/mL) was investigated. Fig. 7 shows J–V curves of the inverted-planar rigid and flexible PSCs, along with a photograph of the flexible PSC without encapsulation. The inverted-planar flexible PSC fabricated with the tBP additive (40 μL/mL) still had an excellent PCE of 13.54% (approxi mately 80% of that of the rigid device) with Voc ¼ 0.98 V, Jsc ¼ 19.74 mA cm 2, and FF ¼ 0.7. The decreases in Jsc and FF are attributed to the higher series resistance of the flexible device. The photovoltaic performance of the flexible PSC after mechanical bending was evaluated and presented in Fig. S5 and Table S1. The good PCE was maintained upon the mechanical bending up to 20 cycles. This indicates that the fabricated flexible PSCs could tolerate repeated mechanical deformation. Fig. S6 and Table S2 shows that our inverted-planar flex ible PSC (without encapsulation) fabricated at an RH larger than 40% still exhibited a high PCE. These results show that high-photovoltaicperformance flexible PSCs can be fabricated without glove-box. The achieved mass manufacture, low-costs, high moisture resistances, and efficiencies of the antisolvent-free flexible and wearable PSCs demon strate the potentials for the commercialization in the future.
Fig. 6. (a) Device stability of fabricated PSCs without encapsulation with CH3NH3PbI3 and CH3NH3PbI3⋅xtBP film. Insets show the water contact angle measurements for each PSC. (b) The XRD patterns of CH3NH3PbI3 film with or without tBP incorporation under the high moisture level (50–60%) for 0 and 7 days, respectively. (c) J–V curves of PSCs with 40 μL/ml tBP additive fabricated at different moisture condition. Inset shows the constituent and structure of tBP functional groups, PbI2 and PbI2-tBP, respectively.
under different moisture conditions. As shown in Fig. 6c, the PSC (without encapsulation) fabricated in the RH range of 40–50% exhibited an excellent PCE of 16.16%. With the increase in the environmental RH to the range of 50–60%, the PCE of the PSC slightly decreased to 14.11%. With the further increase in RH to 60–70%, the average PCE decreased to 10.73%. Table 3 provides the detail information about short-circuit current density, open-circuit voltage and fill factor for each PSC fabricated under different humidity environment. These results indicate that the fabricated PEDOT:PSS-based PSCs with the CH3NH3PbI3⋅xtBP films exhibited high moisture resistances even as the environmental RH was larger than 40%. This implies that the use of a glove-box is not crucial for the fabrication of highly efficient PSCs,
Fig. 7. J–V curves of fabricated PSCs with 40 μL/ml tBP additive for rigid glass and flexible PET substrate. Inset is the photograph of the a fabricated flexible PSC without encapsulation. 7
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
4. Conclusion
[7] W. Zhu, C. Bao, F. Li, T. Yu, H. Gao, Y. Yi, J. Yang, G. Fu, X. Zhou, Z. Zou, Nano Energy 19 (2016) 17–26. [8] S.A. Kulkarni, T. Baikie, P.P. Boix, N. Yantara, N. Mathews, S. Mhaisalkar, J. Mater. Chem. A 2 (2014) 9221–9225. [9] E.T. Hoke, D.J. Slotcavage, E.R. Dohner, A.R. Bowring, H.I. Karunadasa, M. D. McGehee, Chem. Sci. 6 (2015) 613–617. [10] T.M. Michael Saliba, J.-Y. Seo, K. Domanski, J.-P. CorreaBaena, M.K. Nazeeruddin, S.M. Zakeeruddin, W. Tress, A. Abate, A. Hagfeldt, M. Gr€ atzel, Energy Environ. Sci. 9 (2016) 1989–1997. [11] M. Chen, M.-G. Ju, A.D. Carl, Y. Zong, R.L. Grimm, J. Gu, X.C. Zeng, Y. Zhou, N. P. Padture, Joule 2 (3) (2018) 558–570. [12] L.A. Frolova, D.V. Anokhin, A.A. Piryazev, S.Y. Luchkin, N.N. Dremova, K. J. Stevenson, P.A. Troshin, J. Phys. Chem. Lett. 8 (2017) 67–72. [13] J. Lei, F. Gao, H. Wang, J. Li, J. Jiang, X. Wu, R. Gao, Z. Yang, S.(F.) Liu, Sol. Energy Mater. Sol. Cells 187 (2018) 1–8. [14] Z. Lin, J. Chang, H. Zhu, Q.–H. Xu, C. Zhang, J. Quyang, Y. Hao, Sol. Energy Mater. Sol. Cells 172 (2017) 133–139. [15] M. Saliba, T. Matsui, K. Domanski, J.Y. Seo, A. Ummadisingu, S.M. Zakeeruddin, J. P. Correa-Baena, W.R. Tress, A. Abate, A. Hagfeldt, Science 354 (2016) 206. [16] S. Wu, Q. Liu, Y. Zheng, R. Li, T. Peng, J. Power Sources 359 (2017) 303–310. [17] I.S. Yang, M.R. Sohn, S.D. Sung, Y.J. Kim, Y. Yoo, J. Kim, W.I. Lee, Nano Energy 32 (2017) 414–421. [18] Y. Wei, K. Yao, X. Wang, Y. Jiang, X. Liu, N. Zhou, F. Li, Appl. Surf. Sci. 427 (2017) 782–790. [19] S.S. Mali, H. Kim, H.H. Kim, S.E. Shim, C.K. Hong, Mater. Today 21 (5) (2018) 483–500. [20] L.B. Huang, P.Y. Su, J.M. Liu, J.F. Huang, Y.F. Chen, S. Qin, J. Guo, Y.W. Xu, C. Y. Su, J. Power Sources 378 (2018) 483–490. [21] A.K. Jena, H.W. Chen, A. Kogo, Y. Sanehira, M. Ikegami, T. Miyasaka, ACS Appl. Mater. Interfaces 7 (2015) 9817–9823. [22] M. Wang, Z. Zang, B. Yang, X. Hua, K. Sun, L. Sun, Sol. Energy Mater. Sol. Cells 185 (2018) 117–123. [23] T. Duong, Y. Wu, H. Shen, J. Peng, S. Zhao, N. Wu, M. Lockrey, T. White, K. Weber, K. Catchpole, Sol. Energy Mater. Sol. Cells 188 (2018) 27–36. [24] J. Xu, O. Voznyy, R. Comin, X. Gong, G. Walters, M. Liu, P. Kanjanaboos, X. Lan, E. H. Sargent, Adv. Mater. 28 (2016) 2807–2815. [25] W. Zhang, J. Xiong, S. Wang, W. Liu, J. Li, D. Wang, H. Gu, X. Wang, J. Li, J. Power Sources 337 (2017) 118–124. [26] M. Long, T. Zhang, H. Zhu, G. Li, F. Wang, W. Guo, Y. Chai, W. Chen, Q. Li, K. S. Wong, J. Xu, K. Yan, Nano Energy 33 (2017) 485–496. [27] P. Luo, Z. Liu, W. Xia, C. Yuan, J. Cheng, Y. Lu, ACS Appl. Mater. Interfaces 7 (4) (2015) 2708–2714. [28] W. Li, H. Dong, L. Wang, N. Li, X. Guo, J. Li, Y. Qiu, J. Mater. Chem. A 2 (2014) 13587–13592. [29] H. Zhang, J. Mao, H. He, D. Zhang, H.L. Zhu, F. Xie, K.S. Wong, M. Gr€ atzel, W.C. H. Choy, Adv. Energy Mater. 5 (2015) 1501354. [30] Y. Shi, X. Wang, H. Zhang, B. Li, H. Lu, T. Ma, C. Hao, J. Mater. Chem. A 3 (2015) 22191–22198. [31] C. Liu, W. Ding, X. Zhou, J. Gao, C. Cheng, X. Zhao, B. Xu, J. Phys. Chem. C 121 (12) (2017) 6546–6553. [32] Y. Wu, X. Shi, X. Ding, Y. Ren, T. Hayat, A. Alsaedi, Y. Ding, P. Xu, S. Dai, ACS Appl. Mater. Interfaces 10 (4) (2018) 3602–3608. [33] M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y.B. Cheng, L. Spiccia, Angew. Chem., Int. Ed. Engl. 53 (2014) 9898. [34] N.J. Jeon, J.H. Noh, Y.C. Kim, W.S. Yang, S. Ryu, S.I. Seok, Nat. Mater. 13 (2014) 897–903. [35] M. Li, X. Yan, Z. Kang, X. Liao, Y. Li, X. Zheng, P. Lin, J. Meng, Y. Zhang, ACS Appl. Mater. Interfaces 9 (8) (2017) 7224. [36] Y. Wu, A. Islam, X. Yang, C. Qin, J. Liu, K. Zhang, W. Peng, L. Han, Energy Environ. Sci. 7 (2014) 2934. [37] R. Zhu, C. Gao, T. Sun, L. Shen, D. Sun, X. Li, Langmuir 32 (2016) 3294–3299. [38] B.J. Bruijnaers, E. Schiepers, C.H.L. Weijtens, S.C.J. Meskers, M.M. Wienk, R.A. J. Janssen, J. Mater. Chem. A 6 (2018) 6882–6890. [39] D. Liu, M.K. Gangishetty, T.L. Kelly, J. Mater. Chem. A 2 (2014) 19873–19881. [40] W. Nie, H. Tsai, R. Asadpour, J.C. Blancon, A.J. Neukirch, G. Gupta, J.J. Crochet, M. Chhowalla, S. Tretiak, M.A. Alam, H.L. Wang, A.D. Mohite, Science 347 (2015) 522–525. [41] Y. Wu, H. Shen, D. Walter, D. Jacobs, T. Duong, J. Peng, L. Jiang, Y.B. Cheng, K. Weber, Adv. Funct. Mater. 26 (2016) 6807–6813.
The antisolvent-free flexible PSCs with high photovoltaic perfor mances and excellent moisture stabilities were fabricated by a simple strategy, which involved the insertion of the reduction-active PEDOT: PSS polymer underneath the tBP-induced perovskite film. The morphology-modifying effect in the tBP-induced perovskite provided significantly better performances than those on other typical substrates owing to the specific oxidation facilitation by the PEDOT:PSS interlayer. The large (~1 μm) and highly crystalline perovskite grains were ob tained even without antisolvent treatment, which significantly increased the diffusion lengths of the charge carriers, acted as a blocking layer to decrease the electron–hole recombination, and thus improved the FF and the PCE. In addition, the smaller content of tBP suppressed the untuneable band gaps and unfavourable electric resistances, usually emerging at a high concentration of additive, and led to higher photo voltaic performances. For the simplest and unoptimised ITO/PEDOT: PSS/CH3NH3PbI3/PC61BM/Ag PSC, the PCE and the FF increased to 16.16% and 0.79 on the glass substrate and to 13.54% and 0.7 on the flexible PET substrate, respectively. In addition to the morphologymodifying function, the stronger tBP coordination with the perovskite by the reduction-active PEDOT:PSS provided higher moisture re sistances. The flexible PSCs fabrication could be carried out even under ambient air with a small PCE reduction. Acknowledgments This study was supported by the National Natural Science Founda tion of China (51402260), the Zhejiang Provincial Natural Science Foundation of China (LY18F050011, LQ19E030020, LZ16E020002), the Applied Basic Research Project of China National Textile and Apparel Council (J201801), the Science Foundation (17012144-Y), and the Program of the Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education (2017QN02) of the Zhejiang Sci-Tech University. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.jpowsour.2019.227216. References [1] A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 131 (2009) 6050–6051. [2] M.C. Tathavadekar, S.A. Agarkar, O.S. Game, U.P. Bansode, S.A. Kulkarni, S. G. Mhaisalkar, S.B. Ogale, Sol. Energy Mater. Sol. Cells 112 (2015) 12–19. [3] W. Chen, Y. Wu, Y. Yue, J. Liu, W. Zhang, X. Yang, H. Chen, E. Bi, I. Ashraful, M. Gr€ atzel, Science 350 (2015) 944–948. [4] D. Bi, B. Xu, P. Gao, L. Sun, M. Gr€ atzel, A. Hagfeldt, Nano Energy 23 (2016) 138–144. [5] W.S. Yang, B.W. Park, E.H. Jung, N.J. Jeon, Y.C. Kim, D.U. Lee, S.S. Shin, J. Seo, E. K. Kim, J.H. Noh, Science 356 (2017) 1376–1379. [6] National center for photovoltaics (NCPV) at the national renewable energy laboratory (NREL). www.nrel.gov/pv/assets/pdfs/best-research-cell-efficiencies -190416.pdf?tdsourcetag¼s_pctim_aiomsg.
8
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Simple fabrication of perovskite solar cells with enhanced efficiency, stability, and flexibility under ambient air Wei-Hsiang Chen a, b, Linlin Qiu b, Zhishan Zhuang b, Lixin Song b, Pingfan Du b, *, Jie Xiong a, b, **, Frank Ko c a b c
Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education, Hangzhou, 310018, PR China Silk Institute, College of Materials and Textiles, Zhejiang Sci-Tech University, Hangzhou, 310018, PR China Department of Materials Engineering, University of British Columbia, Vancouver, Canada
H I G H L I G H T S
G R A P H I C A L A B S T R A C T
� Positive tBP effect on perovskite can be greatly enhanced via PEDOT:PSS interlayer. � Perovskite grain with 1 μm size and quasi all-in-one structure can be obtained. � Flexible PSCs with excellent efficiency and moisture stability can be fulfilled. � Fabrication procedure can be performed greatly in ambient air condition.
A R T I C L E I N F O
A B S T R A C T
Keywords: 4-tert-butylpyridine Ambient-air fabrication Flexible perovskite solar cells Morphology-modifying Moisture stability
In order to increase the applicability and commercial potential of perovskite solar cells, simple fabrication of high-photovoltaic-performance flexible perovskite solar cells with excellent moisture stabilities without the use of a glove-box and an antisolvent is required. In this paper, we present a simple fabrication strategy involving introduction of 4-tert-butylpyridine into CH3NH3PbI3 and significantly enhancing 4-tert-butylpyridine morphology-modifying effect via a reduction-active flexible poly (3,4-ethylenedioxythiophene):poly (styr enesulfonate) interlayer. Owing to the specific oxidation facilitation by the poly (3,4-ethylenedioxythiophene): poly (styrenesulfonate) polymer, a perovskite film with large (~1 μm) and quasi-all-in-one-structured grains can be obtained, which significantly enhance the efficiency and the stability of the perovskite solar cells. Further more, the high-efficiency flexible perovskite solar cells fabricated by the simple strategy exhibit excellent moisture resistances owing to the stronger coordination and the outside-covering effect of the hydrophobic 4-tertbutylpyridine. The fabrication can be carried out under ambient air (without glovebox, relative humidi ty > 40%), which paves the way for wearable device application and commercialization.
* Corresponding author. ** Corresponding author. Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education, Hangzhou, 310018, PR China. E-mail addresses: [email protected] (P. Du), [email protected] (J. Xiong). https://doi.org/10.1016/j.jpowsour.2019.227216 Received 8 February 2019; Received in revised form 16 September 2019; Accepted 24 September 2019 Available online 1 October 2019 0378-7753/© 2019 Elsevier B.V. All rights reserved.
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
1. Introduction
2. Experimental methods
Organometal halide perovskite solar cells (PSCs) have attracted significant attentions owing to their promising characteristics such as high photovoltaic performances, low-cost solution-processable fabrica tion, flexible device designs, and potentials for commercialization. Over the past few years, the power conversion efficiency (PCE) has been improved amazingly from 3.8% to 25.2% [1–6]. In spite of large po tentials of PSCs for photovoltaic applications, various fundamental and practical issues have to be addressed for commercialization such as the outdoor moisture stabilities and low-cost mass-manufacture of highly efficient flexible and wearable PSCs. In this regard, extensive studies have been carried out including mixed-halide engineering [7–9], cation substitution [10–13], additive incorporation [14–16], high-stability charge-transport material selections [17–19], interface engineering [20–23], encapsulation [24,25] and other techniques [26,27]. Among the above approaches, introducing additive into perovskite is considered the most effective and simple option. However, PSCs are intrinsically limited by the drawback of high concentration additives in perovskite layer, which would lead to untuneable band gaps and poor charge transmission. In addition, so far a glove box is still necessary to avoid contacting moisture for most of high performance PSCs fabrication, which may seriously hamper the mass-manufacture and real applica tions. Therefore, a new and simple strategy that can overcome the lim itation of high concentration additives and improve the stabilities of the PSCs fabricated in ambient air is desperately needed. To simply fabricate PSCs that simultaneously exhibit excellent PCEs and long-term stabilities in ambient air, 4-tert-butylpyridine (tBP) has been used as a morphology-modifying agent to improve the perfor mances of PSCs [28–31]. This additive has a hydrophobic end, which can enhance the moisture resistance of the perovskite [31]. However, the improvements in photovoltaic performance and stability were very limited owing to the adverse effects caused by high concentration tBP in perovskite. In order to overcome this issue, Wu et al. [32] introduced tBP into the antisolvent instead of CH3NH3PbI3 and obtained better perov skite grain size and efficiency. However, the commonly applied anti solvents such as toluene and chlorobenzene [33,34] are highly toxic. In addition, a very large amount of antisolvent (>150 μL cm 2) is required and usually not reclaimed [35] for the PSCs fabrication, which is neither practical nor environment-friendly. Moreover, the antisolvent method is not suitable for a large-scale substrate, leading to PSCs scale-up and commercialization issues. Therefore, an intrinsic improvement in the tBP effect on the perovskite with less amount of tBP and without anti solvent treatment is advantageous. In this study, we propose a simple approach to enhance both tBP coordination with the perovskite and the PSCs flexibility by introducing a reduction-active flexible PEDOT:PSS interlayer. Owing to the specific chemical oxidation, the coordination between tBP and the perovskite could be significantly enhanced. A perovskite with very large (~1 μm) and quasi-all-in-one-structured grains could be obtained without the use of an antisolvent. Moreover, the amount of tBP could be largely reduced avoiding the unfavourable electric resistances and leading to higher short-circuit current density and stability. For the simplest and unopti mised ITO/PEDOT:PSS/CH3NH3PbI3-tBP/PC61BM/Ag PSC, the PCE and the fill factor (FF) were enhanced to 16.16% and 0.79 on a glass sub strate and to 13.54% and 0.7 on a flexible PET substrate, respectively. Furthermore, the PSCs exhibited excellent moisture resistances owing to the stronger coordination with the perovskite and the outside-covering effect of the hydrophobic tBP. The PSCs fabrication could be carried out even at an average relative humidity (RH) above 40% with a small reduction in the PCEs. This indicates that the use of a glove-box, which significantly increases the fabrication difficulties and the costs, is not required.
2.1. Materials CH3NH3I (MAI) and PbI2 were purchased from Kunshan Sunlaite New Energy Technology Co., Ltd. phenyl-C61-butyric acid methyl ester (PC61BM), poly (3,4-ethy-lenedioxythiophene) poly (styrenesulfonate) (PEDOT:PSS) and 4-tert-butylpyridine (tBP) were purchased from Xi’an Polymer Light Technology Corporation. Glass/ITO and PET/ITO sub strates were supplied from Advanced Election Technology Co., Ltd. The other materials were purchased from Aladdin. 2.2. Device fabrication Inverted-planar rigid and flexible perovskite solar cells (PSCs) were fabricated on laser-patterned, indium tin oxide (ITO) coated glass and PET (10 Ω sq 1) substrates, respectively. Both of them were sequentially ultrasonically cleaned with deionised water, acetone, ethyl alcohol, and isopropyl alcohol (IPA) for 15 min followed by a high-temperature drying with clean dry nitrogen and a treatment in an ultraviolet (UV) ozone oven for 20 min. After the cleaning, a thin layer of compact PEDOT:PSS was spin-coated on the clean substrate at 2000 rpm for 10 s and at 5000 rpm for 30 s in succession, and then annealed at 120 � C for 1 h. The PbI2 precursor (1.2 M, dissolved in DMF and stirred for 1 h at 70 � C) without and with different concentrations of tBP were continu ously spin-coated on the PEDOT:PSS substrate at 4500 rpm for 30 s and dried at 70 � C for 15 min. After the formation of the dried PbI2 (or PbI2⋅xtBP) film, CH3NH3I (12 mg/mL, dissolved in the IPA solvent) was spin-coated at 4000 rpm for 30 s to form the perovskite and annealed at 100 � C for 1 h. For the electron-transport layer (ETL), PC61BM (20 mg/ mL in chlorobenzene) was spin-coated on top of the perovskite layer at 2000 rpm for 30 s and annealed at 80 � C for 1 h. The fabrication was completed by thermal evaporation of Ag as an electrode with a thickness of approximately 100 nm and an effective area of 0.06 cm2. The whole fabrication was performed under ambient air (RH > 40%), which in dicates that the use of a glovebox is not necessary. The fabrication is illustrated in Fig. S1. 2.3. Characterisation The photocurrent density–voltage (J–V) characteristics of the cells were measured using a Keithley 2400-SCS source meter under AM 1.5 illumination with an intensity of 100 mW/cm2. The crystal structures of PbI2 and MAPbI3 were analysed by X-ray diffraction (XRD, Thermo ARLX’TRA, America) with Cu Kα radiation (λ ¼ 1.5418 Å). Field-emission scanning electron microscopy (FESEM, Ultra55, ZEISS, Germany) was employed to analyse the morphologies. The incident photon–to–current conversion efficiencies (IPCEs) of the cells were measured with a quantum-efficiency (QE)/IPCE test system (Solar Cell Scan100/Zolix). Electrochemical impedance spectroscopy (EIS) measurements were carried out using an electrochemical workstation (Zennium Pro, Ger many). The contact properties were characterised by a contact-angle instrument (Krüss Optronic, Germany). X-Ray photoelectron spectros copy (XPS) was measured with a K-Alpha instrument. The UV–visible (UV–vis) absorption was measured using a UV–vis spectrophotometer (Lambda 900, America). 3. Results and discussion For achieving high photovoltaic performances, stabilities, low costs, and simple fabrication of flexible perovskite solar cells (PSCs), 4-tertbutylpyridine (tBP) was introduced into the PbI2 precursor solution as a morphology-modifying agent on the reduction-active poly (3,4-ethyl enedioxythiophene):poly (styrenesulfonate) (PEDOT:PSS) interlayer. Fig. 1a–d shows the morphologies of the pristine PbI2, PbI2⋅xtBP, and the formed perovskite films on ITO/PEDOT:PSS, respectively. The insets in 2
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Fig. 1. SEM top-view images of (a) pristine PbI2, (b) CH3NH3PbI3, (c) PbI2⋅xtBP, (d) CH3NH3PbI3⋅xtBP films formed on ITO/PEDOT:PSS substrate. Insets show the theoretical model of each film. (e) The XRD patterns of PbI2 film with and without tBP, insets show the water contact angle measurements for both PbI2 film, respectively.
Fig. 1 show the theoretical model of each film. Fig. 1a shows a top-view SEM image of the pristine PbI2 film. A compact layered dense PbI2 film was formed owing to the high volatility of DMF [36]. This PbI2 film would not be completely converted into perovskite after the following CH3NH3I spin coating, leading to small perovskite grains with large numbers of pinholes and defects, as shown in Fig. 1b. In contrast, a free-standing porous PbI2⋅xtBP film was obtained by introducing tBP into the PbI2 precursor solution, which was then spin-coated on the ITO/PEDOT:PSS substrate, as shown in Fig. 1c. The self-assembled porous structure of the film was obtained owing to the coordination of tBP groups with Pb2þ in PbI2 [30–32,37]. To certify this mechanism, XPS measurements were performed and shown in Fig. S2. The experi mental results demonstrate that the binding energy of Pb 4f7/2 shifted to a lower energy level from 138.98 eV to 138.37 eV for PbI2 treated with tBP, which indicates the coordination of tBP with the center Pb2þ in PbI2 and formed porous PbI2⋅xtBP film. The XRD and water-contact-angle experimental results in Fig. 1e again demonstrate the existence and higher hydrophobicity of the PbI2⋅xtBP film. The complete conversion to the perovskite could be easily obtained by spin-coating CH3NH3I on the porous PbI2⋅xtBP film, owing to the enlarged reaction area for CH3NH3I provided by the higher porosity of the film. Fig. 1d shows a top-view image of the perovskite film, whose formation was facilitated by the porous PbI2⋅xtBP film. A considerably compact and pinhole-free perovskite film with very large grain sizes, even larger than 1 μm, was obtained. This significant change to perovskite by the insertion of tBP could significantly enhance the efficiencies and the long-term stabilities of the PSCs. The lower-magnification SEM image in Fig. S3 demonstrates that CH3NH3PbI3 treated with tBP on the reduction-active PEDOT:PSS interlayer has large grains. In order to increase the efficiencies and the stabilities of the PSCs, large diffusion lengths of the charge carriers, smaller numbers of defects, and low charge recombination are required. For this purpose, perovskite films with high qualities and good morphologies are crucial. We demonstrate that the morphology-modified tBP-induced perovskite through the reduction-active flexible PEDOT:PSS interlayer has better performances than those on a typical FTO/TiO2 and other substrates reported in recent studies [28–31]. For this purpose, we fabricated a pinhole-free dense TiO2 film on glass/FTO and grew a tBP-induced perovskite on its surface under the same conditions, as shown in Fig. 2a and b. The tBP-induced perovskite grains on the FTO/TiO2 substrate had similar sizes (approximately 400 nm) to those in the pre vious studies [28–31]; i.e., a smaller improvement was obtained than that on the PEDOT:PSS layer (~1 μm). Owing to the chemical oxidation of the metal halide perovskite by the reduction-active PEDOT:PSS HTL, the work function and the ionization potential of the perovskite may be
largely changed [38]. This considerable change could sufficiently enhance the coordination between tBP and CH3NH3PbI3, thus improving the morphology-modifying effect on the perovskite. Fig. 2c shows the structure of tBP functional group and a schematic of the operation principle of PEDOT:PSS and CH3NH3PbI3-tBP. As the electrons migrate from CH3NH3PbI3-tBP into the oxidized PEDOT chains, the ionization potential of the perovskite may change enhancing the coordination of Pb2þ in CH3NH3PbI3 with a nitrogen atom (pyridine nitrogen, which works as an electron-donating unit) at the end of tBP. This enhancement could significantly improves the perovskite morphology, the photovol taic performance, and the moisture resistance. In addition, PEDOT:PSS has an excellent flexibility, which is beneficial for flexible and wearable device applications (Fig. 2d–f). In order to optimize the amount of tBP to prevent the untuneable band gaps, unfavourable electric resistances and enhance the photo voltaic performances, different concentrations of tBP were introduced into the PbI2 precursor solution for further investigation. Fig. 3a–d shows SEM images of the CH3NH3PbI3⋅xtBP films with different con centrations of the tBP additive in the range of 20–50 μL/mL. An obvious morphology phase transition of the perovskite was observed. For the tBP content of 20 μL/mL, a particle morphology of the film was observed, similar with that of the perovskite whose formation was facilitated by the pristine PbI2. With the increasing in the concentration of tBP to 30 μL/mL, the perovskite consisted of coexisting particles and big flat grains with some pinholes. The further increasing in the tBP content to 40 μL/mL led to a considerably compact and pinhole-free perovskite film with large grain sizes, even larger than 1 μm. This indicates the signifi cant morphology-modifying effect in the tBP-induced perovskite facili tated by the reduction-active PEDOT:PSS interlayer. It is worth noting that no further considerable morphology changes could be observed at tBP concentrations larger than 40 μL/mL, which indicates that 40 μL/mL is the threshold concentration for the morphology improvement. This threshold concentration of tBP is only approximately 30–40% of those of other systems [28–31], which can be attributed to the stronger coordi nation between the tBP and the perovskite through the reduction-active PEDOT:PSS interlayer. The reduced content of tBP is beneficial to avoid untuneable band gaps and unfavourable electric resistances, improving the electron diffusions in the perovskite. The perovskite compositions and the photovoltaic performances of the inverted-planar PSCs with and without tBP incorporation were investigated. Fig. 4a shows XRD patterns of CH3NH3PbI3 with and without tBP, respectively. For the perovskite with tBP, smaller PbI2 peaks were detected with the increase in the tBP concentration. This implies that PbI2 could be converted into CH3NH3PbI3 more efficiently after the introduction of tBP. In addition, higher peak intensities of 3
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Fig. 2. SEM top-view images of CH3NH3PbI3⋅xtBP film grew on (a) ITO/TiO2, (b) ITO/PEDOT:PSS substrate. (c) The structure of tBP functional group and schematic illustration of the working principle between PEDOT:PSS and CH3NH3PbI3-tBP. Schematic drawings of morphology-modifying effect of perovskite induced by tBP on (d) ITO/TiO2, (e,f) ITO/PEDOT:PSS substrate.
CH3NH3PbI3 with tBP were observed, which indicates the improved crystallization of the perovskite. Fig. 4b shows J–V curves of the different types of inverted-planar rigid PSCs. The PSC fabricated without the tBP additive exhibits a very low PCE of 3.02%. As expected, the devices with incorporated tBP exhibited higher PCEs, as shown in Table 1. The best device (with a tBP concentration of 40 μL/mL) exhibited significantly improved performances with Voc ¼ 0.99 V, Jsc ¼ 20.66 mA cm 2, FF ¼ 0.79, and PCE ¼ 16.16%. The higher shortcircuit current density was achieved mainly by the complete conver sion of PbI2 into CH3NH3PbI3, highly crystalline grains, and lower concentration of tBP. It is worth noting that the PSCs with higher con centrations of tBP (50, 60 μL/mL) exhibited lower short-circuit current densities owing to the unfavourable electric resistances, which usually occurs due to a high content of additive. Further, the UV–vis absorption and corresponding IPCE spectra in Fig. 4c and d demonstrate the in creases in external quantum efficiency (EQE) and IPCE, respectively. The statistical devices PCEs, Jsc, Voc and FF of 2 sets of 15 PSCs devices each with CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) film were
also investigated and shown in Fig. S4. These above results show that the proper incorporation of tBP could largely improve the PCEs and repro ducibility of the PEDOT:PSS-based PSCs. In order to reveal the origin of the large improvement in photovoltaic performances, a cross-sectional SEM analysis of the perovskite was carried out. Fig. 5a and b shows cross-section images of the CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) films on ITO/PEDOT:PSS, respectively. Fig. 5c and d shows theoretical models of the fabricated PSCs with the perovskite films presented in Fig. 5a and b, respectively. As shown in Fig. 5a, a large number of defects could be observed in the small CH3NH3PbI3 grains. These defects and pinholes could lead to penetration of the ETL to the HTL, which might lead to a significant electron–hole recombination at the interface. Therefore, the opencircuit voltage and the FF of the PSCs significantly reduced, which decreased the photovoltaic performances. However, for the CH3NH3PbI3⋅xtBP film (x ¼ 40 μL/mL), considerably compact and very large CH3NH3PbI3⋅xtBP grains were observed, which is consistent with our previous results (Fig. 1d). In addition, the large CH3NH3PbI3⋅xtBP 4
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
The highly crystalline grains provide various merits such as fewer de fects and a large diffusion length of the charge carriers [39,40]. In addition, the quasi-all-in-one structure of the perovskite is suitable as a blocking layer to prevent ETL infiltration, suppress the electron–hole recombination at the interface, and thus significantly improve the open-circuit voltage and the FF of the inverted-planar PSCs, as shown in Fig. 5d. The open-circuit voltage of the PSC was significantly increased from 0.76 to 0.99 V. In addition, the FF for the rigid device fabricated under ambient air increased from 0.43 to 0.79, demonstrating its large potential. Moreover, the defect-free structure of the perovskite could largely suppress the traps at the interfaces of HTL/perovskite and per ovskite/ETL. Fig. 5e shows J–V curves of the inverted-planar PSCs with (40 μL/mL) and without tBP additive; no obvious hysteresis of the Table 1 The detailed photovoltaic parameters for fabricated PSCs with different con centrations of tBP additive. Fig. 3. SEM top-view images of CH3NH3PbI3⋅xtBP film with different concen trations of tBP additive: (a) 20, (b) 30, (c) 40, (d) 50 μL/ml.
grains exhibited the quasi-all-in-one structure with a thickness of approximately 300 nm, as shown in Fig. 5b. This morphology indicates that almost no grain boundaries existed in the perpendicular direction.
μL/mL
J (mA/cm2)
V (V)
FF (%)
PCE (%)
0 20 30 40 50 60
9.26 13.81 16.76 20.66 19.44 18.07
0.76 0.88 0.96 0.99 0.99 0.99
43 58 67 79 79 78
3.02 7.03 10.80 16.16 15.20 13.95
Fig. 4. (a) The XRD patterns of CH3NH3PbI3⋅xtBP film with 0, 20, 30, 40 μL/ml tBP additive. (b) J–V curves; (c) UV–Vis absorbance spectra; (d) IPCEs of fabricated PSCs with different concentrations of tBP additive. 5
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Fig. 5. (a,b) SEM cross-section images of CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (with 40 μL/ml tBP) films on ITO/PEDOT:PSS, respectively. (c,d) Theoretical models of the fabricated PSCs with the perovskite film from (a,b). (e) Typical J–V characteristics of fabricated PSC with or without tBP incorporation. (f) PL spectra of the CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) and conventional CH3NH3PbI3 films grown on PEDOT:PSS hole transport layer. (g) Nyquist plots at V ¼ 0.9 V bias for fabricated PSC with different concentrations of tBP.
photocurrent could be observed for the PSC with a tBP concentration of 40 μL/mL upon the changes in the sweep direction or the rate. The detail information about short-circuit current density, open-circuit voltage and fill factor of each PSC are all provided in Table 2. The results confirm the (almost) absence of traps and defects in this device, which are the main origin of photocurrent hysteresis [41]. The PL quenching efficiency of CH3NH3PbI3⋅xtBP (x ¼ 40 μL/mL) onto PEDOT:PSS films was also investigated in Fig. 5f, which shows higher quenching efficiency than the conventional CH3NH3PbI3-based device; this implies better interfa cial contact and hole transfer in the CH3NH3PbI3⋅xtBP -based devices. In order to again validate our hypothesis about the electron–hole recom bination, EIS was carried out in the range of 1 Hz to 1 MHz under dark conditions at a bias of 0.9 V. Fig. 5g shows the characteristic Nyquist patterns of the fabricated PSCs with the small-particle CH3NH3PbI3 and the large-grain CH3NH3PbI3⋅xtBP (Fig. 5c and d, respectively). The main arcs in the low-frequency region represent the electron–hole recombi nation resistance, which was significantly enhanced in the PSC with the large-grain CH3NH3PbI3⋅xtBP. This further confirms that the quasi-all-in-one morphology of the perovskite could effectively block the electron–hole recombination. The larger morphology-modifying effect in the CH3NH3PbI3⋅xtBP film facilitated by the reduction-active PEDOT:PSS interlayer not only significantly enhanced the PCEs of the PSCs, but also provided good
moisture stabilities. In order to evaluate the moisture resistance of the CH3NH3PbI3⋅xtBP film, a PSC without encapsulation was prepared to analyse the long-term ambient stability. The moisture stabilities of the devices based on CH3NH3PbI3 and CH3NH3PbI3⋅xtBP (40 μL/mL) are presented in Fig. 6a. The PSC with CH3NH3PbI3 exhibited a rapid deterioration in performance; the PCE decreased to 10% after 5 days and became almost zero after 10 days under the ambient air (RH ~ 50%). In contrast, the PSC with CH3NH3PbI3⋅xtBP exhibited a lower deterioration rate; the PCE kept over 80% after 10 days moisture exposure. The experimental results demonstrate the enhanced moisture resistances of the CH3NH3PbI3⋅xtBP film. This enhancement could provide long-term ambient stabilities, improving the properties and the applicability of the PSCs. The two small insets in Fig. 6a show the water-contact-angle measurement results for the CH3NH3PbI3 and CH3NH3PbI3⋅xtBP films. The water contact angle increased from 50.6� to 66.6� upon the tBP incorporation (40 μL/mL). The increase in the contact angle indicates that CH3NH3PbI3⋅xtBP had a higher hydrophobicity than that of the pristine CH3NH3PbI3. This is attributed to the strong coordination be tween the nitrogen atom at the end of tBP and Pb2þ in CH3NH3PbI3 facilitated by the reduction-active PEDOT:PSS interlayer, leading to a surrounding hydrophobic tertiary butyl group of tBP, as shown in the inset in Fig. 6c. The crystalline of CH3NH3PbI3 film with or without tBP incorporation under the high moisture level (50–60%) for several days has also been investigated by XRD measurements as demonstrated in Fig. 6b. For pristine CH3NH3PbI3 film, the characteristic PbI2 peak dramatically increased with the decreasing CH3NH3PbI3 peaks after 7 days moisture exposure. This indicates rapid decomposition of the perovskite into PbI2 by moisture, which might seriously hinder the application of the PSCs. In contrast, strong diffraction peaks character istic of the CH3NH3PbI3 perovskite structure were observed even after 7 days moisture exposure for CH3NH3PbI3⋅xtBP film, demonstrating excellent moisture resistances and potentials for commercialization. In order to further evaluate the moisture resistance, we fabricated PSCs
Table 2 The detailed photovoltaic parameters for fabricated PSCs with or without tBP incorporation. Sample
Scanning directions
Jsc (mA/ cm2)
Voc (V)
FF (%)
PCE (%)
CH3NH3PbI3 w/ tBP CH3NH3PbI3 w/o tBP
Reverse Forward Reverse Forward
20.66 20.47 9.26 8.7
0.99 0.99 0.76 0.94
79 78 43 23
16.16 15.81 3.02 1.99
6
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
Table 3 Detailed photovoltaic parameters for fabricated PSCs with 40 μL/ml tBP incor poration at different moisture condition. RH 40–50% RH 50–60% RH 60–70%
J (mA/cm2)
V (V)
FF (%)
PCE (%)
20.66 19.69 18.02
0.99 0.97 0.87
79 74 65
16.16 14.13 10.19
Table 4 Detailed photovoltaic parameters for fabricated rigid and flexible PSCs with 40 μL/ml tBP incorporation. Glass/ITO rigid substrate PET/ITO flexible substrate
J (mA/cm2)
V (V)
FF (%)
PCE (%)
20.66 19.74
0.99 0.98
79 70
16.16 13.54
demonstrating the potentials for a low-cost mass manufacture (see Table 4). As the reduction-active PEDOT:PSS layer had a high flexibility, we replaced the rigid glass/ITO substrate by a flexible PET/ITO substrate to increase the applicability. The photovoltaic performance of the fabri cated inverted-planar flexible PSC with tBP incorporation (40 μL/mL) was investigated. Fig. 7 shows J–V curves of the inverted-planar rigid and flexible PSCs, along with a photograph of the flexible PSC without encapsulation. The inverted-planar flexible PSC fabricated with the tBP additive (40 μL/mL) still had an excellent PCE of 13.54% (approxi mately 80% of that of the rigid device) with Voc ¼ 0.98 V, Jsc ¼ 19.74 mA cm 2, and FF ¼ 0.7. The decreases in Jsc and FF are attributed to the higher series resistance of the flexible device. The photovoltaic performance of the flexible PSC after mechanical bending was evaluated and presented in Fig. S5 and Table S1. The good PCE was maintained upon the mechanical bending up to 20 cycles. This indicates that the fabricated flexible PSCs could tolerate repeated mechanical deformation. Fig. S6 and Table S2 shows that our inverted-planar flex ible PSC (without encapsulation) fabricated at an RH larger than 40% still exhibited a high PCE. These results show that high-photovoltaicperformance flexible PSCs can be fabricated without glove-box. The achieved mass manufacture, low-costs, high moisture resistances, and efficiencies of the antisolvent-free flexible and wearable PSCs demon strate the potentials for the commercialization in the future.
Fig. 6. (a) Device stability of fabricated PSCs without encapsulation with CH3NH3PbI3 and CH3NH3PbI3⋅xtBP film. Insets show the water contact angle measurements for each PSC. (b) The XRD patterns of CH3NH3PbI3 film with or without tBP incorporation under the high moisture level (50–60%) for 0 and 7 days, respectively. (c) J–V curves of PSCs with 40 μL/ml tBP additive fabricated at different moisture condition. Inset shows the constituent and structure of tBP functional groups, PbI2 and PbI2-tBP, respectively.
under different moisture conditions. As shown in Fig. 6c, the PSC (without encapsulation) fabricated in the RH range of 40–50% exhibited an excellent PCE of 16.16%. With the increase in the environmental RH to the range of 50–60%, the PCE of the PSC slightly decreased to 14.11%. With the further increase in RH to 60–70%, the average PCE decreased to 10.73%. Table 3 provides the detail information about short-circuit current density, open-circuit voltage and fill factor for each PSC fabricated under different humidity environment. These results indicate that the fabricated PEDOT:PSS-based PSCs with the CH3NH3PbI3⋅xtBP films exhibited high moisture resistances even as the environmental RH was larger than 40%. This implies that the use of a glove-box is not crucial for the fabrication of highly efficient PSCs,
Fig. 7. J–V curves of fabricated PSCs with 40 μL/ml tBP additive for rigid glass and flexible PET substrate. Inset is the photograph of the a fabricated flexible PSC without encapsulation. 7
W.-H. Chen et al.
Journal of Power Sources 442 (2019) 227216
4. Conclusion
[7] W. Zhu, C. Bao, F. Li, T. Yu, H. Gao, Y. Yi, J. Yang, G. Fu, X. Zhou, Z. Zou, Nano Energy 19 (2016) 17–26. [8] S.A. Kulkarni, T. Baikie, P.P. Boix, N. Yantara, N. Mathews, S. Mhaisalkar, J. Mater. Chem. A 2 (2014) 9221–9225. [9] E.T. Hoke, D.J. Slotcavage, E.R. Dohner, A.R. Bowring, H.I. Karunadasa, M. D. McGehee, Chem. Sci. 6 (2015) 613–617. [10] T.M. Michael Saliba, J.-Y. Seo, K. Domanski, J.-P. CorreaBaena, M.K. Nazeeruddin, S.M. Zakeeruddin, W. Tress, A. Abate, A. Hagfeldt, M. Gr€ atzel, Energy Environ. Sci. 9 (2016) 1989–1997. [11] M. Chen, M.-G. Ju, A.D. Carl, Y. Zong, R.L. Grimm, J. Gu, X.C. Zeng, Y. Zhou, N. P. Padture, Joule 2 (3) (2018) 558–570. [12] L.A. Frolova, D.V. Anokhin, A.A. Piryazev, S.Y. Luchkin, N.N. Dremova, K. J. Stevenson, P.A. Troshin, J. Phys. Chem. Lett. 8 (2017) 67–72. [13] J. Lei, F. Gao, H. Wang, J. Li, J. Jiang, X. Wu, R. Gao, Z. Yang, S.(F.) Liu, Sol. Energy Mater. Sol. Cells 187 (2018) 1–8. [14] Z. Lin, J. Chang, H. Zhu, Q.–H. Xu, C. Zhang, J. Quyang, Y. Hao, Sol. Energy Mater. Sol. Cells 172 (2017) 133–139. [15] M. Saliba, T. Matsui, K. Domanski, J.Y. Seo, A. Ummadisingu, S.M. Zakeeruddin, J. P. Correa-Baena, W.R. Tress, A. Abate, A. Hagfeldt, Science 354 (2016) 206. [16] S. Wu, Q. Liu, Y. Zheng, R. Li, T. Peng, J. Power Sources 359 (2017) 303–310. [17] I.S. Yang, M.R. Sohn, S.D. Sung, Y.J. Kim, Y. Yoo, J. Kim, W.I. Lee, Nano Energy 32 (2017) 414–421. [18] Y. Wei, K. Yao, X. Wang, Y. Jiang, X. Liu, N. Zhou, F. Li, Appl. Surf. Sci. 427 (2017) 782–790. [19] S.S. Mali, H. Kim, H.H. Kim, S.E. Shim, C.K. Hong, Mater. Today 21 (5) (2018) 483–500. [20] L.B. Huang, P.Y. Su, J.M. Liu, J.F. Huang, Y.F. Chen, S. Qin, J. Guo, Y.W. Xu, C. Y. Su, J. Power Sources 378 (2018) 483–490. [21] A.K. Jena, H.W. Chen, A. Kogo, Y. Sanehira, M. Ikegami, T. Miyasaka, ACS Appl. Mater. Interfaces 7 (2015) 9817–9823. [22] M. Wang, Z. Zang, B. Yang, X. Hua, K. Sun, L. Sun, Sol. Energy Mater. Sol. Cells 185 (2018) 117–123. [23] T. Duong, Y. Wu, H. Shen, J. Peng, S. Zhao, N. Wu, M. Lockrey, T. White, K. Weber, K. Catchpole, Sol. Energy Mater. Sol. Cells 188 (2018) 27–36. [24] J. Xu, O. Voznyy, R. Comin, X. Gong, G. Walters, M. Liu, P. Kanjanaboos, X. Lan, E. H. Sargent, Adv. Mater. 28 (2016) 2807–2815. [25] W. Zhang, J. Xiong, S. Wang, W. Liu, J. Li, D. Wang, H. Gu, X. Wang, J. Li, J. Power Sources 337 (2017) 118–124. [26] M. Long, T. Zhang, H. Zhu, G. Li, F. Wang, W. Guo, Y. Chai, W. Chen, Q. Li, K. S. Wong, J. Xu, K. Yan, Nano Energy 33 (2017) 485–496. [27] P. Luo, Z. Liu, W. Xia, C. Yuan, J. Cheng, Y. Lu, ACS Appl. Mater. Interfaces 7 (4) (2015) 2708–2714. [28] W. Li, H. Dong, L. Wang, N. Li, X. Guo, J. Li, Y. Qiu, J. Mater. Chem. A 2 (2014) 13587–13592. [29] H. Zhang, J. Mao, H. He, D. Zhang, H.L. Zhu, F. Xie, K.S. Wong, M. Gr€ atzel, W.C. H. Choy, Adv. Energy Mater. 5 (2015) 1501354. [30] Y. Shi, X. Wang, H. Zhang, B. Li, H. Lu, T. Ma, C. Hao, J. Mater. Chem. A 3 (2015) 22191–22198. [31] C. Liu, W. Ding, X. Zhou, J. Gao, C. Cheng, X. Zhao, B. Xu, J. Phys. Chem. C 121 (12) (2017) 6546–6553. [32] Y. Wu, X. Shi, X. Ding, Y. Ren, T. Hayat, A. Alsaedi, Y. Ding, P. Xu, S. Dai, ACS Appl. Mater. Interfaces 10 (4) (2018) 3602–3608. [33] M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y.B. Cheng, L. Spiccia, Angew. Chem., Int. Ed. Engl. 53 (2014) 9898. [34] N.J. Jeon, J.H. Noh, Y.C. Kim, W.S. Yang, S. Ryu, S.I. Seok, Nat. Mater. 13 (2014) 897–903. [35] M. Li, X. Yan, Z. Kang, X. Liao, Y. Li, X. Zheng, P. Lin, J. Meng, Y. Zhang, ACS Appl. Mater. Interfaces 9 (8) (2017) 7224. [36] Y. Wu, A. Islam, X. Yang, C. Qin, J. Liu, K. Zhang, W. Peng, L. Han, Energy Environ. Sci. 7 (2014) 2934. [37] R. Zhu, C. Gao, T. Sun, L. Shen, D. Sun, X. Li, Langmuir 32 (2016) 3294–3299. [38] B.J. Bruijnaers, E. Schiepers, C.H.L. Weijtens, S.C.J. Meskers, M.M. Wienk, R.A. J. Janssen, J. Mater. Chem. A 6 (2018) 6882–6890. [39] D. Liu, M.K. Gangishetty, T.L. Kelly, J. Mater. Chem. A 2 (2014) 19873–19881. [40] W. Nie, H. Tsai, R. Asadpour, J.C. Blancon, A.J. Neukirch, G. Gupta, J.J. Crochet, M. Chhowalla, S. Tretiak, M.A. Alam, H.L. Wang, A.D. Mohite, Science 347 (2015) 522–525. [41] Y. Wu, H. Shen, D. Walter, D. Jacobs, T. Duong, J. Peng, L. Jiang, Y.B. Cheng, K. Weber, Adv. Funct. Mater. 26 (2016) 6807–6813.
The antisolvent-free flexible PSCs with high photovoltaic perfor mances and excellent moisture stabilities were fabricated by a simple strategy, which involved the insertion of the reduction-active PEDOT: PSS polymer underneath the tBP-induced perovskite film. The morphology-modifying effect in the tBP-induced perovskite provided significantly better performances than those on other typical substrates owing to the specific oxidation facilitation by the PEDOT:PSS interlayer. The large (~1 μm) and highly crystalline perovskite grains were ob tained even without antisolvent treatment, which significantly increased the diffusion lengths of the charge carriers, acted as a blocking layer to decrease the electron–hole recombination, and thus improved the FF and the PCE. In addition, the smaller content of tBP suppressed the untuneable band gaps and unfavourable electric resistances, usually emerging at a high concentration of additive, and led to higher photo voltaic performances. For the simplest and unoptimised ITO/PEDOT: PSS/CH3NH3PbI3/PC61BM/Ag PSC, the PCE and the FF increased to 16.16% and 0.79 on the glass substrate and to 13.54% and 0.7 on the flexible PET substrate, respectively. In addition to the morphologymodifying function, the stronger tBP coordination with the perovskite by the reduction-active PEDOT:PSS provided higher moisture re sistances. The flexible PSCs fabrication could be carried out even under ambient air with a small PCE reduction. Acknowledgments This study was supported by the National Natural Science Founda tion of China (51402260), the Zhejiang Provincial Natural Science Foundation of China (LY18F050011, LQ19E030020, LZ16E020002), the Applied Basic Research Project of China National Textile and Apparel Council (J201801), the Science Foundation (17012144-Y), and the Program of the Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education (2017QN02) of the Zhejiang Sci-Tech University. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.jpowsour.2019.227216. References [1] A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 131 (2009) 6050–6051. [2] M.C. Tathavadekar, S.A. Agarkar, O.S. Game, U.P. Bansode, S.A. Kulkarni, S. G. Mhaisalkar, S.B. Ogale, Sol. Energy Mater. Sol. Cells 112 (2015) 12–19. [3] W. Chen, Y. Wu, Y. Yue, J. Liu, W. Zhang, X. Yang, H. Chen, E. Bi, I. Ashraful, M. Gr€ atzel, Science 350 (2015) 944–948. [4] D. Bi, B. Xu, P. Gao, L. Sun, M. Gr€ atzel, A. Hagfeldt, Nano Energy 23 (2016) 138–144. [5] W.S. Yang, B.W. Park, E.H. Jung, N.J. Jeon, Y.C. Kim, D.U. Lee, S.S. Shin, J. Seo, E. K. Kim, J.H. Noh, Science 356 (2017) 1376–1379. [6] National center for photovoltaics (NCPV) at the national renewable energy laboratory (NREL). www.nrel.gov/pv/assets/pdfs/best-research-cell-efficiencies -190416.pdf?tdsourcetag¼s_pctim_aiomsg.
8