Enhanced fatigue resistance of suppressed hysteresis in perovskite solar cells by an organic crosslinker

Solar Energy Materials and Solar Cells 176 (2018) 30–35

Contents lists available at ScienceDirect

Solar Energy Materials and Solar Cells journal homepage: www.elsevier.com/locate/solmat

Enhanced fatigue resistance of suppressed hysteresis in perovskite solar cells by an organic crosslinker

T

Chang-Keun Lima,1, Qi Lib,1, Tianmu Zhangc, Tim Thomayc, Alexander N. Cartwrightc, ⁎ ⁎ Mark T. Swihartb, , Paras N. Prasada, a

Department of Chemistry and Institute for Lasers, Photonics, and Biophotonics, University at Buffalo, The State University of New York, Buffalo, NY 14260, United States Department of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, NY 14260, United States c Department of Electrical Engineering, University at Buffalo, The State University of New York, Buffalo, NY 14260, United States b

A R T I C L E I N F O

A B S T R A C T

Keywords: Perovskites Photovoltaics Hysteresis Stability

With record power conversion efficiencies of hybrid perovskite solar cells now exceeding 20% under laboratory conditions, improvements in stability of the cells under real-world working conditions are now key requirements for their commercial success. Here, we present a novel strategy to reduce penetration of humidity and oxygen into perovskite films via incorporation of a diammonium glycol. The two ammonium groups of this molecule allow it to serve as a crosslinker in the structure, bridging two unit cells within a crystallite or even across a grain boundary. In a planar heterojunction solar cell containing PCBM as an electron transport layer, the power conversion efficiency of the cell with ~0.1% diammonium glycol in the absorber layer was 13.96%, slightly exceeding that of the glycol-free device (13.53%). Most importantly, the glycol-free device exhibited the typical growth in hysteresis with performance degradation, but hysteresis remained suppressed in the device doped with diammonium glycol, even as its overall performance deteriorated. Futhermore, the chemical stability of the unpackaged device under continuous AM1.5 G illumination at ambient conditions was substantially improved relative to the glycol-free device. Formation of PbI2 was significantly suppressed, which could minimize release of toxic Pb ions.

1. Introduction Solution processed methylammonium lead halide perovskites have become the most promising new active light harvester for photovoltaic devices over the past few years [1,2]. However, prospects for commercialization of perovskite solar cells (PSC) are still limited by two major challenges: poor stability and hysteresis. Hysteresis is mainly attributable to an imbalance between electron and hole mobilities in the device due to interfacial electronic traps, ionic migration, and unbalanced carrier transport from the perovskite layer [3]. Recently, researchers have successfully suppressed the hysteretic behavior of PSCs by incorporation of a fullerene-based electron-transport material, usually phenyl-C61-butyric acid methyl ester (PCBM), into or upon the perovskite layer [4–6]. This treatment reduces electron trapping and ion migration, and passivates defects at the interface between the electron-transporting layer (ETL) and the perovskite film. The lack of stability of PSCs is largely attributable to the high water solubility of perovskite materials, which leads to degradation at high humidity [7].

⁎

1

Various strategies for improving the stability of the perovskite layer, including replacement of methylammonium with formamidinium [8,9] or Pb with Cs [10,11], introduction of an organic crosslinker [12,13] or a halogen/pseudo halogen (such as thiocyanate (SCN)) anion [14,15], and formation of a 2D hydrophobic interlayer between active 3D perovskite domains [16–18] have shown potential to block diffusion of water into the active layer or to stabilize the cubic phase [19]. However, the crosslinked devices showed hysteretic behavior and the 2D perovskite devices suffered from a loss of efficiency relative to their 3D counterparts. Meanwhile, UV illumination of mesoporous (mp) TiO2 in bulk heterojunction devices has accelerated oxygen-induced degradation through the conversion of molecular oxygen to a reactive radical [7]. Planar heterojunction structures (without mpTiO2) are less susceptible to this problem, but exhibit greater hysteresis compared to bulk heterojunction devices [4]. Therefore, a strategy for improving stability and suppressing hysteresis, simultaneously, without a loss of performance, remains a crucial need. To address this need, here we propose and demonstrate planar

Corresponding authors. E-mail addresses: swihart@buffalo.edu (M.T. Swihart), pnprasad@buffalo.edu (P.N. Prasad). These authors contributed equally to this work.

https://doi.org/10.1016/j.solmat.2017.11.032 Received 15 August 2017; Received in revised form 16 November 2017; Accepted 18 November 2017 0927-0248/ © 2017 Elsevier B.V. All rights reserved.

Solar Energy Materials and Solar Cells 176 (2018) 30–35

C.-K. Lim et al.

Fig. 1. Structure and surface morphology of methylammonium lead iodide (MAPbI3) perovskite thin layer with or without 2,2′-(ethylenedioxy)bis(ethylammonium iodide) (GA) (a) Chemical structure of GA. (b) Schematic chemical structures (upper) in grain boundaries of the perovskite film and surface morphologies by SEM (lower). Scale bars are 1 µm. (c) Full and (d) enlarged (at 8 ppm) 1H NMR spectra of MAI (9.5 mM), MAPbI3 (9.5 mM of MAI + 10 mM of PbI2) with or without GA (1 mM), and GA (1 mM) in DMF-d6 are shown. * and ** represent proton peaks of DMF and water, respectively. The ammonium proton peaks were assigned as –NH3+ in (d).

83 µL 1 M HCl was diluted with another 3 mL anhydrous IPA. Then the diluted acid solution was added dropwise into the Ti-isopropoxide solution with vigorous stirring. The reaction proceeded at room temperature for 2.5 h. Before spin-coating, the solution was filtered through a PTFE membrane syringe filter with 0.2 µm nominal pore size.

heterojunction PSC devices that combine a PCBM ETL with a new approach to sealing crystalline grain boundaries of the perovskite layer, using diammonium crosslinkers to improve device performance and stability. Use of PCBM suppresses hysteresis and improves photovoltaic performance, while crosslinking with diammonium glycol molecules seals the grain boundaries to reduce oxygen and water diffusion into the active layer. With this strategy, we achieved improvements not only in the performance and stability, but also in suppression of hysteresis even during degradation of the devices. The improved chemical stability of the diammonium glycol-doped device can prolong device lifetime, while reducing potential leakage of toxic Pb ions.

2.3. Device fabrication The solar cells were fabricated on prepatterned ITO substrates (Thin Film Devices, Inc., 20 Ω sq−1), which were cleaned sequentially with Hellmanex III solution, acetone, methanol, and isopropyl alcohol with 10 min ultrasonication in each solvent, followed by 40 min ozone treatment. The hole blocking compact layer of TiO2 was deposited on the cleaned substrates by spin-coating at 3000 rpm for 30 s then annealing at 500 °C for 10 min. Substrates were transferred into an inert atmosphere glove box for further fabrication steps. Phenyl-C61-butyric acid methyl ester (PCBM, 20 mg/mL) in chlorobenzene solution was spin-coated on the TiO2 layer and annealed at 100 °C for 10 min. Then, the perovskite precursor solution was spin-coated at 2000 rpm for 30 s and then annealed at 110 °C for 0.5–1 h to remove residual solvent and fully crystallize the film. During the annealing process, hybrid films changed from light yellow to dark brown. Next, a chlorobenzene solution of poly(3-hexylthiophene-2,5-diyl) (P3HT, 15 mg/mL) was deposited by spin-coating at 2500 rpm for 30 s. The top electrode (Au) was thermally evaporated through a shadow mask to achieve a device area of 0.07 cm−2. For comparison, PCBM-free devices were made by directly spin-coating the perovskite precursor solution on the TiO2 layer following the same method.

2. Experimental 2.1. Perovskite precursor preparation Methylammonium iodide (CH3NH3I, MAI) and 2,2′-(Ethylenedioxy) bis(ethylammonium iodide) (GA) were synthesized in our laboratories. Briefly, to prepare MAI and GA, 40 wt% methylamine in water (0.1 mol) and 98% 2,2′-(Ethylenedioxy)bis(ethylamine) (0.05 mol) were each separately dissolved in 20 mL of methanol. 57 wt% HI in water (0.2 mol) was added dropwise to each amine solution at RT with stirring, and the mixtures were stirred at RT for 4 h. The solvent in the resulting mixture was evaporated, and the products were each dissolved in ~10 mL of methanol with mild heating. The product solutions were added dropwise to diethyl ether (200 mL) with stirring to obtain white precipitates. The filtered precipitates were recrystallized from methanol, and the white crystals were dried in a vacuum oven at 90 °C for 24 h. The perovskite precursor solution was prepared by mixing MAI, PbAc2, and GA in dimethylformamide (DMF) at a 3:1 molar ratio of MAI to PbAc2 and 0.1 wt% GA to MAI. The resulting concentration of the precursor solution was 40 wt%. This solution was stirred at 60 °C for 0.5–2 h before spin coating. For comparison, GA-free perovskite precursor solution was prepared by the same method.

2.4. Photovoltaic characterization The current density-voltage (J-V) characteristics were determined using a Keithley 2400 source meter to apply an external potential bias to the solar cells. The generated photocurrent under a simulated AM 1.5G spectrum (Oriel solar simulator) was recorded at a 0.03 V/s scan rate. The measurements were carried out under ambient laboratory conditions (Avg. temp.: 25 °C and Avg. relative humidity: 60%). The solar cell devices were masked with an aperture with 0.07065 cm−2

2.2. TiO2 precursor preparation 438 µL Ti-isopropoxide was diluted with 3 mL anhydrous IPA, while 31

Solar Energy Materials and Solar Cells 176 (2018) 30–35

C.-K. Lim et al.

diffraction patterns were obtained using a Rigaku Ultima IV system with Cu K-α radiation (40 kV, 44 mA), from 10 to 70° (2θ) with a 0.01° step. 2.6. Scanning electron microscopy (SEM) SEM images were taken using a Carl Zeiss AURIGA CrossBeam system at 2 kV accelerating voltage. The cross-sectional images were obtained by milling using the focused ion beam (FIB) in the same microscope. The tilt angle was set at 54° and corrected by 36° when taking images. 2.7. Photoluminescence lifetime measurement Photoluminescence lifetime measurements were performed using an ultrafast laser system as an excitation source and an optical streak camera as a temporal and spectral detection system. Ultrafast pulses with a wavelength of 405 nm and a temporal length of less than 200 fs were directed on the sample, which was placed in an optical cryostat kept at room temperature but evacuated to a pressure of 3.3 × 10−2 mbar to avoid degradation due to humidity. The FWHM diameter of 1 mm of the Gaussian shaped spot was chosen to cover a large area of the sample and achieve a low maximum pulse power density of 0.16 W/ cm2 which corresponds to a total pulse energy of 640 nJ/cm2. The emitted photoluminescence passed through a dichroic beamsplitter (LP526) and a long pass filter (Corning OG) before being focused into a streak camera with a maximum temporal resolution of 2 ps and a spectral resolution of 1 nm. 3. Results and discussion We synthesized 2,2′-(Ethylenedioxy)bis(ethylammonium iodide) (glycol diammonium, GA) (Fig. 1(a)) from 2,2′-(Ethylenedioxy)bis (ethylamine) by treatment with hydroiodic acid, and verified its interaction with MAPbI3 by 1H NMR measurements. We hypothesized that, during the growth of perovskite crystals with methylammonium iodide (MAI), lead acetate (PbAc2), and a small amount of GA, the GA can seal the methylammonium lead iodide (MAPbI3) perovskite layer, in which the two ammonium groups of GA can occupy cation sites in two different unit cells of MAPbI3, connecting them together (Fig. 1(b)) [12]. Before demonstration of this hypothesis, we probed the interaction between GA and MAPbI3 in their mixed solution using 1H NMR (Fig. 1(c) and (d)). The proton peak of the –NH3+ group on MAI was shifted upon formation of MAPbI3 and then mixing with GA. These shifts indicate that GA is interacting with MAPbI3. The proton peak of the –NH3+ group on GA was merged into the broad water peak due to fast proton exchange with water molecules when GA was separately dissolved in DMF-d6. However, when it was mixed with MAPbI3, the water peak became narrower (Fig. 1(c)). This result means the proton exchange was highly suppressed due to an interaction between GA and MAPbI3. Thus, the NMR spectra confirmed the interaction between GA and MAPbI3. Then, we prepared thin films of MAPbI3 perovskite crystals made from PbAc2 and MAI with or without GA. The lead (II) source, PbAc2, was selected to obtain a smooth and pinhole-free perovskite layer [20]. The surface morphologies of the resulting perovskite thin films are shown in Fig. 1(b) and Fig. S1 (Supporting Information). The perovskite layers were generally smooth and free of pinholes. Furthermore, the grains in the film containing GA were fused, with an appearance similar to a plastered wall, while those in the GA-free film were more segregated with much clearer boundaries. These results suggest that incorporating GA into the perovskite layer may help to resist penetration of humidity and oxygen by crosslinking perovskite cubes. Thus, we next studied the role of GA in photovoltaic performance and stability of PSC devices. We prepared solar cells with three different device structures, ITO/ compact(cp)TiO2/MAPbI3/poly(3-hexylthiophene-2,5-diyl)(P3HT)/Au,

Fig. 2. Structure and photovoltaic performance of perovskite solar cells. (a) Structure and energy levels of the PSC device. (b) J-V curves for hysteresis test upon AM1.5G llumination. The scan rate was 30 mV/s for both directions. (c) PCE histograms of the three different types of devices. The curves in the graph are Gaussian distributions for the corresponding colored histograms.

Table 1 Photoluminescence (PL) parameters of the solar cell devices. Devices

PL lifetime (ns)

PL

cpTiO2/MAPbI3 cpTiO2/PCBM/MAPbI3 cpTiO2/PCBM/MAPbI3+GA

91 ± 13.9 39 ± 9.9 38 ± 8.2

765 ± 1.2 765 ± 1.2 768 ± 3.5

max

(nm)

active area. 2.5. X-Ray diffraction The perovskite samples were deposited on glass/ITO/TiO2 with and without PCBM following the procedure described above. The X-ray 32

Solar Energy Materials and Solar Cells 176 (2018) 30–35

C.-K. Lim et al.

Table 2 Photovoltaic parameters of the solar cell devices. Devices cpTiO2/MAPbI3 cpTiO2/PCBM/MAPbI3 cpTiO2/PCBM/MAPbI3+GA

Best Avg. Best Avg. Best Avg.

Voc (V)

Jsc (mA cm−2)

FF

PCE (%)

Hysteresis Factor (%)*

1 0.89 ± 0.13 1.08 0.97 ± 0.09 1.06 0.98 ± 0.03

17.7 14.99 ± 1.76 19.9 16.31 ± 1.74 19.1 16.39 ± 1.59

0.72 0.52 ± 0.1 0.76 0.66 ± 0.09 0.79 0.67 ± 0.07

9.98 6.92 ± 1.99 13.53 10.59 ± 1.91 13.96 10.69 ± 1.59

1.52 17.53 ± 16.59 0.2 3.47 ± 4.09 0 2.79 ± 3.28

* These values were quantified with the equation, Hysteresis factor = Areaforward / Areareverse −1 × 100% [22].

Fig. 3. Stability in photovoltaic performance of solar cells with or without GA in the active layer. J-V curves of (a) the GA-doped (crosslinked) device and (b) the GA-free (pristine) device after periods of continuous AM1.5G illumination. The scan rate was 30 mV/s in both directions. (c) Time-dependent change in hysteresis factor between forward and reverse scans under continuous illumination. (d) Time-dependent PCE attenuation under continuous illumination. All the experiments were conducted under ambient conditions, without encapsulation.

that included a PCBM ETL [with GA (forward (fill factor (FF): 0.70, power conversion efficiency (PCE): 13.69%), reverse (FF: 0.72, PCE: 13.96%)) and without GA (forward (FF: 0.67, PCE: 13.25%), reverse (FF: 0.70, PCE: 13.53%))]. Meanwhile, the PCBM-free device (forward (FF: 0.46, PCE: 8.49%), reverse (FF: 0.58, PCE: 9.98%)) not only showed much lower performance, but also a significant difference between forward and reverse scans. Suppression of hysteresis by PCBM is attributable to passivation of the perovskite interfaces, blocking ion migration and trap generation, as well as to higher electron mobility in PCBM (2 × 10−3 cm2/Vs)[23] compared to TiO2 (3.83 × 10−4 cm2/ Vs) [24]. The three devices showed nearly identical photoluminescence (PL) spectra, but the PL lifetime was considerably shortened for devices containing PCBM (with GA: 39 ns and without GA: 38 ns) compared to those without it (91 ns) (Table 1, and Fig. S3 in Supporting Information), consistent with the superior electron extraction capability of PCBM [4]. Consequently, incorporation of the PCBM layer significantly

ITO/cpTiO2/PCBM/MAPbI3/P3HT/Au, and ITO/cpTiO2/PCBM/ MAPbI3+GA/P3HT/Au and tested their performance (Fig. 2(a)). Because we used only a small amount of insulating GA (less than 0.1 wt% relative to the other precursors) in the active layer, we do not expect any effect on the energy levels of the perovskite film. For the hole transporting layer (HTL), we chose P3HT to exclude any possible interference in stability studies from moisture absorption in the more widely used HTL, Spiro-OMeTAD, which contains polar dopants [21]. Cross sectional images of the devices, like those in Fig. S2 (Supporting information), showed that the thickness of the perovskite layer, 250–300 nm, was similar in all three devices. With these devices, we first compared the hysteretic behaviors with or without PCBM and then studied the GA effect on the photovoltaic performance in combination with the PCBM layer. Fig. 2(b) shows J-V curves of the three devices upon AM1.5G illumination with a solar simulator. Hysteresis was almost completely suppressed in the devices

33

Solar Energy Materials and Solar Cells 176 (2018) 30–35

C.-K. Lim et al.

Fig. 4. Chemical stability of the perovskite layers upon continuous AM1.5G illumination under at ambient conditions, without encapsulation. (a) X-ray diffraction (XRD) pattern (10–17 degrees) of the three different devices before and after 30 or 60 min illumination. (b) XRD peak area ratio of PbI2 to perovskite (110) of the three devices before and after illumination.

gradually increased in the two GA-free perovskite films, through various degradation pathways [7], while the GA-crosslinked perovskite showed negligible PbI2 formation. Results from many films are summarized in Fig. 4(b)) to quantify the trend and demonstrate its reproducibility. Thus, we conclude that crosslinking of perovskite layer by GA can suppress formation of PbI2 precipitates in the film and, hence, prolong the device performances including suppression of hysteresis.

improved the fill factor and therefore power conversion efficiency (PCE) (Table 2). The two PCBM-containing devices showed almost the same photovoltaic performance with respect to all parameters, showing that the small amount of crosslinker (GA) has minimal effect on the baseline device performance. Considering the significant loss of performance in studies where 3D perovskite solar cells were converted to 2D counterparts [16–18], the observation that GA addition does not impact the initial performance is an important and promising result. The PCE for both of the PCBM-containing devices fall significantly short of the world record for perovskite solar cells, but, considering the use of undoped P3HT as the HTL, the efficiency is quite good relative to prior reports [25–27]. To demonstrate the effect of GA addition on device performance, we scanned J-V curves of GA-doped (sealed) and undoped (pristine) perovskite solar cell devices upon continuous AM1.5G illumination under ambient conditions (Avg. temp.: 25 °C, Avg. RH: 60%) without encapsulation. Surprisingly, the GA-doped devices exhibited negligible hysteresis, even when their PCE was substantially degraded, whereas the hysteresis of pristine devices grew continuously as their performance degraded (Fig. 3(a)-(c), and Fig. S4 in Supporting information). As a result, the time-dependent hysteresis factor upon continuous AM1.5G illumination differed dramatically between devices with and without GA (Fig. 3(c)). This strongly suggests that GA crosslinking is suppressing the formation of electron traps as well as the formation of PbI2 precipitates. Furthermore, the illumination time-dependent difference in PCE between the forward and the reverse scan shows complete suppression of hysteresis in the GA-doped perovskite devices (Fig. 3(d)). The PCE of the GA-doped device decreased to 50% of its initial value after 85 min of illumination, while the GA-free device exhibited the same performance degradation after only 23 min (Fig. 3(d), and Fig. S4 in Supporting information). Interestingly, the higher sensitivity of GA-free devices to moisture and oxygen gave rise to higher standard deviations of the initial PCE, with the standard deviation decreasing after 70% attenuation of the performance because of almost complete degradation. Meanwhile, the GA-doped devices showed uniformly lower deviations over the whole process, due to their lower sensitivity to their environment. To evaluate the ability of the GA-crosslinked device to resist humidity and oxygen penetration, we studied the stability of the perovskite layer upon continuous AM1.5G illumination under ambient conditions (Avg. temp.: 25 °C, Avg. relative humidity (RH): 60%) without encapsulation. We assessed the chemical stability by using xray diffraction (XRD) to monitor the formation of PbI2 via degradation of MAPbI3. As shown in Fig. 4(a) and Figs. S5-S7 (Supporting information), for representative films, the PbI2 peaks in the perovskite film on cpTiO2 were slightly higher than others, and those of the GAdoped perovskite were lowest. Under illumination, the PbI2 peaks

4. Conclusions We designed and demonstrated improvement of stability in chemical structure, photovoltaic performance, and suppressed hysteresis of methylammonium lead iodide-based hybrid perovskite solar cells via crosslinking of perovskite films with a small amount of a diammonium glycol molecule (GA). We suppressed hysteresis in the planar perovskite solar cell by inclusion of a PCBM electron transport layer. With GAdoping, the solar cells showed a sustainable suppression of hysteresis even when their performance degraded, and the chemical stability of the hybrid perovskite layer in unpackaged devices under ambient conditions was significantly improved by blocking penetration of humidity and oxygen into grain boundaries of perovskite crystals. This crosslinking or sealing approach can be a robust tool to extend the lifetime of perovskite solar cells and reduce the potential release of toxic Pb2+ ions from them. Acknowledgment This work was supported in part by the New York State Center of Excellence in Materials Informatics. Conflicts of interest None. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.solmat.2017.11.032. References [1] Q. Lin, A. Armin, P.L. Burn, P. Meredith, Organohalide perovskites for solar energy conversion, Acc. Chem. Res. 49 (2016) 545–553. [2] N.-G. Park, M. Grätzel, T. Miyasaka, K. Zhu, K. Emery, Towards stable and commercially available perovskite solar cells, Nat. Energy 1 (2016) 16152. [3] E.L. Unger, E.T. Hoke, C.D. Bailie, W.H. Nguyen, A.R. Bowring, T. Heumüller, M.G. Christoforo, M.D. McGehee, Hysteresis and transient behavior in

34

Solar Energy Materials and Solar Cells 176 (2018) 30–35

C.-K. Lim et al.

[4]

[5]

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16] D.H. Cao, C.C. Stoumpos, O.K. Farha, J.T. Hupp, M.G. Kanatzidis, 2D homologous perovskites as light-absorbing materials for solar cell applications, J. Am. Chem. Soc. 137 (2015) 7843–7850. [17] H. Tsai, W. Nie, J.C. Blancon, C.C. Stoumpos, R. Asadpour, B. Harutyunyan, A.J. Neukirch, R. Verduzco, J.J. Crochet, S. Tretiak, L. Pedesseau, J. Even, M.A. Alam, G. Gupta, J. Lou, P.M. Ajayan, M.J. Bedzyk, M.G. Kanatzidis, Highefficiency two-dimensional Ruddlesden-Popper perovskite solar cells, Nature 536 (2016) 312–316. [18] I.C. Smith, E.T. Hoke, D. Solis-Ibarra, M.D. McGehee, H.I. Karunadasa, A. Layered, Hybrid perovskite solar-cell absorber with enhanced moisture stability, Angew. Chem. Int. Ed. 53 (2014) 11232–11235. [19] Z. Wang, Z. Shi, T. Li, Y. Chen, W. Huang, Stability of perovskite solar cells: a prospective on the substitution of the A cation and X anion, Angew. Chem. Int. Ed. 56 (2017) 1190–1212. [20] W. Zhang, M. Saliba, D.T. Moore, S.K. Pathak, M.T. Horantner, T. Stergiopoulos, S.D. Stranks, G.E. Eperon, J.A. Alexander-Webber, A. Abate, A. Sadhanala, S. Yao, Y. Chen, R.H. Friend, L.A. Estroff, U. Wiesner, H.J. Snaith, Ultrasmooth organicinorganic perovskite thin-film formation and crystallization for efficient planar heterojunction solar cells, Nat. Commun. 6 (2015) 6142. [21] J.L. Yang, B.D. Siempelkamp, D.Y. Liu, T.L. Kelly, Investigation of CH3NH3PbI3 degradation rates and mechanisms in controlled humidity environments using in situ techniques, ACS Nano 9 (2015) 1955–1963. [22] J. Xu, O. Voznyy, R. Comin, X. Gong, G. Walters, M. Liu, P. Kanjanaboos, X. Lan, E.H. Sargent, Crosslinked remote-doped hole-extracting contacts enhance stability under accelerated lifetime testing in perovskite solar cells, Adv. Mater. 28 (2016) 2807–2815. [23] G. Garcia-Belmonte, A. Munar, E.M. Barea, J. Bisquert, I. Ugarte, R. Pacios, Charge carrier mobility and lifetime of organic bulk heterojunctions analyzed by impedance spectroscopy, Org. Electron. 9 (2008) 847–851. [24] D. Yang, X. Zhou, R. Yang, Z. Yang, W. Yu, X. Wang, C. Li, S. Liu, R.P.H. Chang, Surface optimization to eliminate hysteresis for record efficiency planar perovskite solar cells, Energy Environ. Sci. 9 (2016) 3071–3078. [25] J. Xiao, J. Shi, H. Liu, Y. Xu, S. Lv, Y. Luo, D. Li, Q. Meng, Y. Li, Efficient CH3NH3PbI3 perovskite solar cells based on graphdiyne (GD)-modified P3HT holetransporting material, Adv. Energy Mater. 5 (2015) 1401943. [26] W. Yan, Y. Li, W. Sun, H. Peng, S. Ye, Z. Liu, Z. Bian, C. Huang, High-performance hybrid perovskite solar cells with polythiophene as hole-transporting layer via electrochemical polymerization, RSC Adv. 4 (2014) 33039. [27] Q. Lin, A. Armin, R.C.R. Nagiri, P.L. Burn, P. Meredith, Electro-optics of perovskite solar cells, Nat. Photonics 9 (2014) 106–112.

current–voltage measurements of hybrid-perovskite absorber solar cells, Energy Environ. Sci. 7 (2014) 3690–3698. J. Xu, A. Buin, A.H. Ip, W. Li, O. Voznyy, R. Comin, M. Yuan, S. Jeon, Z. Ning, J.J. McDowell, P. Kanjanaboos, J.P. Sun, X. Lan, L.N. Quan, D.H. Kim, I.G. Hill, P. Maksymovych, E.H. Sargent, Perovskite-fullerene hybrid materials suppress hysteresis in planar diodes, Nat. Commun. 6 (2015) 7081. Y. Shao, Z. Xiao, C. Bi, Y. Yuan, J. Huang, Origin and elimination of photocurrent hysteresis by fullerene passivation in CH3NH3PbI3 planar heterojunction solar cells, Nat. Commun. 5 (2014) 5784. C.-H. Chiang, C.-G. Wu, Bulk heterojunction perovskite–PCBM solar cells with high fill factor, Nat. Photonics 10 (2016) 196–200. T.A. Berhe, W.-N. Su, C.-H. Chen, C.-J. Pan, J.-H. Cheng, H.-M. Chen, M.-C. Tsai, L.Y. Chen, A.A. Dubale, B.-J. Hwang, Organometal halide perovskite solar cells: degradation and stability, Energy Environ. Sci. 9 (2016) 323–356. N.J. Jeon, J.H. Noh, W.S. Yang, Y.C. Kim, S. Ryu, J. Seo, S.I. Seok, Compositional engineering of perovskite materials for high-performance solar cells, Nature 517 (2015) 476–480. M.R. Leyden, M.V. Lee, S.R. Raga, Y. Qi, Large formamidinium lead trihalide perovskite solar cells using chemical vapor deposition with high reproducibility and tunable chlorine concentrations, J. Mater. Chem. A 3 (2015) 16097–16103. Q. Ma, S. Huang, X. Wen, M.A. Green, A.W. Ho‐Baillie, Hole transport layer free inorganic CsPbIBr2 perovskite solar cell by dual source thermal evaporation, Adv. Energy Mater. 6 (2016). R.J. Sutton, G.E. Eperon, L. Miranda, E.S. Parrott, B.A. Kamino, J.B. Patel, M.T. Hörantner, M.B. Johnston, A.A. Haghighirad, D.T. Moore, Bandgap‐tunable cesium lead halide perovskites with high thermal stability for efficient solar cells, Adv. Energy Mater. 6 (2016). X. Li, M.I. Dar, C. Yi, J. Luo, M. Tschumi, S.M. Zakeeruddin, M.K. Nazeeruddin, H. Han, M. Gratzel, Improved performance and stability of perovskite solar cells by crystal crosslinking with alkylphosphonic acid omega-ammonium chlorides, Nat. Chem. 7 (2015) 703–711. G. Grancini, C. Roldán-Carmona, I. Zimmermann, E. Mosconi, X. Lee, D. Martineau, S. Narbey, F. Oswald, F. De Angelis, M. Graetzel, One-Year stable perovskite solar cells by 2D/3D interface engineering, Nat. Commun. 8 (2017). Y. Chen, T. Chen, L. Dai, Layer‐by‐layer growth of CH3NH3PbI3−xClx for highly efficient planar heterojunction perovskite solar cells, Adv. Mater. 27 (2015) 1053–1059. Q. Jiang, D. Rebollar, J. Gong, E.L. Piacentino, C. Zheng, T. Xu, Pseudohalide‐induced moisture tolerance in perovskite CH3NH3Pb(SCN)2I thin films, Angew. Chem. 127 (2015) 7727–7730.

35