Author’s Accepted Manuscript Surface Modified Fullerene Electron Transport Layers for Stable and Reproducible Flexible Perovskite Solar Cells Seulki Song, Rebecca Hill, Kyoungwon Choi, Konrad Wojciechowski, Stephen Barlow, Johannes Leisen, Henry J. Snaith, Seth R. Marder, Taiho Park
PII: DOI: Reference:
www.elsevier.com/locate/nanoenergy
S2211-2855(18)30296-9 https://doi.org/10.1016/j.nanoen.2018.04.068 NANOEN2698
To appear in: Nano Energy Received date: 30 January 2018 Revised date: 13 April 2018 Accepted date: 26 April 2018 Cite this article as: Seulki Song, Rebecca Hill, Kyoungwon Choi, Konrad Wojciechowski, Stephen Barlow, Johannes Leisen, Henry J. Snaith, Seth R. Marder and Taiho Park, Surface Modified Fullerene Electron Transport Layers for Stable and Reproducible Flexible Perovskite Solar Cells, Nano Energy, https://doi.org/10.1016/j.nanoen.2018.04.068 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Surface Modified Fullerene Electron Transport Layers for Stable and Reproducible Flexible Perovskite Solar Cells Seulki Songa1, Rebecca Hillb2, Kyoungwon Choia1, Konrad Wojciechowskic3, Stephen Barlowb2, Johannes Leisenb2, Henry J. Snaithc3, Seth R. Marderb2, Taiho Parka1* a
Chemical Engineering, Pohang University of Science and Technology (POSTECH), 77
Cheongam-Ro, Nam-gu, Pohang, Kyoungbuk, Korea. b
Center for Organic Photonics and Electronics and School of Chemistry and Biochemistry,
Georgia Institute of Technology, Atlanta, Georgia 30332-0400, United States.
c
Department of Physics Clarendon Laboratory, Parks Road, Oxford, OX1 3PU, United
Kingdom.
*Corresponding Author.
[email protected]
1
Chemical Engineering, Pohang University of Science and Technology (POSTECH), 77 Cheongam-Ro, Nam-gu, Pohang, Kyoungbuk, Korea. 2
Center for Organic Photonics and Electronics and School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332-0400, United States. 3
Department of Physics Clarendon Laboratory, Parks Road, Oxford, OX1 3PU, United Kingdom. 1
ABSTRACT
We report flexible planar perovskite solar cells with robust electron-transport layers (ETLs) processed at low temperature. A poly(allylamine) (PAA; 0.08 wt%) solution was deposited on a C60 layer and heated at 150 °C for 60 s, resulting in the formation of an insoluble robust C60–PAA electron-transport layer (ETL) on the flexible substrate. The flexible planar perovskite solar cell with the C60–PAA ETL exhibited excellent properties with 83% efficiency retention (= 15.2% without hysteresis) after 600 cycles of bending. This performance is superior to that of the flexible device with a C60 ETL fabricated without the use of PAA (65% efficiency retention; = 9.8% with some hysteresis).
KEYWORDS Low temperature; Planar device; Organic ETL; Flexible device
Introduction
As a result of their outstanding properties, including large absorption coefficients, [1, 2] bipolar charge transport,[3-5] low band gaps,[5] long carrier diffusion lengths,[6] and small exciton-binding energies,[7, 8] perovskites have attracted considerable attention in the field of solar-cell research. The power conversion efficiency (PCE) of perovskite solar cells based on glass substrates has reached approximately 22% in a period of five years.[9-14] However, comparing glass substrate based perovskite solar cell, flexible perovskite solar cells have not been intensively studied despite their various advantages (e.g., low production cost, less thickness, and lightweight) compared to the glass substrate-based devices. Moreover, solar cells based on flexible substrates can be applied to automotive-integrated photovoltaics (AIPV) or portable and indoor electronics.[15, 16] 2
Flexible perovskite solar cells, particularly those with planar n-i-p structures, generally employ flexible plastic substrates, such as polyethylene naphthalate (PEN) and polyethylene terephthalate (PET), rather than conventional glass substrates. These plastic substrates are only stable at temperatures below 150 °C. Therefore, in order to successfully fabricate flexible perovskite solar cells, beyond the usual considerations of energy levels, charge transport, and stability under operating conditions, there are additional limitations placed on the materials chosen for various layers of the solar cell, not only regarding their stability to bending, but also the temperatures at which they are processed. Especially, electron-transporting layers must be carefully selected due to their rigorous preparation process. For example, inorganic TiO2 (processed at >450 °C)[10, 17] and SnO2 (≥ 180 °C)[18] have been widely used as electron-transporting layers (ETLs) in planar solar-cell devices. However, the high-processing temperatures for these materials are incompatible with PEN and PET substrates; moreover, these materials from highly brittle thin layers.[19] Although several examples of room temperature process SnO2 ETL have been reported, complex processes are required and hysteresis behavior is shown. [20-22] The Zn2SnO4 ETL reported by Seok et al. was employed by preheated process (200 oC), showing PCE of 16.5%.[23] Im et al. demonstrated the use of spin-coated ZnO as an ETL material processed at approximately 150 °C, resulting in PCE values of 10.3%–15.6%.[24] However, ZnO is known to degrade the perovskite film.[25] Since now, most n-i-p type flexible solar cells have been fabricated using these inorganic ETLs with PCE of 10 % ~ 18% and most of them have a stability or hysteresis behaviour problem.[26-28] Therefore, a new ETL material that can be deposited at a low temperature, stable and that does not lead to hysteresis is desirable. A fullerene, C60, has high electron mobility (1.6 cm2 V−1 s−1) and conductivity (2.3 × 10−3 S cm−1, as obtained from field-effect transistors), making it an excellent candidate to replace the aforementioned ETL.[29-32]There are many examples of the use of C60 and its 3
derivatives as ETLs in p-i-n-type perovskite devices based on glass substrates.[33-38] However, C60 cannot be easily spin-coated in flexible perovskite solar cells with planar n-i-p structures. It has low solubility in many common organic solvents (we tested the low solubility of C60 and exhibited the result in this manuscript), which limits the choice of solvents from which it can be deposited, but sufficient solubility in many solvents to afford poor solvent resistance during the casting of subsequent layers, in part caused by poor adhesion to substrates. Wojciechowski et al. spin-coated a C60 ETL layer in an n-i-p-type device based on a glass substrate and obtained a stabilized PCE of 13.9%.[39] However, the C60 layer was dissolved by dimethylformamide (DMF) and chlorobenzene, which are commonly used in the depositions of perovskite and hole-transport layers, respectively. The dissolution of C60 resulted in the formation of pinholes, leading to shunting pathways with a concomitant decrease in device efficiency. Therefore, this ETL could not be used to fabricate devices with high reproducibility. One potential solution to this problem is enhancing the interaction and adhesion of C60 with the substrate. However, only one study has attempted to enhance the attraction between C60 molecules and flexible substrates. Yoo et al. employed a bilayer ETL in which C60 was evaporated on ethoxylated polyethylenimine (PEIE) to enhance the adhesion of C60 molecules to the PEIE/ITO–PEN layer, affording a PCE of 13.8% for the flexible substrate-based device.[40] However, the C60 molecules were still partially washed off during the spin-coating process perovskite solution. Wojciechowski et al. employed reactive C60 derivatives that were blended with C60 to increase the attractive interactions between C60 molecules and ultimately the solvent resistance. The resulting glass-substratebased devices exhibited stabilized PCEs of approximately 15%–16%.[41] However, a new ETL that strongly adheres to flexible substrates and does not degrade in common solvents is still needed.
4
To address this need, we attempted to develop a solvent-resistant ETL layer for flexible devices. Here we report a highly stable ETL developed to address this need, in which C 60 is directly treated with 0.08 wt% of poly(allylamine) PAA. As amines are known to react with C60 [42, 43], we heated the PAA-treated C60, and found that this lead to insolubilization of the ETL and sufficient adhesion to the ITO–PEN substrate, while causing the C60 molecules to be robustly fixed on the substrate. In addition, because amine-containing polymers are known to decrease the work function (WF) of the C60 layer,[44] an increased open-circuit voltage (VOC) was expected. Finally, we demonstrate a flexible planar perovskite solar cell with 15.2% PCE and negligible hysteresis. This cell also exhibits excellent flexibility with over 83% PCE retention after 600 cycles of bending. Results and discussion The solvent resistance of a pure C60 ETL was examined by measuring the ultraviolet– visible (UV–Vis) absorption of the films after washing them with a DMF:DMSO (4:1, v/v mixed solvent, which is often used to prepare the perovskite precursor solution. When the neat C60 washed with DMF:DMSO (4:1, v/v mixed solvent), a 50% decrease in the absorbance of C60 was observed, indicating that the C60 molecules did not strongly adhere to the substrate which is well matched with the reported results (Figure 1).[39] However, when 0.04 wt% PAA solution was deposited onto the C60 layer and annealed at 150 °C (see ESI for the detailed procedure), only a slightly smaller (46%) decrease in absorption was observed (Figure 2a). Thus, although PAA itself does not dissolve in the DMF:DMSO mixed solvent, 0.04 wt% PAA is evidently insufficient to insolubilize the C60 film.
5
Figure 1 Ultraviolet–visible (UV–Vis) absorption spectra of fullerene (C60) on FTO glass (black line) and after washing with dimethylformamide (DMF): dimethyl sulfoxide (DMSO) (4:1, v/v; red line). Inset: Top-down high-resolution scanning electron microscopy (HR-SEM) images of neat C60 on FTO substrates before and after washing with DMF:DMSO. On the other hand, no absorption decrease was observed when PAA solutions with concentrations of 0.08 wt%, 0.12 wt%, and 0.16 wt% were used, indicating a solventresistant film on the substrate (Figure 2b and S1). This is attributed to the reaction of the amine groups of PAA with sufficient C60 molecules to form a cross-linked film. We further investigated the films via high-resolution scanning electron microscopy (HR-SEM). The HR-SEM images showed inhomogeneity in the solvent-treated C60 film, whereas the film formed with 0.08 wt% PAA appeared homogeneous (Figure 2c; see Figure S2 for SEM images showing other PAA concentrations).
6
Figure 2 Ultraviolet–visible (UV–Vis) absorption spectra of fullerene C60– poly(allylamine) (PAA) layers on FTO glass (black line) and after washing with dimethylformamide (DMF): dimethyl sulfoxide (DMSO) (4:1, v/v; red line): (a) C 60– PAA (0.04 wt%), (b) C60–PAA (0.08 wt%). (c) Top-down high-resolution scanning electron microscopy (HR-SEM) images of C60–PAA (0.08 wt%) layers on FTO substrates before and after washing with DMF:DMSO. Based on the increased solvent resistance of the C 60 layer, we hypothesized that a reaction between C60 and PAA was occurring, as seen previously for fullerenes and other amines.[42] In addition to nucleophilic reactions, however, endergonic amine to fullerene electron transfer is possible,[45] presumably followed by hydrogen atom transfer between amine radical cations to form ammonium and iminium ions. We investigated the films via XPS using films prepared by spin-coating as done for the device to investigate the nature of the nitrogen species formed. X-ray photoelectron spectroscopy (XPS) indicates a change in the chemical environment of nitrogen in the films between neat PAA and C60-PAA films (Figure 3a), suggesting a chemical reaction of the polymer amine functional groups with the C60. For an FTO/PAA sample, the N 1s XPS feature was modelled as three peaks centered at binding energies (BEs) of 401.5, 400.5, and 399.7 eV (with the full width at half maximum (FWHM) of each constrained at 1.25–1.35 eV). Based on previous research,[45, 7
46]the lower BE peaks can likely be attributed to neutral free amine N, whereas the higher BE peaks are likely attributable to quaternary ammonium and hydrogen-bonded amine species. For the FTO/C60/PAA sample (0.08 wt% PAA), an additional strong N 1s peak centered at 399.1 eV was observed in addition to the features at BEs similar to those observed for the FTO/PAA sample. The BE of this new peak is consistent with a covalently bonded amine, rather than a cationic iminium or ammonium species, and is similar in BE to peaks seen for purified covalent C 60–amine compounds (399.1-400.1 eV). [43, 47]
These results support the hypothesis that a large portion of the nitrogen
atoms in PAA are covalently bound to the C 60 cage (low BE peak), with a smaller portion potentially interacting ionically (high BE peaks).
Figure 3 (a) X-ray photoelectron spectroscopy (XPS) N 1s peaks for PAA (0.08 wt%) and C60–PAA (0.08 wt%) films on FTO substrates. (b) FT-IR spectrum of (from top) 8
1) C60-conc. PAA, a material from a drop-cast C60-PAA film using a ~4 wt% PAA solution; 2) C60-dilute PAA, a material from a drop-cast C60-PAA film using a 0.08 wt% PAA solution; 3) PAA; and 4) C60. (c) LDI spectrum of material from a drop-cast C60-PAA film using a 0.08 wt% PAA solution. Thicker films using drop-casting of the C60 and PAA solutions (varied concentration) were also prepared and analysed by FT-IR, MS, and solid state NMR. The FT-IR showed a decrease in the strength of peaks at 1430 and 1180 cm -1 assigned to pure C60 and an increase in the intensity of the peaks at 2360 and 2330 cm -1 (Figure 3b). For the C60-PAA sample using a more concentrated PAA solution, there was also the appearance of additional peaks at 1460 and 1130 cm-1 (attributable to functionalized fullerene vibrational modes), 1030 cm-1 (in-plane C-H bending), 750 and 660 cm-1 (out-of-plane C-H bending). These observations are consistent with lowering the symmetry of the fullerene cage through nucleophilic reaction with the amine. Further evidence for covalent fullerene-PAA interactions come from mass spectrometry (Figure 3c): lower molecular weight fragments of C60-PAA were observed, with broad m/z distributions with a separation of approximately 700 m/z seen between the center of each broad distribution of peaks, a mass roughly equivalent to that of C60 at 720 (additional spectra are included in the Figure S3). We performed a solid state
13
C CP-MAS NMR experiment to further investigate
the C60-PAA interaction (Figure S4). During the 2 ms contact time employed, a magnetization transfer occurs from 1H to
13
C atoms. This magnetization transfer is
expected to occur to a large extent from the 1H rich area (PAA) to the fullerene moieties, [48] which could be ionically or covalently bound to the PAA or could be unreacted C60 in the immediate proximity (~1 nm) of the PAA. The peak at 143.3 ppm corresponds to the single peak reported for pure C60 (143-144 ppm) [43, 49] and is 9
attributable to unreacted C60 in the immediate proximity of the polymer. New peaks are seen at 164.7, 148.5, 136.4, and 131.3 ppm, arising presumably from covalently functionalized fullerenes where carbon atoms in the fullerene cage are no longer equivalent (additional experimental spectra (Figure S5) and discussion in ESI). The XPS, IR, MS, and NMR spectra all support a covalent reaction of C 60 and PAA, though XPS does not rule out a small amount of quaternary amine being present, potentially ionically bonding with a fullerene anion.
Figure 4 (a) Ultraviolet photoelectron spectroscopy (UPS) secondary electron edge cut-offs and valence peaks for C60 and C60–PAA (0.08 wt%) films on FTO substrates. (b) Effect of PAA on the electronic structures of the surfaces of C60 and C60–PAA (0.08 wt%) films, as deduced from the UPS secondary electron edge cut-offs and valence peaks. As discussed previously, PAA is, by analogy with other work on fullerene-amine systems, expected to decrease the WF of the C60 layer, [44] thereby potentially affecting the VOC value of devices. Thus, the effect of PAA on the WF was investigated using ultraviolet photoelectron spectroscopy (UPS). The introduction of PAA reduced the WF by up to ca. 1 eV [from 4.6 eV ± 0.1 eV for FTO/C60 to 3.7–3.9 eV ± 0.1 eV for FTO/C60/PAA (Figure 4a, S6, and S7)]. A comparison of shifts in the position of the lowest BE valence peak attributable to C 60, at least for low PAA 10
loading (0.08 wt% PAA solution), where this peak is still well defined, suggests that the majority of the total WF shift (0.8 eV) can be attributed to a valence band shift (0.5 eV) associated with n-doping of the C60 by the amine, at least at the surface of the films. This is consistent with previous reports on C 60–amine systems, [50, 51]with the Fermi level for the doped films being pinned close to the C 60 LUMO (Figure 4a and 4b).
Figure 5 (a) Deionized water and glycerol contact angles for bare FTO and C60 and C60–PAA (0.04 and 0.08 wt%) films on FTO substrates and the calculated surface energies. (b) Space-charge-limited current (SCLC) values of C60 and C60–PAA (0.04 wt% and 0.08 wt% PAA) electron-only devices. (c) Time-resolved photoluminescence (TRPL) decay of the perovskite on the bare glass, glass/C 60, and glass/C60–PAA (0.04 wt% and 0.08 wt%) electron-transport layers (ETLs). (d) Nyquist plots of the devices using C60 and C60-PAA 0.08 wt%. Next, we investigated the effects of small amounts of PAA on the surface energy 11
and the crystallinity of the perovskite Cs0.05(MA0.17FA0.83)0.95Pb(I0.83Br0.17). First, we conducted contact-angle measurements using deionized water and glycerol (Figure S8). As the amount of PAA increased, the contact angle decreased, indicating an increase in the surface energy (up to 63.6 mJ m−2; Figure 5a and Table S1). The increased hydrophilicity of the ETL surface improved the wettability by the perovskite solution. [52] Meanwhile, the change in the surface energy was not accompanied by a large change in crystallinity of the perovskite, as shown in Figure S9, which indicates no difference in the X-ray diffraction (XRD) patterns. In addition to the surface properties, the electronic properties of the ETL were examined. To determine the electron mobility in the vertical direction, i.e., the same direction as that of electron transport in solar-cell devices, we measured the electron-transport properties of each layer using the space-charge-limited current (SCLC) method (Figure 5b). We fabricated electron-only devices comprising FTO/C60/PAA/perovskite/LiF/Al (PAA concentration = 0 wt%, 0.04 wt%, 0.08 wt%, 0.12 wt%, and 0.16 wt%). As the concentration of PAA solution used increased from 0 wt% to 0.08 wt%, the electron mobility of the ETL slightly decreased; however, it was similar to that of the neat C60 ETL (8.0 × 10−3 cm2 V−1 s−1, 4.5 × 10−3 cm2 V−1 s−1, and 3.8 × 10−3 cm2 V−1 s−1). However, when the concentration of PAA was increased to 0.12 wt% and 0.16 wt%, the electron mobility decreased to 2.3 × 10 −5 and 1.2 × 10−5 cm2 V−1 s−1, respectively (Figure S10), which is consistent with the dilution of electron-transporting C60 with insulating PAA.
Table 1. Lifetimes of perovskites on different substrates based on photoluminescence decay data. τ1 [ns]a τ2 [ns]b Glass/perovskite 26.6 637.0 Glass/C60/perovskite 8.4 80.5 12
Glass/C60–PAA (0.08 wt%)/perovskite 17.0 162.0 −x/τ1 −x/τ2 Biexponential decay fits of y = y0 + A1e + A2e were used to extract the lifetimes. a,b
To evaluate the electron extraction from the perovskite layer, we performed timeresolved photoluminescence (TRPL) measurements of the C60–PAA ETLs. Mixedcation perovskites were deposited on bare glass/C 60–PAA and the perovskite side was illuminated with 474 nm light. The TRPL traces of all C 60–PAA ETLs are shown in Figure 5c, and the decay lifetimes extracted from the data are listed in Tables 1 and S2. Compared to the neat perovskite on the bare glass, the lifetimes of the samples with neat C60 and C60–PAA (0.04 wt%) were significantly shorter for both fast and slow decay. In contrast, the lifetimes of C 60–PAA (0.08 wt%) were slightly increased (τ1 = 17.0 ns and τ2 = 162.0 ns) compared to the sample with neat C 60, indicating slower but still efficient electron extraction from the perovskite layer. When the PAA concentration was increased to 0.12 wt% and 0.16 wt%, the lifetimes significantly increased, indicating poor electron extraction from the perovskite layer (Figure S11).
Figure 6 Histogram of the parameters of planar FTO-based devices with C60 and C60– PAA (0.08 wt%) ETLs.
13
To further understand the back recombination of electrons in the device, an electrochemical impedance spectroscopy (EIS) measurements were performed in the frequency range of 1 MHz to 1 Hz under bias voltage of VOC under the dark condition (Figure 5d). The EIS data was fitted using the equivalent circuit model which is the inset of Figure 5d. Here, R1 and R2 are related to the resistance originating from the electrode and the electrode/HTL interface, respectively. Since the device structure and HTL are the same in each sample, the recombination resistance (R3) can be varied only from the ETL/Perovskite interface. The recombination resistance was obtained from the low frequency elements of the EIS and the R3 values of C 60 and C60-PAA ETL were 3.72 and 6.09 kΩ, respectively. This result indicates that recombination of electrons in devices employed the C60-PAA ETL is more efficiently suppressed than devices using the neat C60 ETL. To evaluate our ETL, we fabricated a flat device on the FTO substrate (Figure 6). Although the best efficiency obtained by the device employing a C60 ETL was similar to that of the device employing a C 60–PAA ETL, the C60-based devices exhibited wide ranges of the JSC, VOC, FF, and PCE values, which is attributed to the low solvent resistance of the bare C 60 layer. However, the devices with C60–PAA ETLs exhibited a more reproducible PCE due to greatly improved JSC, FF, and VOC values. Figure S12 shows the J–V curves of the best cell (external quantum efficiency measurement of the device in Figure S13). This cell showed a JSC value of 21.5 mA cm−2, a VOC value of 1.07 V, and an FF value of 77.0%, yielding a PCE of 17.7% with an MPP efficiency of 15.7%. This device also exhibited long-term stability, maintaining more than 90% of its initial efficiency for over 1600 h (Figure S14). Based on these promising results, we also fabricated flexible device on an ITO– PEN substrate using the C60–PAA (0.08 wt%) ETL and compared its performance with the device fabricated with a neat C60 ETL (inset of Figure. 8a). 14
Figure 7 Current density−voltage (J-V) curves of the flexible devices employing C 60 and C60-PAA 0.08 wt% In general, due to the flexibility of the flexible substrate, it is well known that the stability of the semiconductor layer constituting the device deteriorates and affects the performance of the device.[53]
Surprisingly, the device with a neat C60 ETL showed
remarkably low efficiency and reproducibility on a flexible substrate. However, the device with C60-PAA(0.08 wt%) ETL showed impressive stability and efficiency (Figure 7). Representative current density (J)–voltage (V) curves are shown in Figure 8 and photovoltaic parameters are summarized in Table 2. Due to the high solvent resistance and adhesion of the C60–PAA layer, the device with the C60–PAA (0.08 wt%) ETL exhibited higher and more reproducible efficiency than the device with the C 60 ETL. Meanwhile, the low solvent resistance and weak adhesion of C 60 on the flexible substrate may have provided pathways for recombination and hindered efficient electron extraction from the perovskite to the cathode. Consequently, the short-circuit current (JSC), VOC, and fill factor (FF) of the device were decreased comparing the 15
C60–PAA (0.08 wt%) ETL. On the other hand, due to the high solvent resistance and adhesion of C60-PAA ETL, electrons were effectively extracted from the perovskite with less recombination result in increased JSC and FF. In addition, the increased VOC value can be attributed to the PAA-induced shift in the WF and the improved electron extraction. As shown in Figure 8a, the best flexible device employing the C 60–PAA (0.08 wt%) ETL exhibited a PCE of 15.2% without hysteresis and a maximum power output (MPP) efficiency of 15.1% (Figure 8b).
Figure 8 (a) Forward and reverse current density (J)–voltage (V) curves of the best flexible device employing the C60–PAA (0.08 wt%) ETL. (b) Stabilized maximum power output (MPP) measurement of the best flexible device. (c) Normalized power 16
conversion efficiencies (PCEs) of flexible devices employing C60 and C60–PAA (0.08 wt%) ETLs as functions of bending cycle (radius of bending = 10 mm). Table 2 Summary of the parameters of flexible perovskite solar cells obtained from the best devices employing C60 and C60–PAA (0.08 wt%) as ETLs. Measurements are performed under AM 1.5 solar illumination. The scan rate is 0.06 V s −1.
C60–PAA (0.08 wt%) reverse C60–PAA (0.08 wt%) forward C60 reverse C60 forward
JSC (mA/cm2) 20.7 20.7 14.6 14.7
VOC (V) 1.04 1.04 1.00 0.99
FF (%) 70.5 70.5 66.8 64.4
PCE (%) 15.2 15.2 9.8 9.4
These values were considerably greater than those obtained from the flexible device employing the C60 ETL (PCE = 9.8% and 9.4% for the reverse and forward scans, respectively; MPP efficiency = 9.6%). In addition, the flexible device employing the C60–PAA (0.08 wt%) ETL also maintained more than 90% of its initial efficiency for over 400 h, which is similar with FTO based device (Figure S15). To confirm the mechanical flexibility of the device under bending stress, we performed a bend test, as shown in Figure 8c. The flexible device was bent with a curvature of 10 mm in one bending cycle (insets of Figure 8c). The device retained 83% of its original efficiency after 600 binding cycles compared to 65% retention for the device with the C60 ETL. The decrease in PCE for the C60-based device is due to significant decreases in both JSC and FF values (Figure S16) and indicate that the C60 ETL is unstable to bending. On the other hand, the device employing C 60–PAA showed less significant decreases in JSC and FF upon bending. The JSC value remained above 96% after 600 cycles of bending, indicating that the C 60–PAA ETL provided a high degree of stability and flexibility.
17
Figure 9 Transient photocurrent measurement of the perovskite device employing TiO2 and the C60–PAA (0.08 wt%) ETL To investigate the reason for this, we measured the transient photocurrent of conventional TiO2 and C60-PAA 0.08 wt% devices (Figure 9). Upon illumination, the conventional TiO2 device took almost 200 μs to stabilize, while the device employing a C60-PAA ETL took less than 30 μs to stabilize. In addition, after illumination terminated, the current of the C60/PAA 0.08 wt% ETL decreased abruptly compared to that of conventional TiO2, which was indicative of superior electron-extraction and collection ability of the C60-PAA ETL. Conclusion In conclusion, we successfully developed and demonstrated low-temperatureprocessed C60–PAA ETLs for use in flexible perovskite solar cells. The composition of these ETLs was optimized by varying the concentration of PAA in the casting solution and the reaction was confirmed from XPS, FT-IR, Mass, and NMR. After processing, the C60–PAA film exhibited excellent solvent resistance and C 60 adhesion while maintaining the electrical properties of C 60. On the flexible substrate, the resulting ETL yielded a PCE of 15.2% and an MPP efficiency of 15.1% without hysteresis, making this ETL one of the best reported to date for flexible planar 18
perovskite solar cells which employed organic ETL. In addition, the flexible device exhibited high stability during bending cycles. After 600 cycles of bending, the device retained over 83% of its original efficiency and the JSC value remained above 96%. The optimized C60–PAA ETL also exhibited excellent solvent resistance and adhesion with good electrical properties, resulting in superior device reproducibility. The results of this study provide insights into the design strategies for efficient charge extraction layers for use in flexible perovskite devices. Supporting Information. Supplementary material related to this article can be found online. ACKNOWLEDGMENT The authors wish to express their gratitude to Dr. Nakita K. Noel and Dr. Maximilian T. Hörantner for training and support in device fabrication, lab instrumentation, and data processing and "The authors also acknowledge the aid of David Botswick at the Georgia Tech Mass Spectrometry Facility.". The authors gratefully acknowledge the support of the United States AFOSR (grant # FA9550-15-1-0115) and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (Code No.2015R1A2A1A10054230). REFERENCES [1] J. De Roo, M. Ibanez, P. Geiregat, G. Nedelcu, W. Walravens, J. Maes, J.C. Martins, I. Van Driessche, M.V. Koyalenko, Z. Hens, ACS Nano, 10 (2016) 2071-2081. [2] N.G. Park, Mater. Today, 18 (2015) 65-72. [3] F.X. Xie, D. Zhang, H.M. Su, X.G. Ren, K.S. Wong, M. Gratzel, W.C.H. Choy, ACS Nano, 9 (2015) 639-646. 19
[4] F.Z. Lv, C.X. Gao, H.A. Zhou, P. Zhang, K. Mi, X.X. Liu, ACS Appl. Mater. Interfaces, 8 (2016) 18985-18990. [5] S. Liu, F. Zheng, N.Z. Koocher, H. Takenaka, F.G. Wang, A.M. Rappe, J. Phys. Chem. Lett., 6 (2015) 693-699. [6] S.D. Stranks, G.E. Eperon, G. Grancini, C. Menelaou, M.J.P. Alcocer, T. Leijtens, L.M. Herz, A. Petrozza, H.J. Snaith, Science, 342 (2013) 341-344. [7] A. Miyata, A. Mitioglu, P. Plochocka, O. Portugall, J.T.W. Wang, S.D. Stranks, H.J. Snaith, R.J. Nicholas, Nat. Phys., 11 (2015) 582-587. [8] K. Galkowski, A. Mitioglu, A. Miyata, P. Plochocka, O. Portugall, G.E. Eperon, J.T.W. Wang, T. Stergiopoulos, S.D. Stranks, H.J. Snaith, R.J. Nicholas, Energy Environ. Sci., 9 (2016) 962-970. [9] M.M. Lee, J. Teuscher, T. Miyasaka, T.N. Murakami, H.J. Snaith, Science, 338 (2012) 643-647. [10] H.S. Kim, C.R. Lee, J.H. Im, K.B. Lee, T. Moehl, A. Marchioro, S.J. Moon, R. Humphry-Baker, J.H. Yum, J.E. Moser, M. Gratzel, N.G. Park, Sci. Rep., 2 (2012) 591. [11] S. Song, G. Kang, L. Pyeon, C. Lim, G.Y. Lee, T. Park, J. Choi, ACS Energy Lett., 2 (2017) 2667-2673. [12] A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J .Am. Chem. Soc., 131 (2009) 60506051. [13] A. Abate, J.P. Correa-Baena, M. Saliba, M.S. Su'ait, F. Bella, Chemistry, 24 (2018) 3083-3100. [14] S. You, S. Bi, J. Huang, Q. Jia, Y. Yuan, Y. Xia, Z. Xiao, Z. Sun, J. Liu, S. Sun, Z. Zhao, Chemistry, 23 (2017) 18140-18145. [15] F. Di Giacomo, A. Fakharuddin, R. Jose, T.M. Brown, Energy Environ. Sci., 9 (2016) 3007-3035. 20
[16] V. Zardetto, G. Mincuzzi, F. De Rossi, F. Di Giacomo, A. Reale, A. Di Carlo, T.M. Brown, Applied Energy, 113 (2014) 1155-1161. [17] J. Burschka, N. Pellet, S.J. Moon, R. Humphry-Baker, P. Gao, M.K. Nazeeruddin, M. Gratzel, Nature, 499 (2013) 316-319. [18] S. Fabiano, S. Braun, X.J. Liu, E. Weverberghs, P. Gerbaux, M. Fahlman, M. Berggren, X. Crispin, Adv. Mater., 26 (2014) 6000-6006. [19] Z. Song, G. Tong, H. Li, G. Li, S. Ma, S. Yu, Q. Liu, Y. Jiang, Nanotechnology, 29 (2018) 025401. [20] Z.L. Chen, G. Yang, X.L. Zheng, H.W. Lei, C. Chen, J.J. Ma, H. Wang, G.J. Fang, Journal of Power Sources, 351 (2017) 123-129. [21] H. Tao, Z.B. Ma, G. Yang, H.N. Wang, H. Long, H.Y. Zhao, P.L. Qin, G.J. Fang, Applied Surface Science, 434 (2018) 1336-1343. [22] G. Yang, C. Chen, F. Yao, Z. Chen, Q. Zhang, X. Zheng, J. Ma, H. Lei, P. Qin, L. Xiong, W. Ke, G. Li, Y. Yan, G. Fang, Adv. Mater., 30 (2018) 1706023. [23] E.H. Anaraki, A. Kermanpur, L. Steier, K. Domanski, T. Matsui, W. Tress, M. Saliba, A. Abate, M. Gratzel, A. Hagfeldt, J.P. Correa-Baena, Energy Environ. Sci., 9 (2016) 31283134. [24] S.S. Shin, W.S. Yang, J.H. Noh, J.H. Suk, N.J. Jeon, J.H. Park, J.S. Kim, W.M. Seong, S.I. Seok, Nat. Commun., 6 (2015) 8. [25] J.H. Heo, M.H. Lee, H.J. Han, B.R. Patil, J.S. Yu, S.H. Im, J. Mater. Chem. A, 4 (2016) 1572-1578. [26] J.L. Yang, B.D. Siempelkamp, E. Mosconi, F. De Angelis, T.L. Kelly, Chem. Mater., 27 (2015) 4229-4236. [27] C.L. Wang, D.W. Zhao, Y. Yu, N. Shrestha, C.R. Grice, W.Q. Liao, A.J. Cimaroli, J. Chen, R.J. Ellingson, X.Z. Zhao, Y.F. Yan, Nano Energy, 35 (2017) 223-232. 21
[28] D. Yang, R.X. Yang, X.D. Ren, X.J. Zhu, Z. Yang, C. Li, S.Z. Liu, Adv. Mater., 28 (2016) 5206-5213. [29] S.S. Shin, W.S. Yang, E.J. Yeom, S.J. Lee, N.J. Jeon, Y.C. Joo, I.J. Park, J.H. Noh, S.I. Seok, J. Phys. Chem. Lett., 7 (2016) 1845-1851. [30] Y.J. Fang, C. Bi, D. Wang, J.S. Huang, ACS Energy Lett., 2 (2017) 782-794. [31] Y.H. Shao, Z.G. Xiao, C. Bi, Y.B. Yuan, J.S. Huang, Nat. Commun., 5 (2014) 5784. [32] Q. Wang, Y.C. Shao, Q.F. Dong, Z.G. Xiao, Y.B. Yuan, J.S. Huang, Energy Environ. Sci., 7 (2014) 2359-2365. [33] J. Xu, A. Buin, A.H. Ip, W. Li, O. Voznyy, R. Comin, M.J. Yuan, S. Jeon, Z.J. Ning, J.J. McDowell, P. Kanjanaboos, J.P. Sun, X.Z. Lan, L.N. Quan, D.H. Kim, I.G. Hill, P. Maksymovych, E.H. Sargent, Nat. Commun., 6 (2015) 7081. [34] J. Seo, S. Park, Y.C. Kim, N.J. Jeon, J.H. Noh, S.C. Yoon, S.I. Sang, Energy Environ. Sci., 7 (2014) 2642-2646. [35] J.Y. Jeng, K.C. Chen, T.Y. Chiang, P.Y. Lin, T.D. Tsai, Y.C. Chang, T.F. Guo, P. Chen, T.C. Wen, Y.J. Hsu, Adv. Mater., 26 (2014) 4107-4113. [36] M.F. Lo, Z.Q. Guan, T.W. Ng, C.Y. Chan, C.S. Lee, Adv. Funct. Mater., 25 (2015) 1213-1218. [37] W. Qiu, J.P. Bastos, S. Dasgupta, T. Merckx, I. Cardinaletti, M.V.C. Jenart, C.B. Nielsen, R. Gehlhaar, J. Poortmans, P. Heremans, I. McCulloch, D. Cheyns, J. Mater. Chem. A, 5 (2017) 2466-2472. [38] Y. Bai, Q.F. Dong, Y.C. Shao, Y.H. Deng, Q. Wang, L. Shen, D. Wang, W. Wei, J.S. Huang, Nat. Commun., 7 (2016) 7081. [39] M. Li, Y.H. Chao, T. Kang, Z.K. Wang, Y.G. Yang, S.L. Feng, Y. Hu, X.Y. Gao, L.S. Liao, C.S. Hsu, J. Mater. Chem. A, 4 (2016) 15088-15094.
22
[40] K. Wojciechowski, T. Leijtens, S. Siprova, C. Schlueter, M.T. Horantner, J.T.W. Wang, C.Z. Li, A.K.Y. Jen, T.L. Lee, H.J. Snaith, J. Phys. Chem. Lett., 6 (2015) 2399-2405. [41] J. Ha, H. Kim, H. Lee, K.G. Lim, T.W. Lee, S. Yoo, Sol. Energy Mater. Sol. Cells, 161 (2017) 338-346. [42] A. Hirsch, M. Brettreich, Fullerenes: chemistry and reactions, Wiley-VCH Verlag GmbH & Co. KGaA2004, pp. 73-99. [43] I.J. Ramirez-Calera, V. Meza-Laguna, T.Y. Gromovoy, M.I. Chavez-Uribe, V.A. Basiuk, E.V. Basiuk, Applied Surface Science, 328 (2015) 45-62. [44] K. Wojciechowski, I. Ramirez, T. Gorisse, O. Dautel, R. Dasari, N. Sakai, J.M. Hardigree, S. Song, S. Marder, M. Riede, G. Wantz, H.J. Snaith, ACS Energy Lett., 1 (2016) 648-653. [45] B. Russ, M.J. Robb, B.C. Popere, E.E. Perry, C.K. Mai, S.L. Fronk, S.N. Patel, T.E. Mates, G.C. Bazan, J.J. Urban, M.L. Chabinyc, C.J. Hawker, R.A. Segalman, Chem. Sci., 7 (2016) 1914-1919. [46] N. Graf, E. Yegen, T. Gross, A. Lippitz, W. Weigel, S. Krakert, A. Terfort, W.E.S. Unger, Surf. Sci., 603 (2009) 2849-2860. [47] J. Chastain, R.C. King, J. Moulder, Handbook of X-ray photoelectron spectroscopy: a reference book of standard spectra for identification and interpretation of XPS data, Physical Electronics Division, Perkin-Elmer Corporation Eden Prairie, Minnesota1992, pp. 42-43. [48] J.H. Waller, D.G. Bucknall, R.A. Register, H.W. Beckham, J. Leisen, K. Campbell, Polymer, 50 (2009) 4199-4204. [49] A.M. Panich, I. Felner, A.I. Shames, S. Goren, P.K. Ummat, W.R. Datars, Solid State Commun., 129 (2004) 81-84. [50] Y.H. Zhou, C. Fuentes-Hernandez, J. Shim, J. Meyer, A.J. Giordano, H. Li, P. Winget, T. Papadopoulos, H. Cheun, J. Kim, M. Fenoll, A. Dindar, W. Haske, E. Najafabadi, T.M. 23
Khan, H. Sojoudi, S. Barlow, S. Graham, J.L. Bredas, S.R. Marder, A. Kahn, B. Kippelen, Science, 336 (2012) 327-332. [51] Y.B. Sun, C.Y. Cao, P.P. Huang, S.L. Yang, W.G. Song, RSC Adv., 5 (2015) 8608286087. [52] Z.R. Liang, S.H. Zhang, X.Q. Xu, N. Wang, J.X. Wang, X. Wang, Z.N. Bi, G. Xu, N.Y. Yuan, J.N. Ding, RSC Adv., 5 (2015) 60562-60569. [53] M. Sakai, Y. Yamazaki, S. Yamaguchi, J. Hayashi, K. Kudo, Phys. Status Solidi A, 211 (2014) 795-799. Author information
Seulki Song obtained his Ph. D degree from the Department of Chemical Engineering at POSTECH, South Korea, under the supervision of Prof. Taiho Park. His research interests are charge transport at the interface of organic-inorganic semiconducting materials and the physics of high efficiency perovskite solar cells.
Rebecca Hill received her Ph.D. from the Department of Chemistry and Biochemistry at Georgia Institute of Technology, USA, under the supervision of Prof. Seth Marder. Her doctoral research focused on surface sensitization and surface modification at organic-inorganic interfaces in next-generation photovoltaics. She is continuing her career as a process engineer in chemical and mechanical planarization at Intel’s Portland Technology Development group. .
24
Kyoungwon Choi is a Ph.D student in Chemical Engineering Department in POSTECH, South Korea under the supervision of Prof. Taiho Park since 2016. His research interests is the charge transport study in organic-inorganic hybrid solar cell and optoelectronic devices.
Konrad Wojciechowski obtained his PhD degree from the University of Oxford, under the supervision of prof. Henry Snaith. His research has primarily focused on understanding the fundamental principles of perovskite device operation and improvement of a solar cell performance. In 2016 Konrad joined Saule Technologies as a Scientific Director, leading the work on perovskite PV commercialisation.
Steve Barlow received a DPhil degree in chemistry from the University of Oxford in 1996. After postdoctoral work with Seth Marder at the California Institute of Technology, he returned to Oxford as a temporary lecturer from 1998–2001. In 2001 he rejoined Seth Marder at the University of Arizona and is now a Principal Research Scientist in the School of Chemistry and Biochemistry at the Georgia Institute of Technology. His research interests are centered on the relationships between the structures of organic and organometallic compounds and their electronic and optical properties.
25
Johannes Leisen obtained his PhD in 1994 at the Max Planck Institute for Polymer Research in Mainz, Germany. After a two year stay at the Fraunhofer Institute for Biomedical Engineering in St. Ingbert, Germany, he moved to the Georgia Institute of Technology where he conducted research with Prof. Haskell Beckham. After 20 years at Georgia Tech Dr. Leisen now holds the rank of Principal Scientist; he is associate manager of the Georgia Tech NMR center and manager of the Magnetic Resonance Core facility. His broad expertise include applications of solid-state NMR for material sciences and the use of MRI techniques in engineering.
Henry J. Snaith is a Professor of Physics at Oxford University, head of the Photovoltaic and Optoelectronic Device Laboratory, and co-founder of Oxford Photovoltaics Ltd. He has received a number of early career research awards including the Institute of Physics Patterson Medal and Prize (2012) the Materials Research Society Outstanding Young Investigator Award (2014) and the EU-40 materials Prize 2016. In addition, he was named one of “Natures Ten” people who mattered most in 2013, and assessed as being the world’s 2nd most influential scientific mind in 2016, based on publication citations, and he was elected as a fellow of the Royal Society in 2015.
Seth R. Marder obtained his Ph.D. from the University of Wisconsin-Madison in 1985. He completed his postdoctoral research at the University of Oxford and at the Jet Propulsion Laboratory, California Institute of Technology. He joined the Georgia Institute of Technology in 2003 where he is currently a Regents’ Professor of Chemistry and Materials Science and Engineering (courtesy) and the Georgia Power Chair in Energy Efficiency. His research interests are in the development of materials for nonlinear optics, applications of organic dyes for photonic, display, electronic and medical applications, and organometallic chemistry.
26
Taiho Park is a professor of in the Department of Chemical Engineering at POSTECH, South Korea, since 2007. Prof. Park received his PhD from the University of Cambridge, UK under the supervision of Prof. Andrew B. Holmes, then worked as a post-doctoral researcher under Professor Steven C. Zimmerman at the University of Illinois (Urbana-Champaign), USA. Current research interests include the material properties and device functions of optoelectronic devices.
Highlights A modified fullerene-based electron transport layer is proposed for stable flexible perovskite device. A poly(allylamine) (PAA; 0.08 wt%) solution was deposited on a C60 layer to form a C60-PAA. C60–PAA electron-transport layer (ETL) is insoluble in perovskite precursor solution. Flexible planar perovskite solar cell with the C60–PAA ETL exhibited excellent properties with 83% efficiency retention.
Graphical abstract
27