Citric Acid Modulated Growth of Oriented Lead Perovskite Crystals for

Jun 23, 2017 - Department of Chemistry, The University of Tokyo 7-3-1 Hongo, Bunkyo-ku, ... the PCE improved from 14.52% (one sun) to 20.4% (0.087 sun...
0 downloads 0 Views 5MB Size
Download PDF
Article pubs.acs.org/JACS

Citric Acid Modulated Growth of Oriented Lead Perovskite Crystals for Efficient Solar Cells Yunlong Guo,*,†,‡ Wataru Sato,‡ Kazutaka Shoyama,‡ Henry Halim,‡ Yuki Itabashi,‡ Rui Shang,‡ and Eiichi Nakamura*,‡ †

Institute of Chemistry, Chinese Academy of Science, Beijing, 100190, PR China Department of Chemistry, The University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

‡

W Web-Enhanced Feature * S Supporting Information *

ABSTRACT: Solar cells made of lead perovskite crystals have attracted much attention for their high performance, but far less attention as a subject of crystal engineering. Here, we report that citric acid (CA) and chloride anion, working together, modulate crystal growth of CH3NH3PbI3, producing sub-mm-sized cuboid crystalsa morphology more suitable for close packing in a thin film than the commonly observed elongated dodecahedral morphology. By using a 15 wt % CA-doped precursor solution, we formed a single layer of large, flat, and oriented cuboid crystals with minimum crystal domain boundaries and maximum contact with neighboring layers, and fabricated an archetypal inverted-structured device of 4 mm2 area, which showed, reproducibly and with little hysteresis, 16.75% power conversion efficiency (PCE), 26% higher than the PCE obtained for a polycrystalline film made without CA doping. Under weaker irradiation of a 1 cm2 device, the PCE improved from 14.52% (one sun) to 20.4% (0.087 suns). Under illumination with white light emitting diode, a 10 wt % CA-doped device showed PCE of 28.1%, suggesting an advantage of PVK-SCs for indoor applications. Further studies on crystal growth modulation will be beneficial for manufacturing efficient and stable lead perovskite solar cells.

■

PCE under one sun (100 mW/cm2) and 20.4% PCE under 0.087 suns (equivalent of rainy day sunlight). Interestingly, the CA-doped perovskite device was found to show 28.1% PCE under illumination with white light emitting diode (LED, 1000 lx), suggesting an advantage of PVK-SCs for indoor applications. Bulk solution experiments indicated that the chloride anion in the precursor mixture is responsible for the cuboid crystal growth (Figure 1c), and CA promotes the growth of large crystals of MA+PbI3−. In the film, CA acted as a solid matrix for growth of large, flat, and oriented crystals. Combined with a chemical mechanism of PVK crystal formation,31 the data illustrate a hitherto unnoticed potential of molecular modulation of PVK crystal growth and its beneficial effects on PVK SC fabrication.

INTRODUCTION Molecular modulation of morphology and habit is a fundamental issue in crystal engineering but has received limited attention in research on solar cells (SCs) using lead perovskite (such as CH3NH3PbI3 or MA+PbI3−; PVK).1−15 Here, a film consisting of a single layer of oriented crystals with minimum crystal domain boundaries and maximum contact with neighboring layers (Figure 1a, left bottom) is required for effective light capture, carrier transport, and device stability.16−22 Although the PVK crystal lattice being tetragonal at room temperature (cubic at > ca. 55 °C), cuboid (Figure 1c) is a morphology most suited for such crystal disposition. In fact, cuboid morphology has a smaller surface area than elongated dodecahedral (ED) morphology (Figures 1d and 2b) that has been commonly observed for PVK crystals.23−30 We describe below that SC fabrication (Figure 1a) by the use of a precursor mixture of CH3NH3+I− (MAI), PbI2 (PI), and PbCl2 (PCl) doped with a small amount of citric acid (CA) forms a single layer of μm-sized cuboid crystals with great size uniformity that has the (100) of the cubic lattice face parallel to the substrate, maintains a broad contact area with the neighboring layers, and can effectively transport carriers to them. A 4 mm2 CA-doped PVK SC device using a simple archetypal device configuration (Figure 1a) composed of a 470-nm-thick film shows 16.75% power conversion efficiency (PCE) with reproducibility of ±0.4% and with very small hysteresis. A reference device consisting of a polycrystalline PVK film obtained without CA doping showed 12.3% PCE. A 1 cm2 device showed 14.52% © 2017 American Chemical Society

■

RESULTS AND DISCUSSION Molecular Modulation of Crystal Habit in Solution. Inspired by our recent studies on the acceleration of PVK formation by zwitterionic sulfamic acid (NH3SO3),32 we conjectured that CA acts as a modulator of PVK crystal growth, and analyzed carefully the morphology of crystals obtained from bulk solution by low-landing voltage scanning electron microscopy (LV-SEM). With this instrument, electrons penetrate shallowly into the surface of organic matter Received: April 17, 2017 Published: June 23, 2017 9598

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604

Article

Journal of the American Chemical Society

without sample charging, and hence we can obtain sub-nm surface information without metal coating. Crystals of PVK (pure MA+PbI3−) have been generally recorded in the ED form;24 we were surprised to find that reactions following a known procedure using a 1:1 molar mixture of MAI and PI (1 M for each) in γ-butyrolactone (GBL) (Figure S1, Supporting Information),25 in our hand, produced mainly stacked small cuboid crystals (16% PCE (Figure 3j): glass/ITO/PEDOT:PSS/15 wt % CA-PVK (470 nm)/PCBM/PEIE/Ag (PCBM = [6,6]-phenyl-C61butyric acid methyl ester; PEIE = polyethylenimine ethoxylated). Here, we see that the crystals contact very well both the top PCBM layer and the bottom PEDOT:PSS layer. The inverse trapezoidal crystal morphology suggests that the crystals have grown from the top surface down to the substrate side,

Figure 1. Molecular modulation of PVK crystal growth for fabrication of SC devices. (a) A single layer of oriented cuboid MA+PbI3− crystal formation by CA and chloride ion doping. (b) Polycrystalline PVK film formation without doping. (c) Cuboid crystal growth by inhibition of rhombic facets of ED morphology by chloride ion. (d) ED crystal growth. For the sake of simplicity, we used cubic-crystalbased indices for the tetragonal crystals here because the lattice parameters of tetragonal PVK crystals at room temperature are close to the cubic values and experimentally indistinguishable for lower angle diffraction peaks. The tetragonal lattice changes to a cubic lattice at > ca. 55 °C.

Figure 2. LV-SEM images showing the morphology of PVK crystals obtained in GBL. (a, b) Mixture of stacked cuboid and isolated ED crystals from 1:1 MAI and PI. The crystals were isolated by filtration and quickly washed with dry methylene chloride before SEM analysis. (c, d) Small cuboid crystals from a 4:1:1 MAI, PI, and PCl in GBL. (e, f) Large cuboid crystals from a 4:1:1 MAI, PI, and PCl containing 15 wt % CA, where we see residual CA as an oil under long-term SEM observation. Wet CA quickly etched the (100) face of the cuboid crystals, as shown in (f). A 5 kV landing voltage was used in these experiments and 1 kV in all others. Large format SEM pictures in Supporting Information (Figure S12).

9599

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604

Article

Journal of the American Chemical Society

XRD signal (2θ ≈ 15.0°) intensity maximized at x = 6−15 and dropped sharply at x = 20. Most notably from a device perspective, the CA doping significantly enhanced the efficiency of hole extraction from photoexcited PVK by the PEDOT:PSS layer,35,36 as shown by the photoluminescence (PL) quenching study illustrated in Figure 3k (red line) and Figure 3l. Thus, 6− 20 wt % CA doping resulted in a dramatic decrease of PL of PVK in a manner complementary to the increase of the XRD signal intensity (blue line). These data agree very well with the broad contact between crystals and the PEDOT:PSS layer seen in Figure 3j. 2-D XRD analysis of CA-doped PVK films on ITO/ PEDOT:PSS (Figure 4) illustrated significant effects on the CA doping and the thermal annealing after spin coating of the 120-nm-thick PVK film. The top row of the reciprocal space maps (RSMs) of the XRD signals in Figure 4a−d for 0 wt % CA-doped series illustrates that an as-spin-coated film shows little sign of crystal lattice orientation, but an annealed film is well oriented (right). The second row for 15 wt % CA-doped film illustrates a high level of crystal orientation already before annealing (left), and, after annealing at 100 °C, shows a prominent orientation of the (100) face parallel to the substrate (right). These effects are more clearly illustrated by signal intensities integrated along ϕ for various diffraction peaks (Figure 4b). For example, the annealing at 100 °C intensifies the (100) peak for both the 0 and 15 wt % CA-PVK films (Figure 4b, top row, red line), as opposed to the virtual lack of this peak before annealing (blue line). It is therefore shown also by the small full-width-at-half-maximum (fwhm) values in Figure 4e−n that the cuboid crystals we saw in Figure 3f (and j) produced by annealing 15 wt % CA-doped film have their (100) face very well oriented and parallel to the substrate. Cuboid PVK Crystals for Efficient Solar Cells. Remarkably, in the above-mentioned visual, XRD and PL trends were reflected very well in the performance of invertedstructured SC devices of a simple archetypal configuration: glass/ITO/PEDOT:PSS/x wt % CA-PVK (120 nm thick)/ PCBM/PEIE/Ag. As shown in Figure 5a and b, the PCE maximized at 15 wt % doping (confirmed among 14, 15, and 16 wt % experiments examined under slightly different conditions). The PCE value of 7.5% PCE for x = 0 increased to 14.0% at x = 15, and decreased to 9.2% at x = 20. We found that Voc and surface coverage by CA-PVK show no correlation (in fact, Voc increased with a film of lower coverage for yet unknown reasons; Figure S7). The same devices but now with 470 nm thickness and an active area of 4 mm2 exhibited a PCE value of 16.75% with little hysteresis (Figure 5c). These data are respectable values in light of the simple device configuration, and the 11% void space is filled by CA (Figure 3j). For 25 devices made from five different batches of the 4:1:1 precursor solution, the PCE data reproducibly fell between 15.5% and 17% with an average value of 16.3% (Figure S8a, Supporting Information). The external quantum efficiency (EQE; Figure 5d) was ≥80% from 450 to 760 nm. A device made under the same conditions without CA doping gave a polycrystalline film and PCE of 13.24%. The observed improvement by 27% from 13.24% to 16.75% is the sum of the improvement of all the device parameters, e.g., Jsc increased from 20.56 to 21.82 mA/cm2, Voc from 0.89 to 0.93 V and FF from 72.1% to 82.7%. Further increase of the thickness of the 15 wt % CA-PVK film over 470 nm showed little improvement

Figure 3. Characterization of CA-doped PVK films. The films were prepared from a 25 wt % CA-doped 4:1:1 precursor solution at a spincoating speed of 3000 rpm and have a thickness of ca. 120 nm. LVSEM images of x wt % CA-PVK film, (a) 0 wt %, (b) 1 wt %, (c) 3 wt %, (d) 6 wt %, (e) 10 wt %, (f) 15 wt %, (g) 20 wt %. (h, i) LV-SEM image of 15 wt % CA-doped PVK with a thickness of 470 nm. (j) The cross-section image of a solar cell device. (k) CA-doping dependence of PL and peak height of XRD (110) signal. (l) CA-dependent change of PL from the PVK film on PEDOT:PSS. Large format SEM pictures in Supporting Information (Figure S13).

probably because solvent loss occurred from the surface during spin coating. Another intriguing feature seen in both Figure 3i and j (red circle) is the dark matter, CA, filling in the space between PVK crystals, over which we see (Figure 3j) the top Ag and PCBM layers are laid. We consider that the added CA (bp 310 °C) acted as a solid matrix for PVK crystal growth, but this insulating material exerts little effect on device performance. The presence of CA in the film was confirmed by attenuated total reflectance (ATR) FT-IR analysis (Figure S4, Supporting Information). Photoelectron yield spectroscopy indicated that the CA doping has little effect on the valence band of CA-PVK (0.03 eV decrease from −5.40 eV to −5.37 eV (Figure S5, Supporting Information). Band-gap of CA-doped PVK films changed little while the doping ratio was varied, as evaluated by UV−vis absorption (see Figure S6). The above visual information on the growth of ordered crystals at ca. 10 wt % CA doping was supported by XRD analysis of the 120 nm film (Figure 3k, blue line). The (100) 9600

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604

Article

Journal of the American Chemical Society

Figure 5. CA-doping effects on devices with 120-nm- and 470-nmthick PVK films. (a) J−V curves for various doping ratios for 120-nmthick PVK film. (b) PCE vs CA-doping percentage. (c) J−V curve of CA-PVK SC with a thickness of 470 nm with 4 mm2 active area using 0 and 15 wt % CA-PVK films (Jsc = 21.8 mA/cm2). (e) Improvement of EQE values by CA doping. Jsc calculated from EQE was 21.4 mA/ cm2.

nitrogen and ambient light decreased by only 5.3%, from 13.3% to 12.6%, after 90 days (Figure 6d). The PCE of a 1 cm2 device under weak irradiation was 20.38% (8.7 mW/cm2 or 0.087 suns), which is significantly higher than the 14.52% PCE value under one sun irradiation (Figure 6e,f). We also found that illumination by 1000-lx white LED results in a PCE value of 28.1% PCE (for a 10 wt % CA-PVK device; Figure S10). In fact, the CA-controlled PVK crystal growth is applicable to fabrication of devices with a PVK film of a variety of thickness (Figure S11a). As compared with the 0 wt % CA-PVK device, the SCs based on a CA-doped film gave 3−5% higher PCE for all thickness examined. Island morphology of PVK crystals such as the one in Figure 3i has been suggested to be useful for semitransparent device for window applications in zero-energy building technology.38 A 180-nm-thick CA-PVK film and a 10 nm Ag electrode is a good compromise between average transmittance (AVT, 400−800 nm) and PCE (Figure S11a-c), showing a PCE value of 12% and an AVT value of 15%. There was found little hysteresis (Figure S11e). Figure S11f shows an AVT/EQE correlation, and Jsc of 16.4 mA/cm2 calculated from the EQE data (Jsc from J−V data in Figure S11e is 3.2% higher). Effects of Various Carboxylic Acids on Morphology. Finally, we describe the comparison among various carboxylic acid (15 wt %) on the film morphology and the device performance as summarized in Figure 7 and Table S1. We first notice that both the morphology and the performance are improved much by α-hydroxy carboxylic acids (TA, CA, and MA) (Figure 7a−c and i). CA shows the highest PCE because of uniformly high Jsc, Voc, and FF values. It is interesting to note that MA-doping resulted in a rather high PCE despite the low coverage seen in Figure 7ca tendency also observed for CA (Figure 3i). Two other polyfunctional acids, LA and ADA, exerted some influence on morphology but little on PCE

Figure 4. XRD analysis of 120-nm-thick PVK films from 4:1:1 precursor doped with x wt % CA with and without annealing at 100 °C for 20 min. See Figure 1 caption for the index assignment. (a−d) 2D mapping of XRD signals. (e−n) ϕ−I plots of RSMs. Blue lines refer to data for as-spin-coated film and red lines for samples without annealing. fwhm values are for films after annealing, while those without annealing could not be determined.

of Jsc, because of decrease of Voc and decrease of the overall performance. We also fabricated 1 cm2 devices (Figure 6a) with the same device configuration based on 15 wt % CA doping, which showed 14.52% PCE. PCE data for 25 devices fell between 13% and 15.0% with an average value of 13.8% (Figure S5b, Supporting Information). The J−V curves showed little hysteresis and scanning-time dependence (Figure 6b and Figure S9, Supporting Information). The photocurrent density and PCE remained unchanged (13.72%) under 0.7 V bias voltage and irradiation, suggesting that our SC is stable under steady-state conditions (Figure 6c).37 The PCE of a 1 cm2 device with a Ag electrode without any encapsulation under 9601

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604

Article

Journal of the American Chemical Society

Figure 7. Acid additive effects (15 wt %) on 120-nm-thick film morphology and performance of PVK solar cells. (a−f) Additive dependent morphology change. (i) Device performance summary. The standard 4:1:1 MAI/PI, PCl precursor was used. (h) Structures of acids and abbrevations. Acids with chiral centers are racemic. Large format SEM pictures in Supporting Information (Figure S14).

hence PCE. Further studies on the chemical mechanism of morphology modulation39−42 and exploration of other modulators to improve morphology and surface coverage would be beneficial for the production of practically useful SC devices.43−46

Figure 6. J−V characteristics of 1 cm2 PVK solar cells. (a) Photograph of the device. (b) J−V curves of 15 wt % CA-PVK (470 nm) SC upon forward and reverse scans. (c) Steady-state power output of a 13.72% PCE device under a bias voltage of 0.7 V. (d) J−V curves of a 1 cm2 device before and after aging under nitrogen for 90 days. (e) J−V curves for SC under different levels of light irradiance. (f) Irradiance dependence of PCE.

■

EXPERIMENTAL SECTION

Preparation of x wt % CA-PVK Precursor Solution. MAI was prepared following previous reports. 47,48 In a glovebox (N 2 atmosphere), MAI, PbI2 (TCI, 99.999%), PbCl2 (Sigma-Aldrich, 99.999%), and citric acid (Wako, >98%) were dissolved in DMF (Tokyo Chemical Industry Co., 99.5%). For 0 wt % CA−PVK precursor solution, a 4:1:1 molar mixture of solid powder MAI, PbI2, and PbCl2 were placed in DMF.49 Then, the solution was heated at 65 °C for about 12 h to form a uniform precursor solution. For CA-added PVK precursor solutions, a 4:1:1 molar mixture of solid powder MAI, PbI2, PbCl2, and additional x wt % (relative to mPbI2+mPbCl2) CA were dissolved in DMF solvent and heated at 65 °C for about 3 h to form the precursor solution. UV−vis and PL Spectra Measurement. The UV−vis spectrum of the PVK thin films on a PEDOT:PSS surface was recorded on a Jasco V-670 spectrophotometer. PL of PV film on glass was performed by a fluorescence spectrophotometer (HITACHI, F-4500). IR Measurement. The ATR−IR spectra were recorded on a JASCO FT-IR-6100 spectrometer. Measurements were performed under a nitrogen atmosphere at room temperature. XRD Measurements. The XRD experiment was performed on a Rigaku SmartLab X-ray diffractometer equipped with a scintillation counter. The measurement employed the Cu Kα line, focused radiation at 9 kW (45 kV, 200 mA) power using a 0.02° 2θ step scan from 3.0° to 60.0° with a scanning speed of 3° min−1. Reciprocal space maps (RSM) analysis was collimated with CBO-f point line (ø < 1 mm) and detected by a HyPix-3000 two-dimensional semiconductor detector. SEM and EDX Measurements. SEM observations were conducted on an FEI Magellan 400L equipped with AMETEK/ EDAX Genesis APEX4 at a landing voltage of 1 kV under a reduced pressure of 5 × 10−5 Pa.

values. Low-boiling monocarboxylic acid (FA and ACA) little affected the quality of crystal growth and device performance. Thus, high-boiling α-hydroxy acids were found to improve crystal morphology and uniformity, and to increase PCE. Possible roles of CA and other α-hydroxy acids may be their ability to bind tightly to a Lewis acidic lead atom and to control crystal nucleation as has been known for CA,39−42 as well as push−pull acceleration of perovskite formation via coordination to Pb−I bonds as has been shown recently for zwitterionic sulfamic acid.32 Slightly increased first dissociation constant of α-hydroxyacids (Table S1) may also be relevant to the improvement of PCE, an issue to be further probed.

■

CONCLUSION In summary, we found that PVK crystals made only from MAI and PI grow into a mixture of stacked cuboid and ED crystals, which are apparently unsuitable for SC device performance. However, the same precursor mixture doubly doped with CA and chloride ion produces large cuboid crystals without significant incorporation of both additives into the crystal lattices. This observation provides a clue to understanding the beneficial effects of minor additives often reported in PVK SC fabrication. The active layer made by the double-doping method shows significant improvement of Jsc, Voc, FF, and 9602

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604

Article

Journal of the American Chemical Society Preparation of SC Devices. The patterned ITO glass was ultrasonically cleaned using a surfactant, rinsed with water, and then given 3 min UV−ozone treatment. A PEDOT:PSS solution (AI4083) was spin-coated on the ITO surface at 500 rpm for 3 s and then 3000 rpm for 30 s in air and was annealed at 130 °C for 25 min under air and 20 min under N2. A solution of 0 to 20 wt % CA−PVK precursor solution was spin-coated on the PEDOT:PSS surface under nitrogen atmosphere. During this process, a modified spin-coating method was used (please see the Video 1). Before annealing, the film stayed 20 min and was then annealed at 100 °C/30 min for 0 wt % CA-PVK and 100 °C/20 min for 1 to 20 wt % CA-PVK film. An electron-transporting layer (PCBM, 20 mg/mL in chlorobenzene for thin x wt % CA-PVK film (below 250 nm), 30 mg/mL for thick film (>250 nm) in chlorobenzene) was deposited by spin-coating (1000 rpm/45 s). PEIE (0.02 wt %, in methanol), an interface-modifying layer, was spincoated at 6000 rpm for 30 s.50,51 The top electrode (Ag, 150 nm) was deposited via a metal shadow mask. Evaluation of PV Devices. Current−voltage sweeps were taken on a Keithley 2400 source measurement unit controlled by a computer. The light source used to determine the PCE was an AM1.5G solar simulator system (Sumitomo Heavy Industries Advanced Machinery) with intensity of 100 mW/cm2. For weak light, different filters were used to tune the irradiance. The SCs were masked with a metal aperture to define the active area.

■

(4) Mitzi, D.; Field, C.; Harrison, W.; Guloy, A. Nature 1994, 369, 467−469. (5) Stoumpos, C.; Malliakas, C.; Kanatzidis, M. Inorg. Chem. 2013, 52, 9019−9038. (6) Wakamiya, A.; Endo, M.; Sasamori, T.; Tokitoh, N.; Ogomi, Y.; Hayase, S.; Murata, Y. Chem. Lett. 2014, 43, 711−713. (7) Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. Science 2015, 348, 1234−1237. (8) Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S. I. Nat. Mater. 2014, 13, 897−903. (9) Zhou, H. P.; Chen, Q.; Li, G.; Luo, S.; Song, T.; Duan, H.; Hong, Z. R.; You, J. B.; Liu, Y. S.; Yang, Y. Science 2014, 345, 542−546. (10) Malinkiewicz, O.; Yella, A.; Lee, Y. H.; Espallargas, G. M.; Grätzel, M.; Nazeeruddin, M. K.; Bolink, H. J. Nat. Photonics 2013, 8, 128−132. (11) Liu, D.; Kelly, T. L. Nat. Photonics 2013, 8, 133−138. (12) Nazeeruddin, M. K.; Gao, P.; Grätzel, M. Energy Environ. Sci. 2014, 7, 2448−2463. (13) Manser, J. S.; Christians, J. A.; Kamat, P. V. Chem. Rev. 2016, 116, 12956−13008. (14) Saparov, B.; Mitzi, D. B. Chem. Rev. 2016, 116, 4558−4596. (15) Stoumpos, C. C.; Kanatzidis, M. G. Acc. Chem. Res. 2015, 48, 2791−2802. (16) Yang, B.; Dyck, O.; Poplawsky, J.; Keum, J.; Puretzky, A.; Das, S.; Ivanov, I.; Rouleau, C.; Duscher, G.; Geohegan, D.; Xiao, K. J. Am. Chem. Soc. 2015, 137, 9210−9213. (17) Li, X.; Ibrahim Dar, M.; Yi, C.; Luo, J.; Tschumi, M.; Zakeeruddin, S. M.; Nazeeruddin, M. K.; Han, H.; Grätzel, M. Nat. Chem. 2015, 7, 703−711. (18) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Grätzel, M. Nature 2013, 499, 316−319. (19) Liu, M. Z.; Johnston, M. B.; Snaith, H. J. Nature 2013, 501, 395−398. (20) Nie, W.; Tsai, H.; Asadpour, R.; Blancon, J.; Neukirch, A. J.; Gupta, G.; Crochet, J.; Chhowalla, M.; Tretiak, S.; Alam, M. A.; Wang, H.; Mohite, A. D. Science 2015, 347, 522−525. (21) Wang, S.; Jiang, Y.; Juarez-Perez, E.; Ono, L.; Qi, Y. Nat. Energy 2016, 2, 16195. (22) Quan, L. N.; Yuan, M. J.; Comin, R.; Voznyy, O.; Beauregard, E. M.; Hoogland, S.; Buin, A.; Kirmani, A. R.; Zhao, K.; Amassian, A.; Kim, D. H.; Sargent, E. H. J. Am. Chem. Soc. 2016, 138, 2649−2655. (23) Shi, D.; Adinolfi, V.; Comin, R.; Yuan, M. J.; Alarousu, E.; Buin, A.; Chen, Y.; Hoogland, S.; Rothenberger, A.; Katsiev, K.; Losovyj, Y.; Zhang, X.; Dowben, P. A.; Mohammed, O. F.; Sargent, E. H.; Bakr, O. M. Science 2015, 347, 519−522. (24) Dong, Q.; Fang, Y.; Shao, Y.; Mulligan, P.; Qiu, J.; Cao, L.; Huang, J. Science 2015, 347, 967−970. (25) (a) Liu, Y.; Yang, Z.; Cui, D.; Ren, X.; Sun, J.; Liu, X.; Zhang, J.; Wei, Q.; Fan, H.; Yu, F.; Zhang, X.; Zhao, C.; Liu, S. F. Adv. Mater. 2015, 27, 5176−5183. (b) Williams, S.; Zuo, F.; Chueh, C.; Liao, C.; Liang, P.; Jen, A. ACS Nano 2014, 8, 10640−10654. (26) Lian, Z.; Yan, Q.; Gao, T.; Ding, J.; Lv, Q.; Ning, C.; Li, Q.; Sun, J. L. J. Am. Chem. Soc. 2016, 138, 9409−9412. (27) Yakunin, S.; Dirin, D. N.; Shynkarenko, Y.; Morad, V.; Cherniukh, I.; Nazarenko, O.; Kreil, D.; Nauser, T.; Kovalenko, M. V. Nat. Photonics 2016, 10, 585−589. (28) Náfrádi, B.; Náfrádi, G.; Forró, L.; Horváth, E. J. Phys. Chem. C 2015, 119, 25204−25208. (29) Zhou, Y.; You, L.; Wang, S.; Ku, Z.; Fan, H.; Schmidt, D.; Rusydi, A.; Chang, L.; Wang, L.; Ren, P.; Chen, L.; Yuan, G.; Chen, L.; Wang, J. Nat. Commun. 2016, 7, 11193. (30) Xie, L.; Chen, L.; Nan, Z.; Lin, H.; Wang, T.; Zhan, D.; Yan, J.; Mao, B.; Tian, Z. J. Am. Chem. Soc. 2017, 139, 3320−3323. (31) Guo, Y.; Shoyama, K.; Sato, W.; Matsuo, Y.; Inoue, K.; Harano, K.; Liu, C.; Tanaka, H.; Nakamura, E. J. Am. Chem. Soc. 2015, 137, 15907−15914. (32) Guo, Y.; Sato, W.; Shoyama, K.; Nakamura, E. J. Am. Chem. Soc. 2016, 138, 5410−5416.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b03856. Details of EDX data, PYS analysis, IR analysis, and statistics of devices performance, and large format LVSEM pictures (PDF) W Web-Enhanced Feature *

Video showing modified spin-coating method with and without citric acid is available in the online version of this paper.

■

AUTHOR INFORMATION

Corresponding Authors

*[email protected]. *[email protected]. ORCID

Rui Shang: 0000-0002-2513-2064 Eiichi Nakamura: 0000-0002-4192-1741 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Mitsubishi Chemical Corporation and Drs. Izuru Takei and Toshifumi Kawano for the data under LED illumination, and MEXT for financial support [KAKENHI 15H05754 and CREST, JST to E.N., the Strategic Promotion of Innovative Research, JST to R.S.], and Rigaku Company for RMS analysis.

■

REFERENCES

(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am. Chem. Soc. 2009, 131, 6050−6051. (2) Bergmann, V. W.; Weber, S. A. L.; Ramos, F. J.; Nazeeruddin, M. K.; Grätzel, M.; Li, D.; Domanski, A. L.; Lieberwirth, I.; Ahmad, S.; Berger, R. Nat. Commun. 2014, 5, 5001. (3) Kim, H. S.; Lee, C. R.; Im, J. H.; Lee, K. B.; Moehl, T.; Marchioro, A.; Moon, S. J.; Humphry-Baker, R.; Yum, J. H.; Moser, J. E.; Grätzel, M.; Park, N. G. Sci. Rep. 2012, 2, 591. 9603

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604

Article

Journal of the American Chemical Society (33) (a) Chung, J.; Granja, I.; Taylor, M.; Mpourmpakis, G.; Asplin, J.; Rimer, J. Nature 2016, 536, 446−450. (b) Bi, C.; Wang, Q.; Shao, Y.; Yuan, Y.; Xiao, Z.; Huang, J. Nat. Commun. 2015, 6, 7747. (34) You, J.; Hong, Z.; Yang, Y.; Chen, Q.; Cai, M.; Song, T.; Chen, C.; Lu, S.; Liu, Y.; Zhou, Z.; Yang, Y. ACS Nano 2014, 8, 1674−1680. (35) Zuo, L.; Gu, Z.; Ye, T.; Fu, W.; Wu, G.; Li, H.; Chen, H. J. J. Am. Chem. Soc. 2015, 137, 2674−2679. (36) Xing, G.; Mathews, N.; Sun, S.; Lim, S.; Lam, Y.; Grätzel, M.; Mhaisalkar, S.; Sum, T. Science 2013, 342, 344−347. (37) Snaith, H.; Abate, A.; Ball, J.; Eperon, G.; Leijtens, T.; Noel, N.; Stranks, S.; Wang, J.; Wojciechowski, K.; Zhang, W. J. Phys. Chem. Lett. 2014, 5, 1511−1515. (38) http://www.m-kagaku.co.jp/english/aboutmcc/RC/special/ feature1. (39) Yin, Y.; Alivisatos, P. Nature 2005, 437, 664−670. (40) Wang, G.; Li, D.; Cheng, H.; Li, Y.; Chen, C.; Yin, A.; Zhao, Z.; Lin, Z.; Wu, H.; He, Q.; Ding, M.; Liu, Y.; Huang, Y.; Duan, X. Sci. Adv. 2015, 1, e1500613. (41) Ruan, L.; Ramezani-Dakhel, H.; Chiu, C.; Zhu, E.; Li, Y.; Heinz, H.; Huang, Y. Nano Lett. 2013, 13, 840−846. (42) Chiu, C.; Li, Y.; Ruan, L.; Ye, X.; Murray, C.; Huang, Y. Nat. Chem. 2011, 3, 393−399. (43) Son, D.; Lee, J.; Choi, Y.; Jang, I.; Lee, S.; Yoo, P.; Shin, H.; Ahn, N.; Choi, M.; Kim, D.; Park, N. Nat. Energy 2016, 1, 16081. (44) Walsh, A.; Scanlon, D.; Chen, S.; Gong, X.; Wei, S. Angew. Chem., Int. Ed. 2015, 54, 1791−1794. (45) Jeon, N.; Noh, J.; Yang, W.; Kim, Y.; Ryu, S.; Seo, J.; Seok, S. Nature 2015, 517, 476−480. (46) Zhou, Z.; Wang, Z.; Zhou, Y.; Pang, S.; Wang, D.; Xu, H.; Liu, Z.; Padture, N.; Cui, G. Angew. Chem., Int. Ed. 2015, 54, 9705−9709. (47) Lee, M.; Teuscher, J.; Miyasaka, T.; Murakami, T.; Snaith, H. Science 2012, 338, 643−647. (48) Ito, S.; Tanaka, S.; Manabe, K.; Nishino, H. J. Phys. Chem. C 2014, 118, 16995−17000. (49) Xue, Q.; Hu, Z.; Sun, C.; Chen, Z.; Huang, F.; Yip, H.; Cao, Y. RSC Adv. 2015, 5, 775−783. (50) Zhou, Y.; Fuentes-Hernandez, C.; Shim, J.; Meyer, J.; Giordano, A. J.; Li, H.; Winget, P.; Papadopoulos, T.; Cheun, H.; Kim, J.; Fenoll, M.; Dindar, A.; Haske, W.; Najafabadi, E.; Khan, T.; Sojoudi, H.; Barlow, S.; Graham, S.; Brédas, J.; Marder, S.; Kahn, A.; Kippelen, B. Science 2012, 336, 327−332. (51) Guo, Y.; Shoyama, K.; Sato, W.; Nakamura, E. Adv. Energy Mater. 2016, 6, 1502317.

9604

DOI: 10.1021/jacs.7b03856 J. Am. Chem. Soc. 2017, 139, 9598−9604