Conventional dye-sensitized solar cells consist of liquid. CH 3 NH 3 Sn x Pb (1 x) I 3 Perovskite Solar Cells Covering up to 1060 nm

Transcription

1 pubs.acs.org/jpcl CH 3 NH 3 Sn x Pb (1 x) I 3 Perovskite Solar Cells Covering up to 1060 nm Yuhei Ogomi,*, Atsushi Morita, Syota Tsukamoto, Takahiro Saitho, Naotaka Fujikawa, Qing Shen,, Taro Toyoda,, Kenji Yoshino,, Shyam S. Pandey, Tingli Ma, and Shuzi Hayase*,, Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, 2-4 Hibikino, Wakamatsu-ku, Kitakyushu, Fukuoka , Japan Graduate School of Informatics and Engineering, University of Electro-Communications, Chofugaoka, Chofu, Tokyo , Japan Department of Electrical and Electronic Engineering, University of Miyazaki, 1-1, Gakuen Kibanadai Nishi, Miyazaki , Japan CREST, Japan Science and Technology Agency (JST), Honcho Kawaguchi, Saitama , Japan *S Supporting Information ABSTRACT: We report photovoltaic performances of all-solid state Sn/Pb halidebased perovskite solar cells. The cell has the following composition: F-doped SnO 2 layered glass/compact titania layer/porous titania layer/ch 3 NH 3 Sn x Pb (1 x) I 3 /regioregular poly(3-hexylthiophene-2,5-diyl). Sn halide perovskite itself did not show photovoltaic properties. Photovoltaic properties were observed when PbI 2 was added in SnI 2. The best performance was obtained by using CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 perovskite. 4.18% efficiency with open circuit voltage 0.42 V, fill factor 0.50, and short circuit current ma/cm 2 are reported. The edge of the incident photon to current efficiency curve reached 1060 nm, which was 260 nm red-shifted compared with that of CH 3 NH 3 PbI 3 perovskite solar cells. SECTION: Energy Conversion and Storage; Energy and Charge Transport Conventional dye-sensitized solar cells consist of liquid electrolytes that provide well-organized electronic contact between titania and the electrolyte in nanoporous titania layer. Although these liquid electrolytes have a lot of advantages for high-efficiency solar cells, developments of all-solid state dyesensitized solar cells with high efficiency have also been one of the major research items. 1 All solid-state solar cells consisting of perovskite have recently attracted interest because of the high efficiency reaching 12 16%. 1 9 Perovskite solar cells consist of compact titania layer, porous metal oxide layer, perovskite layer, and p-type organic semiconductor layer. Two mechanisms on the electron collection have been reported depending on which porous layer (porous titania or porous alumina) was employed. When porous titania layer was used, electrons are collected by the porous titania layer. Park and coworkers have reported perovskite solar cells with 9.7% efficiency, where the cell is composed of TiO 2 /CH 3 NH 3 PbI 3 /(2,2,7,7 -tetrakis[n,n-di (4-methoxy phenyl) amino]-9,9 -spirobifluorene) (spiro-ome- TAD). 3 The efficiency has further increased to 14.14% (certified efficiency) by using two-step perovskite fabrication process. 4 When the perovskite layer was fabricated on porous alumina layer, electrons are collected by the perovskite layer itself covering the porous alumina surface. 5,6 Recently, perovskite solar cells with flat heterojunction structure prepared by a coevaporation process under vacuum have been reported, and the efficiency reached 15.4%. 9 Electronic absorption spectra for Pb halide perovskite shift to longer wavelength by changing the halide from Cl to Br and I. 2,10 15 CH 3 NH 3 PbI 3 perovskite has the most red-shifted edge of 800 nm. Light harvesting in the area longer than 800 nm (near-infrared region: NIR region) is necessary to realize high-efficiency tandem perovskite solar cells. It has been reported that Sn halide perovskites have electronic absorption up to 1000 nm However, there was no report on the NIR perovskite solar cells. We have succeeded in harvesting energy in the NIR region by using Sn halide perovskite. Figure 1 shows X-ray diffraction patterns for glass/porous TiO 2 /CH 3 NH 3 Sn x Pb (1 x) I (2θ) (Figure 1A) and 28.5 (2θ) (Figure 1B) assigned to 110 and 220 crystal plane, respectively, were clearly observed for all CH 3 NH 3 Sn x Pb (1 x) I 3 compositions, leading to the conclusion that all compositions have similar crystal units. In addition, as shown in Figure 1, these perovskite materials did not contain the starting materials such as CH 3 NH 3 I, PbI 2, and SnI 2. Figure 2 shows the SEM images of CH 3 NH 3 Sn x Pb (1 x) I 3 layers fabricated on the porous TiO 2 layer. CH 3 NH 3 PbI 3 had needlelike structures, which were the same as those reported in the previous paper. 4 When SnI 2 was added in the CH 3 NH 3 PbI 3, Received: January 29, 2014 Accepted: March 3, 2014 Published: March 3, American Chemical Society 1004

2 The Journal of Physical Chemistry s Figure 1. XRD patterns for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite coated on porous TiO 2 along with individual components. the needlelike crystals disappeared and flowerlike crystals appeared. The results of XPS were summarized in Figure 3 and Table 1. CH 3 NH 3 Sn x Pb (1 x) I 3 has peaks assigned to Sn3d, Pb4f, and I3d. Final compositions of Sn:Pb:I for CH 3 NH 3 Sn x Pb (1 x) I 3 were CH 3 NH 3 SnI 1.96 (expected: CH 3 NH 3 SnI 3 ), CH 3 NH 3 Sn Pb 0.26 I 2.47 (expected: CH 3 NH 3 Sn 0.7 Pb 0.3 I 3 ), CH 3 NH 3 Sn Pb 0.44 I 2.77 (expected: CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 ), CH 3 NH 3 Sn 0.37 Pb 0.63 I 2.95 (expected: CH 3 NH 3 Sn 0.3 Pb 0.7 I 3 ), and CH 3 NH 3 PbI 3.07 (expected: CH 3 NH 3 PbI 3 ) when x was varied from 1 to 0. These final compositions were almost the same as those for initial blending ratios. Reaction product of CH 3 NH 3 I with SnI 2 has CH 3 NH 3 SnI 1.96 composition, which has less I content than expected (CH 3 NH 3 SnI 3 ), probably because of oxidation during the measurement. These results were similar to those reported in the previous report. 22 It has been reported that Sn 4+ and Sn 2+ signals appear at and ev, respectively. 23 In our result of XPS, these two separate peaks were not observed. However, the peak assigned to Sn shifted from higher to lower binding energy when x in CH 3 NH 3 Sn x Pb (1 x) I 3 decreased, suggesting that the content of Sn 4+ increased as x decreased. It has been reported that carrier concentration of CH 3 NH 3 SnI, CH 3 NH 3 Sn 0.5 Pb 0.5 I 3, and CH 3 NH 3 PbI 3 is /cm 3, /cm 3, /cm 3, and drift mobility for them is 2320, 270, and 66 cm 2 /(V s), respectively. 22 Because the presence of Sn 4+ increases carrier concentrations, 22 the addition of PbI 2 decreases the Sn 4+ content and decreases the carrier concentration. Namely, the oxidation of Sn 2+ to Sn 4+ seems to be somehow retarded by the addition of Pb, which increases the stability of these Sn/Pb perovskites in the air. Actually, the peak height of electronic absorption of CH 3 NH 3 SnI 3 decreased to 90% in 1 min and that of CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 became longer (50 min) The electronic absorption spectrum edge were also varied with x of CH 3 NH 3 Sn x Pb (1 x) I 3. The edge shifted from 1000 to 1200 nm when x increased from 0.3 to 1.0, as shown in Figure 4. Valence band for Sn/Pb perovskite (CH 3 NH 3 Sn x Pb (1 x) I 3 ) shifted from 5.12 to 4.73 ev when x increased from 0.3 to 1.0. Conduction band shifted from 3.81 to 3.63 ev in the same way as shown in Figure 5. The valence band shift was larger than that of the conduction band shift on x change. Figure 6 summarizes an energy level diagram. These conduction bands of Sn/Pb perovskites were shallower than that of titania ( 0.40 ev), suggesting that electron can be injected from these perovskites to titania. In addition, these valence bands are deeper than that of poly(3-hexylthiophene- 2,5-diyl) (P3HT). Therefore, holes can be injected from these perovskite layers to P3HT. spiro-ometad has been employed for the hole-transporting layer in previous reports. 2 9 However, the valence band level of CH 3 NH 3 Sn x Pb (1 x) I 3 (expect for CH 3 NH 3 PbI 3 ) is shallower than that of spiro-ometad ( 5.2 ev), suggesting that hole injection from CH 3 NH 3 Sn x - Pb (1 x) I 3 perovskite to spiro-ometad would not occur. Therefore, P3HT was employed for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite solar cell. Figure 7 shows the schematic diagram of the perovskite solar cell. A compact titania layer (35 nm) and a porous titania layer (200 nm) were fabricated on a transparent conductive oxide layer (F-doped SnO 2 glass: FTO). The thickness of P3HT fabricated on the perovskite layer was 100 nm. On the P3HT layer, Ag (10 nm) and Au (55 nm) layers were fabricated by vacuum deposition method. Figure 8 shows I V curves of perovskite solar cells. The relationship between Sn/Pb ratio and photovoltaic performances is summarized in Figures 9 and 10 and Table 2. CH 3 NH 3 SnI 3 perovskite/p3ht solar cells did not show any photovoltaic performances. When Pb was added in the Sn perovskite, photovoltaic properties appeared. The maximum photovoltaic performance was observed when x was 0.5 (CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 ). Open circuit voltage (V oc ), fill factor (FF), short circuit current (J sc ), and efficiency for CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 perovskite solar cell were around 0.32 V, 0.37, ma/cm 2, and 2.37%. As the ratio of Sn decreased further from 0.5, the photovoltaic performance decreased again. The efficiency for CH 3 NH 3 PbI 3 /P3HT solar cell (0.18%) was extremely lower than that of CH 3 NH 3 PbI 3 /spiro-ometad solar cells (12%) in our previous report. 24 The photovoltaic performance for CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /spiro-ometad HTM was Jsc: 0.04 ma/cm 2, V oc : 0.13 V, FF: 0.32, and efficiency: %, which was lower than that employing P3HT as the hole transport layer. HOMO level for spiro-ometad HTM is deeper than that of perovskites, as shown in Figure 6, which would make the hole injection from the hole-transport layer to perovskite layer difficult. R s (series resistance) of CH 3 NH 3 PbI 3 /P3HT was 477 Ω, which decreased to 50 Ω as the Sn content increased. R sh (shunt resistance) also had a tendency to decrease as the Sn content increases. It has been reported that Sn perovskite has high carrier concentration and good conductivities. 21 The high carrier concentration is suitable for decreasing R s. However, the high carrier concentration may increase the opportunity of charge recombination for CH 3 NH 3 Sn x Pb (1 x) I 3 / P3HT perovskite with high Sn ratio. The low efficiency for CH 3 NH 3 PbI 3 /P3HT can be explained by extremely high R s (477Ω). The high R s suggests that the contact between CH 3 NH 3 PbI 3 and P3HT layers was not optimized. The best

3 The Journal of Physical Chemistry s Figure 2. SEM images of CH3NH3SnxPb(1 x)i3 perovskite coated on porous TiO2. In Figure 8, supposing that the Voc max is determined by the difference between titania conduction band ( 4.0 V) and HOMO of P3HT ( 4.67 V), the expected Voc should be 0.67 V. However, the Voc of CH3NH3Sn0.5Pb0.5I3 was 0.3 V, which is extremely lower than expected. This may be explained by large opportunity of charge recombination between electrons in titania and oxidized perovskite or between electrons in titania and P3HT. Surface passivation of traps on surface of titania layer would be needed to improve the efficiency, as we reported in the previous paper Figure 11 shows dark current for Sn/Pb perovskite solar cells. Dark current was suppressed as the x of CH3NH3SnxPb(1 x)i3 decreased, and the diode properties was improved probably because the carrier concentration of the Sn/Pb perovskite was reduced. It has been reported that the edge of electronic absorption spectra is shifted to longer wavelength on the addition of Sn perovskite into Pb perovskite.22 The edge of the incident photon to current efficiency (IPCE) curves for CH3NH3SnxPb(1 x)i3 actually shifted from 800, 1030, 1050, and 1060 nm as the Sn content (x) increased from 0 to 70%, as shown in Table 2 and Figure 12. This shows the possibility that IPCE edge can be tuned by the ratio of (x) and (1 x) in CH3NH3SnxPb(1 x)i3. Figure 13 shows photocurrent curves calculated from IPCE curves. In CH3NH3Sn0.5Pb0.5I3, the calculated photocurrent (17.57 ma/cm2) was a little lower than those actually measured (19.88 ma/cm2) probably because of Figure 3. XPS spectra of CH3NH3SnxPb(1 x)i3 perovskite coated on porous TiO2. efficiency of CH3NH3Sn0.5Pb0.5I3 would be explained by the balance of high Rsh and low Rs. 1006

4 The Journal of Physical Chemistry s Table 1. XPS Analysis of CH 3 NH 3 Sn x Pb (1 x) I 3 Perovskite Coated on Porous TiO 2 a binding energy/ev CH 3 NH 3 Sn x Pb y I z Sn 3d3/2 Sn 3d5/2 Pb 4f5/2 Pb 4f7/2 I 3d3/2 I 3d5/2 x y z CH 3 NH 3 PbI 3 N/A N/A CH 3 NH 3 Sn 0.3 Pb 0.7 I CH 3 NH 3 Sn 0.5 Pb 0.5 I CH 3 NH 3 Sn 0.7 Pb 0.3 I CH 3 NH 3 Sn 0.9 Pb 0.1 I CH 3 NH 3 SnI N/A N/A a Ratio of x, y, and z were obtained from integration of the peak area of corresponding XPS peaks. Figure 4. Electronic absorption spectra of CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite coated on porous TiO 2. Figure 6. Energy diagram of titania, P3HT, spiro-ometad, and CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite. Figure 5. Energy diagram for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite. easier oxidation of these perovskites with large x ratio during IPCE measurements. Figure 14 and Table 3 show the optimized characteristics for CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /P3HT and CH 3 NH 3 PbI 3 /spiro-ometad with 12% efficiency in our previous report. 24 V oc, FF, J sc, and efficiency for CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /P3HT were 0.42 V, 0.50, ma/cm 2, and 4.18%. CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /P3HT can harvest light in the area from 800 to 1060 nm, where previous Pb-based solar cells were not able to. R sh of CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 / P3HT was 1402 Ω, which was seriously lower than that of Ω for CH 3 NH 3 PbI 3 -based solar cell. As discussed in the previous session, retardation of charge recombination is needed to increase the efficiency further. Figure 15 shows the energy level diagrams for dye-sensitized solar cells with iodide/triiodide redox species, conventional 1007 Figure 7. Structure of CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite solar cells. organic thin film solar cells and perovskite solar cells. In each solar cell, observed V oc is lower than that expected from the light absorption edge. The difference between the observed V oc and that expected from the light-absorption spectrum edge (ΔG = 1240/light absorption edge (nm)) is called V loss. The V loss for dye-sensitized solar cells with iodide/triiodide and organic thin film solar cells is 0.7 V. 29 For example, in the case of dye-sensitized solar cell consisting of iodide/triiodide redox species, V loss (0.7 V) was estimated by summarizing ΔG1(energy loss caused by electron injection from dye to titania), ΔG2 (energy loss caused by electron injection from iodide to oxidized dye), and ΔGx (energy loss caused by charge recombination). V loss for perovskite solar cells is expected to be 0.5 V, which is lower than that with iodide/triiodide redox species. 29 Figure 16 shows the relationship between efficiency and electronic absorption spectrum edge when FF and average

5 The Journal of Physical Chemistry s Figure 8. I V curves for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite solar cells. Composition: CH 3 NH 3 Sn x Pb (1 x) I 3 /P3HT. Solar cells were measured under 100 mw/cm 2 (AM 1.5G) light illumination with masked area 0.4 mm 0.4 mm. Figure 10. Relationship between FF, V oc, and Sn content (x) for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite solar cells (I V curves and experimental conditions; see Figure 8). Figure 9. Relationship between J sc,efficiency, and Sn content (x) for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite solar cells (I V curves and experimental conditions; see Figure 8). IPCE were fixed to be 0.7 and 0.8. When V loss is varied, the relationship between maximum efficiency and the electronic absorption spectrum edge changed. The relationship between efficiency (PCE: power conversion efficiency) and wavelength was calculated in the following way. J scmax was obtained from integration of each photocurrent at each wavelength by using solar spectrum listed in ASTM G (American Society of Testing and Materials). 28 ΔG was obtained by band gap (ΔG = λ/1240, λ: edge of light absorption Figure 11. Dark current for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite/p3ht solar cells (I V curves and experimental conditions: see Figure 8). spectrum). Therefore, PCE can be obtained by the following equation. PCE (power conversion efficiency) = ( ΔG V ) FF ( J IPCE) loss scmax where V loss stands for loss of voltage from ΔG. (See Figure 15.) The PCE at each wavelength was plotted against the wavelength to give Figure 16. For example, supposing that V loss for dye-sensitized solar cells is 0.7 V, the maximum efficiency Table 2. Photovoltaic Performances for CH 3 NH 3 Sn x Pb (1 x) I 3 Perovskite/P3HT Solar Cell a PbI 2 ratio [%] SnI 2 ratio [%] J sc [ma/cm 2 ] V oc [V] FF efficiency [%] R s [Ω] R sh [Ω] IPCE (at 500 nm) IPCE (at 900 nm) IPCE edge [nm] E g [ev] Jsc a [ma/cm 2 ] N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A a Solar cells were measured under 100 mw/cm 2 (AM 1.5G) light illumination with masked area 0.4 mm 0.4 mm. Jsc a represents the short-circuit current densities calculated from the integration of IPCE (Figure 12) at each wavelength and light dose at each wavelength obtained from solar spectrum (ASTM G ). 1008

6 The Journal of Physical Chemistry s Figure 12. IPCE curves for CH 3 NH 3 Sn x Pb (1 x) I 3 perovskite/p3ht solar cells (I V curves and experimental conditions: see Figure 8). Figure 15. Energy diagrams for (A) dye-sensitized solar cells with iodine/iodide, (B) organic thin film solar cell (OPV), and (C) perovskite solar cell that are focused on voltage losses. Figure 13. IPCE spectrum for CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 perovskite solar cell. Calculated Jsc a represents the short-circuit current densities calculated from the integration of IPCE (Figure 12) at each wavelength and light loss at each wavelength obtained from solar spectrum (ASTM G ). See the experimental section in the Supporting Information. (13%) is obtained by using dyes, which can harvest the light up to 800 nm (0.7 (V loss ) in Figure 15). In the case of perovskitesensitized solar cells, supposing that V oc loss is 0.5 V, maximum efficiency (16.6%) can be obtained by 900 nm light absorption edge (0.5 (V loss ) in Figure 15). The value of 0.5 V loss was taken from the previous paper. 29 When the light absorption edge is further shifted to longer wavelength, the efficiency decreased again, as shown in Figure 16. The edge of the IPCE curve for CH 3 NH 3 PbI 3 -sensitized solar cells is limited to 800 nm. If the light absorption edge is optimized to be 900 nm by adjusting Pb/Sn ratio, further enhancement of solar cell efficiency is expected. Of course, these Sn/Pb solar cells having sensitivity in the area of IR regions are useful for bottom cells for all-solid-type perovskite tandem cells. Figure 14. Optimized photovoltaic performance (A) I V and (B) IPCE curves for CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /P3HT and CH 3 NH 3 PbI 3 /spiro-ometad perovskite solar cells. (See Figure 8 for experimental conditions.) Table 3. Photovoltaic Performances for Perovskite Solar Cells Using CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /P3HT and CH 3 NH 3 PbI 3 /spiro- OMeTAD a J sc [ma/cm 2 ] V oc [V] FF efficiency [%] R s [Ω] R sh [Ω] J a sc [ma/cm 2 ] CH 3 NH 3 PbI 3 /spiro-ometad CH 3 NH 3 Sn 0.5 Pb 0.5 I 3 /P3HT a Solar cells were measured under 100 mw/cm 2 (AM 1.5G) light illumination with masked area 0.4 mm 0.4 mm. J sc a represents the short circuit current densities calculated from the integration of IPCE (Figure 14B) at each wavelength and light dose at each wavelength obtained from solar spectrum (ASTM G ). These data were obtained from Figure

7 The Journal of Physical Chemistry s Figure 16. Relationship between light harvesting edge (nm) and solar cell efficiency when voltage loss was varied from 1.0 to 0 V. Photovoltaic performances for all solid-state Sn/Pb based perovskite solar cells were reported for the first time. It was proved that the cell can harvest the light in the area up to 1060 nm. The Sn/Pb-based perovskite solar cells are useful for bottom cells for tandem perovskite solar cells. To increase the efficiency, we concluded that retardation of charge recombina- and increase in the R sh are needed. tion ASSOCIATED CONTENT *S Supporting Information Experimental methods. This material is available free of charge via the Internet at AUTHOR INFORMATION Corresponding Authors * ogomi@life.kyutech.ac.jp (Y.O.). * hayase@life.kyutech.ac.jp (S.H.). Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS This research was supported by JST and CREST. REFERENCES (1) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H. Dye- Sensitized Solar Cells. Chem. Rev. 2010, 110, (2) Noh, J. H.; m, S. H.; Heo, J. H.; Mandal, T. N.; Seok, S. I. Chemical Management for Colorful, Efficient, And Stable Inorganic organic Hybrid Nanostructured Solar Cells. Nano Lett. 2013, 13, (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.; Graetzel, M.; Park, N.-G. Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (4) Burschka, J.; Pellet, N.; Moon, S. J.; Baker, R. H.; Gao, P.; Nazeeruddin, M. K.; Graetzel, M. Sequential Deposition As a Route to High-Performance Perovskite-Sensitized Solar Cells. Nature 2013, 499, (5) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Efficient Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide Perovskites. Science 2012, 338, (6) Ball, J. M.; Lee, M. M.; Hey, A.; Snaith, H. J. Low-Temperature Processed Mesosuperstructured to Thin-Film Perovskite Solar Cells. Energy Environ. Sci. 2013, 6, (7) Liu, D.; Kelly, T. L. Perovskite Solar Cells with a Planar Heterojunction Structure Prepared Using Room-Temperature Solution Processing Techniques. Nat. Photonics 2014, 8, (8) Snaith, H. J. Perovskites: The Emergence of a New Era for Low- Cost, High-Efficiency Solar Cells. J. Phys. Chem. Lett. 2013, 4, (9) Liu, M.; Johnston, M. B.; Snaith, H. J. Efficient Planar Heterojunction Perovskite Solar Cells by Vapour Deposition. Nature 2013, 501, (10) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131, (11) Im, J.-H.; Lee, C.-R.; Lee, J.-W.; Park, S.-W.; Park, N.-G. 6.5% Efficient Perovskite Quantum-Dot-Sensitized Solar Cell. Nanoscale 2011, 3, (12) Tanaka, K.; Takahashi, T.; Ban, T.; Kondo, T.; Uchida, K.; Miura, N. Comparative Study on the Excitons in Lead-Halide-Based Perovskite-Type Crystals CH 3 NH 3 PbBr 3 CH 3 NH 3 PbI 3. Solid State Commun. 2003, 127, (13) Calabrese, J.; Jones, N. L.; Harlow, R. L.; Herron, N.; Thorn, D. L.; Wang, Y. Preparation and Characterization of Layered Lead Halide Compounds. J. Am. Chem. Soc. 1991, 113, (14) Mosconi, E.; Amat, A.; Nazeeruddin, M. K.; Graẗzel, M.; Angelis, F. D. First-Principles Modeling of Mixed Halide Organometal Perovskites for Photovoltaic Applications. J. Phys. Chem. C 2013, 117, (15) Onoda-Yamaura, N.; Matsuo, T.; Suga, H. Dielectric Study of CH 3 NH 3 PbX 3 (X = Cl, Br, I). J. Phys. Chem. Solids 1992, 53, (16) Mitzi, D. B.; Feild, C. A.; Harrison, W. T. A.; Guloy, A. M. Conducting Tin Halides with a Layered Organic-Based Perovskite Structure. Nature 1994, 369, (17) Mitzi, D. B.; Wang, S.; Feild, C. A.; Chess, C. A.; Guloy, A. M. Conducting Layered Organic-Inorganic Halides Containing <110>- Oriented Perovskite Sheets. Science 1995, 267, (18) Mitzi, D. B.; Feild; Schlesinger, Z.; Laibowitz, R. B. Transport, Optical, and Magnetic Properties of the Conducting Halide Perovskite CH 3 NH 3 SnI 3. J. Solid State Chem. 1995, 114, (19) Clark, S. J.; Donaldson, J. D.; Harvey, J. A. Evidence for the Direct Population of Solid-State Bands by Non-Bonding Electron Pairs in Compounds of the Type CsMIIX 3 (MII= Ge, Sn, Pb; X = Cl, Br, I). J. Mater. Chem. 1995, 5, (20) Song, I.; J., H.; Im, J.; Androulakis, J.; Malliakas, C. D.; Li, H.; Freeman, A. J.; Kenney, J. T.; Kanatzidis, M. G. CsSnI3: Semiconductor or Metal? High Electrical Conductivity and Strong Near- Infrared Photoluminescence from a Single Material. High Hole Mobility and Phase-Transitions. J. Am. Chem. Soc. 2012, 134, (21) Takahashi, Y.; Hasegawa, H.; Takahashi, Y.; Inabe, T. Hall Mobility in Tin Iodide Perovskite CH3NH3SnI3: Evidence for a Doped Semiconductor. J. Solid State Chem. 2013, 205, (22) Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Semiconducting Tin and Lead Iodide Perovskites with Organic Cations: Phase Transitions, High Mobilities, And near-infrared Photoluminescent Properties. Inorg. Chem. 2013, 52, (23) Moulder, J. F. Handbook of X-ray Photoelectron Spectroscopy; Perkin-Elmer Corp.: Eden, Prairie, (24) Kukihara, K.; Ogomi, Y.; Shen. Q.; Toyota, T.; Yoshino, K.; Pandey, S. S.; Hayase, S. All-Solid Dye-Sensitized Solar Cells Consisting of Passivated Metal Oxide Semiconductor. MRS Fall Meeting and Exhibit, Boston, MA, Dec 16, 2013; No. Y5.01. (25) Kukihara, K.; Ogomi, Y.; Shen, Q.; Saito, T.; Yoshino, K.; Pandey, S. S.; Toyoda, T.; Momose, H.; Hayase, S. All Solid

8 The Journal of Physical Chemistry s Dye-Sensitized Solar Cells with Perovskite Material. PVSEC-23 Proceedings, Taipei, Taiwan, Oct 28 Nov 1, 2013, 5-O-11. (26) Ogomi, Y.; Kukihara, K.; Shen, Q.; Toyoda, T.; Yoshino, K.; Pandey, S. S.; Momose, H.; Hayase, S. Control of Charge Dynamics by Charge Separation Interface for All-Solid Perovskite Sensitized Solar Cells. ChemPhysChem. 2014, in press. (27) Ogomi, Y.; Sakaguchi, S.; Kado, T.; Kono, M.; Yamaguchi, Y.; Hayase, S. Ru Dye Uptake under Pressurized CO 2 Improvement of Photovoltaic Performances for Dye-Sensitized Solar Cells. J. Electrochem. Soc. 2006, 153, A2294 A2297. (28) ASTM G Reference Spectra Derived from SMARTS v ASTMG173.html. (29) Nayak, P. K.; Cahen, D. Updated Assessment of Possibilities and Limits for Solar Cells. Adv. Mater. 2014, DOI: / adma

9 1 Supporting Information CH 3 NH 3 Sn x Pb (1-x) I 3 Perovskite Solar Cells Covering up to 1060 nm Yuhei Ogomi, *, Atsushi Morita, Syota Tsukamoto, Takahiro Saitho, Naotaka Fujikawa, Shen Qing,, Taro Toyoda,, Kenji Yoshino,, Shyam S. Pandey, Tingli Ma, and Shuzi Hayase, *,, Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, 2-4 Hibikino, Wakamatsu-ku, Kitakyushu, Fukuoka , Japan Graduate School of Informatics and Engineering, University of Electro-Communications, Chofugaoka, Chofu, Tokyo , Japan. Department of Electrical and Electronic Engineering, University of Miyazaki, 1-1, Gakuen Kibanadai Nishi, Miyazaki, , Japan CREST, Japan Science and Technology Agency (JST), Honcho Kawaguchi, Saitama , Japan ogomi@life.kyutech.ac.jp; hayase@life.kyutech.ac.jp

10 2 Experimental Methods PbI2 (Purity: %), SnI2, and regioregular poly(3-hexylthiophene-2,5-diyl) (P3HT) were purchased from Sigma Aldrich without further purification. Cells were fabricated by the following process. F-doped SnO2 layered glass (FTO glass, Nippon Sheet Glass Co. Ltd) was patterned by using Zn and 6N HCl aqueous solution. On this patterned FTO glass, titanium diisopropoxide bis(acetylacetonate) solution in ethanol were sprayed at 300 C to prepare a compact TiO2 layer. The substrate was dipped in 40 mm solution of TiCl4 in water for 30 min. followed by baking at 500 C for 20 min. A porous TiO2 layer was fabricated by spin-coating TiO2 paste (PST-30NRD, JGC Catalysts and Chemicals Ltd.) in ethanol (TiO2 paste: ethanol = 2:7 weight ratio), followed by heating the substrate at 550 C for 30 min. Perovskite layer and hole conductor layer were prepared in N2 atmosphere. A mixture of SnI2, PbI2, and CH3NH3I (x : 1-x : 1.0 / mol ratio) in dimethylformamide (DMF) (40 wt %) was spin-coated, followed by baking the substrate at 70 C for 30 min. In this report, perovskite materials are abbreviated as CH3NH3SnxPb(1-x)I3 depending of the mixing ratio. Then, a solution of regioregular poly(3- hexylthiophene-2,5-diyl) (P3HT) in chlorobenzene 15 mg/ ml) was spin coated on the perovskite layer and the substrate was hold at room temperature for 1 h under N2 atmosphere. Finally, Ag and Au electrodes were fabricated by a vacuum deposition method. Ag was evaporated at 0.5 Å/sec under 2.0 x 10-5 Pa followed by Au evaporation at Å/sec on a hole transport layer without exposed to air. These layer thicknesses were measured by focus variation with white light interferometer Nikon BW-502. Ag and Au thicknesses were monitored by quartz crystal microbalance. Cells without sealing were used for evaluation of photovoltaic performances. Current-voltage curves were measured in air by using AM 1.5G 100mW/cm 2 irradiance two light sources solar simulator (CEP-2000SRR, Bunkoukeiki Inc). Photomask was used in all cells for measurement of photovoltaic performances. The cell and mask sizes were described in each figure and table. I-V curves varied depending on the sweep direction and scan speed of voltage. When voltage

11 3 was swept from + to -, charge-up occurred and fill factor increased, leading to higher efficiency. Therefore, we swept the voltage from to + direction with 100 msec delay so as to retard efficiency increase caused by the charge-up phenomena. Incident photon to current efficiency (IPCE) was measured by using CEP-2000SRR (Bunko Keiki) equipped with 300 W Xe lamp. Monochromatic light was exposed by DC mode and the current was taken every 100msec after the light exposure (10 nm interval). The monochromatic light was adjusted to 1 x mw/cm 2 which was monitored by Si photodiode. IPCE was calculated by the following equation. IPCE = Ne/Np where Ne and Np stand for exposure light dose and photocurrent respectively. The series resistance (Rs) and shunt resistant (Rsh) were measured from current-voltage curves for solar cells. Rs (V/I) was obtained by the slope near I=0 and Rsh (V/I) was taken from the slope near V=0. Least-squares method was used for the calculation. The X-ray diffraction patterns for glass/tio2/ch3nh3snxpb(1-x)i3 samples were measured by RINT-Ultima III, Rigaku. The valence band level for the Sn/Pb perovskite on porous TiO2 and P3HT film on a glass substrate (not device structure) were measured by using Photo-Electron Spectroscopy in Air (PESA) ( AC3, RIKEN-KEIKI ). Conduction band level was calculated by taking the absorption spectrum edge as the band gap and the band gap energy was added to the valence band level measured by AC3. Energy levels for spiro-ometad was taken from the previous paper. 1 References (1) 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. Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591.