Layered Mixed Tin-Lead Hybrid Perovskite Solar Cells with High Stability

Transcription

1 Supplementary information for: Layered Mixed Tin-Lead Hybrid Perovskite Solar Cells with High Stability Daniel Ramirez, a Kelly Schutt, b Zhiping Wang, b Andrew J. Pearson, c Edoardo Ruggeri, c Henry J. Snaith, b Samuel D. Stranks, c and Franklin Jaramillo a* a. Centro de Investigación, Innovación y Desarrollo de Materiales CIDEMAT, Facultad de Ingeniería, Universidad de Antioquia UdeA, Calle 70 No , Medellín, Colombia. b. Department of Physics, University of Oxford, Clarendon Laboratory, Parks Road, Oxford, OX13PU, United Kingdom. c. Cavendish Laboratory, Department of Physics, University of Cambridge, JJ Thomson Avenue, Cambridge CB3 0HE, UK Experimental Methods (t-ba) 2 (FA (1-x) Cs x ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 and (t-ba) 2 (FA 0.85 Cs 0.15 ) n-1 (Pb 0.6 Sn 0.4 ) n I 3n+1 precursor solutions. The (t-ba) 2 (FA (1-x) Cs x ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 precursor solution was prepared by mixing Formamidinium iodide (FAI) (Dyesol), cesium iodide (CsI) (Alfa Aesar, 99.99%), lead iodide (PbI 2 ) (TCI, %) tin iodide (SnI 2 ) (Sigma-Aldrich, Anhydrous, 99.99%) and tin fluoride (SnF 2 ) with a molar ratio of 2:2(1- x):2x:1.8:1.2:0.24 = t-bu:fai:csi:pbi 2 :SnI 2 :SnF 2. Note that the 0.24 moles of SnF 2 corresponds to a 20% excess respect to SnI 2, in order to prevent solution oxidation. The mixed powders were dissolved in anhydrous N,N-dimethylformamide (DMF) (Sigma-Aldrich, 99.8%) to give a 69 %w/v solution when x=0. The solutions were stirred overnight in a nitrogen-filled glove box, prior use. (t-ba) 2 (FA 0.85 Cs 0.15 ) n-1 (Pb 0.6 Sn 0.4 ) n I 3n+1 precursor solution was prepared the same solution concentration was used, but as the FA/Cs ration was kept fixed to 85/15, then the powders mixing was modulated according to the molar ratio of 2:0.85(n-1):0.15(n-1):0.6n:0.4n = t- BU:FAI:CsI:PbI 2 :SnI 2. 20% excess of SnF 2 respect to SnI 2 was also added to all solution. These solutions were also stirred overnight in a nitrogen-filled glove box, prior use. t-ba/fa/cs Solar cell fabrication ITO substrates were sonicated in water (5 min), Acetone (5 min) and Isopropanol (IPA) (5 min), then dried and O 2 -plasma treated (10 min). PEDOT:PSS (Clevios, P VP AI 4083) was diluted 2:1 by volume (methanol: PEDOT:PSS) and then spin coated in a drybox at 4000rpm, with a 4000rpm/s ramp during 40s. The substrates were then annealed at 150 C for 10 min and then transferred to a glove box, where they were annealed again at 120 C for 10 min before depositing the perovskite. The substrates were allowed to cool down before depositing the precursor solutions using a two-step program, 1000 rpm and 4000 rpm for 10 and 15 s, respectively. 18 s after the program started, 200 µl of anhydrous toluene (Sigma-Aldrich, 99%) was dropped, forming immediately a dark brown perovskite film. The substrates were then annealed at 150 C for 20 minutes. Filtered PCBM solution (20 mg/ml in 3:1 by volume Chlorobenzene:(1,2- Dichlorobenzene)) was deposited by dynamic spin coating at 2000 rpm during 30 s. The substrates where then place on a hotplate at 80 C for 5 minutes. Then a solution of 0.5mg BCP in 1ml of IPA was deposited at 4000 rpm. Finally, to complete the devices, 100 nm silver electrode was thermally evaporated.

2 t-ba/fa/cs film preparation Films for SEM and AFM were deposited on top of PEDOT and ITO, while samples for Absorption, PL and TRPL were deposited on top of glass. The same procedure as the one for the solar cell fabrication was carried out, including cleaning and deposition of the precursor solutions. Films were fabricated at different annealing temperatures (100 C, 130 C and 150 C) for 20 minutes, depending on the characterization performed. 3D (FA 0.85 Cs 0.15 )(Pb 0.6 Sn 0.4 )I 3 Solar cell fabrication The same procedure as for the t-ba/fa/cs solar cell was followed. But in this case the precursor solution did not contain t-ba. A mixture of DMF:DMSO (4:1 volume) instead of only DMF was used, because the films containing only DMF formed cracks after toluene was dropped. Current-voltage measurements Solar cells were measured with an Abet Class AAB solar simulator under simulated AM 1.5 sunlight at 100 mw cm -2 irradiance, calibrated by an NREL-calibrated KG5 filtered silicon reference cell. The mismatch factor was calculated at < 1%. J-V curves were recorded with a 2400 Series Sourcemeter by Keithly Instruments. The solar cell active area was cm -2. External quantum efficiency (EQE) EQE spectra were evaluated via custom-built Fourier transform photocurrent spectroscopy based on the Bruker Vertex 80v Fourier transform spectrometer. A Newport AAA sun simulator was used as the light source and the light intensity was calibrated with a Newport-calibrated reference silicon photodiode Ultraviolet-visible absorption (Uv-vis) The steady-state absorption spectra were acquired with a Perkin-Elmer Lambda 1050 UV/Vis/NIR spectrophotometer. Photoluminescence spectroscopy Steady-state and time-resolved PL measurements were performed using a time resolved singlephoton counting setup (FluoTime 300, PicoQuant GmbH) using a 634 nm pulsed laser as excitation at frequencies between 2 MHz and 5MHz. Scanning electron microscopy (SEM) The morphology of perovskite films was investigated using a SEM (Hitachi S-4300) X-ray diffraction (XRD) For the Mixed tin-lead perovskites a Rigaku SmartLab X-ray diffractometer with Cu Kα radiation was used. GIWAXS Samples were measured in a grazing-incidence geometry at beamline I07 of the Diamond Light Source (Harwell, UK). A beam energy of 10 kev was used with samples housed in a custom-built

3 chamber during measurement. Samples were tilted at 0.3 into the path of the incident X-rays. X- ray scatter was measured using a Pilatus 2M detector, calibrated using silver behenate powder. Collected data were analyzed using the DAWN software package ( AFM AFM images were obtained in a MFP-3D infinity from Oxford Instruments. The samples were sealed in a hermetic Fluid cell inside the glove box in order to have an inert atmosphere during the measurements. Stability test Unencapsulated devices were kept inside a nitrogen-filled glovebox over 910 h for the operational performance test. These devices were air exposed for around 10 min during photovoltaic characterization. Other set of unencapsulated devices were aged under open-circuit conditions, under full-spectrum simulated AM1.5, 76mA/cm 2 irradiance, using an Atlas SUNTEST XLS+ (1,700W air-cooled xenon lamp). We note that the light source is pulsed at 100Hz frequency and we do not apply any additional ultraviolet filter during the ageing process. The chamber is air-cooled to have a temperature around 60 C as indicated by a black standard temperature control unit mounted inside. We do not have control on the humidity but monitored the laboratory humidity, which around 50% relative humidity at room temperature, during the course of the ageing.

4 Experimental results To optimize the FA/Cs ratio as basic structure for latter investigations on the dimensionality induced by the organic cation we prepared films with Cs as partial replacement of FA to form (t- BA) 2 (FA (1-x) Cs x ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 films ranging from x=0.05 to x=0.20. As shown in Figure S1a-c, when more cesium was incorporated, the grain size increased to around 600 nm for x=0.15, while preserving the same film morphology. This was refelcted in an improved absorption, steady state PL and TRPL lifetime. In contrast, the evident cracks in x=0.20 had a negative impact in both, steady state PL and TRPL lifetime. The latter suggests that adding 15%Cs has a benefitial effect for both, improving film morphology and reducing non-radiative recombination. Figure S1: (t-ba) 2 (FA (1-x) Cs x ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 thin films. SEM images for a) x=0.05, b) x=0.15 and c) x=0.20. Scale bar corresponds to 2µm. d) Absorption, e) steady state PL and f) TRPL for x=0, 0.05, 0.10, 0.15 and 0.20.

5 Figure S2. Comparative morphology of a-c) (t-ba) 2 (FA) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 and d-f) (t- BA) 2 (FA 0.95 Cs 0.05 ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 thin films annealed at different temperatures. Scale bar corresponds to 4μm. Intensity (a.u.) 150 C 130 C 100 C θ (degree) Figure S3. Comparative XRD pattern for (t-ba) 2 (FA 0.95 Cs 0.05 ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 thin films annealed at 100 C, 130 C and 150 C.

6 Figure S4. (t-ba) 2 (85FA15Cs) n-1 (0.6Pb0.4Sn) n I 3n+1 thin films. SEM images for a) n=2, b) n=3, c) n=5 and d) n=9 in backscattered mode.

7 Figure S5. Tauc plots of (t-ba) 2 (FA 0.85 Cs 0.15 ) n-1 (Pb 0.6 Sn 0.4 ) n I 3n+1 thin films

8 Figure S6. Forward and reverse scans for hybrid perovskite solar cells using (t-ba) 2 (FA 0.85 Cs 0.15 ) n- 1(Pb 0.6 Sn 0.4 ) n I 3n+1 as active layer.

9 Figure S7. Complementary photovoltaic data for hybrid perovskite solar cells using (t- BA) 2 (FA 0.85 Cs 0.15 ) n-1 (Pb 0.6 Sn 0.4 ) n I 3n+ as active layer. a) J sc, b) V oc and c) FF.

10 In Figure S8a, we show current density voltage (J-V) characteristics for all the cesium dependent samples. Average PCE in Figure S6b went from 3.2% for x=0 up to 6.3% for both, x=0.10 and x=0.15 with the highest efficiency of 8.5% in the latter case, result of a short-circuit current (J sc ) of 21.3 ma/cm 2, fill factor (FF) of 0.60, and open-circuit voltage (V oc ) of 0.64V. The previous findings for x=0.20 suggested a low quality of the film, which led to a lower average PCE of 3.9%, limited by the low photocurrent. Other photovoltaic parameters are summarized in Table S1. Figure S8. Photovoltaic performance of cesium dependent solar cells using (t-ba) 2 (FA (1-x) Cs x ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 as active layer. a) JV curves, b) power conversion efficiency box plot, c) SPO and d) EQE. Table S1. Cesium dependent solar cell performance parameters determined from J V curves Device PCE (%) Jsc (ma/cm 2 ) V oc (V) FF (t-ba) 2 (FA) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 (x = 0) Average 3.2± ± ± ±0.03 Best (t-ba) 2 (FA 0.95 Cs 0.05 ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 (x = 0.05) Average 5.3± ± ± ±0.04 Best (t-ba) 2 (FA 0.90 Cs 0.10 ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 (x = 0.10) Average 6.3± ± ± ±0.04 Best (t-ba) 2 (FA 0.85 Cs 0.15 ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 (x = 0.15) Average 6.3± ± ± ±0.04 Best (t-ba) 2 (FA 0.80 Cs 0.20 ) 2 (Pb 0.6 Sn 0.4 ) 3 I 10 (x = 0.20) Average 3.9± ± ± ±0.06 Best

11 Figure S9. Evolution of the photovoltaic parameters a) J sc, b) Fill Factor and c) V oc over time for unencapsulated devices stored under nitrogen atmosphere.

12 Figure S10. a) Photovoltaic parameters and b) SPO of hybrid perovskite solar cells with 3D (FA 0.85 Cs 0.15 )(Pb 0.6 Sn 0.4 )I 3 as active layer.