Quantum-Cutting Ytterbium-Doped CsPb(Cl 1-x Br x ) 3 Perovskite Thin Films with Photoluminescence Quantum Yields over 190%

Transcription

1 Supporting Information for: Quantum-Cutting Ytterbium-Doped CsPb(Cl 1-x Br x ) 3 Perovskite Thin Films with Photoluminescence Quantum Yields over 190% Daniel M. Kroupa, Joo Yeon Roh, Tyler J. Milstein, Sidney E. Creutz, and Daniel R. Gamelin* Department of Chemistry, University of Washington, Seattle, Washington *Gamelin@chem.washington.edu EXPERIMENTAL DETAILS General Remarks: All materials were used as received unless otherwise specified. All procedures were carried out under ambient conditions unless otherwise noted. Materials: Cesium chloride (CsCl; >99%) was purchased from TCI. Cesium bromide (CsBr; %), lead(ii) chloride (PbCl 2 ; 99%), lead(ii) bromide (PbBr 2 ; 98+%), and ytterbium(iii) bromide hydrate (YbBr 3 xh 2 O; 99.99%) was purchased from Alfa Aesar. Ytterbium(III) chloride hexahydrate (YbCl 3 6H 2 O; 99.9%) was purchased from Strem Chemicals. Methanol (MeOH; 99.9%) was purchased from Sigma Aldrich. Dimethylsulfoxide (DMSO; 99.9%) was purchased from Fischer Scientific. Concentrated detergent solution was purchased from Starna Cells, Inc. and diluted in deionized water. Substrate Preparation: 12.5 mm x 12.5 mm glass substrates (1 mm thick) were cut from 75 x 25 mm microscope slides (VWR International). Substrates were cleaned by sequential rounds of 5 min sonication in i) warm, aqueous detergent solution, ii) isopropyl alcohol, and iii) acetone. Cleaned substrates were then treated with O 2 plasma for 15 min and used immediately. Precursor Solution Preparation: PbCl 2 /DMSO: 2.4 g (8.6 mmol) PbCl 2 was dissolved in 10 ml DMSO with heating and stirring. The solution was passed through a 0.2 µm PTFE filter prior to deposition. CsCl/MeOH: CsCl was completely dissolved in MeOH with different concentrations ranging from 0.05 M to 0.30 M. CsCl/YbCl 3 6H 2 O/MeOH: CsCl and YbCl 3 6H 2 O were dissolved in MeOH at a fixed 0.25 M CsCl concentration, with [Yb 3+ ]:[Cs + ] ratios ranging from At the largest [Yb 3+ ]:[Cs + ] ratios, some of the metal-halide salt remained undissolved after prolonged heating and sonication, likely resulting in an increasing deficiency of dissolved cesium with increasing nominal [Yb 3+ ]:[Cs + ] ratio. This is an unavoidable consequence of the poor solubility of CsCl in MeOH under these specific experimental conditions. Mixed-halide solutions were prepared in a similar fashion as the chloride samples, but the CsX, PbX 2, and YbX 3 xh 2 O salts were mixed in the desired [Cl - ]:[Br - ] ratio prior to dissolution. Thin-Film Deposition: Glass substrates and metal-halide salt solutions were heated to 70 C on a hot plate. A substrate was quickly transferred to a spincoater, and 30 µl of the PbCl 2 /DMSO solution was dynamically spun at 6000 rpm for 35 sec. The substrate/pbcl 2 film was annealed at 100 C for 5 min. PbCl 2 films were 80 ± 2.5 nm thick by profilometry. To prepare undoped CsPbCl 3 films, both the substrate/pbcl 2 film and CsCl/MeOH solution were heated to 70 C on a hot plate. The substrate/pbcl 2 film was quickly transferred to a spincoater, and 100 µl of the CsCl/MeOH solution was dynamically spun at 6000 rpm for 35 sec. The substrate/pbcl 2 /CsCl film was annealed at 250 C for 10 min. The S-1

2 undoped CsPbCl 3 films were 97 ± 9 nm thick by profilometry. To prepare Yb 3+ -doped CsPbCl 3 films, both the substrate/pbcl 2 film and CsCl/YbCl 3 6H 2 O/MeOH solution were heated to 70 C on a hot plate. The substrate/pbcl 2 film was quickly transferred to a spincoater, and 100 µl of the CsCl/YbCl 3 6H 2 O/MeOH solution was dynamically spun at 6000 rpm for 35 sec. The substrate/pbcl 2 /CsCl/YbCl 3 6H 2 O film was annealed at 250 C for 10 min. The Yb 3+ -doped CsPbCl 3 films were 150 ± 31 nm thick by profilometry. Mixed-halide solutions were prepared in a similar fashion as the chloride samples, except that the corresponding mixed [Cl - ]:[Br - ] precursor solutions were used. All films were stored under dry conditions in the dark. Absorption: Absorption spectra were measured using an Agilent Cary 5000 spectrometer operating in transmission mode. Photoluminescence (PL): Steady-state, room-temperature photoluminescence (PL) data were measured using a 375 nm Thorlabs LED excitation source and a LN 2 -cooled silicon CCD for detection in front-face sample-excitation geometry. All steady-state PL spectra were corrected for instrument response and sample optical density at 375 nm. Absolute PL quantum-yield measurements (PLQY) were performed with the thin films positioned at the 180 port of a 5.3 inch teflon-based integrating sphere. The samples were directly excited with a 375 nm LED, and the samples were slightly angled so that the primary and secondary reflections were positioned on the integrating sphere wall. Light from the sphere was fiber-coupled to a home-built luminescence spectrometer equipped with a LN 2 -cooled silicon CCD for detection. All spectra were corrected for integrating sphere, fiber, lens, grating, and detector spectral response using a radiometric calibration lamp (Ocean Optics, LS-1-Cal). The emission correction curve generated using the calibrated lamp was further reinforced and extended in the UV-Vis spectral region using secondary emission standards. 1 PLQYs were calculated using: PLQY = N!" N!"# = I!"#$%& λ I!"# λ dλ E!"# λ E!"#$%& λ dλ where I indicates the spectrally corrected intensity of the emitted light, E indicates the spectrally corrected intensity of the excitation light, sample indicates measurements of samples, and ref indicates measurements of a reference glass substrate. We continuously calibrated our PLQY setup using well-characterized dye emission standard solutions and found good agreement with literature values: Rhodamine 6G Measured (91.9%); Literature (90-92%). 2 IR140 Measured (19.9%); Literature (20.0%). 3 Transient Absorption (TA): TA data were collected using an Ultrafast Systems Helios spectrometer. The light source consisted of a Coherent Libra amplified Ti:Sapphire laser operating at 1 khz and 800 nm, with ~3.9 mj maximum pulse energy and ~100 fs pulse width. Pump pulses (365 nm) were generated using a Coherent OPerA Solo Ultrafast optical parametric amplifier and passed through a 365nm +/- 10nm bandpass filter. Pump and probe beams were directed into the entrance ports of the Helios spectrometer, and the pump beam was attenuated using neutral density filters. Broadband probe pulses ( nm) were generated by focusing the 800 nm pump beam onto a translating CaF 2 crystal. Time delays up to ~5.5 ns were achieved via an optical delay line. X-ray Diffraction (XRD): X-ray diffraction of thin films on glass substrates was measured using a Bruker D8 Discover equipped with a Pilatus 100K large-area 2D detector. Samples were irradiated using Cu Kα radiation (50000 mw). S-2

3 Scanning Electron Microscopy (SEM): Images were captured using a FEI Sirion SEM operating at 5 kv. Profilometry: Film thickness was measured using a Bruker OM-DektakXT profilometer. ADDITIONAL DATA Figure S1. (a) Absorption spectra of PbCl 2 (gray) and CsPbCl 3 thin films made with increasing Cs + loading (from red to violet). (b) Corresponding steady-state photoluminescence spectra; λ ex = 375 nm and (c) X- ray diffraction patterns for the undoped films. Absorption spectra and XRD patterns are offset for clarity. Arrows indicate increasing CsCl deposition. Red = 0.05 M CsCl/MeOH; orange = 0.10 M CsCl/MeOH; yellow = 0.15 M CsCl/MeOH; green = 0.20 M CsCl/MeOH; blue = 0.25 M CsCl/MeOH; violet = 0.30 M CsCl/MeOH. Figure S2. Representative absorption (dashed lines) and PL (solid lines) spectra of (a) undoped and (b) Yb 3+ -doped CsPbCl 3 thin films. S-3

4 Figure S3. Representative low-magnification (left) and high-magnification (right) SEM images of (a) a preformed PbCl2 film, (b) an undoped CsPbCl3 film, (c) an Yb3+-doped CsPbCl3 film with 0.4 [Yb3+]:[Cs+] loading, and (d) an Yb3+-doped CsPbCl3 film with 0.7 [Yb3+]:[Cs+] loading, highlighting a two-dimensional phase. S-4

5 Figure S4. High-resolution SEM images of (a) an undoped CsPbCl3 film and (b) an Yb3+-doped CsPbCl3 film with 1.0 [Yb3+]:[Cs+] loading. Table S1. Elemental analysis data for undoped and Yb3+-doped CsPbCl3 perovskite thin films collected using energy-dispersive X-ray spectroscopy (EDS). Data are shown as atomic %. %Yb3+ doping calculated as [Yb3+]/([Yb3+]+[Pb2+]). Nominal [Yb3+]:[Cs+] % Cs % Pb % Cl % Yb ND S-5 % Yb Doping

6 Figure S5. Near-infrared photoluminescence quantum yield (NIR PLQY) plotted as a function of % Yb 3+ doping calculated as [Yb 3+ ]/([Yb 3+ ]+[Pb 2+ ]) using the analytical data in Table S1. S-6

7 Figure S6. Transient-absorption spectra of (a) undoped, (b) 0.3 [Yb 3+ ]:[Cs + ] loading, and (c) 0.8 [Yb 3+ ]:[Cs + ] loading CsPbCl 3 films averaged over the pump-probe delay ranges indicated in the legend in panel (b). Black traces are the corresponding normalized ground-state absorption spectra of the same films. Figure S7. Transient absorption kinetics for undoped (black), 0.3 [Yb 3+ ]:[Cs + ] loading (green), and 0.8 [Yb 3+ ]:[Cs + ] loading (blue) CsPbCl 3 films plotted as Δα/α. The inset shows a representative transient absorption spectrum overlaid with the ground-state absorption spectrum for an undoped CsPbCl 3 film. S-7

8 Figure S8. XRD patterns for (a) CsPb(Cl 1-x Br x ) 3 and (b) Yb 3+ -doped CsPb(Cl 1-x Br x ) 3 thin films with varying ratios of nominal [Cl - ]:[Br - ]. Panel (a) is reproduced from the main text. The anion content value, x, for each film is estimated from the lattice parameter using Vegard s law, and error bars are on the order of ±0.03 based on analyzing the lattice parameter at different reflection angles. XRD patterns are offset for clarity. S-8

9 Figure S9. (a) Absorption and (b) excitonic PL spectra of CsPb(Cl 1-x Br x ) 3 thin films prepared with varying nominal ratios of [Cl - ]:[Br - ] during synthesis. (c) Absorption and (d) excitonic PL of Yb 3+ -doped CsPb(Cl 1- xbr x ) 3 thin films prepared with varying nominal ratios of [Cl - ]:[Br - ] and a fixed [Yb 3+ ]:[ Cs 3 ] of 0.8 during synthesis. Panels (a) and (b) are reproduced from the main text. Absorption spectra are offset for clarity. S-9

10 Figure S10. PL spectra showing photoinduced anion segregation in mixed-halide CsPb(Cl 1-x Br x ) 3 thin films with varying [Cl - ]:[Br - ] ratios. Increasing exposure to high-power excitation (red to violet traces, λ ex = 375 nm) leads to a PL peak shift and in some cases splitting. The nominal [Cl - ]:[Br - ] ratio used during film synthesis is indicated in each panel. For reference, the black traces show magnified PL spectra of the same films collected using very low excitation fluence; these spectra do not change with time. Supporting Information References 1. Gardecki, J. A.; Maroncelli, M., Set of Secondary Emission Standards for Calibration of the Spectral Responsivity in Emission Spectroscopy. Appl. Spec., 1998, 52, Würth, C.; González, M. G.; Niessner, R.; Panne, U.; Haisch, C.; Genger, U. R., Determination of the Absolute Fluorescence Quantum Yield of Rhodamine 6G with Optical and Photoacoustic Methods Providing the Basis for Fluorescence Quantum Yield Standards. Talanta 2012, 90, Hatami, S.; Wurth, C.; Kaiser, M.; Leubner, S.; Gabriel, S.; Bahrig, L.; Lesnyak, V.; Pauli, J.; Gaponik, N.; Eychmuller, A.; Resch-Genger, U., Absolute Photoluminescence Quantum Yields of IR26 and IR-Emissive Cd 1-x Hg x Te and PbS Quantum Dots - Method- and Material-Inherent Challenges. Nanoscale 2015, 7, Li, J.; Zhang, H.; Wang, S.; Long, D.; Li, M.; Guo, Y.; Zhong, Z.; Wu, K.; Wang, D.; Zhang, T., Synthesis of All-Inorganic CsPb2Br5 Perovskite and Determination of its Luminescence Mechanism. RSC Adv. 2017, 7, Kim, M. K.; Jo, V.; Ok, K. M., New Variant of Highly Symmetric Layered Perovskite with Coordinated NO3 Ligand: Hydrothermal Synthesis, Structure, and Characterization of Cs2PbCl2(NO3)2. Inorg. Chem. 2009, 48, S-10