Supporting Information

Size: px

Start display at page:

Download "Supporting Information"

Anastasia Wells
6 years ago
Views:

1 Supporting Information Eaton et al /pnas Additional Characterization and Simulation of CsPbX 3 Nanowires and Plates Atomic Force Microscopy Measurements. Atomic force microscopy (AFM) images of the nanowires were acquired using an Asylum MFP 3D in tapping mode. Energy-Dispersive Spectroscopy Measurements. Quantification of the elemental ratios in a CsPbBr 3 was studied using an FEI Titan microscope operated at 200 kv at the National Center for Electron Microscopy in the Molecular Foundry at Lawrence Berkeley National Laboratory. Signal was collected using an FEI Super-X Quad windowless detector based on silicon drift technology controlled by Bruker Esprit software. Low-Temperature PL Measurements. Samples were held under vacuum ( torr) in a Janis Research Co. ST-500 microscopy cryostat and cooled to 6 K with liquid helium. The samples were excited with the fourth harmonic (266 nm) of a Spectra-Physics Quanta Ray INDI-40 neodymium-doped yttrium aluminum garnet laser (10-Hz, 8-ns pulse width). The PL emission was collected by a 50, 0.55-N.A. air objective (Nikon CF Plan) and the spectra were measured with the same UV-vis spectrometer as described in Materials and Methods, but using a 150-mm 1, 300-nm blaze grating. CsPbBr 3 Refractive Index Calculations. The calculations were based on density functional theory (42) within the generalized gradient approximation formulated by Perdew, Burke, and Ernzerhof (43) as implemented in the Vienna ab initio simulation package (VASP) code (44). The projector augmented wave pseudopotentials were used (45), and the valence wavefunctions were expanded using a plane-wave basis with an energy cutoff of 300 ev. AprimitivecellofCsPbBr 3 (20 atoms) in the orthorhombic phase was used (26), and the Brillouin zone was sampled with a k-point mesh. To calculate the frequency-dependent dielectric function, 200 empty bands were used. The calculated static dielectric constant (ω = 0) agrees with the result from the density functional perturbation theory calculation implemented in VASP. The frequency-dependent dielectric function was used to calculate the refractive indices, using the equations (46) nðνþ = 1 pffiffi e 1 ðνþ + e 1 ðνþ 2 + e 2 ðνþ 2 1=2 1=2, [1] 2 kðνþ = 1 pffiffi e 1 ðνþ + e 1 ðνþ 2 + e 2 ðνþ 2 1=2 1=2, [2] 2 where n and «1 are the real parts of the refractive index and relative dielectric constant, respectively, k and «2 are the imaginary parts of the refractive index and relative dielectric constant, respectively, and all are a function of the frequency, ν. The resulting refractive index values were used directly for the finite-difference eigenmode and finite-difference time-domain simulations below. Graphs of the real and imaginary refractive indices may be found in Fig. S11. Finite-Difference Eigenmode Simulations. The Lumerical MODE Solutions 7.6 software package was used to simulate the waveguide modes of 535-nm light in a CsPbBr 3 nanowire. The nanowire (n = 2.53) was simulated on fused silica (n = 1.46) (47) under vacuum (n = 1.00). The side length ratio was kept at 2:1 to approximate the rectangular cross-section of the as-grown nanowires. The nanowire size was incremented by 25 nm (based on width), and the effective refractive index was calculated for all sustainable modes. Finite-Difference Time-Domain Simulations. The Lumerical FDTD Solutions 8.12 software package was used to estimate the absorption cross-section of the 483 nm 265 nm 11.1-μm CsPbBr 3 nanowire studied in Fig. 2. As in the finite-difference eigenmode (FDE) mode simulation, the nanowire was simulated laying horizontal on fused silica surrounded by vacuum. The 400-nm totalfield scattered-field light source was positioned above the nanowire and the transverse electric and transverse magnetic polarizations averaged to acquire the absorption cross-section. Combined with the excitation fluence and beam profile, the nanowire carrier density was determined. Discussion of the Nanowire and Nanoplate Growth Mechanism During the initial stages of the reaction using 6 10 mg/ml CsBr in methanol, the film color transitioned rapidly to red, then gradually to yellow. We hypothesize that an intermediate CsPbI x Br 3 x forms initially, then evolves to pure CsPbBr 3 as confirmed by XRD patterns at different growth times (Fig. S2). The peaks corresponding to PbI 2 disappear within the first 2 min, indicating that PbI 2 dissolves/reacts rapidly. For the intermediate products from 2 10 min, the (001) peak was shifted 0.2 below that of pure CsPbBr 3 ; this is indicative of lattice expansion caused by the formation of CsPbI x Br 3 x. However, the final CsPbBr 3 nanowires produced by this method contain a minimal amount of iodide as the iodide content of the nanowires tested by energy-dispersive spectroscopy (EDS) was found to be below the 1% detection limit of the instrument. Notably, Zhu et al. reported intermediate CH 3 NH 3 PbI 2 Br products during the growth of single-crystal CH 3 NH 3 PbBr 3 nanostructures (20). Therefore, inorganic perovskite nanowires likely possess similar growth dynamics to hybrid perovskites. 1of7

$(B) The calculated effective refractive index for each mode as a function of nanowire width.$ $Light is no longer confined to the nanowire once the effective refractive index reaches that of the quartz substrate (n = 1.46).$

2 Fig. S1. FDE simulations of a CsPbBr 3 nanowire. (A) The six sustainable transverse modes for 540-nm light in a CsPbBr 3 nanowire with width 500 nm. (B) The calculated effective refractive index for each mode as a function of nanowire width. Light is no longer confined to the nanowire once the effective refractive index reaches that of the quartz substrate (n = 1.46). These simulations suggest that a CsPbBr 3 nanowire must have a width greater than 180 nm to act as an emission waveguide. Fig. S2. Composition and morphology evolution during nanowire growth at room temperature. (A) XRD patterns at different reaction times. The 24-h pattern is indexed to the CsPbBr 3 orthorhombic phase. SEM images at corresponding reaction states: (B) 1 min, (C) 10min,(D) 24h,(E) 34 h. (Scale bar, 1 μm.) 2of7

Fig. S3. CsPbBr 3 height and width distribution indicates rectangular cross-sections are favored. Data points determined by AFM (height) and SEM (width). Fig. S4.

3 Fig. S3. CsPbBr 3 height and width distribution indicates rectangular cross-sections are favored. Data points determined by AFM (height) and SEM (width). Fig. S4. Elemental distribution determined from EDS of a single-crystalline CsPbBr 3 nanowire. (A) Image of CsPbBr 3 nanowire and corresponding (B) Cs, (C)Pb, and (D) Br elemental maps. (E) EDS spectrum from nanowire in A. No detectable levels of iodide were observed indicating the nanowire is not alloyed. 3of7

4 Fig. S5. Full PL data from the CsPbBr 3 nanowire power-dependent study presented in Fig. 2F.(A) Emission spectra on a log scale, and (B) the peak center of the most intense mode as a function of excitation fluence. Fig. S6. Q-factor analysis of a CsPbBr 3 nanowire. (A) A linear plot of the emission spectrum at 10.3 μj cm 2 from Fig. 2E is shown with Gaussian fits of each lasing mode. The spontaneous emission has been subtracted for clarity and ease of fitting. (B) The same spectrum and fit plotted on a log scale showing the quality of the fit. Peak annotations correspond to the Q factor for the specific peak. 4of7

Fig. S7. Low-energy emission band in CsPbBr 3 nanowires. (A) Growth of the low-energy band near 530 nm with increasing femtosecond pulsed excitation fluence.

5 Fig. S7. Low-energy emission band in CsPbBr 3 nanowires. (A) Growth of the low-energy band near 530 nm with increasing femtosecond pulsed excitation fluence. (B) CsPbBr 3 nanowire emission at room temperature (290 K) and low temperature (6 K) also shows the growth of the low-energy band. (C) The lowtemperature emission spectrum may be fit to a pair of Gaussians (red) that are 69 mev apart. Fig. S8. SEM image of CsPbBr 3 nanowire after 1 h of continuous pulsed excitation above the lasing threshold while under nitrogen. (Scale bar, 1 μm.) 5of7

Fig. S9. Characterization of as-grown CsPbCl 3 nanowires. (A) SEM image of an individual CsPbCl 3 nanowire. (Scale bar, 5 μm.) (B) Magnified SEM image depicting rectangular end facets.

6 Fig. S9. Characterization of as-grown CsPbCl 3 nanowires. (A) SEM image of an individual CsPbCl 3 nanowire. (Scale bar, 5 μm.) (B) Magnified SEM image depicting rectangular end facets. (Scale bar,1 μm.) (C) XRD pattern of the growth substrate indexed to the tetragonal phase of CsPbCl 3.(D) Narrow PL emission centered at 418 nm observed from a single CsPbCl 3 nanowire. Fig. S10. PL spectrum of the CsPbBr 3 nanoplate in Fig. 4B at 56 μj cm 2 fit to 12 Gaussian functions. The spontaneous emission has been subtracted for clarity and ease of fitting. The mode spacing suggests Fabry Pérot-type lasing. 6of7

7 Fig. S11. Frequency-dependent refractive indices for the orthorhombic phase of CsPbBr 3.(A) Real and (B) imaginary parts of the refractive index. 7of7

Structural, optical, and electrical properties of phasecontrolled cesium lead iodide nanowires

Electronic Supplementary Material Structural, optical, and electrical properties of phasecontrolled cesium lead iodide nanowires Minliang Lai 1, Qiao Kong 1, Connor G. Bischak 1, Yi Yu 1,2, Letian Dou