Designing complex Pb3SBrxI4−x chalcohalides: tunable emission semiconductors through halide-mixing

Chalcohalides are desirable semiconducting materials due to their enhanced light-absorbing efficiency and stability compared to lead halide perovskites. However, unlike perovskites, tuning the optical properties of chalcohalides by mixing different halide ions into their structure remains to be explored. Here, we present an effective strategy for halide-alloying Pb3SBrxI4−x (1 ≤ x ≤ 3) using a solution-phase approach and study the effect of halide-mixing on structural and optical properties. We employ a combination of X-ray diffraction, electron microscopy, and solid-state NMR spectroscopy to probe the chemical structure of the chalcohalides and determine mixed-halide incorporation. The absorption onsets of the chalcohalides blue-shift to higher energies as bromide replaces iodide within the structure. The photoluminescence maxima of these materials mimics this trend at both the ensemble and single particle fluorescence levels, as observed by solution-phase and single particle fluorescence microscopy, respectively. These materials exhibit superior stability against moisture compared to traditional lead halide perovskites, and IR spectroscopy reveals that the chalcohalide surfaces are terminated by both amine and carboxylate ligands. Electronic structure calculations support the experimental band gap widening and volume reduction with increased bromide incorporation, and provide useful insight into the likely atomic coloring patterns of the different mixed-halide compositions. Ultimately, this study expands the range of tunability that is achievable with chalcohalides, which we anticipate will improve the suitability of these semiconducting materials for light absorbing and emission applications.

Synthesis.All syntheses were performed in air under standard atmospheric conditions.Pb3SBrI3 was prepared by stirring Pb(SCN)2 (0.2 mmol), PbI2 (0.6 mmol) and PbBr2 (0.2 mmol) in a mixture of ODE (10 mL, 31 mmol), oleic acid (0.25 mL, 0.8 mmol) and oleylamine (0.25 mL, 0.8 mmol) in a 100 mL round bottom flask at 110 °C for 5 min until the solids were completely dissolved or well dispersed in solution.The temperature was then raised to 180 °C and the reaction mixture remained at that temperature for 90 min before cooling to room temperature by removing the heating mantle.Various mixed-halide compositions were prepared in a similar manner by adjusting the relative concentrations and conditions to: 0.4 mmol PbI2, 0.2 mmol PbBr2, 180 °C, 60 min; 0.2 mmol PbI2, 0.2 mmol PbBr2, 180 °C, 60 min; or 0.2 mmol PbI2, 0.4 mmol PbBr2, 200 °C, 60 min.Purification.Crude solutions of the chalcohalides were first suspended in hexanes (5 mL) and methanol (5 mL), then centrifuged at 4500 rpm for 5 min.After discarding the supernatant, the pellet was resuspended in hexanes and methanol (5 mL of each) and centrifuged again to remove excess oleylamine and oleic acid ligands.This process was repeated until the remaining supernatant was colorless after centrifugation.
Structural Characterization.Powder X-ray diffraction (XRD) was measured on a Rigaku Ultima IV diffractometer (40 kV, 44 mA) using Cu Kα radiation on a zero-background quartz sample holder.Rietveld refinements of the XRD patterns were performed using the GSAS-II software package. 1 Scanning electron microscopy (SEM) images were acquired on a JEOL JSM-IT200 scanning electron microscope.Transmission electron microscopy (TEM) imaging was performed on a JEOL 2100 scanning transmission electron microscope.Samples were prepared by drop casting dilute solution in hexanes onto a carbon-coated 200 mesh copper grid.
Solid-State 207 Pb NMR Spectroscopy.A majority of solid-state NMR experiments were performed on a Bruker 14.1 T [n0( 1 H) = 600 MHz] wide bore magnet equipped with a Bruker AVANCE NEO console and a 2.5 mm HX Magic Angle Spinning (MAS) probe.Solid-state NMR experiments on Pb3SBr1.2I2.8 were performed on a Bruker wide-bore 9.4 T [n0( 1 H) = 400 MHz] NMR spectrometer equipped with a Bruker Advance III HD console and a Bruker 2.5 mm broadband HX MAS probe.The probes were configured in 1 H-207 Pb mode. 207Pb chemical shifts were referenced by using the published indirect referencing scale and that relates the 207 Pb reference frequency to the 1 H reference frequency (Larmor frequency ratio 207 Pb and 1 H is 20.920599).S2 1 H chemical shifts were referenced to neat tetramethylsilane (TMS) by using adamantane [diso( 1 H) = 1.76 ppm] as a secondary standard.The 207 Pb spectra shown in the main text were obtained by using the variable offset cumulative spectra (VOCS) approach, S3 where the transmitter offset was incremented across the spectral range in steps of 300 ppm (37.6 kHz) until no signal was observed.All of the sub-spectra were then co-added to form the total spectrum.Each 207 Pb NMR sub-spectrum was acquired using a spin echo pulse sequence, a MAS frequency of 25 kHz, and 67 kHz radiofrequency field 207 Pb pulses (3.74 µs p/2 and 7.48 µs p pulse durations), a recycle delay of 1 s, and 12800 scans.For each sample, recycle delays of 1 s, 2 s and 4 s were tested, and the 1 s recycle delay was found to provide the best sensitivity.Generally, fewer than 10 individual sub-spectra were acquired and co-added to form the total 207 Pb NMR spectrum of each sample.Peak fitting of 207 Pb Solid-state NMR was done using the solid line shape analysis module (SOLA) in topspin version 3.6.5.
Optical Characterization.Solution photoluminescence (PL) spectra were collected on a Horiba Jobin Yvon Fluorolog-3 spectrofluorometer (Slit width = 5 nm; λexc= 400 nm).The crude samples were suspended in hexanes for all solution PL measurements.Diffuse-reflectance spectra were collected using a SL1 Tungsten Halogen lamp (vis-IR), a SL3 Deuterium Lamp (UV), and a BLACK-Comet C-SR-100 spectrometer (200-1080 nm).The band gap values were estimated by extrapolating the linear slope of Tauc plots by plotting (Ahν) r versus hν (A = absorbance, hν = incident photon energy in eV, and r = 2 for indirect band gap semiconductors).S4 Single Particle Fluorescence Microscopy.Single particle photoluminescence (PL) was performed on an inverted microscope operated in epi-fluorescence mode (Nikon Eclipse TE2000U Melville, NY).Quaternary mixed-halides were diluted in hexanes and sonicated for 90 min before depositing 50 µL on a glass microscope coverslip (Fisher Scientific, Pittsburgh, PA).The solvent was removed under vacuum for 15 min.An Xcite Series 120 PC mercury lamp was used for excitation.A filter set from Omega Optical (Brattleboro, VT) was equipped with XF1009 (425DF45) excitation and XF3304 (605WB20) emission filter (λexc = 425±25 nm and λem = 605±15 nm).The dichroic filter used was XF2007 (475DRLP).A 100× Plan Apo, 1.49 numerical oilimmersion objective, was used for all experiments.Single particle PL images were collected on an Andor iXon Ultra EMCCD camera (Oxford Instruments, Abingdon, UK) with 40 ms exposure time and 100× electron multiplication (EM) gain.Each movie was 60 s in duration, and 5 movies were collected per sample.ImageJ was used to analyze the PL intensity versus time for five selected particles and backgrounds.The reported data represents an average intensity and background.
Calculations.Relative energy calculations were performed using the Vienna Ab initio Simulation Package (VASP).S5 Electronic exchange-correlation was treated using the Perdew-Burke-Ernzerhof (PBE) functional.S6 The cut-off energy for the plane wave basis functions was 500 eV and projected augmented-wave (PAW) pseudopotentials were used.During the structural optimizations the volume, atomic positions and cell shape were allowed to fully relax until the convergence energy was less than 1×10 -4 eV.The relative total energies were calculated over a 16×16×16 k-point grid using the tetrahedron method.Density of states (DOS) were calculated after converging the total energy on a k-mesh of 20×20×20 in the irreducible wedge of the Brillouin zone also using the tetrahedron method.Unit cell representations and atomic coloring patterns were generated using VESTA.

Figure S1 .
Figure S1.Crude solutions (prior to purification) of mixed-halide lead chalcohalides prepared using various concentrations of PbI2 and PbBr2.

Figure S2 .
Figure S2.Powder XRD patterns of mixed-halide lead chalcohalides compared to those of common standards.

Figure S4 .
Figure S4.Powder XRD pattern obtained from a reaction of 38 mM PbI2 and 19 mM PbBr2 with lattice planes visualized for the Pb3SBrI3 (P21/m) unit cell.

Figure S5 .
Figure S5.Peak shift of different XRD reflections as a function of relative halide synthetic loading (%Br). 1

Figure S6 .
Figure S6.Rietveld refinement of mixed-halide chalcohalide XRD patterns with respect to the P21/c Pb3SBrI3 standard pattern.

Figure S7 .
Figure S7.Comparison of lattice parameters determined from Equation 1 and Rietveld refinement and those reported in the literature S8 as a function of relative halide synthetic loading (%Br).

Figure S20 .
Figure S20.Calculated relative energies of Pb3SBrI3 with different atomic coloring patterns.

Figure S22 .
Figure S22.Calculated relative energies of Pb3SBr2I2 with different atomic coloring patterns.

Figure S26 .
Figure S26.Total DOS of (a) Pb4S3Br2 and (b) Pb5S2I6 calculated using VASP S5,S6 (both structures were first allowed to fully relax to their lowest energy structures).

Table S1 .
Representative precursor concentrations used in the synthesis of mixed-halide chalcohalides.a

Table S2 .
Structure parameters of mixed-halide chalcohalides determined from raw XRD data and equation S1.

Table S3 .
Lattice parameters obtained from the Rietveld refinement of mixed-halide chalcohalide XRD patterns.

Table S5 .
Solubility product constants of binary lead compounds.S9

Table S6 .
Experimental 207 Pb ssNMR chemical shift values for mixed-halide lead chalcohalides compared to reported values for potential binary lead impurity phases.