Giant emission enhancement from Cs₃Bi₂Br₉via oxygen-induced optimization of radiation channels

Abstract

Among Pb-free perovskite candidates, Bi-based perovskite derivatives show promise as they meet eco-friendly standards and warrant further development. However, the fundamental question of how atmospheric conditions affect the optical performance of Bi-based perovskites, which is closely related to their practical applications, is seldom researched. Here, we observed a significant increase in the light emission from Cs₃Bi₂Br₉ crystals when the reaction was carried out in an oxygen-filled environment, boosting the photoluminescence quantum yield (PLQY) by about 10 times. Through a detailed analysis of the optical properties using X-ray photoelectron spectroscopy (XPS) and density functional theory (DFT) calculations, it was determined that the oxygen molecules can effectively passivate vacancy defects and suppress nonradiative recombination. In addition, it was found that Cs₃Bi₂Br₉ exhibits a very intense exciton–phonon interaction, with a low thermal activation energy and a high Huang–Rhys factor (S), leading to phonon emission. However, by passivating with oxygen, the exciton–phonon interaction could be regulated, and the radiation channels were optimized for self-trapping emission (STE). These findings could be valuable in gaining a deeper understanding of the impacts of reaction atmosphere, which are often overlooked, on the optical properties of perovskite derivatives.

---

Introduction

In the last decade, lead halide perovskites have made significant strides in the field of light-emitting diodes (LEDs), boasting an impressive external quantum efficiency of over 25%.^1–4 This is largely attributed to their exceptional electrical and optical properties. Specifically, the unique Pb²⁺-6s² lone-pair configuration in lead-based perovskites (APbX₃, A = CH₃NH₃ (MA), CHNH₂NH (FA), and Cs, X = Cl, Br, and I) has been identified as a crucial factor in achieving outstanding photoelectric properties.^5–7 However, there are significant drawbacks associated with lead halide perovskites, including the toxicity of lead and their susceptibility to environmental factors, such as moisture, heat, and ultraviolet radiation.⁸ Therefore, pure eco-friendly lead-free halide perovskite alternatives are desirable to address the issue of toxicity and instability and enable further development. It is known that the valence band is made up of the p-orbitals of halides and the s-orbitals of lead, while the conduction band is primarily derived from the p-orbitals of lead.⁹ Hence, elements with electronic configuration similar to the lone-pair Pb²⁺-s² are essential in potential substitutes.¹⁰ Though Sn and Ge, belonging to the same main group as Pb, have a comparable electronic structure for substitution,^11,12 the instability of Sn²⁺ and Ge²⁺ makes them prone to oxidation to Sn⁴⁺ and Ge⁴⁺.⁸ Besides, many other alternatives, such as Ag⁺, Cu⁺, Zn²⁺, Ge²⁺, Mg²⁺, Mn²⁺, Bi³⁺, and Sb³⁺, have also been explored.^13–24 Here, we focus our discussion on Bi-based perovskites. Being an element of group VA, Bi has a similar electronic configuration and a comparable ionic radius to Pb.⁸ In addition, the abundant storage and non-toxicity of Bi make Bi-based perovskites the best replacement candidates.

In Bi-based perovskites, the antibonding interaction at the valence band maximum (VBM), which originates from the active ns² electronic configuration, effectively restricts the defects to shallow energy levels near the band edges.⁹ Moreover, the octahedron-coordinated construction of A₃Bi₂X₉ compounds tend to exhibit enhanced tolerance towards defects and longer carrier diffusion lifetimes, and the low-dimensional crystal structure and electron occupancy play important roles in boosting the exciton binding energy.^25,26 In the past, various strategies have been explored to enhance the luminescence performance of Bi-based perovskites. For example, Han et al.²⁷ found that the trap states in Cs₃Bi₂Br₉ could be effectively passivated by adding oleic acid (OA) as the surfactant, leading to an increase in PLQY from 0.2% to 4.5%. Additionally, Tang et al.²⁶ demonstrated the crucial role of octylamine (OLAm) and OA in ligand passivation for Cs₃Bi₂X₉ (X = I, Br, Cl). Besides, the use of Cl passivation²⁸ has been investigated, and the results show that Cl preferred to attach to the surface of MA₃Bi₂Br₉ quantum dots (QDs), and Cl-passivated MA₃Bi₂Br₉, with a higher activation energy, promoted radiative recombination. While Bi-based perovskites exhibit better air stability, lower moisture sensitivity, and higher thermal tolerance than traditional Pb-based perovskites, they still suffer challenges related to low PLQYs.^26–30

It is well-established that the atmospheric conditions during synthesis significantly influence the structural and photoelectric properties of Pb-based perovskites.^31–34 Studies have shown that gaseous molecules and moisture can interact with perovskite crystals through physical adsorption or chemical binding, offering a deeper understanding of Pb-based perovskites.^35,36 Therefore, it is crucial to understand how atmospheric conditions affect the photoelectric properties of Bi-based perovskites. Early research has suggested that water molecules can interact with Cs₃Bi₂Br₉ to form a perovskite hydrate of BiOBr, which can passivate surface defects.^26,27 In addition, Han et al.²⁷ found that the Cs₃Bi₂Br₉ sample stored in high moisture conditions had significantly higher PLQYs compared with those stored in low-moisture conditions. Nevertheless, the interaction between atmospheric molecules and perovskite crystals remains a subject of debate. Therefore, understanding the role of atmospheric conditions in determining the luminescence properties of Bi-based perovskites is of importance for further practical applications.

In this work, we investigated the radiative recombination of self-trapped emission (STE) of Cs₃Bi₂Br₉ by taking different atmospheric conditions into consideration. Our findings show that the PLQY of the Cs₃Bi₂Br₉ solution synthesized in an oxygen-filled environment was approximately 10-fold higher than that synthesized in a flask full of nitrogen. Through a thorough analysis of the structural properties, density functional theory (DFT) calculations, and optical performance analysis, we discovered that the absorbed oxygen molecules can optimize the radiation channels for highly efficient emission. On the one hand, the Br⁻ vacancies could be effectively passivated to suppress nonradiative recombination. On the other hand, the exciton–phonon interaction was relaxed to suppress nonradiative and phonon emission. The non-toxic nature and air stability of the Cs₃Bi₂Br₉ crystals make them a promising alternative to Pb-based perovskites. Coupled with the broad emission spectrum, which is ideal for high-quality lighting and wide color-gamut displays, these advantages pave the way for their usage in various optoelectronic applications, including solid-state lighting and display backlight systems.

Experimental methods

Materials

CsBr was purchased from Advanced Election Technology CO., Ltd, China. BiBr₃, dimethyl sulfoxide (DMSO, anhydrous, 99.9%), n-octylamine (OLAM, 99%), n-dodecylamine (DLAM, 99%), n-butylamine (BLAM, 99%), and anhydrous ethanol were purchased from Shanghai Macklin Biochemical Co., Ltd. Oleic acid (OA) was purchased from Alfa Aesar.

Synthesis of Cs₃Bi₂Br₉ crystals

For the synthesis of Cs₃Bi₂Br₉, 120 μmol CsBr and 80 μmol BiBr₃ were co-dissolved in DMSO under stirring for 30 min at 50 °C. After that, 8 μL of OLAM was added and stirred for another 30 min to form a clear precursor solution. Then, 0.2 mL of the precursor solution was injected into a mixed solution of 2 mL ethanol anhydrous and 0.2 mL OA, followed by continuous stirring for 10 min at room temperature, and a cloudy yellow reaction mixture was obtained. The reaction mixture was then centrifuged at 8000 rpm for 1 min to precipitate the large particles, and a clear supernatant containing Cs₃Bi₂Br₉ with blue emission was obtained.

Structural and compositional characterization

The X-ray diffraction (XRD) data were collected using a Smart Lab 9 kW (Rigaku Ltd) diffractometer (D/teX Ultra 250) with Cu Kα radiation. Transmission electron microscopy (TEM) images were obtained using a Talos F200 X (FEI Ltd) microscope using Cs₃Bi₂Br₉ samples deposited on copper-coated copper grids. X-ray photoelectron spectroscopy (XPS) experiments were performed using an ESCALAB250Xi (THERMO SCIENTIFIC Ltd).

Optical characterizations

The steady photoluminescence emission (PL) and excitation (PLE) measurements were carried out using a PL spectrometer (Horiba FluoroMax-4, Horiba Ltd) under the excitation of an Xe lamp. For optical measurements under different atmospheric conditions, the Cs₃Bi₂Br₉ sample was placed in an airtight container. The PL spectra were recorded under different power intensities by exciting the samples using a monochromatic pulsed laser of 365 nm wavelength. The temperature-related PL and decay lifetime were measured using a Quanta Master 8000 spectrometer (Horiba Ltd), with the temperature ranging from 100 to 300 K using a liquid helium cooler. The excitation sources were a Xe lamp (365 nm) and a pulsed spectral LED (340 nm) for PL and PL decay measurements, respectively. The absorption spectra were measured using the UV-vis spectrophotometer model U-3900 (Hitachi Ltd). The PLQYs were determined using the fluorescence spectrometer FLS1000 (Edinburgh Ltd).

Calculations

First-principles calculations were performed using the density functional theory; the Perdew–Burke–Ernzerhof (PBE) functional with generalized gradient approximation was chosen to consider the exchange interaction of the electrons. The semi-empirical dispersion-corrected D3 scheme was used, as the long-range van der Waals interaction between the atomic layers was taken into consideration. O₂ absorption on the surface of Cs₃Bi₂Br₉ was simulated for 80 atoms with a vacuum gap of 20 Å. The energy cutoff for the plane wave basis was set at 500 eV, and for Brillouin zone integration, the Monkhorst–Pack scheme with a 7 × 7 × 1 k-mesh was chosen. The convergence criteria were set to 10⁻⁵ eV and 0.001 eV Å⁻¹ for self-consistent calculation and geometry optimization, respectively.

Results and discussion

Structure and photoluminescence properties

Cs₃Bi₂Br₉ was synthesized using the ligand-assisted precipitation method, in which the precursor solution was dropped into an antisolvent. As shown in Fig. 1a, CsBr and BiBr₃ were initially dissolved in DMSO to form Solution A. Then, octylamine (OLAM) was added to form Solution B. A specific aliquot of Solution B was added dropwise to ethanol, and continuous magnetic stirring induced the nucleation and growth of Cs₃Bi₂Br₉ crystals, accompanied by a distinct color transition of the reaction system to muddy yellow. The crystal structure of Cs₃Bi₂Br₉ was confirmed using X-ray diffraction (XRD) analysis. As exhibited in Fig. 1b, the XRD pattern of the sample containing 30 µmol OLAM exhibited the characteristic hexagonal structure of Cs₃Bi₂Br₉ with P3m1 symmetry. The diffraction peaks at 12.83°, 15.68°, 22.13°, 27.40°, 31.68°, and 39.11° were indexed to the (100), (101), (102), (201), (022), and (212) crystal planes, respectively.²⁶ In comparison, the crystallinity of Cs₃Bi₂Br₉ gradually improved as the OLAM content in the precursor increased, and optimal crystallinity was observed at an OLAM concentration of 60 µmol. However, when OLAM was increased to 70 µmol, several additional diffraction peaks appeared, indicating potential disruption of Cs₃Bi₂Br₉. Notably, as illustrated in Fig. S1, these additional diffraction peaks could be ascribed to Bi₂Br_1.67 and BiBr (Fig. S1a, SI), and the samples containing higher OLAM exhibited more pronounced diffraction peaks of Bi_xBr_y. The XPS spectra were then recorded to compare the chemical states of the obtained samples, as shown in Fig. S2. The Bi-4f peaks observed at binding energy values of 164.4 and 159.1 eV corresponded to the 4f_5/2 and 4f_7/2 bonding of Bi³⁺, respectively. Notably, both peaks exhibited a positive shift as the OLAM content increased from 30 µmol to 90 µmol, indicating electron loss during the formation of Bi_xBr_y. Simultaneously, with the introduction of Bi_xBr_y, the color of the Cs₃Bi₂Br₉ solution shifted from light yellow to white as the amount of OLAM in the precursor solution increased (Fig. S3, SI). The crystal structure of Cs₃Bi₂Br₉ is illustrated in Fig. S4, showing Bi atoms coordinated with Br atoms to form an octahedral structure. However, only two-thirds of the octahedron are fully occupied, giving rise to a two-dimensional layered structure. The transmission electron microscopy (TEM) images in Fig. S4b depict the distribution of Cs₃Bi₂Br₉, with a lattice space of 3.8 Å in the high-resolution transmission electron microscopy (HRTEM) image, corresponding to the (003) crystal plane. Fig. S5 displays the size distribution histogram of the Cs₃Bi₂Br₉ crystals, as obtained from the TEM analysis, demonstrating an average diameter of 5.20 ± 0.09 nm.

Fig. 1 (a) Schematic of the synthesis process of Cs₃Bi₂Br₉. (b) XRD patterns of Cs₃Bi₂Br₉ with different contents of OLAM. (c) Trend plots of the integrated PL intensity of the Cs₃Bi₂Br₉ solution (solid) and Cs₃Bi₂Br₉ powder (hollow) under different atmospheric conditions.

Then, we designed different atmospheric conditions for the formation of Cs₃Bi₂Br₉, including ambient air, dry oxygen (O₂), dry nitrogen (N₂) and moist N₂ with 40% RH. As shown in Fig. S6, gases, such as air, O₂, and N₂, were introduced to control the reaction atmosphere, while water vapor was used to adjust the humidity of the reaction environment. The temporal evolution of PL emission of the Cs₃Bi₂Br₉ solution and Cs₃Bi₂Br₉ powder was monitored under various atmospheric conditions. As shown in Fig. 1c, both the Cs₃Bi₂Br₉ solution and powder exhibited closely matching trends. The PL intensity showed no significant enhancement under N₂ or moist N₂ (40% RH). In contrast, exposure to dry O₂ resulted in a marked increase in PL intensity, which surpassed that observed with air. These results indicate that both dry N₂ and moist N₂ (40%RH) atmospheres have a minimal impact on the PL emission of Cs₃Bi₂Br₉, whereas oxygen plays a distinct beneficial role in enhancing its PL performance. Furthermore, photographs of both the solution and powder under UV illumination visually confirmed the enhanced blue emission intensity.

The luminescence of Cs₃Bi₂Br₉ is critically dependent on the presence of both OA and OLAM. This was evidenced by the negligible PL intensity observed for samples synthesized without these ligands (Fig. S8). Furthermore, the PL intensity was highly sensitive to the amount of OA added, and the optimal volume was identified as 0.2 mL in the antisolvent (Fig. S9). Furthermore, we compared the luminescence properties of Cs₃Bi₂Br₉ prepared in an oxygen atmosphere with different contents of OLAM. The PL spectra shown in Fig. 2a and b illustrate that the intensity of the PL signal increased with the OLAM content, reaching a maximum at 60 µmol. When excited at 365 nm, Cs₃Bi₂Br₉ prepared within 2 min, referred to as the control sample, exhibited broad emissions from 380 to 420 nm, with peaks at about 434 nm. The large full-width at half maximum (FWHM) and significant Stokes shift indicated the presence of self-trapped excitons (STEs) in the excited states of Cs₃Bi₂Br₉, where excitons are trapped due to strong electron–phonon coupling.³⁷ Despite the control sample exhibiting a broader PLE spectrum, O₂-absorbed Cs₃Bi₂Br₉ (prepared with an 80 min treatment) showed a significantly enhanced PL intensity, as seen in Fig. 2b. This enhancement indicates an improvement in light conversion efficiency. Fig. 2d obviously exhibits the PL intensity comparison between the control and O₂-absorbed samples, revealing a 10-fold increase in strength for four samples with different OLAM contents as the reaction time was long enough for sufficient O₂ absorption. Meanwhile, the striking luminescence images captured under UV light excitation (Fig. 2c) revealed that the control samples appeared dim, while the O₂-absorbed samples were bright. Notably, the sample containing 60 µmol OLAM emitted the most intense blue light. The recorded PLQYs displayed in Fig. 2e illustrate the changes in Cs₃Bi₂Br₉ samples after longer reaction times, and the PLQY of the 60 µmol-OLAM sample increased from 6.2% to 63%. Notably, the Cs₃Bi₂Br₉ samples prepared with n-butylamine (BLAM) and n-dodecylamine (DLAM) also exhibited a significant enhancement in luminescence performance upon O₂ absorption, as shown in Fig. S10. In addition, we monitored the PL spectra of Cs₃Bi₂Br₉ ethanol solution mixed with deionized water to evaluate the stability of Cs₃Bi₂Br₉ in water. The statistical analysis presented in Fig. S11 reveals a gradual decrease in the normalized PL intensity, yet it remained at ∼88% of the initial value, demonstrating remarkable water resistance.

Fig. 2 PLE and PL spectra of Cs₃Bi₂Br₉ with different OLAM content: (a) control and (b) O₂-absorbed samples. (c) Luminescence photographs of the samples under UV light. (d) Statistics of the peak intensity of the PL spectra. (e) PLQYs of the control and O₂-absorbed Cs₃Bi₂Br₉ solutions with different content of OLAM.

Defect passivation

To uncover the mechanism underlying oxygen-boosted PL, we investigated the interaction between oxygen and Cs₃Bi₂Br₉. In general, two possibilities may be considered for the enhanced PL performance in oxygen compared to the control sample. (1) The crystal structure of Cs₃Bi₂Br₉ might have changed, leading to the formation of products that possibly impact the emission enhancement. Similar cases have been reported in Bi-based NCs, in which the presence of trace water resulted in the formation of BiOBr on the surface, which helped in passivating the undesirable structural defects.^38,39 (2) The suppression of non-radiative recombination facilitates exciton recombination for STE. In order to validate these hypotheses, XRD measurement was performed on Cs₃Bi₂Br₉. The comparative XRD patterns of the Cs₃Bi₂Br₉ crystal are presented in Fig. 3a. Obviously, O₂-absorbed Cs₃Bi₂Br₉ showed similar patterns to the control sample, with no additional peaks or shift in peak positions. This suggests that the improvement in luminescence properties may not be attributed to changes in the crystal structure. Additionally, XPS analysis was also carried out to explore the surface chemical states of exposed Cs₃Bi₂Br₉. Due to spin–orbit coupling, the energy levels of Cs 3d and Br 3d split into two peaks, as shown in Fig. S12. The Cs 3d spectrum (Fig. S12a) was deconvoluted into 3d_5/2 and 3d_3/2 peaks at 738.2 and 724.3 eV, respectively. The fitting peaks in the Br 3d spectrum (Fig. S12b) were observed at 69.3 and 68.3 eV for the 3d_5/2 and 3d_3/2 orbitals, respectively. Notably, these XPS spectra indicated that the peak positions of Cs and Br remained unchanged, suggesting similar crystal and electronic structures in both control and O₂-absorbed Cs₃Bi₂Br₉ samples. This finding is consistent with the XRD results. Additionally, in Fig. 3b, the deconvoluted peaks of Bi 4f at 164.4 and 159.1 eV are assigned to Bi 4f_7/2 and Bi 4f_5/2, respectively. The O 1s spectrum in Fig. 3c shows three separate peaks corresponding to the C–OH bond at 533.3 eV, the C=O bond at 531.8 eV, and Bi–O at 530.0 eV.³⁵ It is worth noting that the peaks of Bi 4f had shifted toward higher binding energy values, while the peak of Bi–O exhibited a negative shift in the opposite direction. This observation indicates that the absorbed oxygen may react with Bi ions by filling the bromine defect sites on the surfaces of the Cs₃Bi₂Br₉ crystals. In addition, the minimal binding energy shift also demonstrates that the attractive force between Bi and adsorbed oxygen is governed by electrostatic interactions, which do not lead to the formation of new compounds. This conclusion is well aligned with the XRD results. It was further supported by the FTIR spectra (Fig. S13), which showed no discernible features indicative of chemical interactions with the oxygen species (e.g., O₂⁻ or O₂²⁻).

Fig. 3 (a) XRD patterns of control and O₂-absorbed Cs₃Bi₂Br₉. XPS spectra of (b) Bi-4f and (c) O-1s for control and O₂-absorbed samples. DFT calculations for the band structure of (d) Br-defect-free Cs₃Bi₂Br₉, (e) Cs₃Bi₂Br₉ with Br defects, and (f) Cs₃Bi₂Br₉ with oxygen molecule repair.

In order to further investigate defect passivation based on oxygen molecules in Cs₃Bi₂B₉, we performed first-principles calculations using the density functional theory (DFT) to explore the role of absorbed oxygen on the electronic structure of Cs₃Bi₂Br₉, particularly in the presence of bromine defects. The band structures of Cs₃Bi₂Br₉ in different states are illustrated in Fig. 3d–f. In the absence of defects, Cs₃Bi₂Br₉ exhibits both the valence band maximum (VBM) and the conduction band minimum (CBM) at the Γ point, with minimal trapping states near the bottom of the conduction band. However, the presence of Br defects introduces new shallow energy levels at the bottom of the CBM. Here, trap states refer to electronic states with energies exceeding K₀T (≈25 meV) below the Fermi level, thus enabling charge localization and carrier trapping, which ultimately leads to non-radiative decay.⁴⁰ Previous works on Pb-based perovskites have reported that halide vacancies easily induce trapping states in the bandgap space.^31,41 When O₂ is absorbed at the site of Br vacancy, the trap states are noticeably eliminated, as exhibited in Fig. 3f. These results indicate that O₂ absorption can restore full octahedral coordination with Bi and effectively remove the trap states, thereby enhancing radiative recombination for PL enhancement in the air environment.

Photophysical mechanism

The origin of emission was investigated by analyzing the changes in PL intensity as the excitation power intensity varied, as shown in Fig. 4a. The linear relationship observed between excitation power density and the PL intensity suggests that emission is not assisted by permanent defects, and the transition is related to excitonic STE.^42,43 Subsequently, excitation-dependent PL spectra were recorded for both control and O₂-absorbed Cs₃Bi₂Br₉, as shown in Fig. 4b and c, respectively, in the excitation wavelength range from 315 to 375 nm, allowing for a comprehensive analysis of the emission characteristics. For the control sample, when the excitation wavelength was lower than 335 nm, a broad emission spectrum emerged at around 380 nm from the mixed solution used in the experiment (Fig. S15), while the intrinsic emission of Cs₃Bi₂Br₉ was barely detected. For the O₂-absorbed sample, the intensity of the typical Cs₃Bi₂Br₉ emission initially increased and then decreased, in accordance with the PLE spectra shown in Fig. 2b. These corroborate the trapping of excitons in the defect states.⁴⁴ Furthermore, the exciton absorption peak of control Cs₃Bi₂Br₉ was not readily visible (Fig. 4d). However, it was more prominent in the O₂-absorbed sample, as well as in samples with varying OLAM content, as shown in Fig. S16. This suggests that the exciton binding energy increased, and the excitons became more stable in aged Cs₃Bi₂Br₉. To gain a deeper understanding, we compared the time-resolved PL spectra of the control and O₂-absorbed Cs₃Bi₂Br₉ samples. As shown in Fig. 4e, the PL decay curves were fitted using a bi-exponential function, comprising a short-time (τ₁) component and a long-time (τ₂) component, and the fitting parameters are summarized in Table S1. In line with previous reports,³⁹ the long-time component is assigned to radiative charge-carrier recombination (process 1 in Fig. 4f), whereas the short-time component is attributed to non-radiative quenching pathways involving trap-state transitions and phonon scattering (processes 2 and 3 in Fig. 4f). In O₂-absorbed Cs₃Bi₂Br₉, which exhibited a higher PLQY, the amplitude (A₁) of the short-time component was reduced compared to the control. This indicates that the nonradiative recombination channels, primarily those mediated by trap states, are effectively suppressed by oxygen passivation. As a result, the radiative recombination efficiency was enhanced, and the average lifetime (τ) increased from 1.61 ns in the control samples to 7.19 ns in the O₂-absorbed samples. This process mitigates the detrimental influence of trap states, thereby reducing charge-carrier trapping and suppressing nonradiative recombination, as depicted in Fig. 4f.

Fig. 4 (a) Integrated PL intensity of the control and O₂-absorbed Cs₃Bi₂Br₉ samples under different excitation power intensities. PL spectra of (b) control and (c) O₂-absorbed Cs₃Bi₂Br₉ under different excitation wavelengths. (d) UV-vis absorption spectra and (e) PL decay curves of control and O₂-absorbed Cs₃Bi₂Br₉. (f) Schematic of the proposed recombination channels based on oxygen passivation in Cs₃Bi₂Br₉.

In addition, the photophysical mechanisms underlying the improved luminescence characteristics of O₂-absorbed Cs₃Bi₂Br₉ were further explored through temperature-dependent PL spectroscopy (Fig. 5a–f). As previously mentioned, the PL intensity of the control Cs₃Bi₂Br₉ (Fig. 5a) was extremely low, with faint emission occurring at about 435 nm. As the temperature decreased from 300 K to 80 K, the PL intensity showed a consistent increase, indicating a reduction in nonradiative recombination at lower temperatures.⁴⁴ Similarly, the evolution of O₂-absorbed Cs₃Bi₂Br₉ emission followed a similar pattern, displaying increased PL intensity as the temperature decreased, as shown in Fig. 5b. Based on the temperature-dependent PL intensity, the activation energy was determined using eqn (1) ¹⁸

where I₀ represents the PL intensity at 0 K, A is the free parameter, K_B is the Boltzmann constant, and E_a is the activation energy. By analyzing the Arrhenius fitting curves depicted in Fig. 5c and d, the E_a values of control and O₂-absorbed Cs₃Bi₂Br₉ were calculated to be 29.1 and 52.3 meV, respectively. The E_a of the control sample was approximately 26 meV, close to the thermal ionization energy,⁴⁵ leading to exciton annihilation by nonradiative centers. For passivated Cs₃Bi₂Br₉, this value increased to 52.3 meV, demonstrating enhanced exciton stability for radiative recombination and enabling brighter emission at room temperature. This phenomenon explains the improved luminescence properties of oxygen-passivated Cs₃Bi₂Br₉. As the temperature decreased, the PL peaks experienced a slight red shift, attributed to the effect of lattice thermal expansion on the interaction between valence orbitals, a phenomenon previously observed in Pb-based perovskites and PbS QDs.^46,47

Fig. 5 Temperature-dependent PL spectra of (a) control and (b) O₂-absorbed Cs₃Bi₂Br₉ in the temperature range of 80–300 K. Integrated PL intensity and FWHM of (c) control and (d) O₂-absorbed Cs₃Bi₂Br₉versus reciprocal temperature. PL decay curves of (e) control and (f) O₂-absorbed Cs₃Bi₂Br₉ at different temperatures.

Significantly, the FWHM of the PL spectra increased gradually as the temperature rose, as illustrated in Fig. 5a–d. This trend is indicative of exciton–phonon interactions in Cs₃Bi₂Br₉ with a flexible two-dimensional lattice structure, likely leading to lattice distortion and the formation of polarons, which couple with carriers. When the interaction between electrons and phonons reaches a certain level, polarons will trap the carriers at lattice sites, creating a self-trapping state emission. In this study, we investigated the impact of exciton–phonon coupling on the optical properties of Cs₃Bi₂Br₉ prepared in oxygen, and the Huang–Rhys factor (S) was calculated using eqn (2),⁴² which takes into account the temperature-dependent FWHM of the PL spectra.

The Boltzmann constant (K_B) and phonon energy (ħω_phonon) are also considered in the calculation. The phonon broadening plots in Fig. 5c and d show that the S values of control and O₂-absorbed Cs₃Bi₂Br₉ were 31.09 and 28.75, respectively. Compared with CdSe,⁴⁸ ZnSe,⁴⁹ and CsPbBr₃,⁵⁰ the larger Huang–Rhys factor indicates that Cs₃Bi₂Br₉ is more prone to the formation of STEs. However, systems with large S indicate the strong coupling between excitons and phonons, which can lead to the dissipation of excited-state energy. This means that some excited excitons may relax to the ground state through a process known as curve crossing of the excited and ground states, accompanied by nonradiative recombination and phonon emission, as illustrated in Fig. 4f. It has been shown that the photoluminescence performance is directly related to the value of S.⁴² Therefore, the smaller S observed for the O₂-absorbed Cs₃Bi₂Br₉ samples demonstrates the effectiveness in reducing exciton–phonon coupling. This optimization ultimately leads to enhanced radiative recombination and improved PL efficiency. In this context, the Huang–Rhys parameter can be considered as a key metric for regulating the efficient emission of excited states.

Moreover, the temperature sensitivity of PL lifetime is directly linked to nonradiative recombination. As illustrated in Fig. 5e, the lifetime of control Cs₃Bi₂Br₉ decreased gradually as the temperature rose from 100 K to 300 K, reflecting the occurrence of a rapid nonradiative process. In contrast, the lifetime of O₂-absorbed Cs₃Bi₂Br₉ remained relatively constant at various temperatures (Fig. 5f), indicating successful oxygen passivation, which effectively suppresses nonradiative recombination. In conclusion, the enhanced luminescence performance stems from two factors: passivation of the Br⁻ vacancies to suppress non-radiative recombination, and the relaxation of exciton–phonon interaction to reduce energy loss.

Photoluminescence and structural stability

Material stability is of paramount importance for practical applications. We, therefore, investigated the stability of the as-synthesized O₂-absorbed Cs₃Bi₂Br₉ by monitoring its PL under ambient air with 30–50% relative humidity over an extended period. As shown in Fig. 6a and b, the PL intensity showed no significant decay over 20 days. Even after two months of air exposure, the intensity decreased only slightly, with less than 20% loss, demonstrating excellent retention of the luminescence properties. Notably, the PL peak position remained unchanged through the two-month storage period, confirming the robust optical stability of the material.

Fig. 6 (a) Pseudocolor map of the time-dependent PL spectra of Cs₃Bi₂Br₉ and (b) the corresponding statistics of peak intensity and peak position. (c) XRD spectra of the as-synthesized Cs₃Bi₂Br₉ crystals stored in open air for different days. (d) Illustration of water–oxygen-induced degradation of Cs₃Bi₂Br₉ and the reduction of luminescence when placed in the air.

Additionally, to better understand the reason for the decrease in PL intensity, we examined the structural evolution of the material using X-ray diffraction (XRD) after ambient exposure. Fig. 6c displays the XRD patterns of Cs₃Bi₂Br₉ samples stored in the air for 20, 40, and 60 days. Besides the characteristic peaks of Cs₃Bi₂Br₉, new diffraction peaks emerged and were identified as those of tetragonal BiOBr (●) and complex BixBry (▲) compounds. The intensities of these impurity peaks became more pronounced with prolonged exposure, confirming that the formation of these non-perovskite decomposition products is responsible for the observed decline in luminescence. This phenomenon underscores the direct structure–property relationship governing the material's photoluminescence, as schematically illustrated in Fig. 6d. The Cs₃Bi₂Br₉ crystal inherently contains surface dangling bonds and uncoordinated atoms. Upon air exposure, atmospheric H₂O and O₂ molecules preferentially attack these vulnerable sites, leading to the formation of non-perovskite by-products, such as BiOBr and Bi_xBr_y. This reaction disrupts the structural integrity of the [Bi₂Br₉]³⁻ clusters, which are the primary luminescence centers. The resulting decomposition products not only lack the desired emissive properties but also function as charge-trapping states. Consequently, prolonged exposure increases the density of these non-radiative recombination centers, which is the fundamental reason for the observed decline in luminescence efficiency.

Conclusions

In summary, our study delves into the impact of oxygen on the photoluminescence of Cs₃Bi₂Br₉, offering valuable foundational insights into Pb-free perovskite derivatives. Through both experimental and computational analyses, we examined the optical properties of control and O₂-absorbed Cs₃Bi₂Br₉. Our findings indicate that oxygen absorption can repair defect states, while effective oxygen passivation can enhance radiative recombination and improve light emission efficiency. Additionally, the temperature-dependent spectra revealed that oxygen passivation can mitigate the strong exciton–phonon interaction in Cs₃Bi₂Br₉, optimizing STE channels. This work emphasizes the importance of considering the interaction between Cs₃Bi₂Br₉ and the environmental conditions as it can enhance optical performance of Pb-free perovskite derivatives fabricated under different conditions. Furthermore, as an environmentally friendly and non-toxic alternative, Bi-based perovskites often exhibit superior chemical stability, suitable for better public acceptance and device lifetimes. Coupled with the broad emission spectrum, which is ideal for high-quality lighting and wide color-gamut displays, Bi-based perovskites are expected to have highly promising optoelectronic applications, including solid-state lighting and display backlight systems.

Author contributions

The manuscript was written through the contributions of all authors.

Conflicts of interest

The authors declare that they have no conflict of interests.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). The SI provides Cs₃Bi₂Br₉ crystal structural data (TEM, XRD), optical properties (PL, UV-vis spectra) and measurement details. See DOI: https://doi.org/10.1039/d5nr01915j.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 62104169, 52272166, 12274334), Hubei Key Laboratory of Optical Information and Pattern Recognition, Wuhan Institute of Technology (No. 202301), Youth Science and Technology Talent Support Project of Jiangsu Provincial Association of Science and Technology (JSTJ-2024-654) and Key Laboratory of Advanced Electrode Materials for Novel Solar Cells for Petroleum and Chemical Industry of China, and the School of Chemistry and Life Sciences, Suzhou University of Science and Technology, Suzhou City, Jiangsu Province 215009 (No. 2024A065).

References

1. S. Q. Sun, J. W. Tai, W. He, Y. J. Yu, Z. Q. Feng, Q. Sun, K. N. Tong, K. Shi, B. C. Liu, M. Zhu, G. Wei, J. Fan, Y. M. Xie, L. S. Liao and M. K. Fung, Enhancing Light Outcoupling Efficiency via Anisotropic Low Refractive Index Electron Transporting Materials for Efficient Perovskite Light–Emitting Diodes, Adv. Mater., 2024, 36(24), 2400421.
2. Y. Feng, H. Li, M. Zhu, Y. Gao, Q. Cai, G. Lu, X. Dai, Z. Ye and H. He, Nucleophilic Reaction–Enabled Chloride Modification on CsPbI₃ Quantum Dots for Pure Red Light–Emitting Diodes with Efficiency Exceeding 26%, Angew. Chem., Int. Ed., 2024, 63(11), 202318777.
3. Y. Nong, J. Yao, J. Li, L. Xu, Z. Yang, C. Li and J. Song, Boosting External Quantum Efficiency of Blue Perovskite QLEDs Exceeding 23% by Trifluoroacetate Passivation and Mixed Hole Transportation Design, Adv. Mater., 2024, 36(27), 2402325.
4. J. S. Kim, J.-M. Heo, G.-S. Park, S.-J. Woo, C. Cho, H. J. Yun, D.-H. Kim, J. Park, S.-C. Lee, S.-H. Park, E. Yoon, N. C. Greenham and T.-W. Lee, Ultra-bright, efficient and stable perovskite light-emitting diodes, Nature, 2022, 611(7937), 688–694.
5. W. Bai, T. Xuan, H. Zhao, H. Dong, X. Cheng, L. Wang and R. J. Xie, Perovskite Light–Emitting Diodes with an External Quantum Efficiency Exceeding 30%, Adv. Mater., 2023, 35(39), 2302283.
6. J. Yuan, D. Zhang, B. Deng, J. Du, W. C. H. Choy and J. Tian, High Efficiency Inorganic Perovskite Solar Cells Based On Low Trap Density and High Carrier Mobility CsPbI₃ Films, Adv. Funct. Mater., 2022, 32(47), 2209070.
7. K. R. Hansen, J. S. Colton and L. Whittaker-Brooks, Measuring the Exciton Binding Energy: Learning from a Decade of Measurements on Halide Perovskites and Transition Metal Dichalcogenides, Adv. Opt. Mater., 2023, 12(3), 2301659.
8. Z. Jin, Z. Zhang, J. Xiu, H. Song, T. Gatti and Z. He, A critical review on bismuth and antimony halide based perovskites and their derivatives for photovoltaic applications: recent advances and challenges, J. Mater. Chem. A, 2020, 8(32), 16166–16188.
9. S. Attique, N. Ali, S. Ali, R. Khatoon, N. Li, A. Khesro, S. Rauf, S. Yang and H. Wu, A Potential Checkmate to Lead: Bismuth in Organometal Halide Perovskites, Structure, Properties, and Applications, Adv. Sci., 2020, 7(13), 1903143.
10. T. Li, S. Luo, X. Wang and L. Zhang, Alternative Lone–Pair ns2–Cation–Based Semiconductors beyond Lead Halide Perovskites for Optoelectronic Applications, Adv. Mater., 2021, 33(32), 2008574.
11. Z. Shi, J. Guo, Y. Chen, Q. Li, Y. Pan, H. Zhang, Y. Xia and W. Huang, Lead–Free Organic–Inorganic Hybrid Perovskites for Photovoltaic Applications: Recent Advances and Perspectives, Adv. Mater., 2017, 29(16), 1605005.
12. L. Romani, A. Speltini, R. Chiara, M. Morana, C. Coccia, C. Tedesco, V. Armenise, S. Colella, A. Milella, A. Listorti, A. Profumo, F. Ambrosio, E. Mosconi, R. Pau, F. Pitzalis, A. Simbula, D. Ricciarelli, M. Saba, M. Medina-Llamas, F. De Angelis and L. Malavasi, Air- and water-stable and photocatalytically active germanium-based 2D perovskites by organic spacer engineering, Cell Rep. Phys. Sci., 2023, 4(1), 101214.
13. T. Jun, K. Sim, S. Iimura, M. Sasase, H. Kamioka, J. Kim and H. Hosono, Lead–Free Highly Efficient Blue–Emitting Cs₃Cu₂I₅ with 0D Electronic Structure, Adv. Mater., 2018, 30(43), 1804547.
14. D. Huang, H. Xiao, D. Liu, Q. Ouyang, Y. Kong, B. Wang, H. Lian and J. Lin, All-inorganic tin-doped Cs₂BiAgCl₆ double perovskites with stable blue photoluminescence for WLEDs, J. Mater. Chem. C, 2021, 9(28), 8862–8873.
15. H. Shankar, S. Ghosh and P. Kar, Highly stable blue fluorescent lead free all-inorganic Cs₂ZnX₄ 2D perovskite nanocrystals, J. Alloys Compd., 2020, 844, 156148.
16. H. Zhao, Y. Wei, Y. Chen, C. Liu, F. Jiang, Y. Liu and M. Hong, Blue-emitting 0D Cs₃ZnX₅ (X = Cl, Br) perovskite nanocrystals based on self-trapped excitons, J. Lumin., 2022, 249, 119048.
17. Z. Ma, Z. Shi, D. Yang, F. Zhang, S. Li, L. Wang, D. Wu, Y. Zhang, G. Na, L. Zhang, X. Li, Y. Zhang and C. Shan, Electrically-Driven Violet Light-Emitting Devices Based on Highly Stable Lead-Free Perovskite Cs₃Sb₂Br₉ Quantum Dots, ACS Energy Lett., 2019, 5(2), 385–394.
18. J. Zhang, Y. Yang, H. Deng, U. Farooq, X. Yang, J. Khan, J. Tang and H. Song, High Quantum Yield Blue Emission from Lead-Free Inorganic Antimony Halide Perovskite Colloidal Quantum Dots, ACS Nano, 2017, 11(9), 9294–9302.
19. L. Wang, Z. Shi, Z. Ma, D. Yang, F. Zhang, X. Ji, M. Wang, X. Chen, G. Na, S. Chen, D. Wu, Y. Zhang, X. Li, L. Zhang and C. Shan, Colloidal Synthesis of Ternary Copper Halide Nanocrystals for High-Efficiency Deep-Blue Light-Emitting Diodes with a Half-Lifetime above 100 h, Nano Lett., 2020, 20(5), 3568–3576.
20. S. Yue, S. C. McGuire, H. Yan, Y. S. Chu, M. Cotlet, X. Tong and S. S. Wong, Synthesis, Characterization, and Stability Studies of Ge-Based Perovskites of Controllable Mixed Cation Composition, Produced with an Ambient Surfactant-Free Approach, ACS Omega, 2019, 4(19), 18219–18233.
21. J. Liu, X. Meng, K. Liu, Z. Chen, X. Ma, S. Liu, W. Fan, H. Liu, Z. Cheng and J. Wu, Optimizing the performance of Ge-based perovskite solar cells by doping CsGeI₃ instead of charge transport layer, Sol. Energy, 2023, 259, 398–415.
22. Y. Shen, J. Yin, B. Cai, Z. Wang, Y. Dong, X. Xu and H. Zeng, Lead-free, stable, high-efficiency (52%) blue luminescent FA₃Bi₂Br₉ perovskite quantum dots, Nanoscale Horiz., 2020, 5(3), 580–585.
23. Z. Ji, Y. Liu, W. Li, C. Zhao and W. Mai, Reducing current fluctuation of Cs₃Bi₂Br₉ perovskite photodetectors for diffuse reflection imaging with wide dynamic range, Sci. Bull., 2020, 65(16), 1371–1379.
24. S. Yan, W. Tian, H. Chen, K. Tang, T. Lin, G. Zhong, L. Qiu, X. Pan and W. Wang, Synthesis of 0D Manganese–Based Organic–Inorganic Hybrid Perovskite and Its Application in Lead–Free Red Light–Emitting Diode, Adv. Funct. Mater., 2021, 31(26), 2100855.
25. K. M. McCall, C. C. Stoumpos, S. S. Kostina, M. G. Kanatzidis and B. W. Wessels, Strong Electron–Phonon Coupling and Self-Trapped Excitons in the Defect Halide Perovskites A₃M₂I₉ (A = Cs, Rb; M = Bi, Sb), Chem. Mater., 2017, 29(9), 4129–4145.
26. M. Leng, Y. Yang, K. Zeng, Z. Chen, Z. Tan, S. Li, J. Li, B. Xu, D. Li, M. P. Hautzinger, Y. Fu, T. Zhai, L. Xu, G. Niu, S. Jin and J. Tang, All–Inorganic Bismuth–Based Perovskite Quantum Dots with Bright Blue Photoluminescence and Excellent Stability, Adv. Funct. Mater., 2017, 28(1), 1704446.
27. B. Yang, J. Chen, F. Hong, X. Mao, K. Zheng, S. Yang, Y. Li, T. Pullerits, W. Deng and K. Han, Lead–Free, Air–Stable All–Inorganic Cesium Bismuth Halide Perovskite Nanocrystals, Angew. Chem., Int. Ed., 2017, 56(41), 12471–12475.
28. M. Leng, Y. Yang, Z. Chen, W. Gao, J. Zhang, G. Niu, D. Li, H. Song, J. Zhang, S. Jin and J. Tang, Surface Passivation of Bismuth-Based Perovskite Variant Quantum Dots To Achieve Efficient Blue Emission, Nano Lett., 2018, 18(9), 6076–6083.
29. H. Yang, W. Shi, T. Cai, K. Hills-Kimball, Z. Liu, L. Dube and O. Chen, Synthesis of lead-free Cs₄(Cd_1−xMn_x)Bi₂Cl₁₂ (0 ≤ x ≤ 1) layered double perovskite nanocrystals with controlled Mn–Mn coupling interaction, Nanoscale, 2020, 12(45), 23191–23199.
30. M. Leng, Z. Chen, Y. Yang, Z. Li, K. Zeng, K. Li, G. Niu, Y. He, Q. Zhou and J. Tang, Lead–Free, Blue Emitting Bismuth Halide Perovskite Quantum Dots, Angew. Chem., Int. Ed., 2016, 55(48), 15012–15016.
31. Y. Wang, Y. Ren, S. Zhang, J. Wu, J. Song, X. Li, J. Xu, C. H. Sow, H. Zeng and H. Sun, Switching excitonic recombination and carrier trapping in cesium lead halide perovskites by air, Commun. Phys., 2018, 1(1), 96.
32. K. Park, S. Tan, T. Kodalle, D. K. Lee, M. Abdelsamie, J. S. Park, J. H. Lee, S. K. Jung, J. H. Ko, N. G. Park, C. M. Sutter-Fella, Y. Yang and J. W. Lee, Atmospheric Humidity Underlies Irreproducibility of Formamidinium Lead Iodide Perovskites, Adv. Mater., 2023, 36(14), 2307265.
33. Y. Zou, Y. Li, X. Pang, Y. Song, W. Xu and B. Sun, Unraveling Deposition Atmosphere Impact on Reproducibility of Perovskite Light-Emitting Diodes, J. Phys. Chem. Lett., 2023, 14(21), 5025–5032.
34. H. Li, J. Wu, S. Hu, W. Zhang, J. Gao, Y. Guo, P. Guo, C. Liu, G. Zhao, M. Fu, H. Bian, T. Liu, H. Wang and H. Wang, Air–Processed Blue Perovskite Light–Emitting Diodes Enabled by Manipulation of Adsorbed–Moisture–Dominated Crystallization Kinetics, Adv. Funct. Mater., 2023, 33(44), 2303787.
35. Y. Ju, X. G. Wu, S. Huang, G. Dai, T. Song and H. Zhong, The Evolution of Photoluminescence Properties of PEA₂SnI₄ Upon Oxygen Exposure: Insight into Concentration Effects, Adv. Funct. Mater., 2021, 32(2), 2108296.
36. H. Wang, Y. Tong, H. Qi, K. Wang, M. Fu and H. Wang, Water assisted chlorination enables ambient air fabrication of mixed halide perovskite films for blue light emitting diodes, J. Lumin., 2023, 263, 120114.
37. M. Leng, Y. Yang, Z. Chen, W. Gao, J. Zhang, G. Niu, D. Li, H. Song, J. Zhang, S. Jin and J. Tang, Surface Passivation of Bismuth-Based Perovskite Variant Quantum Dots To Achieve Efficient Blue Emission, Nano Lett., 2018, 18(9), 6076–6083.
38. B. Yang, J. Chen, F. Hong, X. Mao, K. Zheng, S. Yang, Y. Li, T. Pullerits, W. Deng and K. Han, Lead-Free, Air-Stable All-Inorganic Cesium Bismuth Halide Perovskite Nanocrystals, Angew. Chem., Int. Ed., 2017, 56(41), 12471–12475.
39. Z. Z. Ma, Z. F. Shi, L. T. Wang, F. Zhang, D. Wu, D. W. Yang, X. Chen, Y. Zhang, C. X. Shan and X. J. Li, Water-induced fluorescence enhancement of lead-free cesium bismuth halide quantum dots by 130% for stable white light-emitting devices, Nanoscale, 2020, 12(6), 3637–3645.
40. R. Brenes, C. Eames, V. Bulovic, M. S. Islam and S. D. Stranks, The Impact of Atmosphere on the Local Luminescence Properties of Metal Halide Perovskite Grains, Adv. Mater., 2018, 30(15), 1706208.
41. J. Haruyama, K. Sodeyama, L. Han and Y. Tateyama, Surface Properties of CH₃NH₃PbI₃ for Perovskite Solar Cells, Acc. Chem. Res., 2016, 49(3), 554–561.
42. J. Luo, X. Wang, S. Li, J. Liu, Y. Guo, G. Niu, L. Yao, Y. Fu, L. Gao, Q. Dong, C. Zhao, M. Leng, F. Ma, W. Liang, L. Wang, S. Jin, J. Han, L. Zhang, J. Etheridge, J. Wang, Y. Yan, E. H. Sargent and J. Tang, Efficient and stable emission of warm-white light from lead-free halide double perovskites, Nature, 2018, 563(7732), 541–545.
43. Z. Fang, H. He, L. Gan, J. Li and Z. Ye, Understanding the Role of Lithium Doping in Reducing Nonradiative Loss in Lead Halide Perovskites, Adv. Sci., 2018, 5(12), 1800736.
44. L. Zhou, J. F. Liao, Y. Qin, X. D. Wang, J. H. Wei, M. Li, D. B. Kuang and R. He, Activation of Self–Trapped Emission in Stable Bismuth–Halide Perovskite by Suppressing Strong Exciton–Phonon Coupling, Adv. Funct. Mater., 2021, 31(31), 2102654.
45. B. R. Sutherland and E. H. Sargent, Perovskite photonic sources, Nat. Photonics, 2016, 10(5), 295–302.
46. P. Dey, J. Paul, J. Bylsma, D. Karaiskaj, J. M. Luther, M. C. Beard and A. H. Romero, Origin of the temperature dependence of the band gap of PbS and PbSe quantum dots, Solid State Commun., 2013, 165, 49–54.
47. H. C. Woo, J. W. Choi, J. Shin, S. H. Chin, M. H. Ann and C. L. Lee, Temperature-Dependent Photoluminescence of CH₃NH₃PbBr₃ Perovskite Quantum Dots and Bulk Counterparts, J. Phys. Chem. Lett., 2018, 9(14), 4066–4074.
48. Y. Wang, X. Zhu, H. Jia, D. Xu, X. Liu, Y. Bai, X. Bian, F. Wang, M. Li and Z. A. Tan, In situ fabrication of perovskite nanocrystals with Dion–Jacobson phase for high-performance blue light-emitting diodes, Appl. Phys. Rev., 2024, 11(3), 031413.
49. H. Zhao and H. Kalt, Energy-dependent Huang-Rhys factor of free excitons, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 68(12), 125309.
50. X. Lao, Z. Yang, Z. Su, Z. Wang, H. Ye, M. Wang, X. Yao and S. Xu, Luminescence and thermal behaviors of free and trapped excitons in cesium lead halide perovskite nanosheets, Nanoscale, 2018, 10(21), 9949–9956.

---

Footnote

† These authors contributed equally to this work.

---