Whether or Not Emission of Cs ₄ PbBr ₆ Nanocrystals: High-Pressure Experimental Evidence

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Apr 2020Whether or Not Emission of Cs4PbBr6 Nanocrystals: High-Pressure Experimental Evidence Zhiwei Ma, Fangfang Li, Dianlong Zhao, Guanjun Xiao and Bo Zou Zhiwei Ma State Key Laboratory of Superhard Materials, College of Physics, Jilin University, Changchun 130012 (China) , Fangfang Li Key Laboratory of Physics and Technology for Advanced Batteries (Ministry of Education), College of Physics, Jilin University, Changchun 130012 (China) , Dianlong Zhao State Key Laboratory of Superhard Materials, College of Physics, Jilin University, Changchun 130012 (China) , Guanjun Xiao *Corresponding authors: E-mail Address: [email protected], E-mail Address: [email protected] State Key Laboratory of Superhard Materials, College of Physics, Jilin University, Changchun 130012 (China) and Bo Zou *Corresponding authors: E-mail Address: xguan[email protected], E-mail Address: [email protected] State Key Laboratory of Superhard Materials, College of Physics, Jilin University, Changchun 130012 (China) https://doi.org/10.31635/ccschem.020.201900086 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail The origin of green emission in the zero-dimensional (0D) perovskite Cs4PbBr6 nanocrystals (NCs) remains a considerable debate. Herein, an approach involving a combination of high-pressure experiments and theoretical simulation was employed to elucidate the controversial origin of photoluminescence from emissive Cs4PbBr6 NCs (E416). Results obtained from first-principles density functional theory (DFT) calculations, as implemented in the Vienna ab initio simulation package codes, implied that the photoluminescence energies from bromine vacancy decreased persistently with pressure. Experimentally, the photoluminescence energies tended to decrease in the low-pressure region, followed by an increase beyond ∼1.4 GPa. While the emergent disagreement between the first-principles calculation and high-pressure experiment excludes the possibility of vacancy-tuning, the consistent change observed in the pressure-dependent emission between E416 and CsPbBr3 NCs offered a reliable interpretation for the occurrence of green emission from a CsPbBr3 impurity embedded in the Cs4PbBr6 matrix. Further comprehensive analysis demonstrated that the strong green emission of E416 NCs originated from the impurity CsPbBr3 NCs embedded in Cs4PbBr6 matrix. Our study represents a significant step forward to a deeper understanding of the emissive origins of Cs4PbBr6 NCs and promotes the application of this novel strategy in light-emitting devices. Download figure Download PowerPoint Introduction The all-inorganic cesium lead halide perovskite (Cs4PbX6, where X = Cl, Br, I) is a typical zero-dimensional (0D) perovskite material, in which the octahedra are isolated completely by cation bridges, and charge carriers are localized within the ordered metal halide component.1–4 Recent works have demonstrated that Cs4PbBr6 possesses a higher exciton binding energy and photoluminescence (PL) quantum yield (PLQY) both in its single crystals and nanocrystalline forms.5–8 In a few cases, the PL intensity is found to be much higher than that of CsPbBr3.9 Therefore, these perovskite materials are expected to be highly useful in light-emitting applications.7,9 However, some very recent reports have shown that Cs4PbBr6 nanocrystals (NCs) do not exhibit any PL in the visible region because of seldom display of excited broad bandgap.1–4 Accordingly, whether or not emissive Cs4PbBr6 NCs occur sparked a sudden great discussion. On one hand, some investigators hypothesize that the emission of 0D perovskite Cs4PbBr6 NCs arises from the midband gap states formed due to the halogen vacancy.10,11 On the other hand, continued efforts demonstrated that the pure Cs4PbBr6 NCs do not exhibit any emission within the visible light spectrum region, and thus, the green emission is ascribed to an impurity CsPbBr3-like phase embedded in Cs4PbBr6 matrix.1,5,6,12 Although many attempts to reveal the origin of PL from Cs4PbBr6 NCs were made by several independent groups; still, the debate has hardly subsided. Through various investigations, we have noted that the state of localized state vacancies depends on the host media and their transition energy, which could be dependent weakly on the change of the host's bandgap.13,14 From this point, we envisioned that the study of the correlation between the nanostructure and optical properties could indeed distinguish between point defects and embedded nanostructures to reveal the origin of the green emission derived from Cs4PbBr6 NCs. Pressure application, as a unique thermodynamic variable, provides a powerful means of studying the structural and emissive behaviors of nanomaterials.15–17 In view of this, three-dimensional (3D) lead halide perovskite NCs have been employed to explore if the pressure dependence and the bandgap alignment of CsPbX3 NCs could be finely tuned successfully via utilization of pressure.18–20 Besides, pressure-sintered CsPbBr3 NCs have been shown to exhibit a 1.6-fold enhancement in PL intensity and display longer emission lifetimes than untreated NCs.21 Furthermore, significant progress has been made in the studies of pressure-induced emission of non-emissive Cs4PbBr6 (N416) NCs in a recent work.22 However, so far, adopting the pressure-response PL behaviors of emissive E416 NCs to unravel the origin of the emission has never been considered. Here, we have systematically investigated high-pressure-induced optical changes of E416 NCs through first-principles density functional theoretical (DFT) and high-pressure experimental methods. The dependence of the charge-transition level for Cs4PbBr6 with bromine vacancy on the external pressure was obtained via the first-principles calculations, which displayed a persistent decrease in the separation of the defect states and conduction band minimum (CBM) upon compression. We deduced from this calculated result that the PL from bromine vacancy exhibits a persistent redshift wavelength with increasing pressure. However, a high-pressure PL peak shift was detected upon compression as follows: a gradual redshift initially occurred below 1.4 GPa, followed by a persistent blueshift with further increasing pressure. The observed pressure-dependent PL behaviors contradicted with the calculated result, but well coincidental with the PL evolution of CsPbBr3 NCs under high pressure. Upon further compression, a distinct broad emission appeared at 3.02 GPa, consistent with the PL behavior of N416. By comparing the theoretical and experimental results, we illustrated that the green emission of Cs4PbBr6 NCs mainly derived from the impurity of CsPbBr3 NCs embedded in Cs4PbBr6 matrix rather than its intrinsic emission from a halogen vacancy. With an increase in pressure, we deduced that the initial behavior of PL was attributable to an impurity in the CsPbBr3 NCs, whereas, the sudden appearance of a broad emission derived from pure Cs4PbBr6 NCs. Meanwhile, the absorption spectrum from a mixture of the CsPbBr3 and N416 NCs revealed that the abnormal absorption tail of E416 NCs originated from the superposition of the absorption band from N416 NCs and the impurity CsPbBr3 NCs. Our findings offer a clue critical for the emissive causes of E416 NCs and facilitate the understanding and establishment of a novel optical phenomenon. Experimental Methods Sample preparation The E416 NCs were synthesized according to a modified reverse microemulsion approach. Typically, a dimethylformamide (DMF) solution of PbBr2, hydrogen bromide (HBr), oleic acid (OA), and oleylamine (OAm) was injected into an n-hexane solution of cesium oleate (CsOL) and OA under vigorous stirring. The CsPbBr3 NCs were prepared by the transitional hot-inject method. The NCs were obtained by reacting Cs-oleate with a Pb(II)-halide in a high-boiling solvent (octadecene) containing a 1∶1 mixture of OAm and OA at 140−200 °C. The N416 NCs were prepared by modifying the synthesis approach for CsPbBr3 NCs using an excess of CsOL and lowering the reaction temperature (for details, see the ). General procedure for generation of high-pressure E416 NCs High-pressure experiments for the fabrication of the E416 NCs were carried out in symmetric diamond anvil cell (DAC). The sample, together with a small ruby ball, were loaded into the 150 µm diameter chamber of the DAC, constructed from a T301 steel gasket indented to a thickness of 45 µm. A pressure calibration was performed using the standard ruby fluorescent technique (23). In the high-pressure experiments, silicon oil was utilized as the pressure transmitting medium for PL, whereas argon was employed as a pressure transmitting medium for Raman measurements. All measurements were conducted at room temperature. Results and Discussion The morphology and structure of the synthesized E416 NCs were investigated using transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) characterizations. Figures 1a and 1b show the Cs4PbBr6 NCs displaying nanosphere morphology before compression with good crystallinity. The NCs had an average diameter of 18.1 nm, with a standard deviation of 1.9 nm. The inset shows that the E416 NCs display a bright green color under UV light irradiation. Figure 1c illustrates the crystal structure of the 0D perovskite Cs4PbBr6 in its rhombohedral phase along the perpendicular c axis. The Cs4PbBr6 consisted of the isolated [PbBr6]4− octahedra as 0D perovskite structure, where [PbBr6]4− octahedra were surrounded by Cs+ cations. The absorption spectrum of synthesized Cs4PbBr6 NCs before compression exhibited a long absorption tail accompanied by a strong absorption peak at 316 nm (3.92 eV), as shown in Figure 1d, whereas the PL spectrum was centered at 517 nm with a standard Gaussian profile and a full width at half maximum (FWHM) of 25 nm (Figure 1d), consistent with previous reports.7,8 The position of the green emission showed a large Stokes shift value (∼200 nm) relative to the anticipated band edge position. Figure 1 | (a) Representative TEM image of Cs4PbBr6 nanocrystals (NCs). Inset shows the size distribution histogram of Cs4PbBr6 NCs, featuring a diameter of 18.1 nm and the PL photographs under UV irradiation. (b) The High-resolution TEM (HRTEM) image of Cs4PbBr6 NCs before compression. The scale bars in the TEM image are 100 nm and 20 nm in the HRTEM image. (c) Schematic crystal structure of Cs4PbBr6 along and perpendicular to [001] direction. (d) Steady-state absorption and PL spectra. Download figure Download PowerPoint We also investigated the structure and optical properties of CsPbBr3 NCs. As shown in , the CsPbBr3 NCs exhibited a nearly monodisperse nanocube morphology. Within the CsPbBr3 NCs, the [PbBr6]4− octahedra formed an extended 3D network by an all-corner-connected type. The Cs+ filled the hole among the octahedra, which balanced the charge of the whole network (the inset of ). The steady-state absorption and PL spectra of the synthesized CsPbBr3 NCs are shown in , which are consistent with a previous report.24 Notably, the sharp PL peak was centered at 519 nm, which was highly close to the PL position of E416 NCs. The similarity in the crystallographic components and PL properties propelled more scientists to hypothesize that the origin of the PL for E416 NCs might be related to the impurity in the CsPbBr3 NCs. The CsPbBr3 impurity was thought to be embedded possibly in Cs4PbBr6 matrix, as reported by Zhong and co-workers.5 However, the conventional characterizations of E416 NCs did not exhibit any signal assigned to CsPbBr3. The same difficulty appears to have been encountered by several independent groups.1,5 It is presumed that due to the tiny size of the CsPbBr3 embedded in Cs4PbBr6 matrix, it results in an amorphous formation, which leads to a broad width and low intensity of X-ray diffraction (XRD) signals. Besides, the low doping level of CsPbBr3 in Cs4PbBr6 NCs is another critical factor for the absence of XRD signals from CsPbBr3 NCs.1,5 Consequently, the CsPbBr3 NCs embedded in Cs4PbBr6 NCs are difficult to observe by conventional methods. Nevertheless, Nikl et al.25 first illustrated CsPbBr3 impurity embedded in Cs4PbBr6 film by absorption, emission, and decay kinetics characteristics. Recently, Riesen et al.26 also confirmed that the green emission originated from the CsPbBr3 impurity based on cathodoluminescence imaging results. Note that, usually, the extremely small size of CsPbBr3 NCs theoretically exhibits blue emission due to dielectric mismatch between NCs and organic ligands, which improves the confinement effect. However, within Cs4PbBr6 matrix, the wide bandgap Cs4PbBr6 NCs could be regarded as a finite quantum well surrounding CsPbBr3 NCs, which is different from organic ligands possessing an infinite quantum well model. According to the calculation of finite quantum well models,27 although the size of CsPbBr3 impurity was so small, the E416 was still able to exhibit green emission. Likewise, several recent reports hypothesized that the origin of PL was attributable to the midband gap states derived from the halogen vacancy.10,11 The bromide vacancies in 0D perovskite Cs4PbBr6 have low formation energy and an appropriate defect level to contribute to the midgap radiative state. Additionally, first-principles calculations also present a valuable methodology for the clarification of the structural effects on the defect and energetic landscape. However, a proper calculation on the correlation between the defect states and its structural evolution in Cs4PbBr6 lacks so far. We adopted the first-principles calculations by building halogen vacancy in our fabricated E416 NCs to reveal the response of the defect states to the external pressure application. The dependence of the electronic structure on the external pressure could be well-understood by considering the projected density of states (PDOS) shown in Figure 2a. The total density of states (DOS) is separated to show contributions by Pb-6s and -6p states, and Br-4s and -4p states, yielding a calculated band gap of pristine perovskite as 3.86 eV, consistent with the experimental results (Figure 1d). Besides, a 2 × 1 × 1 rhombohedral supercell with 12 formula units of Cs4PbBr6 was used to optimize the structures of the halogen vacancies, simulated by removing a single bromine atom from the supercell ().28,29 The lattice constants at different pressure were derived by in situ high-pressure angle-dispersive X-ray diffraction (ADXRD) experiment (), typically used in structural optimization. During this process, the lattice constants were fixed and the symmetrical structures assumed at varying pressures were disabled, while atomic positions remained relaxed until the calculated Hellmann–Feynman forces on the atoms were lower than 0.02 eV/Å. Within the defective structure, the signature of the defect states became evident and appeared inside the bandgap below the CBM level with the electron captured at Pb-6p (Figure 2b). The defect state was populated by an electron, and therefore, the bromine vacancy acted as an n-type dopant and introduced negative charge carriers near the conduction band, thereby acting as an electron donor and creating a hole trap near the conduction band. Accordingly, the defect PL was indeed dependent on transitions between the defect states and the CBM. Figure 2 | PDOS for (a) pristine Cs4PbBr6 and (b) Cs4PbBr6 with halogen vacancy defects obtained for the optimized geometry. (c) The pressure-dependent total density of states for Cs4PbBr6 with halogen vacancy defects. (d) The separation of transition energy levels for defects (halogen vacancy) and the CBM (ΔE) with increasing pressure. Download figure Download PowerPoint Figure 2c displays a high-pressure behavior of the total DOS of Cs4PbBr6 with the bromine vacancy. As could be seen, there is no sudden change with increasing pressure. Also, the DOS in Figure 2c could not represent the evolution of the defect state in Cs4PbBr6. The precise calculations on the charge-transition level for halide vacancy, intuitively, exhibited the evolutionary trend of the defect state of halide vacancy in Cs4PbBr6 with increasing pressure (see the for details). Figure 2d displays the energy differences (ΔE) between the transition energy levels of the bromine vacancy and the CBM. With the increase in pressure, the ΔE exhibited a persistent narrowing by ∼0.1 eV from an initial 2.37 eV, suggesting that the PL peak underwent a continuous redshift. Further, the valence band maximum (VBM) consists mostly of Br-4p and Pb-6s states, whereas the Pb-6p states dominate the CBM. With an increase in pressure, the energy of VBM underwent a persistent increase until attaining a GPa of 2.37 (Figure 2c). During this process, the optimized supercell structures exhibited a persistent reduction in the six bond lengths of Pb–Br within the octahedra upon compression. The decrease in the Pb–Br bonds enhanced coupling between the Br-4p and Pb-6s orbitals, increasing the VBM energy.30 However, the CBM was less sensitive to the contraction of the Pb–Br bond length and changed only slightly upon compression owing to the nonbonding characteristic of Pb-6p orbitals (Figure 2c). Therefore, the evolutionary trends of the calculation demonstrated a persistent decrease in bandgap energy (). We considered that if the PL of E416 originated from the midband gap states formed due to the halogen vacancy, both the PL and absorption peak of E416 should exhibit a continuous redshift, consistent with the theoretical predictions mentioned above. Therefore, further exploring of the in situ high-pressure absorption measurement of E416 NCs was necessary. Hence, we performed a modulation process by UV–vis absorption spectroscopy up to 18.64 GPa () to trace the bandgap. Our results showed that under high pressure, the strong absorption peak experienced a slight redshift before reaching ∼3.09 GPa. When the pressure was increased beyond ∼3.09 GPa, the profile of the absorption peak underwent an abrupt change and then developed into two absorption peaks (). This distinct change in the absorption spectrum was associated with a pressure-induced structural phase transformation.22 The bandgap of E416 NCs was estimated from the extrapolation of the linear region to the intercept of the energy axis by plotting the linear trend of the spectral dependency (αhν)2 over the variations of photon energies (hν) of the band-to-band transition signals, known as a Tauc plot, where α is the absorption coefficient, (αhν)2 is defined as y axis, and the photon energy hν is defined as x axis. Detailed evolution of the bandgap energy for E416 NCs with increasing pressure is displayed in . The electronic band exhibited a linear narrowing by 0.02 eV from the initial 3.59 eV, which confirmed the validity of the evolutionary tendency derived from the theoretical calculations (). However, we found that the absorption edge attributable to the defect states was very weak. Therefore, to thoroughly understand the nature of the optical properties, an in situ high-pressure PL measurement for E416 NCs was performed. The high-pressure-dependent PL spectra of E416 NCs were recorded up to 18.26 GPa, as shown in Figure 3a and . With increasing pressure, the PL intensity decreased sharply and gradually became broad upon compression to 1.62 GPa, after which the fluorescence was quenched abruptly. The pressure dependence of the PL wavelength was recorded up to 2.01 GPa (Figure 3b). The PL peak displayed a gradual redshift in wavelength. However, beyond 1.41 GPa, the PL peak showed a sudden blueshift, as shown in Figure 3b. Upon further compression, a broad emission band with a full-width half max (FWHM) of ∼220 nm suddenly appeared () 3.02 GPa, which is in with the high-pressure compression result of N416 NCs. The origin of this broad emission was attributable to the radiative of associated with structural phase We presumed that if the green PL originated from the vacancy state of the E416 NCs, based on its much energy, a state than the bandgap of the Cs4PbBr6 should as a result of the fabricated NCs very and persistent spectral redshift any point, and should the first-principles calculations, as shown in Figure the evolutionary trend of PL position with pressure is an the Figure | (a) PL spectra of E416 NCs. (b) PL peak position of E416 NCs as a of pressure. (c) High-pressure PL properties of CsPbBr3 NCs. (d) PL peak position of CsPbBr3 NCs at various pressure Download figure Download PowerPoint We that the ligands possibly the optical and properties of NCs. However, the absorption and PL spectra of perovskite NCs were by electronic transitions within the components with the very of ligands on energy and of these typical ligands as the and used in our this is indeed the as energies are from perovskite NCs states, strong to the very optical properties, the of optical behavior with increasing pressure between CsPbBr3 and Cs4PbBr6 NCs be an for the origin of emission from E416 NCs. Therefore, we further performed the high-pressure PL experiments for CsPbBr3 NCs with the results shown in Figure The PL peak were detected as follows: a gradual redshift occurred the GPa followed by a persistent blueshift (Figure In the PL intensity also decreased upon compression to GPa, after which the fluorescence was The sudden changes of the PL at GPa and the of PL intensity were attributable to the phase phase transition is associated with the same and considered to have originated from the electronic structural change in a Although the remained upon compression, the evolution of PL and the absorption spectra that occur under high pressure was a significant the in situ high-pressure PL measurement that both the Cs4PbBr6 and CsPbBr3 NCs emissive behavior with increasing pressure. localized states as vacancies, their state was dependent on the host Therefore, upon compression, a PL behavior of E416 NCs should to its structural This to the between the optical properties and structure of E416 NCs. In recent the in situ high-pressure and Raman of N416 NCs revealed that these materials did not any change below We also in situ high-pressure of E416 NCs as shown in . The measurements that the E416 NCs did not a structural phase transition this The in situ high-pressure Raman spectra of E416 NCs and N416 NCs were also conducted as shown in . E416 and N416 NCs exhibited the same The of their Raman with compression, as illustrated in , show a continuous any the Raman and measurements confirmed that the E416 NCs did not a structural phase transition this Additionally, there was no sudden change in the pressure-dependent steady-state absorption spectra (). Therefore, both the pressure-induced structural and phase transitions in Cs4PbBr6 NCs could be out as to the PL emission from the E416 NCs. based on the high-pressure measurements the is that the PL from the CsPbBr3 impurity rather than the intrinsic emission of Cs4PbBr6 NCs. Therefore, in this our experiment provides to that the of defect states from the bromine vacancy is not the factor for the emission of Cs4PbBr6 NCs. The non-emissive Cs4PbBr6 (N416) NCs in the pressure of GPa have been investigated in previous work.22 The N416 NCs initially exhibited no PL response to external pressure GPa. this a broad emission band with an of nm suddenly the pressure-dependent PL of CsPbBr3 NCs has also been by The PL intensity of CsPbBr3 NCs decreased sharply upon compression to GPa, after which the PL quenched abruptly. In our present we also PL spectra upon application of much higher pressure but did not observe any PL peak appearance with increasing pressure, consistent with the results of previous By comparing the emissive behavior of E416 NCs, N416 NCs, and CsPbBr3 NCs, we that the pressure-induced emission beyond GPa is assigned to the structural phase transition of Cs4PbBr6 NCs (the large of [PbBr6]4− octahedra this However, in the low-pressure region, the Cs4PbBr6 NCs did not any significant structural and exhibited an abnormal green emissive behavior upon compression. it did not before the phase Consequently, we that with an increase in pressure, the initial behavior of PL could be attributable to the impurity CsPbBr3 NCs, whereas, the sudden appearance of the broad emission derived from pure Cs4PbBr6 NCs. This high-pressure experiment has to that the emission from E416 NCs originated from the CsPbBr3 NCs embedded in the Cs4PbBr6 matrix. Furthermore, with the of a the optical photographs also revealed the trend of the green PL in E416 and CsPbBr3 NCs, as shown in Figure due to the shift of the PL peak in the two below 2 GPa, the pressure-dependent color change could not be by the The optical also demonstrated the trend of the broad emission in E416 NCs beyond GPa, accompanied by color changes from bright to (Figure It is that the loaded are rather than completely NCs in results in a of NCs in We further the PL spectra of the and NCs, as illustrated in . We observed that the PL peaks from the NCs were slightly and than their that the states of these NCs underwent a This possibly from the between the ligands and NCs, as well as and the photon process in NCs. Figure | The high-pressure PL photographs of E416 NCs (a) and CsPbBr3 NCs (b) under UV = nm) in a diamond anvil Download figure Download PowerPoint The [PbBr6]4− octahedra within Cs4PbBr6 completely in as a result of electronic between octahedra Therefore, the unique perovskite structure an at the bandgap of eV, shown in Figure 2a. The octahedra in Cs4PbBr6 NCs valence and conduction (Figure which in band However, the absorption spectrum of synthesized E416 NCs before compression exhibited a long absorption tail accompanied by a strong absorption peak at 316 nm (3.92 eV), as shown in which is with the calculated The absorption spectrum of N416 NCs before compression was also as shown in with the absorption spectrum of E416 NCs, the N416 NCs exhibited a absorption peak at nm the long which is more consistent with the results of the

Topics & Concepts