Revealing the role of a unique local structure in lanthanide-doped Cs₂LiInCl₆ in boosting visible and NIR-II luminescence

Abstract

Local structure engineering is one of the most useful strategies for tunable down-shifting and upconversion emissions in halide double perovskites (DPs). However, the roles of the local structure, including local symmetry, coordination structure, and active sites, in these DPs remain largely unexplored. Herein, we studied the local structure-correlated up/down-conversion luminescence in Yb³⁺/Er³⁺/Ho³⁺ co/tri-doped Cs₂LiInCl₆ with a unique local structure, which is different from its typical Cs₂NaInCl₆ counterpart. At optimal concentrations, the emission intensities of Cs₂LiInCl₆:Yb³⁺,Er³⁺ were 6 and 110 times higher than those of Cs₂NaInCl₆ at 552 and 1540 nm, respectively. The evolution of the local site symmetry calculated by the first-principles density functional theory demonstrates that the lower local symmetry of Cs₂LiInCl₆:Yb³⁺,Er³⁺ is responsible for luminescence enhancement at a low-doping level of Er³⁺. In addition, we verified that the concentration quenching effect can be effectively suppressed by two independent In³⁺ sites in Cs₂LiInCl₆, leading to increased visible and near-infrared (NIR-II) emissions close to the optimal concentration with higher Er³⁺ contents. Moreover, time-resolved photoluminescence (PL) measurements and steady-state rate equations are used to thoroughly understand the mechanism underlying the local-structure-dependent PL properties. Our results demonstrate that both the local symmetry breaking and presence of multiple independent active sites are beneficial for luminescence enhancement, and accordingly provide meaningful guidance for designing highly efficient visible and NIR-II emitters.

---

Introduction

Lead-free halide double perovskites (DPs) as promising luminescent materials have received increased interest from researchers for potential applications in various materials science and biomedical fields owing to their low toxicity and high chemical-, thermal-, and photo-stability.^1–8 However, the photoluminescence (PL) performance is very poor for most DPs in view of the indirect bandgaps or parity-forbidden transitions, limiting their further practical photonic and biological applications. Fortunately, DPs have the general formula A₂M⁺M³⁺X₆ (where A = Rb⁺ and Cs⁺; M⁺ = Ag⁺ and Na⁺; M³⁺ = In³⁺ and Bi³⁺; and X = Cl⁻ and Br⁻) that provides abundant opportunities for the material composition adjustment. For example, Luo et al. used Na⁺ for the partial substitution of Ag⁺ in Cs₂AgInCl₆, resulting in the break of the parity-forbidden transition.¹ Similarly, Pei et al. found the Na⁺-induced breakdown of the local site symmetry in Cs₂(Na/Ag)BiCl₆ leading to an obviously improved near-infrared (NIR) luminescence.⁹ In particular, lanthanide (Ln) elements are generally suitable for incorporation into DPs for favorable optical properties owing to the same valence state as that of In³⁺ or Bi³⁺, proper octahedral coordination environment, and special ladder-like 4f electronic states. Consequently, DP hosts with Ln-ion-doping often display enhanced visible or NIR emissions with sharp peaks.^10–13

The second near-infrared (NIR-II, 1000–1700 nm) spectral region has shown great application prospects in NIR-based lighting devices, biosensing, optical fiber communication, high-resolution imaging, information storage, etc.^14–18 Among Ln ions, Er³⁺ has a good spectral response of approximately 1540 nm with a relatively large absorption cross-section and is therefore commonly used as either an NIR-II sensitizer in up-conversion (UC) materials or NIR-II emitting center in down-conversion (DC) systems. Unfortunately, the photoluminescence quantum yields (PLQY) of Er-doping UC/DC materials are still low because of the lack of appropriate host materials or powerful design strategies. To date, one of the most useful strategies for tunable down-shifting and upconversion emissions is local structure engineering.^13,19,20 For example, Er³⁺–Sb³⁺ co-doped Cs₂NaInCl₆ has been demonstrated as a potential host for 1540 nm emission by partially alleviating the parity-forbidden rule within 4f intraconfigurational transitions of Er³⁺.¹⁰ However, most of the previous works proved that the visible and NIR-II emission enhancement cannot only be attributed to the change of the local structure as the synergistic effect of the improved crystallization also works.^1,9,10,21 The role of the local structure alone, such as local symmetry, coordination structure, and interionic distance, in these DPs, remains largely unexplored. Therefore, it requires in-depth corroboration of the exact effect of the local structure on rational modulation of both visible and NIR emissions in the UC/DC systems.

In this work, we have, for the first time, synthesized lead-free Yb³⁺/Er³⁺/Ho³⁺ co/tri-doped Cs₂LiInCl₆ double perovskite with a unique local structure, including low unit cell symmetry (trigonal), two independent In³⁺ active sites, proper In³⁺ separation distance, and large crystal-field strength. By comparing with the conventional Cs₂NaInCl₆ counterpart, we provide direct evidence that local symmetry breaking and the presence of multiple active sites are two priority factors for emission enhancement and elucidate the local-structure-correlated up/down-conversion luminescence in terms of emission intensity, transition selectivity, wavelength shift, and lifetime.

Results and discussion

We selected Cs₂LiInCl₆, a double perovskite, as a prototype to uncover the role of local structure, because of its different local structure compared with the commonly used DPs, such as Cs₂AgInCl₆, Cs₂NaBiCl₆, and Cs₂NaInCl₆ (in our study for comparison). Meanwhile, other factors, including phonon energy, crystallization quality, and crystal size, can be excluded, because they are very similar for both Cs₂NaInCl₆ and Cs₂LiInCl₆. Both Cs₂LiInCl₆ and Cs₂NaInCl₆ DPs were synthesized using a solid-state method (Fig. S1†). As shown in Fig. 1a, the unique local structure of Cs₂LiInCl₆ DP has two points: one is the trigonal-phase crystallographic structure compared with the cubic-phase structure of Cs₂NaInCl₆, and the second is that Cs₂LiInCl₆ has two independent In³⁺ active sites (In1 and In2) with a lower site symmetry of D_3d, whereas Cs₂NaInCl₆ has only one In³⁺ active site with the highest site symmetry of O_h.⁶ The distances of In1–In1 (or In2–In2) and In1–In2 are 7.3184 and 7.3712 Å, respectively, confirming that the two In sites are not crystallographically equivalent. The difference could allow us to thoroughly understand the mechanism underlying the local-structure-dependent PL properties. As a proof of concept, Yb³⁺/Er³⁺/Ho³⁺ ions were doped into the host for both upconversion luminescence and NIR emissions. As shown in the X-ray diffraction (XRD) patterns in Fig. 1b, it was found that almost all the main peaks match well with the standard pattern of Cs₂LiInCl₆ (PDF#79-0775), indicating the retaining trigonal phase. We then carefully compared the XRD results with the standard pattern of Cs₂LiInCl₆, and found that there are some impurity peaks belonging to Cs₂InCl₅(H₂O), due to the easy hygroscopicity of Li-based materials.²² Fortunately, such impurities have no impact on our results. In addition, the peak positions shifted toward a smaller angle, as a result of lattice expansion by the substitution of In³⁺ by Yb³⁺, Er³⁺, and/or Ho³⁺ with larger ionic radii according to Bragg's law. In a Yb–Er coupled system, Yb³⁺ is widely used as the sensitizer for 980 nm light, and enables efficient energy transfer to Er³⁺ (Fig. 1c). Therefore, green and 1540 nm NIR-II emissions can be obtained simultaneously under 980 nm excitation. The scanning electron microscopy (SEM) images show that the products are micro-sized crystals with uniform distribution of all the elements of Cs, In, Cl, Yb, Er, and Ho using energy dispersive X-ray spectroscopy (EDS mappings in Fig. 1d). The actual doping content of lanthanide ions in Cs₂LiInCl₆ was determined by EDS and the induced coupled plasma optical emission spectrometry, as shown in Table S1 and Fig. S2.†

Fig. 1 (a) The crystalline structure of trigonal-phase Cs₂LiInCl₆ with two independent In³⁺ active sites: In1 and In2. (b) XRD patterns of Cs₂LiInCl₆ and Yb³⁺/Er³⁺/Ho³⁺ co/tri-doped Cs₂LiInCl₆ DPs. * are the impurity peaks of Cs₂InCl₅(H₂O). (c) A schematic illustration of the energy transfer process from Yb³⁺ to Er³⁺ for the green UC emission and 1540 nm DC luminescence. (d) The SEM image and EDS distribution of the elements in Yb³⁺/Er³⁺/Ho³⁺ tri-doped Cs₂LiInCl₆.

For a better understanding of the local environment, Raman spectroscopic measurements were performed. According to the factor group analysis (Raman and hyper-Raman scattering database on the Bilbao Crystallographic Server, cryst.ehu.es), the following number of Raman modes can be found for Cs₂LiInCl₆, Γ_Raman = 7A_1g + 9E_g. Although all these modes are Raman-active, there are only three main Raman modes, as shown in Fig. 2a. Comparing with Cs₂NaInCl₆ DP,^6,10 the lowest frequency mode (∼174 cm⁻¹) of pure Cs₂LiInCl₆ belongs to the E_g mode, due to the asymmetric stretching of In–Cl bonds in the octahedral cages, and the other two frequency modes of 289 and 321 cm⁻¹ can be attributed to the A_1g mode related to the octahedrons. After Yb/Er co-doping, both A_1g and E_g modes were slightly shifted to longer wavenumbers due to the sublattice distortion of the octahedrons. The UV-vis absorption spectra show that the host absorption band edge located at 235 nm is almost unchanged after doping with lanthanide ions (Fig. 2b). Some new strong absorption in the 270–550 nm could be associated with the presence of Yb and Er. The optical band gaps of Cs₂LiInCl₆ and Cs₂NaInCl₆ DPs were evaluated to be 4.7 and 4.9 eV, respectively (Fig. S3†). The calculated electronic bandgaps were approximately 3.2 and 2.9 eV (Fig. S4†), respectively, which were underestimated using the first-principles method. Thermogravimetric analysis (TGA) was also conducted for both pure and Yb/Er co-doped Cs₂LiInCl₆ samples. As shown in Fig. 2c, there were two stages of mass loss, 100 °C and 540 to 560 °C. The mass change at 100 °C should be ascribed to the desorption of physically or chemically absorbed H₂O, because of the easy hydration of Li-based materials. The mass loss in the high-temperature range indicates that the samples begin to decompose gradually. These results indicate that the thermal stability of Cs₂LiInCl₆ DP is as high as those of other halide double perovskites. Fig. 2d shows the UC and NIR emission spectra of Cs₂LiInCl₆ DP with different doping concentrations under 980 nm excitation. Three emission peaks, located at 526, 552, and 669 nm, can be assigned to ²H_11/2, ⁴S_3/2, and ⁴F_9/2 to ⁴I_15/2 transitions of Er³⁺. The short-wave infrared emission shows multiple peak features at 1540, 1546, and 1561 nm with ultra-narrowband full width at half maximum (FWHM) of 6 nm, corresponding to the ⁴I_13/2 → ⁴I_15/2 transition of Er³⁺.

Fig. 2 (a) Raman spectra of Cs₂LiInCl₆ and Cs₂LiInCl₆:18%Yb³⁺,2%Er³⁺. (b) Absorbance spectra of Cs₂LiInCl₆, Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ and Cs₂LiInCl₆:18%Yb³⁺,10%Er³⁺,2.5%Ho³⁺. (c) TGA results of Cs₂LiInCl₆ and Cs₂LiInCl₆:18%Yb³⁺,2%Er³⁺. (d) PL spectra of Cs₂LiInCl₆:15%Yb³⁺,2%Er³⁺, Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺, and Cs₂LiInCl₆:20%Yb³⁺,8%Er³⁺.

In principle, a high dopant concentration will definitely benefit the overall brightness of UC and NIR emissions. However, the concentration quenching effect of lanthanide ions always limits their optimal concentration.^13,23 To determine the optimal concentrations of Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs, the emission intensities of both visible and NIR regions were measured as a function of the total doping content (% Yb³⁺ + % Er³⁺). The comparison can be clearly seen in Fig. 3a and b. At a relatively low doping level of Er³⁺ (blue region at 552 nm), the optimal concentration of Cs₂NaInCl₆ DP was 18% Yb³⁺ + 2% Er³⁺, while the intensity of Cs₂LiInCl₆ was 3 times stronger than that of Cs₂NaInCl₆. At a high doping concentration of Er³⁺ (yellow region), the concentration quenching effect inevitably occurred, leading to decreased emission intensity for Cs₂NaInCl₆. Whereas the trend was different for Cs₂LiInCl₆, the UC luminescence can be amplified continuously until reaching the maximum intensity at 18% Yb³⁺ + 5% Er³⁺. At this time, Cs₂LiInCl₆ DP was nearly 6 times brighter than Cs₂NaInCl₆ DP with the highest concentration of Er³⁺. The DC emissions at 1540 nm showed similar trends for both Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs. The optimal concentration of Cs₂NaInCl₆ was 15% Yb³⁺ + 5% Er³⁺, in comparison, with a higher optimal concentration for Cs₂LiInCl₆ was 15% Yb³⁺ + 10% Er³⁺ (Fig. 3b). Meanwhile, the enhancement factors were 90 times and 110 times at these two concentrations, respectively. Fig. 3c and d show the compared PL spectra of the UC and NIR emissions of Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs at the corresponding optimal concentrations of Cs₂LiInCl₆. The origin of luminescence enhancement is highly desirable for study in detail. Besides, the larger optimal concentration for Cs₂LiInCl₆ indicates that the concentration quenching effect can be effectively suppressed, thus resulting in further increased emission, which deserves to be explored further for designing highly efficient DP-based emitters.

Fig. 3 (a) The UC emission intensity of Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs as a function of the total doping content. (b) The DC emission intensity of Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs as a function of the total doping content. (c) and (d) PL spectra of Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs at the corresponding optimal concentrations of Cs₂LiInCl₆.

To explore the difference in the luminescence enhancement between Cs₂NaInCl₆ and Cs₂LiInCl₆ DPs, we first used first-principles density functional theory (DFT) calculations to track the evolution of local site symmetry at different doping concentrations (Fig. 4). At a low doping concentration of Er³⁺, when Yb³⁺ and Er³⁺ ions are doped into the In lattice of Cs₂NaInCl₆, the local site symmetry will be reduced from O_h to D_2h, whereas the local site symmetry of Cs₂LiInCl₆ will descend from D_3d to C_3v. According to the branching rules,²⁴ Cs₂LiInCl₆ should have a lower symmetry than that of Cs₂NaInCl₆. On the basis of the Judd–Ofelt (J–O) theory,²⁵ the probability of spontaneous radiative electric dipole (ED) transitions can be expressed as:

where e is the electron charge, λ is the mean wavelength, n is the refractive index, and h is the Planck's constant. The electric-dipole line strength S_ed is given by eqn (2):where Ω_t (t = 2, 4, 6) are J–O intensity parameters, which depend on the local environment and symmetry around the Ln³⁺ ions. That means breaking the local site symmetry is a useful strategy to increase the ED transitions of rare-earth ions. Several previous studies have demonstrated that the lower symmetry usually contributes to the larger Ω_t,^26–28 and therefore leads to high probabilities of radiative ED transitions, 552 nm visible and 1540 nm NIR-II emissions in our case. Hence, considering the lower local site symmetry of Cs₂LiInCl₆, it is not surprising that the superior luminescence intensity can be obtained for Cs₂LiInCl₆. The increased distortion of Yb³⁺ and Er³⁺ sites in Cs₂LiInCl₆ should cause the observed luminescence enhancement in both low-doping and high-doping levels of Er³⁺, as seen in the blue and yellow regions in Fig. 3a and b. At a high doping level, namely close to the optimal concentration of Cs₂LiInCl₆, the local site symmetry of Cs₂LiInCl₆ maintained the same as C_3v, whereas the symmetry of Cs₂NaInCl₆ will increase from D_2h to D_4h. Consequently, the effect of the site symmetry still works, and much larger enhancement factors for NIR-II emission in Cs₂LiInCl₆ mean that local symmetry breaking can bring a greater impact on NIR emission than visible luminescence.

Fig. 4 (a and b) The calculated local site symmetry and sublattice distortions of Cs₂LiInCl₆:Yb³⁺,Er³⁺ at the low and high doping levels, respectively. (c and d) The calculated local site symmetry and sublattice distortions of Cs₂NaInCl₆: Yb³⁺,Er³⁺ at the low and high doping levels, respectively.

The PL decay curves of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺, and Cs₂NaInCl₆:18%Yb³⁺,5%Er³⁺ for both visible and NIR emissions were then measured (Fig. 5a and b). It is interesting that Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ has a shortened green emission lifetime (2.73 ms) compared to its Cs₂NaInCl₆ counterpart (4.23 ms). The PL decay of Cs₂NaInCl₆:18%Yb³⁺,5%Er³⁺ can be fitted by three components: a fast lifetime of τ₁: 0.67 ms, a middle lifetime of τ₂: 2.54 ms, and a slow lifetime of τ₃: 8.55 ms (Table S2†); while the PL decay of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ is fitted by a mono-exponential function with a slow lifetime of 2.73 ms (Fig. 5a). The measured lifetime (τ) in the PL decay curves is connected to the radiative (τ_r) and nonradiative lifetimes (τ_nr) from the relation 1/τ = 1/τ_r + 1/τ_nr. On one hand, distortions of the local symmetry would facilitate electronic transitions, and increase the radiative transition rate (k_r = 1/τ_r); on the other hand, the improved crystallization results in a reduction of the number of defects for less nonradiative relaxation, leading to a prolonged emission lifetime (τ_nr). Thus, if the main contribution of emission enhancement comes from the improved crystallization, the measured average τ should be longer for Cs₂LiInCl₆ than Cs₂NaInCl₆. Instead, Cs₂LiInCl₆ showed a shortened lifetime, indicating that the radiative transition rate k_r plays a predominant role in luminescence, directly suggesting that the main contribution of the emission enhancement comes from the breaking of local symmetry. The shortened slow lifetime of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ is also consistent with DFT results, due to lower local symmetry. Based on the fitting results, we also infer that the cross-relaxation process can be effectively suppressed in Cs₂LiInCl₆ compared to Cs₂NaInCl₆ with high-doping concentrations of Er³⁺. Furthermore, we investigated the decay lifetimes of the ⁴I_13/2 state (1540 nm) under 980 nm excitation. The ⁴I_13/2 lifetime of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ is also fitted by a mono-exponential function, which is much longer than that of the Cs₂NaInCl₆ counterpart (Fig. 5b and Table S3†), indicating that the nonradiative decay processes are largely reduced.

Fig. 5 (a and b) The PL decay curves of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ and Cs₂NaInCl₆:18%Yb³⁺,5%Er³⁺ for both visible and NIR emissions under 980 nm excitation. (c) High-resolution PL spectra of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ and Cs₂NaInCl₆:18%Yb³⁺,5%Er³⁺ for the ⁴I_9/2 → ⁴I_15/2 transition of Er³⁺. (d) The energy level splitting of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ from the ⁴I_9/2 crystal field.

Basically, the emission wavelength of Ln³⁺ shows almost no shift, because of the shield of the 4f energy levels by exterior shells. However, as shown in Fig. 5c, both Cs₂LiInCl₆ and Cs₂NaInCl₆ DPs have the same crystal-field strength, and the center wavelength of Cs₂LiInCl₆ slightly shifts to 820 nm compared with 813 nm of Cs₂NaInCl₆. The emission band at approximately 800 nm belongs to the ⁴I_9/2 → ⁴I_15/2 transition of Er³⁺,²⁹ and the ⁴I_9/2 level can split into 5 stark levels as long as the crystal field symmetry is lower than the cubic symmetry.³⁰ The large difference in the ⁴I_9/2 energy level between Cs₂LiInCl₆ DP and commercial NaYF₄ nanoparticles (808 nm) supports that the ⁴I_9/2 intermediate state could be a good indicator of the crystal-field environment. Fig. 5d exhibits the energy level splitting of Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺ from the ⁴I_9/2 crystal field. The crystal-field strength of Er³⁺ from the ⁴I_9/2 crystal field was calculated to be 400 cm⁻¹ for Cs₂LiInCl₆ and Cs₂NaInCl₆ (Fig. S5†), demonstrating that the ⁴I_9/2 crystal field is insensitive to the variation in local symmetry.³¹ On the basis of a similar coordination environment, the wavelength shift in the ⁴I_9/2 state should be attributed to the population redistribution among the crystal-field-split manifolds with thermal fluctuations. Meanwhile, Cs₂LiInCl₆ shows excellent transition selectivity under 1550 nm excitation. A record green-to-red ratio of up to 281 can be seen in Fig. S6† for Cs₂LiInCl₆:18%Yb³⁺,10%Er³⁺,2.5%Ho³⁺, suggesting that Cs₂LiInCl₆ is a good candidate for single-band UC applications.³² Fig. S6 and S7† prove that the high green-to-red ratio is closely related to the excitation wavelength because of different UC populating mechanisms, further confirming the weak electron–phonon coupling in Cs₂LiInCl₆.

To obtain a deep insight into the boosting mechanism, pump-power-dependent PL measurements were performed for visible and NIR emissions. As shown in Fig. 6a, the slopes are fitted to 2.49, 2.32, and 3.10 for ²H_11/2, ⁴S_3/2, and ⁴F_9/2 emissions, respectively, following the two-photon and three-photon conversion mechanisms. The general UC mechanism under 980 nm excitation is depicted in Fig. 6b. The three-photon process could increase the cross-relaxation (CR) rate at a high doping level of Er³⁺, leading to an intensified output of red emission. Considering that the ⁴I_9/2 state is closely related to other excited states of Er³⁺, it is desirable to investigate the relationship of this emission band with the cross-relaxation process.^33–35 A set of rate equations were established based on the UC process, as shown in Fig. 6b:

where N₀, N₁, N₂, and N₃ are the population densities of ⁴I_15/2, ⁴I_13/2, ⁴I_11/2, and ⁴I_9/2 states of Er³⁺, respectively. β₁, β₂, and β₃ are the non-radiative rates of ⁴I_13/2, ⁴I_11/2, and ⁴I_9/2 states of Er³⁺. A₁, A₂, and A₃ are the radiative rates of ⁴I_13/2, ⁴I_11/2, and ⁴I_9/2 levels of Er³⁺, respectively. T₂ denotes the possible thermal pumping rate from the ⁴I_11/2 to ⁴I_9/2 state. W₁, W₂, and W₃ are the energy transfer rates from the ²F_5/2 state of Yb³⁺ to the ⁴F_9/2, ⁴I_11/2, and ⁴F_5/2 levels of Er³⁺. C₁ represents the cross-relaxation process of ⁴I_13/2 + ⁴I_13/2 → ⁴I_15/2 + ⁴I_9/2 (CR3). The CR4 process can be omitted due to less contribution of ⁴I_9/2 → ⁴S_3/2. In eqn (5), the term T₂N₂ is commonly ignored at room temperature. In such case, the following expression is obtained:

Fig. 6 (a) The power density dependence of Er³⁺ emissions of ²H_11/2 (526 nm), ⁴S_3/2 (552 nm), and ⁴F_9/2 (669 nm) for Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺. (b) UC emission mechanism of Cs₂LiInCl₆:Yb³⁺,Er³⁺ under 980 nm excitation. (c) The power density dependence of the Er³⁺ emission of ⁴I_9/2 (820 nm) for Cs₂LiInCl₆:18%Yb³⁺,5%Er³⁺. The inset shows the power density dependence of 1540 nm emission under 980 nm excitation. (d) The NIR emission mechanism of Cs₂LiInCl₆: Yb³⁺,Er³⁺ under 980 nm excitation.

From eqn (6), we can observe that N₃∞N₁²∞P², indicating that the two NIR photons at 980 nm are needed to populate the ⁴I_9/2 excited state, which is in accordance with the pump-power dependent PL result in Fig. 6c. As a result of the intrinsic low phonon energy (320 cm⁻¹) and weak electron–phonon coupling for Cs₂LiInCl₆ DP, much weaker non-radiative transition from ⁴I_9/2 → ⁴I_13/2 can be neglected (β₃), the intensity of the ⁴I_9/2 state (I₈₂₀) can be derived from eqn (6):

I₈₂₀ = N₃A₃ = C₁N₁N₁ (7)

We then obtain I₈₂₀∞C₁. As the intensity of Cs₂NaInCl₆ is 3-fold higher than that of Cs₂LiInCl₆ (Fig. 5c), we can conclude that the CR rate of Cs₂LiInCl₆ is 3 times weaker than the Cs₂NaInCl₆ counterpart, demonstrating that the concentration quenching effect can be effectively suppressed for Cs₂LiInCl₆ at a high-doping level of Er³⁺. We infer that the distribution of Er³⁺ is statistical onto the two In³⁺ sites, and the distance of two adjacent Er³⁺ ions can be directly increased in Cs₂LiInCl₆, blocking the CR process at a high doping level of Er³⁺ (Fig. S8†).^36–40 Thus, the higher tolerance of Er³⁺ content in Cs₂LiInCl₆ is definitely beneficial to the UC emission, consistent with the results in Fig. 3a. The NIR-II emission originated from an energy transfer from Yb³⁺ to Er³⁺ because the emission intensity of ⁴I_13/2 shows a linear relationship with the pump power of 980 nm laser (the inset of Fig. 6c).¹ Together with the PL decay and PL results shown in Fig. 3b and 5b, respectively, we suppose that the Er³⁺–Er³⁺ energy transfer is the main nonradiative channel, and the presence of the two independent In³⁺ sites in Cs₂LiInCl₆ can decrease undesirable nonradiative energy transfer (CR1, CR2, and CR3 in Fig. 6d), accounting for the obvious NIR-II luminescence enhancement.

Conclusions

In summary, we provided a comprehensive understanding of the exact effect of local structure on both down-shifting and upconversion emissions in double perovskites by means of the successful synthesis of Cs₂LiInCl₆. Taking advantage of its unique local structure, including low local symmetry (trigonal), two In³⁺ active sites, and large crystal-field strength, Cs₂LiInCl₆:Yb³⁺ and Er³⁺ show 6-fold and 110-fold emission enhancement compared with Cs₂NaInCl₆ counterpart at 552 and 1540 nm, respectively_. Experimental results and theoretical calculations directly demonstrate that local symmetry breaking and the presence of multiple independent active sites are very important for luminescence enhancement. Our results provide a novel optical design through the engineering of local structures for desired luminescent properties of double perovskites in both visible and NIR-II regions.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The work was financially supported by the National Natural Science Foundation of China-Yunnan Joint Fund (U1902222), the National Natural Science Foundation of China (12064021 and 51862020), and the Foundation of Yunnan Province (202001AT070037, 202101AT070104 and 202101AT070097), and Yunnan Major Scientific and Technological Projects (grant no. 202202AG050016). The authors would like to thank Kang Jiao from Shiyanjia Lab (https://www.shiyanjia.com) for the SEM analysis.

References

1. 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, 541–545.
2. P. Saghy, A. M. Brown, C. Chu, L. C. Dube, W. Zheng, J. R. Robinson and O. Chen, Lanthanide double perovskite nanocrystals with emissions covering the UV-C to NIR spectral range, Adv. Opt. Mater., 2023, 11, 2300277.
3. Y. Liu, A. Nag, L. Manna and Z. Xia, Lead-free double perovskite Cs₂AgInCl₆, Angew. Chem., Int. Ed., 2021, 60, 11592–11603.
4. X. Xu, P. Han, D. Zheng, K. Du, C. Li, F. Liu, R. Zhang and K. Han, Boosting the upconversion and near-infrared emission via alloying Bi³⁺ in Cs₂NaErCl₆ double perovskite, Laser Photonics Rev., 2022, 16, 2200318.
5. S. Jin, R. Li, H. Huang, N. Jiang, J. Lin, S. Wang, Y. Zheng, X. Chen and D. Chen, Compact ultrabroadband light-emitting diodes based on lanthanide-doped lead-free double perovskites, Light: Sci. Appl., 2022, 11, 52.
6. Z. Rao, Z. Li, X. Zhao and X. Gong, Targeted high-precision up-converting thermometer platform over multiple temperature zones with Er³⁺, Mater. Horiz., 2023, 10, 1816–1824.
7. F. Igbari, Z.-K. Wang and L.-S. Liao, Progress of lead-free halide double perovskites, Adv. Energy Mater., 2019, 9, 1803150.
8. P. Han, X. Zhang, C. Luo, W. Zhou, S. Yang, J. Zhao, W. Deng and K. Han, Manganese-doped, lead-free double perovskite nanocrystals for bright orange-red emission, ACS Cent. Sci., 2020, 6, 566–572.
9. Y. Pei, D. Tu, C. Li, S. Han, Z. Xie, F. Wen, L. Wang and X. Chen, Boosting near-infrared luminescence of lanthanide in Cs₂AgBiCl₆ double perovskites via breakdown of the local site symmetry, Angew. Chem., Int. Ed., 2022, 61, e202205276.
10. S. Saikia, A. Joshi, H. Arfin, S. Badola, S. Saha and A. Nag, Sb³⁺–Er³⁺-codoped Cs₂NaInCl₆ for emitting blue and short-wave infrared radiation, Angew. Chem., Int. Ed., 2022, 61, e202201628.
11. Y. Shi, X. Zhang, X. Wang and Y. Zhang, Pure green upconversion from a multicolor downshifting perovskite crystal, Adv. Opt. Mater., 2023, 11, 2202704.
12. M. N. Tran, I. J. Cleveland, J. R. Geniesse and E. S. Aydil, High photoluminescence quantum yield near-infrared emission from a lead-free ytterbium-doped double perovskite, Mater. Horiz., 2022, 9, 2191–2197.
13. H. Arfin, J. Kaur, T. Sheikh, S. Chakraborty and A. Nag, Bi³⁺-Er³⁺ and Bi³⁺-Yb³⁺ codoped Cs₂AgInCl₆ double perovskite near-infrared emitters, Angew. Chem., 2020, 132, 11403–11407.
14. B. Zhou, L. Yan, J. Huang, X. Liu, L. Tao and Q. Zhang, NIR-II-responsive photon upconversion through energy migration in an ytterbium sublattice, Nat. Photonics, 2020, 14, 760–766.
15. X. Cheng, Y. Pan, Z. Yuan, X. Wang, W. Su, L. Yin, X. Xie and L. Huang, Er³⁺ sensitized photon upconversion nanocrystals, Adv. Funct. Mater., 2018, 28, 1800208.
16. X. Feng, L. Lin, R. Duan, J. Qiu and S. Zhou, Transition metal ion activated near-infrared luminescent materials, Prog. Mater. Sci., 2022, 129, 100973.
17. G. Chen, T. Y. Ohulchanskyy, A. Kachynski, H. Agren and P. N. Prasad, Intense visible and near-infrared upconversion photoluminescence in colloidal LiYF₄:Er³⁺ nanocrystals under excitation at 1490 nm, ACS Nano, 2011, 5, 4981–4986.
18. X. Lei, R. Li, D. Tu, X. Shang, Y. Liu, W. You, C. Sun, F. Zhang and X. Chen, intense near-infrared-II luminescence from NaCeF₄: Er/Yb nanoprobes for in vitro bioassay and in vivo bioimaging, Chem. Sci., 2018, 9, 4682–4688.
19. H. Dong, L.-D. Sun and C.-H. Yan, Local structure engineering in lanthanide-doped nanocrystals for tunable upconversion emissions, J. Am. Chem. Soc., 2021, 143, 20546–20561.
20. H. Fu, P. Peng, R. Li, C. Liu, Y. Liu, F. Jiang, M. Hong and X. Chen, A general strategy for tailoring upconversion luminescence in lanthanide-doped inorganic nanocrystals through local structure engineering, Nanoscale, 2018, 10, 9353–9359.
21. D. Avram, B. Cojocaru, I. Tiseanu, M. Florea and C. Tiseanu, Down-/up-conversion emission enhancement by Li addition: improved crystallization or local structure distortion, J. Phys. Chem. C, 2017, 121, 14274–14284.
22. Q. Ge, R. Zheng, J. Lin and H. Chen, Cs₂InCl₅(H₂O): A moisture-stable defective double halide perovskite analogue with broadband emission, Mater. Lett., 2020, 277, 128280.
23. L. Yan, B. Zhou, N. Song, X. Liu, J. Huang, T. Wang, L. Tao and Q. Zhang, Self-sensitization induced upconversion of Er³⁺ in core–shell nanoparticles, Nanoscale, 2018, 10, 17949–17957.
24. D. Tu, Y. Liu, H. Zhu, R. Li, L. Liu and X. Chen, Breakdown of crystallographic site symmetry in lanthanide-doped NaYF₄ crystals, Angew. Chem., Int. Ed., 2013, 52, 1128–1133.
25. B. R. Judd, Optical absorption intensities of rare-earth ions, Phys. Rev., 1962, 127, 750–761.
26. J. Hao, Y. Zhang and X. Wei, Electric-induced enhancement and modulation of upconversion photoluminescence in epitaxial BaTiO₃: Yb/Er thin films, Angew. Chem., Int. Ed., 2011, 50, 6876–6880.
27. Y. Wang, X. Wang, Y. Mao and J. A. Dorman, Impact of Sc³⁺-modified local site symmetries on Er³⁺ ion upconversion luminescence in Y₂O₃ nanoparticles, J. Phys. Chem. C, 2022, 126, 11715–11722.
28. S. Tanabe, T. Ohyagi, N. Soga and T. Hanada, Compositional dependence of Judd-Ofelt parameters of Er³⁺ ions in alkali-metal borate glasses, Phys. Rev. B: Condens. Matter Mater. Phys., 1992, 46, 3305–3310.
29. E. M. Chan, E. S. Levy and B. E. Cohen, Rationally designed energy transfer in upconverting nanoparticles, Adv. Mater., 2015, 27, 5753–5761.
30. P. A. Loiko, E. V. Vilejshikova, N. M. Khaidukov, M. N. Brekhovskikh, X. Mateos, M. Aguilo and K. V. Yumashev, Judd–Ofelt modeling, stimulated-emission cross-sections and non-radiative relaxation in Er³⁺: K₂YF₅ crystals, J. Lumin., 2016, 180, 103–110.
31. A. Li, D. Xu, H. Lin, L. Yao, S. Yang, Y. Shao, Y. Zhang and Z. Chen, A novel anion doing strategy to enhance upconversion luminescence in NaGd(MoO₄)₂:Yb³⁺/Er³⁺ nanophosphors, Phys. Chem. Chem. Phys., 2017, 19, 15693–15700.
32. L. Lei, J. Zhang and S. Xu, The progress of single-band upconversion nanomaterials, RSC Adv., 2016, 6, 81076–81084.
33. L. Li, F. Qin, Y. Zhou, J. Mao and Z. Zhang, Origin of the giant thermal enhancement of the Er³⁺ ion's ⁴I_9/2-⁴I_15/2 photoluminescence, Spectrochim. Acta, Part A, 2020, 229, 117862.
34. X. Yin, W. Xu, G. Zhu, Y. Ji, Q. Xiao, X. Dong, M. He, B. Cao, N. Zhou, X. Luo, L. Guo and B. Dong, Towards highly efficient NIR II response upconversion phosphor enabled by long lifetimes of Er³⁺, Nat. Commun., 2022, 13, 6549.
35. L. Peng, F. Qin, C. Wang, L. Li and Z. Zhang, A high-precision thermometry strategy by replacing the infrared with visible light for detection, Opt. Lett., 2023, 48, 4061–4064.
36. C. Ma, X. Xu, F. Wang, Z. Zhou, D. Liu, J. Zhao, M. Guan, C. I. Lang and D. Jin, Optical sensitizer concentration in single upconversion nanocrystals, Nano Lett., 2017, 17, 2858–2864.
37. A. Aebischer, M. Hostettler, J. Hauser, K. Kramer, T. Weber, H. U. Gudel and H.-B. Burgi, Structural and spectroscopic characterization of active sites in a family of light-emitting sodium lanthanide tetrafluorides, Angew. Chem., Int. Ed., 2006, 45, 2802–2806.
38. Y.-X. Yang, Z. Xu, S.-L. Zhao, Z.-Q. Liang, W. Zhu and J.-J. Zhang, Shape controllable synthesis and enhanced upconversion photoluminescence of β-NaYF₄:Yb³⁺ and Er³⁺ nanocrystals by introducing Mg²⁺, Chin. Phys. B, 2017, 26, 087801.
39. E. M. Chan, D. J. Gargas, P. J. Schuck and D. J. Milliron, Concentrating and recycling energy in lanthanide codopants for efficient and spectrally pure emission: the case of NaYF₄: Er³⁺/Tm³⁺ upconversion nanocrystals, J. Phys. Chem. B, 2012, 116, 10561–10570.
40. A. N. C. Neto, R. T. Moura and O. L. Malta, On the mechanisms of non-radiative energy transfer between lanthanide ions: centrosymmetric systems, J. Lumin., 2019, 210, 342–347.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3qi01952g

---