Wan
Wu
,
Chunyu
Zhao
,
Mingyou
Hu
,
Aizhao
Pan
*,
Wei
Xiong
and
Yinghao
Chen
Department of Chemistry, School of Chemistry, Xi'an Jiaotong University, Xianning West Road, 28, Xi'an 710049, China. E-mail: panaizhao2017032@xjtu.edu.cn
First published on 16th February 2023
All-inorganic cesium lead halide (CsPbX3, X = Cl, Br and I) perovskite quantum dots (QDs) have received enormous research interest because of their exceptional optoelectronic properties, but their low chemical stability under ambient conditions from inevitable defects restricts their practical applications. In an effort to enhance the stability of QDs, in this study, novel functional nanocomposites were fabricated by encapsulating perovskite QDs with zeolite X doped with iron ions. Focusing on the as-obtained nanocomposites labeled with QDs@Fe/X-n, doping a reasonable amount of Fe3+ ions can tremendously improve the order of perovskite lattices and reduce the halide vacancies. The results of stability improvement in nanocomposites with an optimal Fe3+ load (QDs@Fe/X-3) are presented. After storage in air for 100 days, the emission-peak position of the composites can remain almost unchanged, and the photoluminescence (PL) intensity can reach ∼98% of the original intensity. Additionally, the PL intensity of QDs@Fe/X-3 can decrease immediately when exposing it to a NH3 atmosphere at room temperature. The PL intensity can be linearly varied with a change in the NH3 concentration. The original value of the PL can be rapidly recovered by separating the sample from the NH3 environment. This work enables the QDs@Fe/X composite to be an ideal active material for ammonia sensing.
Owing to the increasing release of volatile ammonia from the chemical, pharmaceutical and agricultural industries, detection of ammonia has been extensively researched.13,14 Various materials have been devoted to fabricating ammonia gas sensors, such as quantum dots, MOFs, and stimuli-responsive polymers, which have demonstrated excellent response.15–17 However, the poor gas sensitivity and selectivity and low yields limit their practical applications. Recently, fluorescence perovskite-based sensors have achieved a superior response to NH3 gas,18–20 indicating fluorescence perovskite to be a potent material for fabricating highly sensitive gas sensors for ammonia. Generally speaking, a neat perovskite phase can easily bear the defect originating from some elements being removed from the crystal surface during annealing, such as Pb and X vacancy defects, reducing its performance and stability (vulnerable to various aging stresses such as oxygen, moisture and ultraviolet (UV) irradiation).21–23 After surface encapsulation/modification or UV filtration, the lifetime of perovskite can be prolonged by a temporary separation.24,25 Porous zeolite is an ideal candidate for encapsulating and stabilizing perovskite QDs due to its regularly arranged pores and channels.26–28 Perovskite QDs embedded in zeolite can remain much more stable than the neat QDs under atmospheric moisture. Sun et al. reported a two-step synthesis of CsPbX3 QDs embedded in zeolite Y.29 The results showed that the stability of the QDs embedded in zeolite Y is improved compared to that of the neat QDs under ambient conditions. On this basis, Kim further found that zeolite X is a better host, because it has a higher Al content in its framework (more extra framework cations may stabilize the QD guests).30
Moreover, doping metal ions is an effective strategy for reducing defects. A small number of di- or tri-valent metal cations can favour the nucleation of perovskite grains, the reduction of grain boundary defects, a decrease in trap-state density, an increase in the charge-carrier lifetime and an improvement of perovskite's properties.31,32 For examples, Liu et al. revealed that Mn2+ can be easily inserted into the interstices of octahedral [PbI6]4− to restrain the formation of vacancy defects to favour perovskite crystallization. Eventually, the efficiency of PSCs by excessively doping MnI2 (1%) reaches 19.09%, which is superior to that of methylamine plumbum iodine (MAPbI3)-based devices (17.68%).31 Zhou et al. discovered that Eu3+, Y3+ and Fe3+ have a positive impact on the power conversion efficiency (PCE) and device stability of PSCs.32
Inspired by metal ion doping and surface coating for improving the stability, we would like to propose multiple protection strategies that show that encapsulating perovskite within a metal ion-doped zeolite can effectively suppress the photo-induced regrowth and deterioration for an improvement of long-term storage stability of CsPbBr3 perovskite QDs. In this work, we have introduced CsPbBr3 QDs into iron-doped zeolite X (labelled with QDs@Fe/X) to achieve uniform QD dispersion, and tunable photoluminescence (PL) properties are induced via controlling the Fe dosage. The defect characteristics of QDs@Fe/X samples with different Fe contents were studied via time-resolved photoluminescence (TRPL) spectra and decay lifetime tests. Interestingly, the QDs@Fe/X nanocomposites can rapidly respond to ammonia gas with reversibility at room temperature. This finding is very encouraging for developing perovskite sensors using NH3-responsive QDs@Fe/X matter in the future.
The Fe/X-n (n = 1, 2, 3 and 4) samples with different Fe dosages were synthesized by a molar ratio of n(Al2O3):
n(SiO2)
:
n(Na2O)
:
n(H2O)
:
n(Fe(NO3)3·9H2O) = 1
:
3.2
:
7.36
:
441.6
:
x (x = 3.27 × 10−3, 4.96 × 10−3, 2.20 × 10−2 and 3.97 × 10−2, respectively, and doping dosages of the metal were confirmed by adopting inductively coupled plasma (ICP) mass spectrometry). The mixture was vigorously stirred for 0.5 h for homogenization (800 rpm), followed by ageing for 8 h at room temperature. Finally, the solid was obtained by filtration and washed thoroughly several times with deionized water until the pH value reached 8–9, followed by drying at 105 °C and then milling.
The QDs in Fe/X-n solution were prepared by the procedures reported by Sun29et al. First, Cs+–Fe/X (Fe/X with Cs+via partial exchange) was prepared. Second, PbBr2 solution was prepared. Finally, ODE (5.0 mL) and Cs+–Fe/X (0.5 g) were blended, transferred into a three-neck flask (100 mL), and vacuum degassed for 30 min at 120 °C. Then, the temperature was increased to 150 °C under N2 protection, followed by injection of the PbBr2 solution. The mixture was stirred for 15 min and then cooled to room temperature. Ultimately, the product was washed (with n-hexane first followed by using isopropanol) and further centrifuged. The product was dried at 60 °C for 12 h. The resulting composites were labeled with QDs@Fe/X-n (n = 1, 2, 3 and 4 respectively). Various n values reflect different Fe dosages during preparation.
![]() | ||
Fig. 1 SEM (a) and high-resolution TEM (b) images of QDs@Fe/X-3 and SEM image (c) along with elemental mapping profiles (d–j) of Si, O, Al, Cs, Pb, Br and Fe of the surveyed area of QDs@Fe/X-3. |
The crystal structures of QDs@Fe/X-n and Fe/X samples were determined by the PXRD method as shown in Fig. 2a and b. In Fig. 2a, compared with Fe/X (in blue), QDs@Fe/X-3 (in red) exhibits both the primary Fe/X diffraction peaks and an additional new peak (12.3°), originating from partial Cs+ exchange into Fe/X (well matched with the reported Cs+–Fe/X),30,35,36 Moreover, the intensity of characteristic peak (6.1°) reduces significantly, which is clarified as follows. During Cs+ exchange, the framework of Al might be slightly deprived due to a low hydrothermal stability of zeolite X, leading to damage to the framework structure of zeolite X and decreased crystallinity.37 Besides, in Fig. S2,† a successful Cs+ exchange can be verified based on a slight movement of all characteristic diffraction peaks to lower angles. A Cs+ ion (a radius of 167 pm) has a larger covalent radius than a Na+ ion (a radius of 102 pm), resulting in an increase in lattice spacing. These peak variations further confirm the formation of QDs within zeolite.
![]() | ||
Fig. 2 PXRD patterns of Fe/X, QDs/X and QDs@Fe/X-n (a and b), XPS results of Fe/X and QDs@Fe/X-3 (c), and Br 3d XPS core-level spectra of QDs/X and QDs@Fe/X-3 (d). |
In Fig. 2b, XRD peaks from CsPbBr3 (JCPDS #00-018-0364) QDs cannot be observed in QDs@Fe/X-n. The volume of crystal domains of the embedded QDs should be a few nm3 at most, and thus the QDs diffraction signals are weaker than the micrometre-sized zeolite host. Furthermore, with an increase in the Fe dosage, the peaks from iron species cannot be detected in QDs/Fe@X-n, ascribed to a rather low iron content (highly scattered iron species).
Additionally, XPS technology (Fig. 2c) was utilized to verify the elemental composition of QDs@Fe/X-3 compared to that of Fe/X. Peaks of Fe/X from Al 2s, O 1s and Si 2p are observed ascribed to zeolite X.38 Additional peaks of QDs@Fe/X-3 from Cs 3d, Pb 4f and Br 3d can confirm the presence of CsPbBr3.9,39,40 The iron content is too low to be detected. Besides, XPS core-level spectra (Fig. 2d) are shown for verifying the chemical states of Br 3d affected by Fe3+.
Br− in CsPbBr3 QDs can be in two classes of chemical environments, namely, the higher band energy regions and the lower band energy regions (assigned to Pb–Br and Cs–Br complexes, respectively).41,42 Compared to QDs/X, QDs@Fe/X-3 can have number-stable Cs–Br complexes, number-decreased Pb–Br species and number-increased Fe–Br species. A transformation from Pb–Br species to Fe–Br species is caused by the fact that a small number of trivalent iron ions can be easily embedded in the interstices of the perovskite via chemical bonding with Br− ions to reduce bromine vacancy defects.43 In total, after incorporating Fe3+ ions into CsPbBr3 QDs, the number of Br− ions can remarkably increase.
TRPL (Fig. 3c) tests were further conducted to study the defect effect of QDs/X and QDs@Fe/X-n. These decay data were fitted using a two-exponential decay model,48 and the parameters are listed in Table S1.† There are two parts in PL decay curves (τ1 and A1 represent the decay time and percentage of intrinsic radiative recombination, respectively; τ2 and A2 represent the decay time and percentage of nonradiative recombination, respectively).49,50 The two radiative times, namely, τ1 and τ2, may stem from structural defects of vacancies and surface states, respectively. τavg refers to the weighted-average PL decay time. The QDs@Fe/X-3 sample can have the maximum percentage of radiative recombination and the minimum percentage of nonradiative recombination (an average lifetime of 38.39 ns), which is much longer than other hybrid halide perovskites. This indicates that the QDs@Fe/X-3 specimen has a higher ratio of exciton recombination and less transition at defect states. Herein, self-passivation plays an important role in surface nonradiative recombination. The results suggest that Fe3+ doping with zeolite can suppress the defect recombination to extend the carrier lifetime of perovskites. Nonetheless, excess Fe3+ ions may increase the trap states and act as additional radiative relaxation channels, leading to the decrease in radiative decay rates and charge-carrier lifetime.1,41,46,51,52 This agrees well with the conclusions drawn in Fig. 3a and b.
To study the environmental stability, the QDs@Fe/X-3 sample was exposed to ambient air for 100 days. In Fig. 4a, after 100 days of air exposure, the cyan brightness of QDs@Fe/X-3 is well maintained and the relative PLQY only drops by 2%, showing a high stability. By inference, this high stability may not be totally due to zeolite encapsulation. Thus, we measured the acidity of the samples (Fig. 4b). An obvious decrease in acid strength in zeolite X after the QD formation (2.941 mmol g−1 for Fe/X; 0.904 mmol g−1 for QDs@Fe/X-3) is verified, attributed to a partial deprivation of the framework of Al during the QD formation (resulting in the reduced acid densities). Furthermore, the Brunauer–Emmett–Teller (BET) results demonstrated the significant decrease in the surface area (891.0 to 166.1 m2 g−1) and pore volume (0.299 to 0.051 cm3 g−1) of Fe/X after QD growth, implying that the original zeolite framework structures were filled and collapsed after the formation of quantum dots (Table S2†). This is basically consistent with the XRD results.
Moreover, Fe/X itself is dominated by Lewis acid37 and has no strong acid sites. However, QDs@Fe/X-3 has strong acid sites after introducing QDs with a decrease in Lewis acid and an increase in Brønsted acid. This may be caused by the fact that CsPbBr3 QDs are confined in the interrupted nano-spaces of zeolite-X, where there are extensive dangling functional groups including Al–OH and Si–OH. Complex hydrogen bonds can be formed between the zeolite framework and perovskite halide anions.44 This causes the electrons of silicon hydroxyl and aluminium hydroxyl groups to move into the cage (increasing the electron cloud density in the cage, making the hydroxyl group behave like a strong Brønsted acid). These strong interactions not only enhance the cohesion of the perovskite with zeolite-X, but also passivate imperfections and defects of the perovskite.
All results together demonstrate that embedding CsPbBr3 QDs in Fe-doped zeolite X can obtain the ultra-stable perovskite/zeolite composite, thanks to encapsulation and in situ passivation.
To confirm the change in luminescence properties by observation with the naked-eye, PL spectra of the composites in the presence of NH3/air gases were achieved (Fig. 5c). After exposure to NH3 gas, the cyan-emission peak at 522.7 nm can disappear (black line) and a broad emission band from 450 to 513 nm can be seen (red line). When air was introduced, the characteristic cyan emission peak reappeared (a small blue shift; gray line), and the PL intensity of the sample basically remained unchanged. This indicates a superior stability and reversibility of this perovskite sample. Later, we investigated the transient response of QDs@Fe/X-3 via an exposure to different concentrations of NH3 gases (0–10 mL/10 mL), as shown in Fig. 5d. The PL intensity of QDs@Fe/X-3 quickly decreased with an increase in the NH3 concentration. Fig. 5e reveals a linear response of the sample PL intensity with respect to the NH3 concentration varying from 0 mL to 10 mL based on the following fitting equation: Y = −71929X + 1
231
473 (R2 = 0.954). The aforementioned results can suggest that the QDs@Fe/X-3 composite has a fast response, good reversibility and high stability under NH3 exposure, which enables the QDs@Fe/X-n materials to become promising candidates for the construction of state-of-the-art gas sensors.
Footnote |
† Electronic supplementary information (ESI) available: Materials and characterization details, TEM and SEM elemental mapping profiles, and PL decay parameters. See DOI: https://doi.org/10.1039/d2nr06923g |
This journal is © The Royal Society of Chemistry 2023 |