Förster Resonance Energy Transfer and Enhanced Emission in Cs₄PbBr₆ Nanocrystals Encapsulated in Silicon Nano-Sheets for Perovskite Light Emitting Diode Applications

Abstract

Encapsulating Cs₄PbBr₆ quantum dots in silicon nano-sheets not only stabilizes the halide perovskite, but also takes advantage of the nano-sheet for a compatible integration with the traditional silicon semiconductor. Here, we report the preparation of un-passivated Cs₄PbBr₆ ellipsoidal nanocrystals and pseudo-spherical quantum dots in silicon nano-sheets and their enhanced photoluminescence (PL). For a sample with low concentrations of quantum dots in silicon nano-sheets, the emission from Cs₄PbBr₆ pseudo-spherical quantum dots is quenched and is dominated with Pb²⁺ ion/silicene emission, which is very stable during the whole measurement period. For a high concentration of Cs₄PbBr₆ ellipsoidal nanocrystals in silicon nano-sheets, we have observed Förster resonance energy transfer with up to 87% efficiency through the oscillation of two PL peaks when UV excitation switches between on and off, using recorded video and PL lifetime measurements. In an area of a non-uniform sample containing both ellipsoidal nanocrystals and pseudo-spherical quantum dots, where Pb²⁺ ion/silicene emissions, broadband emissions from quantum dots, and bandgap edge emissions (515 nm) appear, the 515 nm peak intensity increases five times over 30 min of UV excitation, probably due to a photon recycling effect. This irradiated sample has been stable for one year of ambient storage. Cs₄PbBr₆ quantum dots encapsulated in silicon nano-sheets can lead to applications of halide perovskite light emitting diodes (PeLEDs) and integration with traditional semiconductor materials.

1. Introduction

Solution-processing halide perovskites such as APbX₃ (A = Cs, MA, FA, and PEA (FA, formamidinium; MA, methylammonium; PEA, phenylethylammonium); X = Cl, Br, and I) are very attractive due to their low-cost production and potential commercial application in pure color LED and photovoltaic devices [1,2,3,4]. A solar-cell efficiency of up to 26.1% has been reported since their discovery [5]. The thermal stability and photostability of halide perovskites pose challenges for their commercial application. Therefore, the surface passivation and encapsulation of halide perovskites in nanotubes or polymers is necessary in order to improve their stability [2,6,7,8,9,10,11,12,13,14,15]. Sargent and Grätzel’s groups have recently applied fluorinated aniliniums for an interfacial passivation for triple cation Cs_0.05MA_0.05FA_0.9Pb (I_0.95Br_0.05)₃ perovskite films [16]. The accelerating aging tests of halide perovskites under high temperatures and high humidity have reported a power-conversion efficiency of 24.09% and a 1560 h T85 at the maximum power point under 1 sun illumination, operating at 85 °C and 50% relative humidity [16]. For perovskite light emitting diode (PeLED) applications, the quantum efficiency of bicomponent perovskite nanocomposites can reach near 100% [17] based on the Förster resonance energy transfer (FRET) from the core CsPbBr₃ to the CsPbI₃ shell. FRET [18,19] has further been reported between CsPbBr₃ and CsPbCl₃ [20], between (PEA)₂PbI₄ and MAPbBr₃ [21], and between 2D perovskites [22].

Cs₄PbBr₆ is a zero-dimensional perovskite that can be easily prepared as quantum dots. Quantum dots of Cs₄PbBr₆ have been stabilized in glass, where the photoluminescence (PL) of perovskites shifts from 519 to 503 nm [23]. The superior thermal stability [24] and photostability [25] have been reported in CsPbBr₃-in-Cs₄PbBr₆ quantum dots. Highly emissive blue (470 nm) quantum dots have resulted in a high photoluminescence quantum yield of >90% in CsPbBr₃ + Cs₄PbBr₆ material systems [24]. Silicon nano-sheets are 2D layers of silicon with a tunable band gap. CsPbBr₃ isolated in silicene has been realized for PeLED applications [26]. On the other hand, Cs₄PbBr₆ can have a very high photoluminescence quantum yield (PLQY), which can be two orders of magnitude higher than its 3D counterpart [27]. Furthermore, it is also desired to stabilize Cs₄PbBr₆ in silicene and use the silicene for integration with traditional semiconductors. FRET can also be incorporated to further improve the PLQY for PeLED applications [17].

In this paper, nanomaterial systems of Cs₄PbBr₆ ellipsoidal nanocrystals and pseudo-spherical quantum dots in silicon nano-sheets have been prepared, and their PL emissions have been studied. For Cs₄PbBr₆ quantum dots in silicon nano-sheets, broadband emissions have been observed due to the surface trap states of Cs₄PbBr₆. They are replaced by stable Pb²⁺ ion emissions. For the Cs₄PbBr₆ nanocrystals in silicon nano-sheets, we have observed FRET of up to 87% efficiency. A three-month-old sample of Cs₄PbBr₆ nanocrystals in silicon nano-sheets has shown an enhanced emission five times higher during 30 min of UV laser irradiation.

2. Materials and Methods

Preparation of Cs₄PbBr₆ nanocrystals: 71.2 mg CsBr and 24.8 mg PbBr₂ was dissolved in 0.6 mL DMSO at 80 °C in a vial. The mixture was cooled down slowly while putting it in the methanol chamber. Green Cs₄PbBr₆ crystals were obtained at the bottom of the vial.

Preparation of Cs₄PbBr₆ nanocrystals and quantum dots in silicon nano-sheets: 45 mg of Cs₄PbBr₆ crystals was dissolved in 0.1 mL of DMSO and mixed with a given amount of silicon nano-sheets. In low concentrations of Cs₄PbBr₆ in silicon nano-sheets, the prepared Cs₄PbBr₆ solution was diluted by 10 times. The silicon nano-sheets were prepared following the method described in the reference [28]. In brief, the silicon nano-sheets were obtained by removing calcium ions from CaSi₂ in concentrated HCl. As examined by TEM, a different concentration of Cs₄PbBr₆ in silicon nano-sheets results in different crystal sizes and shapes. Cs₄PbBr₆ crystals in silicon nano-sheets smaller than 10 nm form a pseudo-spherical shape. We call them Cs₄PbBr₆ quantum dots hereafter. Cs₄PbBr₆ crystals in silicon nano-sheets larger than 10 nm form an ellipsoid shape with an elongation in the silicon nano-sheet plane. We call them Cs₄PbBr₆ nanocrystals hereafter.

The material characterizations and crystal structures of Cs₄PbBr₆ crystals, silicon nano-sheets, and Cs₄PbBr₆ nanocrystals in silicon nano-sheets were analyzed using Rigaku Miniflex 600 X-ray diffraction (XRD) with CuKα radiation (λ = 0.15405 nm).

Structural characterization was carried out with transmission electron microscopy (TEM) and energy-dispersive X-ray (EDX) in JEOL JEM-2100 at 200 kV. A carbon-coated 200-mesh copper grid was used for the preparation of TEM samples. The solution of Cs₄PbBr₆ nanocrystals in silicon nano-sheets was drop-casted on the copper grid and dried in a vacuum.

The PL of the Cs₄PbBr₆ crystals and Cs₄PbBr₆ nanocrystals in silicon nano-sheets was excited using a 375 nm UV laser (CrystaLaser, Reno, NV, USA). The PL of the Cs₄PbBr₆ quantum dots in silicon nano-sheets and pure silicon nano-sheets was excited using a He-Cd 325 nm laser (Kimmon, Tokyo, Japan). The PL spectra were analyzed using a BaySpec SuperGamut fiber-coupled UV-NIR spectrometer.

Fluorescence lifetime imaging (FLIM), PL decay lines, and lifetime histograms were obtained using a MicroTime 200 time-resolved confocal fluorescence microscope. The measurement was recorded using a PicoQuant PicoHarp 300 time-correlated single-photon-counter. Data acquisition was performed using SymPho Time 64 software. A Picoquant 405 nm picosecond laser was used with a repetition rate of 40 MHz. A cut-off filter of 435 nm was used before the detectors. Neutral density filters were used to reduce photon counts below 10⁶. A 20× objective lens (numerical aperture NA = 0.4) was used in the confocal microscope. For the FRET measurements, filters of 490 ± 5 nm (i.e., 485–495 nm) and 530 ± 20 nm (i.e., 510–550 nm) were placed in front of the detectors. The intensity FRET was calculated using software in SymPho Time 64.

3. Results

3.1. Material and Structural Characteristics

The prepared solutions of silicon nano-sheets and Cs₄PbBr₆ nanocrystals encapsulated in silicon nano-sheets were drop-casted onto a Si wafer and dried. Figure 1 shows the X-ray diffraction (XRD) of the silicon nano-sheet, the Cs₄PbBr₆ nanocrystals encapsulated in silicon nano-sheets, and the bulk Cs₄PbBr₆ crystals. The XRD peaks at 33.1° in Figure 1a,b belong to the Si wafer, as confirmed in the supporting information in reference [26]. Silicon nano-sheets have XRD peaks at 17.4°, 28.6°, 37.8°, 44.0°, 47.5°, 49.2°, and 56.5° (Figure 1a) [26]. These peaks appear in the XRD of Cs₄PbBr₆ nanocrystals (high concentration) encapsulated in silicon nano-sheets in Figure 1b, as labeled with yellow squares. The XRD peaks from Cs₄PbBr₆ nanocrystals in Figure 1b are located at 12.6°, 12.9°, 20.1°, 22.4°, 25.4°, 27.5°, 28.6°, and 30.3°, and are assigned to the low diffraction orders of the (012), (110), (113), (300), (024), (131), (214), and (223) planes, respectively. Figure 1c shows the XRD pattern of bulk Cs₄PbBr₆ for comparison. The XRD pattern in Figure 1c can be fitted to the hexagonal phase of Cs₄PbBr₆ with a = 13.73 Å, b = 13.73 Å, c = 13.73 Å, α = 90.0°, β = 90.0°, and γ = 120.0° (XRD database Card No.: 1538416). The XRD pattern of Cs₄PbBr₆ quantum dots encapsulated in silicon nano-sheets shows peaks from silicon nano-sheets only due to the low concentration. However, their presence is confirmed by TEM (vide infra).

Figure 2a shows a zoomed-in image (enlarged view) of a TEM image of Cs₄PbBr₆ ellipsoidal nanocrystals encapsulated in silicon nano-sheets. The dashed arrow in the figure indicates the location of one Cs₄PbBr₆ nanocrystal (dark region). All Cs₄PbBr₆ nanocrystals are elongated with the same orientation, as indicated by a dashed blue line, assuming they are in the same plane as a given silicon nano-sheet. From Figure 2a, we can observe that the nanocrystal size is not uniform; the elongated nanocrystals have a larger size in the middle than these at both sides of elongation. The stretching of the nanocrystals and different sizes along the long nanocrystals will be modeled for the explanation of the multiple peaks in the PL measurements.

Figure 2b shows a TEM image of one piece of sample for TEM-EDX study. The scale bars in2(b–f are the same. The dashed blue line indicates the nano-sheet orientation. One of the dark regions is indicated by an arrow. These nanocrystals in the dark regions are in the same orientation. Over 10 of them are revealed on the surface using TEM. Inside the bulk sample, TEM-EDX maps can reveal more information. The distribution of elements associated with the Cs₄PbBr₆ nanocrystals encapsulated in silicon nano-sheet perovskites (i.e., Si, Pb, Br, and Cs) is shown in Figure 2c–f, respectively. The elements of Si, Pb, Br, and Cs cover the whole piece of the sample in Figure 2b, as judged by the map shape. It is understandable that the map intensity of Si in Figure 2c is uniform due to the role of the host. The map intensity of Pb, Br, and Cs is not uniform. Especially in Figure 2e, it is very clear to see that the strong intensity regions (indicated by dashed white arrows, for example) correspond to the dark regions in Figure 2b.

The diffraction peaks from Cs₄PbBr₆ quantum dots did not show up in the XRD pattern due to their low concentration (diluted by 10 times). TEM allowed us to observe the quantum dots. Figure 3a shows a typical TEM image of Cs₄PbBr₆ pseudo-spherical quantum dots encapsulated in silicon nano-sheets. The size of the quantum dots is not uniform. Figure 3b shows the size distribution of the quantum dots in a range of 1 to 16 nm. The mean size is 4.67 nm. Figure 3c shows a high-resolution TEM (HRTEM) image of quantum dots and fast Fourier transform (FFT) patterns. A d-value of 0.31 nm was obtained, corresponding to the space between the (214) plane of hexagonal Cs₄PbBr₆. Figure 3d shows a TEM image of one piece of the sample and its distribution of elements associated with the Cs₄PbBr₆ quantum dots encapsulated in silicon nano-sheet perovskites in Figure 3e,f for Si and Br as an example, respectively. Si covers the whole sample in Figure 3e, while the map intensity of Br is not as dense as Si due to the low concentration of quantum dots.

3.2. PL Spectra and Förster Resonance Energy Transfer

Firstly, we measured the absorption of bulk Cs₄PbBr₆ Cs₄PbBr₆ nanocrystals and quantum dots in DMSO solution before they were mixed with silicon nano-sheets, where narrow peaks were observed at 295, 287, and 272 nm, respectively. We then measured the PL for bulk Cs₄PbBr₆, its ellipsoidal nanocrystals, and pseudo-spherical quantum dots encapsulated in silicon nano-sheets. Figure 4b shows normalized PLs, with a peak position at 520 nm for bulk Cs₄PbBr₆, two peaks at 512 and 490 nm for Cs₄PbBr₆ nanocrystals, and a broad peak centered at 480 nm for Cs₄PbBr₆ quantum dots. With a small crystal size, the PL peak is blueshifted, similar to what has been observed in CsPbBr₃ in silicene and bulk CsPbBr₃. The broad peak of Cs₄PbBr₆ quantum dots in Figure 4b can be due to the un-passivated surfaces and its surface state. Similar to other results [29], Cs₄PbBr₆ had a very small spectral overlap of absorption and PL emission with a Stokes shift of up to 1.82 eV. Defect states due to Br-poor vacancy or local deformation of [PbBr₆]⁴⁻ octahedra were used to explain the large Stokes shift [29,30,31,32]. Although the shift of the narrow peak with the crystal size can be observed in Figure 4a, such a large Stokes shift poses a complexity in explaining the FRET below in Cs₄PbBr₆ in silicon nano-sheets.

The PL in Figure 4b for Cs₄PbBr₆ nanocrystals in silicon nano-sheets was measured after several minutes of UV exposure. In the short period of UV 375 nm laser exposure, the peak at 512 nm is very weak initially, then increases its intensity from 0 to 28 s, and finally its intensity is higher than the peak at 490 nm, as shown in Figure 4c. We also recorded a Video S1, “PL peak intensity oscillation”. We can see the increase in peak intensity at 512 nm, turning off the UV laser at 36–42 s, the dropping of the peak intensity, and its increase again after the UV laser is turned on. Figure 4d shows the peak intensity as a function of UV exposure time, where we see an increase in intensity by more than two times for 512 nm emission in 20 s.

In order to understand both the intensity oscillation of the 512 nm emission peak and the exchange of peak intensity of both PL lines in Cs₄PbBr₆ nanocrystals encapsulated in silicon nano-sheets, we measured the lifetime of PL peaks and FRET in the next steps. Due to the intensity change and shift of the PL peaks with UV laser exposure time, we measured the PL after the UV laser had been turned on for several minutes (typical time for PL lifetime imaging). We measured the PL for six times of the irradiation hardening, starting from yellow to dark blue as shown in Figure 5a. The irradiation-hardened Cs₄PbBr₆ nanocrystals had PL peaks at 490 and 512 nm. We used bandpass filters of 490 ± 5 nm (i.e., 485–495 nm) and 530 ± 20 nm (i.e., 510–550 nm) and measured the FLIM. After fitting and calculation, Figure 5b,c show the average lifetime histogram for the PL peaks at 490 and 512 nm, respectively. The lifetime events were controlled below 10⁶ for measurement accuracy. As seen from Figure 5b,c, the average lifetime is centered at 6.8 ns for 490 nm and 5.0 ns for 512 nm. Their PL decay lines are shown in Figure 5d. As the nanocrystal size becomes smaller, the PL peak usually blueshifts [33]. The PL of 490 nm comes from a smaller nanocrystal than that of 512 nm. The more surface localized charges [34] (also the more surface trap states) in smaller nanocrystals, the higher the PL lifetime, which also makes the lifetime histogram curve broader in Figure 5b than that in Figure 5c.

Using bandpass filters of 490 ± 5 nm and 530 ± 20 nm, we measured the FLIM and calculated FRET events and efficiency between 490 nm and 512 nm as shown in Figure 5e. The FRET efficiency E [19] is defined in Equation (1):

$$E={\displaystyle \frac{1}{{\left(\frac{R}{R_0}\right)}^6+1}}$$

(1)

where R_o is the distance between the Cs₄PbBr₆ donor (490 nm emission) and the Cs₄PbBr₆ acceptor (512 nm emission) with a FRET efficiency of 50%, and R is the distance between the donor and the acceptor. Non-uniformity in the FRET efficiency is clearly observed. Based on the scale bar, most of areas are covered by a green color with approximately 66% efficiency with a distance of 0.89 R_o, as also calculated by Equation (1). For the two red regions, we processed with the “region of interest” and obtained a FRET efficiency histogram in Figure 5f. Both red regions have a FRET efficiency of 86–87%, corresponding to a distance of 0.738 R_o, as indicated by the red arrow in Figure 5g, where the occurred FRET events over the map region are plotted with respect to R. Figure 5h shows a schematic of the proposed model of Cs₄PbBr₆ nanocrystals encapsulated in silicon nano-sheets with horizontal elongation, following the information that was obtained using TEM in Figure 2a,b for Cs₄PbBr₆ ellipsoidal nanocrystals in the dark regions. The circles in Figure 5h are for visual indication only. We have a schematic of wide (blue lines) and narrow (green lines) bandgaps for the Cs₄PbBr₆ nanocrystals above. FRET can occur between these wide and narrow bandgaps.

3.3. Stability of Un-Passivated Cs₄PbBr₆ and Enhanced PL Emissions

We present the stability of un-passivated Cs₄PbBr₆ quantum dots encapsulated in silicon nano-sheets, their non-uniformity, their Pb²⁺ ion emission, and their enhancement of 512 nm emission. Figure 6a shows the PL spectra of Cs₄PbBr₆ quantum dots encapsulated in silicon nano-sheets after continuous UV laser exposure for 0, 1, 2, 3, 4, 5, 10, 25, and 90 min. The broad peak centered at 483 nm drops in intensity very quickly within 5 s. The intensity of Pb²⁺ ion emission [31,32] at 410 nm increases quickly with increasing exposure. Figure 6b shows the PL intensities as a function of UV laser exposure time for these two peaks. At 12 min, the intensity of the 483 nm peaks reaches almost zero, while the Pb²⁺ ion emission reaches the maximum at 12 min and is stable up to the end of measurement (90 min), indicating that un-passivated Cs₄PbBr₆ quantum dots are not stable, and isolated Pb²⁺ ions are formed and are stable. After a longer UV exposure time, Pb²⁺ ion emission becomes stronger. Figure 6c shows the PL of Pb²⁺ ion emission after the UV degradation of Cs₄PbBr₆ quantum dots for 5 h. The peak below 400 nm is the emission tail of the 375 nm laser after a band cutoff filter. To confirm this, we excited the PL of degraded Cs₄PbBr₆ quantum dots using a UV 325 nm laser. Figure 6d shows the PL of the degraded Cs₄PbBr₆ quantum dots with one peak only, at 410 nm. We also show the PL spectrum of pure silicene, excited using a UV 325 nm laser for comparison. The silicene (silicon nano-sheets) has no PL between 350 and 550 nm.

We further studied the stability of Pb²⁺ ion emission. We picked up degraded Cs₄PbBr₆ quantum dots and monitored the intensity stability of the 410 nm peak (Pb²⁺ ion emission [31,32]) as shown in Figure 7a on day 1. The red circles in the figure indicate that the laser power is very stable. The intensity of Pb²⁺ ion emission changed slightly at the very beginning and then became stable. We measured the intensity of Pb²⁺ ion emission on days 3, 4, 10, and 22 for 2 h. Figure 7b plots the intensity of Pb²⁺ ion emission on day 22 as a representative example, and Figure 7c shows the intensity ratio of Pb²⁺ ion emission over the laser tail power. Overall, the intrinsic Pb²⁺ ion emission is stable.

With the above understanding of the stability of Pb²⁺ ions and the instability of Cs₄PbBr₆ quantum dots, we further studied a region of the sample where Cs₄PbBr₆ nanocrystals have three PL peaks at 515 nm, 475 nm, and 410 nm, as shown in Figure 8a. The shift of the 490 nm peak in Figure 5a to 475 nm in Figure 8a indicates smaller nanocrystal sizes in Figure 8a than that in Figure 5a. From Figure 8a,b, we can learn that the intensity of the PL peak of 515 nm is initially weaker than that of 475 nm, it increases at the very beginning of UV 375 nm exposure, it becomes stronger than that of 475 nm, and then it decreases in intensity with increasing exposure time. For the 475 nm peak, its intensity drops quickly at the very beginning of UV exposure and drops at a slower rate together with the 515 nm peak in Figure 8b. The dropping of intensity of both PL peaks can be caused by the instability of Cs₄PbBr₆ nanocrystals related to the 475 nm peak.

We waited for 3 months for the disappearance of the 475 nm peak. As shown in Figure 8c, the 3-month-old sample has a weak peak at 460 nm. However, the Pb²⁺ ion emission becomes relatively strong due to the degradation of Cs₄PbBr₆ nanocrystals. In such an aged sample, the intensity of PL 515 nm increases five times upon exposing the sample to a UV 375 nm laser for 15 min, and reaches the maximum after 30 min, as shown in Figure 8c,d. The Pb²⁺ ion emission at 417 nm was stable, and the intensity of 460 nm varied slightly. The stable PL of Pb²⁺ ion emission can lead to the possibility of photon recycling (the process of photon re-absorption and internal re-emission from Pb²⁺ to 515 nm) for the enhancement of 515 nm. Future research can be performed on the passivation of Cs₄PbBr₆ nanocrystals, then further encapsulated in silicon nano-sheets for photoelectric devices.

After 12 months of ambient storage of Cs₄PbBr₆ nanocrystals in silicon nano-sheets, we were still able to observe the FRET, as shown in Figure 9. This time, the peak intensity at 487 nm was stable during the measurement period from 0 to 120 min. The peak intensity at 515 nm was stronger than the 487 nm peak. Its intensity increased from 6400 to 7600 in a short time, then stayed almost constant. By the comparison of two peak intensities in Figure 9a,b, we can see that FRET plays a crucial role in the PL enhancement of 515 nm for PeLED application.

4. Discussion

Cs₄PbBr₆ was reported to be stable under alpha-particle excitation [35]. We first discussed the UV irradiation effect on Cs₄PbBr₆ nanocrystals encapsulated in silicon nano-sheets. After 12 months, we rechecked XRD for the sample as shown in Figure 10. As we can see from the figure, XRD peaks from silicon nano-sheet were still observed, indicating silicon nano-sheets were stable for 12 months. We can still observe the XRD peaks from Cs₄PbBr₆ nanocrystals at 12.6°, 12.9°, 22.4°, 25.4°, 27.5°, 28.6°, and 30.3° due to the low diffraction orders of the (012), (110), (300), (024), (131), (214), and (223) planes, respectively. Diffraction from the (113) plane is missing and XRD peaks are weaker from the aged sample than the fresh one, corresponding to the fact that the broad peak around 460 nm from the Cs₄PbBr₆ became weak and Pb²⁺ emission appeared after UV irradiation in Figure 6 and Figure 8.

We further discussed why the PL peak from Cs₄PbBr₆ quantum dots is broader than those from nanocrystal or bulk material. Usually, the PL peak becomes narrower when the nanocrystal size becomes smaller due to the quantum discrete state [33]. We assume that there are many surface trap states in the interface of silicon nano-sheets and un-passivated quantum dots, leading to a broad PL peak. To confirm this assumption, we compared our results for Cs₄PbBr₆ with those for CsPbBr₃. We prepared CsPbBr₃ encapsulated in silicon nano-sheets with decreasing concentrations (the quantity of the concentration was ignored) and measured their PL as shown in Figure 11a. Similarly, a broad peak was observed at the low concentration of CsPbBr₃ in silicon nano-sheets. The intensity of the broad peak dropped quickly during the UV 375 nm laser irradiation, as shown in Figure 11b,c. The only difference between the Cs₄PbBr₆ and CsPbBr₃ samples is that there is no Pb²⁺ ion emission in the CsPbBr₃ sample.

5. Conclusions

Silicon nano-sheets have been used to encapsulate Cs₄PbBr₆ nanocrystals for their potential integration with traditional semiconductors. For low concentrations of un-passivated Cs₄PbBr₆ in silicon nano-sheets, pseudo-spherical quantum dots were formed with a broad PL peak, which was not stable upon irradiation with a UV 375 nm laser. The Pb²⁺ ion emission was stable upon the degradation of Cs₄PbBr₆ quantum dots. For high concentrations of un-passivated Cs₄PbBr₆ in silicon nano-sheets, ellipsoidal nanocrystals were observed. The intensity oscillation of two PL peaks was observed upon the UV laser being switched on and off after exposure, and explained by FRET with an efficiency of up to 86–87%. In a sample with three PL peaks at 515, 475, and 410 nm, the degraded (aged for 3 months) sample showed stable Pb²⁺ ion/silicene emission and enhanced intensity by five times for 515 nm, presumably due to photon recycling from Pb²⁺ ion/silicene emission to 515 nm.