Observation of electronic spectra modulation in a CH₃NH₃PbBr₃ crystal by utilizing transient absorption microscopy

CH₃NH₃PbBr₃ crystals were prepared by a previously reported method. ⁸⁾ Aqueous lead(II) acetate solution (100 mg ml⁻¹) was dropped onto a glass slide and dried at 60 °C for 30 min to obtain a film of lead acetate. Microcrystals of CH₃NH₃PbBr₃ were then obtained by immersing the film in methylammonium bromide solution (5 mg ml⁻¹) with isopropanol as a solvent for 20 h at room temperature in air.

Transient absorption microscopy set-up was used in previously reported apparatus. ^27–29) For transient absorption measurements, a regenerative amplified femtosecond pulse (Spectra-Physics, Solstice, 795 nm, 1 kHz) was split in two; the second harmonic of one beam was used as the excitation light, and the white light generated with the other beam by introducing it into an optical parametric amplifier (Light Conversion, TOPAS) and focusing the obtained signal light (1.3 μm) into a CaF₂ plate was used as the probe light.

The thin-film interferometry calculations were done using Eq. (1) which expresses general transmittance (T) light interferometry

$$\begin{eqnarray}&&T=\frac{16{n}_{0}{n}_{m}({n}^{2}+{k}^{2})}{{b}_{1}{e}^{2\gamma }+{b}_{2}{e}^{-2\gamma }+{b}_{3}\cos 2\delta +{b}_{4}\sin 2\delta },\end{eqnarray} \tag{ 1 }$$

where n, n ₀ , and n_m are the refractive index of the sample, air, and glass. b₁ , b₂ , b₃ , and b₄ are the amplitude parameters. γ and δ are the phase parameters.

The Franz–Keldysh calculation was done using the Eq. (2)

$$\begin{eqnarray}&&\frac{\unicode{x02206}R}{R}={\rm{\Delta }}F{\cdot }A(F),\end{eqnarray} \tag{ 2 }$$

where R and ΔR are reflectances and ΔF represents a transient electronic field and A(F) is the combination of Airy functions.

Figure 1(a) shows the emission image of lead halide perovskite crystal excited with a femtosecond laser at 400 nm. Excitation intensity was 4.2 μJ cm⁻². The pump light was sufficiently broader than the crystal to excite the whole crystal. Emission was observed over the crystal. Figure 1(b) shows the emission image of the same crystal excited at 11.4 μJ cm⁻². Strong emissions were observed at both ends of the crystal. These results suggested that amplified spontaneous emission (ASE) or lasing occurred in the crystal, and as a result, these strong emissions have been observed to leak from the crystal edges.

Zoom In Zoom Out Reset image size

To see the carrier dynamics in the crystal, the transient absorption microscopic technique was utilized. Figure 2(a) shows the transient absorption spectra in a lead halide perovskite excited at the intensities below the lasing threshold. The spectral signal was not observed before the time origin (−1 ps). The modulated spectral shape was significantly observed after the time origin (≥0 fs). The spectral bleaching around 540 nm indicates carrier generation by the excitation light, and the relaxation of the bleaching indicates the recombination of electrons and holes.

Zoom In Zoom Out Reset image size

Figure 2(b) shows the transient absorption spectra in a lead halide perovskite excited at the intensities over the lasing threshold. A negative signal appeared around 557 nm, which was not present in the transient absorption spectrum excited at the intensity below the lasing threshold.

The differences in the spectra appearing with the excitation light intensity suggest the generation of stimulated emission signals. In fact, a strong luminescence signal is also observed in the emission image observations as shown in Fig. 1(b) under the condition over the threshold, supporting the idea that the appearing new band in Fig. 2(b) is due to ASE or lasing.

To reveal the different dynamics of ASE or lasing, Fig. 3 shows time profiles of transient absorbance monitored at 557 nm. When the excitation intensity was 11.4 μJ cm⁻², in contrast to the result at the excitation intensity of 4.2 μJ cm⁻², a notable decay with a time constant of 2 ps was observed. These results suggest the ASE or lasing process progressing in 2 ps. The time constant of ASE or lasing is consistent with previous reports. ¹¹⁾

Zoom In Zoom Out Reset image size

Spectral modulation is often observed as interference of light propagating in thin-film systems. The thickness of the perovskite crystals is around 200–300 nm, which is suitable to observe the optical interference. However, generally, this interference tends to become lower in frequency as one observes longer wavelengths, which is essentially different from the modulation observed here.

To discuss the optical interference effect, the actual optical interference pattern was simulated by using Eq. (1).

Figure 4 shows the calculation results by the thin-film interference model. The least mean-square fitting result was quite different from the experimental result. We tried several model calculations, including induced Raman signals and splitting signals with exciton-polaritons, but none showed good agreement with our results. If the modulation occurs during the time when light and matter are interacting with stimulated emission, the signal should disappear at 2 ps as shown in Fig. 3.

Zoom In Zoom Out Reset image size

The spectra modulation appeared immediately after the carrier is generated, and the modulation continues for up to 1 ns. These results suggest that the modulation is caused by the electric field effect of the generated carriers on the electron spectrum. Franz–Keldysh theory is often used for spectral modulation in the presence of an electric field. In reflectance transient absorption spectroscopy, the electric field modulated spectrum has been explained in the GaInP₂ system by the Franz–Keldysh theory in systems where charges are generated on the semiconductor surfaces. ³⁰⁾ The Franz–Keldysh theory could express this spectral modulation in high-frequency regions as shown in Fig. 5. These results indicate the spectral modulation mainly occurred from the charges trapping on the surface of the crystal. However, because of the complicated spectra shift and modulation, it is difficult to explain quantitatively how much electric field is generated. These results suggest that our transient absorption microscope measures the internal carrier dynamics differently from the surface by averaging the surface and internal charges, respectively. Transient absorption imaging measurements will clarify these differences in carrier dynamics between the edge and internal carrier, and we will report on unraveling ASE and lasing from the viewpoint of light–matter interactions in near the future. This technique will be also available for detecting charge accumulation to suppress lasing threshold for small laser media and provide design guidelines for more efficient and miniaturized lasing materials.

Zoom In Zoom Out Reset image size