Transient Absorption Spectroscopy of Films: Impact of Refractive Index

Abstract

Transient absorption spectroscopy is a powerful technique to study the photoinduced phenomena in a wide range of states from solutions to solid film samples. It was designed and developed based on photoinduced absorption changes or that photoexcitation triggers a chain of reactions with intermediate states or reaction steps with presumably different absorption spectra. However, according to general electromagnetic theory, any change in the absorption properties of a medium is accompanied by a change in the refractive properties. Although this photoinduced change in refractive index has a negligible effect on solution measurements, it may significantly affect the measured response of thin films. In this Perspective paper, we examine why and how the measured responses of films differ from their expected "pure" absorption responses. The effect of photoinduced refractive index change can be concluded and studied by comparing the transmitted and reflected probe light responses. Another discussed aspect is the effect of light interference on thin films. Finally, new opportunities of monitoring the photocarrier migration in films and studying nontransparent samples using the reflected probe light response are discussed. Most of the examples provided in this article focus on studies involving perovskite, TiO₂, and graphene-based films, but the general discussion and conclusions can be applicable to a wide range of semiconductor and thin metallic films.

Figure 1

Model spectra of (a) real and imaginary parts of dielectric function ε, (b) n and k, and (c) transmittance (T) and reflectance (R) spectra of 300 nm (dashed line) and 500 nm (solid line) thick perovskite film. The dashed lines in (b) show experimental n and k spectra of perovskite film.

Figure 2

Model differential spectra of the perovskite excited state calculated for (a) △n and △k, (b) transmittance (△T) and reflectance (△R), and (c) corresponding absorbance change for standard transient transmittance △A_T and reflectance △A_R measurements. The green dashed line in plot (c), △A_T corr., represents the △A_T calculated using eq 4, see text for details.

Figure 3

2D presentation of the TA response of 30 nm annealed TiO₂ thin film deposited via ion beam sputtering excited at 320 nm measured in both (a) transmittance and (b) reflectance modes. (c) Calculated true transient absorbance of the film. The time scale is linear until 1 ps delay time and logarithmic after that, and (d) extracted △n and △k components at delay times (from darker to lighter) 1, 10, 100, and 1000 ps from the TT and TR spectra. Figure (a–c) reprinted with permission from ref (36). Copyright (2021) Royal Society of Chemistry, and Figure (d) reprinted with permission from ref (39). Copyright (2024) Ramsha Khan.

Figure 4

Measured (a) steady-state absorbance spectrum and (b) 2D presentation of the TA response obtained from 320 nm excitation of as-deposited 30 nm TiO₂ thin film deposited via atomic layer deposition. Reprinted with permission from ref (37). Copyright (2022) American Chemical Society.

Figure 5

(a) Diagram of graphene TAS data analysis used in ref (28) and obtained dielectric function spectra for (b) sub-picosecond (“first wave”) TAS response and (b) slower nanosecond (“second wave”) response. Reprinted with permission from ref (28). Copyright (2023) American Chemical Society.

Figure 6

Schematics of the Kramers–Kronig-based analysis technique reported by Ashoka et al. Reprinted with permission from ref (17). Copyright (2010) Springer Nature BV.

Figure 7

(a) Model change in time of the photocarrier distribution across 500 nm thick perovskite films. Sample absorbance at the excitation wavelength is A = 2, and the carrier diffusion coefficient is D = 2 cm² s^–1. Corresponding TA responses for (b) transmitted and (c) reflected signals.

Figure 8

Measured and modeled CsMAFA perovskite TR signal (a) at 480 and 550 nm probes after excitation at 400 nm and (b) at 1125 nm probe after excitation at different pump wavelengths. In the TFI model the △ñ(λ) per carrier is kept the same, only the carrier distribution changes due to diffusion and recombination. By comparison, the SCC tracks the measured signal only after 1 to 10 ps depending on the wavelength. The signal at 550 nm was normalized to match the 480 nm signal at late delay times. Reprinted with permission from ref (58). Copyright (2021) Royal Society of Chemistry.

Figure 9

(a) Charge carrier distribution in TiO₂ thin film after excitation at 320 nm (red) or after charge transfer from underlying silicon (blue) and (b) the corresponding measured and modeled TR signals. Reprinted with permission from ref (60). Copyright (2023) Elsevier BV.