Ultrafast Characterization of Indium Tin Oxide Grating

Indium Tin Oxide (ITO) has a metallic response in the near- mid- infrared but becomes dielectric in the visible [1]. When exposed to laser pulses, the distribution of conduction band electron induces changes in the real and imaginary parts of the permittivity [2]. When an electromagnetic wave couples with the electrons in the conduction band of metallic nanostructures, thanks to the nonlinear plasmonic response we can provide optical modulation that can be exploited and used to create integrated optical components [3].<br/>Here, we present a photophysical study on ITO grating, realized using the femtosecond micromachining technology, and comparing its properties with the one of an ITO thin film. The geometries, dimensions, and pitch of the various gratings analyzed are obtained using direct ablation in a controlled atmosphere of a homogeneous thin layer of ITO deposited on a glass substrate. This solution guarantees us precision, repeatability, and extreme manufacturing flexibility.<br/>Ultrafast pump-probe spectroscopy has proven to be a powerful technique to study out-of-equilibrium phenomena, being applicable over a broad range of photon energies from THz to x rays. In pump-probe, a medium is first excited with a short pump pulse and the photoinduced dynamics are probed by a time-delayed broad-band probe pulse.<br/>We characterize both the plasmon and inter-band temporal dynamics. To excite the plasmon of ITO we tuned our Optical Parametric Amplifier at the resonance in the NIR (1550nm) and to probe the modulation of the transmitted spectra from the grating we generate a broad-band probe in the visible range by focusing our fundamental beam into a thin sapphire plate.<br/>We observe a modulation of the visible spectra of the probe induced by the excitation of the material plasmon in the NIR. Exciting the plasmon induces a modulation of the permittivity of ITO, this change in permittivity changes the refractive index that changes the transmission efficiency of the grating.<br/>When using ultrashort laser pulses we can have the generation of several Coherent Artifacts (CAs) of considerable intensity that can completely or partially distort the first hundreds of femtoseconds of relaxation dynamics, causing loss of information about early electronic processes under investigation. In particular, in our measurement there is a strong Cross Phase Modulation artifact (XPM)[4], limiting our capabilities of extracting information of the first 200fs.<br/>Here, we introduce a novel AI-driven method to remove XPM artifacts from ultrafast pump-probe dynamics. We propose XPMnet, a Neural Network (NN) model able to operate directly on raw pump-probe data and efficiently retrieve the embedded electronic dynamics of physical interest in material systems. We designed and developed XPMnet as a supervised ML model. We structured the model architecture as a Convolutional Neural Network trained on a labeled dataset, which consisted of simulated pairs of XPM-affected pump-probe instances and related electronic time dynamics, the latter carrying the physically relevant information about the sample under investigation.<br/>[1] Mryasov, O. N., and A. J. Freeman. "Electronic band structure of indium tin oxide and criteria for transparent conducting behavior." <i>Physical Review B</i> 64.23 (2001): 233111.<br/>[2] Roxworthy, Brian J., et al. "Understanding and controlling plasmon-induced convection." <i>Nature communications</i> 5 (2014): 3173.<br/>[3] Guo, P., et al (2016). Ultrafast switching of tunable infrared plasmons in indium tin oxide nanorod arrays with large absolute amplitude. <i>Nature Photonics</i>, <i>10</i>(4), 267.<br/>[4] M. Lorenc, M. Ziolek, R. Naskrecki, J. Karolczak, J. Kubicki, and A. Maciejewski, “Artifacts in femtosecond transient absorption spectroscopy,” Appl. Phys. B 74, 19–27 (2002).<br/>[5] F. Emmert-Streib, Z. Yang, H. Feng, S. Tripathi, and M. Dehmer, “An introductory review of deep learning for prediction models with big data,” Front. Artif.Intell. 3, 4 (2020).