(Digital Presentation) Tailored Rare Earth-Doped Nanomaterials Toward Information Storage and Deep Learning Decoding

Abstract

Lanthanide-doped nanoparticles have been considered as one of the most promising luminescent materials due to their excellent properties such as high photochemical stability, long-lived (μs-ms) luminescence, narrow emission band, and low toxicity.Moreover, benefiting from a unique electronic structure (4fn5s25p6 , n = 0-14), lanthanides have discrete energy levels and exhibit practical wavelength conversion via downshifting and upconversion processes. Hence, their emissions cover the spectral regions from ultraviolet (UV) to near-infrared (NIR).[1,2] Here, my talk is mainly devoted to our recent developments, including (1) recently, we present a new composition of Er3+-based upconversion nanoparticles with color-switchable output under irradiation with 980, 808, or 1535 nm light for information security. The variation of excitation wavelengths changes the intensity ratio of visible (Vis)/near-infrared 1535 nm (NIR-II) emissions. Taking advantage of the Vis/NIR-II multi-modal emissions of upconversion nanoparticles and deep learning, we successfully demonstrated the storage and decoding of visible light information in pork tissue.[3] (2) we construct heterostructured nanocomposites based on upconversion nanoparticles and EuSe semiconductors by using cation exchange method. It is generally considered that epitaxial growth is difficult when the lattice mismatch is large between two materials. In this case, the cation exchange of Eu3+ ions and other rare-earth ions could promote the formation of buffer layers to reduce the lattice mismatch and promote the heterogeneous epitaxial growth of EuSe on the upconversion nanoparticles. The heterostructured nanocomposites can emit tunable multicolor fluorescence under excitation of UV, continuous NIR, and pulsed NIR light. Based on the advantage of multiple tunable luminescence, the nanocomposites are designed as optical modules to load optical information. This work enables multi-dimensional storage of information and provides new insights into the design and fabrication of next-generation storage materials.