Spatially Resolved Band Gap and Dielectric Function in Two-Dimensional Materials from Electron Energy Loss Spectroscopy

Edge-induced excitations in Bi2Te3 from spatially-resolved electron energy-gain spectroscopy

Among the many potential applications of topological insulator materials, their broad potential for the development of novel tunable plasmonics at THz and mid-infrared frequencies for quantum computing, terahertz detectors, and spintronic devices is particularly attractive. The required understanding of the intricate relationship between nanoscale crystal structure and the properties of the resulting plasmonic resonances remains, however, elusive for these materials. Specifically, edge- and surface-induced plasmonic resonances, and other collective excitations, are often buried beneath the continuum of electronic transitions, making it difficult to isolate and interpret these signals using techniques such as electron energy-loss spectroscopy (EELS). Here we focus on the experimentally clean energy-gain EELS region to characterise collective excitations in the topologically insulating material Bi2Te3 and correlate them with the underlying crystalline structure with nanoscale resolution. We identify with high significance the presence of a distinct energy-gain peak around -0.8eV, with spatially-resolved maps revealing that its intensity is markedly enhanced at the edge regions of the specimen. Our findings illustrate the reach of energy-gain EELS analyses to accurately map collective excitations in quantum materials, a key asset in the quest towards new tunable plasmonic devices.

Alignment-invariant signal reality reconstruction in hyperspectral imaging using a deep convolutional neural network architecture

The energy resolution in hyperspectral imaging techniques has always been an important matter in data interpretation. In many cases, spectral information is distorted by elements such as instruments' broad optical transfer function, and electronic high frequency noises. In the past decades, advances in artificial intelligence methods have provided robust tools to better study sophisticated system artifacts in spectral data and take steps towards removing these artifacts from the experimentally obtained data. This study evaluates the capability of a recently developed deep convolutional neural network script, EELSpecNet, in restoring the reality of a spectral data. The particular strength of the deep neural networks is to remove multiple instrumental artifacts such as random energy jitters of the source, signal convolution by the optical transfer function and high frequency noise at once using a single training data set. Here, EELSpecNet performance in reducing noise, and restoring the original reality of the spectra is evaluated for near zero-loss electron energy loss spectroscopy signals in Scanning Transmission Electron Microscopy. EELSpecNet demonstrates to be more efficient and more robust than the currently widely used Bayesian statistical method, even in harsh conditions (e.g. high signal broadening, intense high frequency noise).