Martin Borstad Eriksen

arXiv2025

Artificial Intelligence-Assisted Workflow for Transmission Electron Microscopy: From Data Analysis Automation to Materials Knowledge Unveiling

(Scanning) transmission electron microscopy ((S)TEM) has significantly advanced materials science but faces challenges in correlating precise atomic structure information with the functional properties of devices due to its time-intensive nature. To address this, we introduce an analytical workflow for the holistic characterization, modelling, and simulation of device heterostructures. This workflow automates the experimental (S)TEM data analysis, providing an in-depth characterization of crystallographic information, 3D orientation, elemental composition, and strain distribution. It reduces a process that typically takes days for a trained human into an automatic routine solved in minutes. Utilizing a physics-guided artificial intelligence model, it generates representative descriptions of materials and samples. The workflow culminates in creating digital twins, 3D finite element and atomic models of millions of atoms, enabling simulations that provide crucial insights into device behaviour in practical applications. Demonstrated with SiGe planar heterostructures for scalable spin qubits, the workflow links digital twins to theoretical properties, revealing how atomic structure impacts materials and functional properties such as spatially-resolved phononic or electronic characteristics, or (inverse) spin orbit lengths. The versatility of our workflow is demonstrated through its application to a wide array of materials systems, device configurations, and sample morphologies.

cond-mat.mtrl-sci↗

arXiv2025

Enhancing atomic-resolution in electron microscopy: A frequency-domain deep learning denoiser

Atomic resolution electron microscopy, particularly high-angle annular dark-field scanning transmission electron microscopy, has become an essential tool for many scientific fields, when direct visualization of atomic arrangements and defects are needed, as they dictate the material's functional and mechanical behavior. However, achieving this precision is often hindered by noise, arising from electron microscopy acquisition limitations, particularly when imaging beam-sensitive materials or light atoms. In this work, we present a deep learning-based denoising approach that operates in the frequency domain using a convolutional neural network U-Net trained on simulated data. To generate the training dataset, we simulate FFT patterns for various materials, crystallographic orientations, and imaging conditions, introducing noise and drift artifacts to accurately mimic experimental scenarios. The model is trained to identify relevant frequency components, which are then used to enhance experimental images by applying element-wise multiplication in the frequency domain. The model enhances experimental images by identifying and amplifying relevant frequency components, significantly improving signal-to-noise ratio while preserving structural integrity. Applied to both Ge quantum wells and WS2 monolayers, the method facilitates more accurate strain quantitative analyses, critical for assessing functional device performance (e.g. quantum properties in SiGe quantum wells), and enables the clear identification of light atoms in beam sensitive materials. Our results demonstrate the potential of automated frequency-based deep learning denoising as a useful tool for atomic-resolution nano-materials analysis.

cond-mat.mtrl-sci↗