Search

In situ scanning transmission electron microscopy enables observation of the domain dynamics in ferroelectric materials as a function of externally applied bias and temperature. The resultant data sets contain a wealth of information on polarization switching and phase transition mechanisms. However, identification of these mechanisms from observational data sets has remained a problem due to a large variety of possible configurations, many of which are degenerate. Here, an approach based on a combination of deep learning‐based semantic segmentation, rotationally invariant variational autoencoder (VAE), and non‐negative matrix factorization to enable learning of a latent space representation of the data with multiple real‐space rotationally equivalent variants mapped to the same latent space descriptors is introduced. By varying the size of training sub‐images in the VAE, the degree of complexity in the structural descriptors is tuned from simple domain wall detection to the identification of switching pathways. This yields a powerful tool for the exploration of the dynamic data in mesoscopic electron, scanning probe, optical, and chemical imaging. Moreover, this work adds to the growing body of knowledge of incorporating physical constraints into the machine and deep‐learning methods to improve learned descriptors of physical phenomena. [ABSTRACT FROM AUTHOR]

Size, morphology, and surface sites of electrocatalysts have a major impact on their performance. Understanding how, when, and why these parameters change under operating conditions is of importance for designing stable, active, and selective catalysts. Herein, we study the reconstruction of a Cu‐based nanocatalysts during the startup phase of the electrochemical CO2 reduction reaction by combining results from electrochemical in situ transmission electron microscopy with operando X‐ray absorption spectroscopy. We reveal that dissolution followed by redeposition, rather than coalescence, is the mechanism responsible for the size increase and morphology change of the electrocatalyst. Furthermore, we point out the key role played by the formation of copper oxides in the process. Understanding of the underlying processes opens a pathway to rational design of Cu electro (re)deposited catalysts and to stability improvement for catalysts fabricated by other methods. [ABSTRACT FROM AUTHOR]

Electrochemical liquid-phase transmission electron microscopy (TEM) is showing excellent promise in fundamental studies of energy-related processes including lithium-ion battery (LIB) cycling. A key requirement to accurately interpret the measurements and acquire quantitative information is the implementation of a reliable reference electrode. Quasi-reference electrodes (QRE) remain commonly used due to microfabrication constraints of the electrochemical cell, however, they typically yield dramatic potential drifts making the electrochemical results inconclusive. Here, we present a method of producing a stable and readily interpretable lithium-gold alloy micro-reference electrode, which exhibits a reference potential of 0.1 V vs Li/Li+. We first examine the feasibility of electrochemically alloying a pristine gold electrode, patterned on a chip for in situ TEM, using a benchtop setup, and investigate various sources to support the lithiation. We confirm the presence of the Li-Au alloy using chronopotentiometry (CP) and open circuit voltage (OCV) measurements, and by scanning electron microscopy (SEM), electron energy loss spectroscopy (EELS) and high-resolution (HR) TEM. Finally, we apply this methodology in situ and use LiFePO4 as a model cathode material to demonstrate the merit of the Li-Au alloy reference electrode for obtaining reproducible cyclic voltammetry (CV) measurements on a liquid cell microelectrode system [ABSTRACT FROM AUTHOR]