Publications
(Identifying 2.2nm Au nanoparticles in a high-resolution TEM image)
For the most updated list of publications, please see my Google Scholar page here.
If you have trouble accessing any of these papers, please send me an email.
2024
Generalization Across Experimental Parameters in Neural Network Analysis of High-Resolution Transmission Electron Microscopy
K. Sytwu, L.R. DaCosta, M.C. Scott. Microscopy and Microanalysis (2024)
Neural networks are promising tools for high-throughput TEM analysis of nanomaterials, but are known to generalize poorly on data that is "out-of-distribution" from their training data. In this paper, we provide curated, experimentally collected high-resolution TEM image datasets of nanoparticles under various imaging and material parameters. We then investigate how the choice of metadata features in the training dataset affects neural network performance on the example task of nanoparticle segmentation.
2023
Kinetics and Mechanism of Light-Induced Phase Separation in a Mixed-Halide Perovskite
S. Peng, Y. Wang, M. Braun, Y. Yin, A. Meng, W. Tan, B. Saini, K. Severson, A. Marshall, K. Sytwu, J. Baniecki, J. Dionne, W. Cai, P.C. McIntyre. Matter (2023)
Halide ion phase separation is a barrier to the application of halide perovskites in optoelectronics, where the presence of large populations of photogenerated or -injected carriers cause undesirable changes in the local band gap. This paper investigates the mechanisms of phase separation in inorganic mixed-halide perovskite single crystals driven by light, visualizing the process using cryogenic scanning transmission electron microscopy and cathodoluminescence.
2022
Understanding the Influence of Receptive Field and Network Complexity in Neural Network-Guided TEM Image Analysis
K. Sytwu, C. Groschner, M.C. Scott. Microscopy and Microanalysis (2022)
Trained neural networks have consistently outperformed traditional image analysis methods at a variety of nanoparticle analysis tasks, but it is unclear how to best customize these networks for the unique features in TEM images. By systematically varying neural network architecture across various controlled nanoparticle datasets, we identify the importance of receptive field for low nanoparticle contrast datasets.
2021
Driving energetically-unfavorable dehydrogenation dynamics with plasmonics
K. Sytwu, M. Vadai, F. Hayee, D.K. Angell, A. Dai, J. Dixon, J. Dionne. Science (2021)
Nanoparticle structure and geometry generally dictate where chemical transformations occur. Here, we demonstrate how we can use optical excitation of plasmons to enable spatially-modified phase transformations at the nanoscale, activating otherwise energetically-unfavorable sites. Using optically-coupled in situ environmental TEM, we track the dehydrogenation of individual Au-PdHx antenna-reactor pairs under various illumination and hydrogen pressure conditions.
2019
Bimetallic Nanostructures: Combining Plasmonic and Catalytic Metals for Photocatalysis
K. Sytwu, M. Vadai, J. Dionne. Advances in Physics: X (2019)
By combining plasmonically active metals with traditional catalytic metals, bimetallic nanostructures are promising photocatalysts that promote simultaneous light conversion and strong molecular adsorption. In this review, we focus on how intermetallic interactions contribute to a nanoparticle's plasmonic and catalytic response in three bimetallic geometries: antenna-reactor, core-shell, and alloyed nanoparticle systems. We review both state-of-the-art bimetallic photocatalysts as well as emerging research opportunities.
2018
In-situ observation of plasmon-controlled photocatalytic dehydrogenation of individual palladium nanoparticles
M. Vadai, D.K. Angell, F. Hayee, K. Sytwu, J.A. Dionne. Nature Communications (2018)
Plasmonic nanoparticle catalysts offer improved light absorption and carrier transport compared to traditional photocatalysts, but it remains unclear how plasmonic excitation affects multi-step reaction kinetics and promotes site-selectivity. Here, we visualize the plasmon-induced dehydrogenation of individual Pd nanocubes coupled to Au nanoparticles at the sub-nanoparticle level using an environmental TEM combined with light excitation.
Visualizing facet-dependent hydrogenation dynamics in individual palladium nanoparticles
K. Sytwu, F. Hayee, T.C. Narayan, A. Koh, R. Sinclair, J.A. Dionne. Nano Letters (2018)
Surface faceting can affect the rate and selectivity of chemical transformations, but the precise role of surface termination can be challenging to elucidate from ensemble measurements of nanoparticles. Here, we compare the kinetics and mechanism of the palladium hydride phase transition in {100}-terminated cubes and {111}-terminated octahedra using a combination of diffraction, electron energy loss spectroscopy, and dark-field contrast in an environmental TEM. Regardless of surface faceting, both particle morphologies nucleate the new phase from particle corners, and support a phase-front that propagates linearly with time.
2013
Weakly explosive percolation in directed networks
S. Squires, K. Sytwu, D. Alcala, T.M. Antonsen, E. Ott, M. Girvan. Physical Review E (2013)
Percolation, or the formation of a macroscopic connected component, is a key feature in the description of complex networks. Recent studies have shown that if network edges are added "competitively" in undirected networks, the onset of percolation is abrupt or "explosive". Here we generalize this network growth process from undirected networks to directed networks and find that this growth is not as sudden as in undirected networks.