Poster Session 2 Details
Date: May 28th, 2025
Time: 2:30 PM – 3:00 PM
Location: Earth and Space Sciences, Room 145
Autonomous Discovery of Dealloying Transitions in Thin-Film Solid-State Metal Dealloying
Presenter: Cheng-Chu Chung
Abstract: The thin-film solid-state metal dealloying (SSMD) process is emerging as a novel method for fabricating nanoarchitectured materials. Unlike liquid metal dealloying, SSMD leverages solid-state processing to create finer feature sizes in bi-continuous nanostructures with lower temperature treatments and shorter processing times. This efficiency opens new opportunities for applications involving thin-film processes. However, the exploration of the materials library to design new dealloyed nanostructures remains inefficient, often relying on experimental serendipity, which limits the ability to select appropriate engineering parameters linked to fundamental physical and chemical properties. In this study, we present a comprehensive method to fabricate machine-learning (ML)-predicted potential systems, specifically Nb-Al/Sc and Nb-Al/Cu, using a thermal gradient treatment ranging from 100 to 800 °C via laser heating. With an in situ setup for exploring time durations, the high-dimensional thin-film samples were rapidly characterized using a suite of multimodal synchrotron X-ray techniques, including Grazing Incidence Wide-/Small-Angle X-ray Scattering (GIWAXS/GISAXS) and X-ray Absorption Spectroscopy (XAS). This characterization was integrated with an autonomous ML approach for decision-making in the experimental search process. The results revealed critical phase and morphology transitions across a broad thermal and temporal space, indicating a potential dealloying process for nanostructure formation. These findings provide valuable insights into the design of new dealloyed nanostructures, elucidating key processing conditions and enhancing our understanding of the dealloying mechanism. This includes insights into phase transitions, chemical bonding statuses, and morphological changes, thereby paving the way for more efficient and targeted development of advanced nanoarchitectured materials for future applications.
Elucidating a dissolution-deposition reaction mechanism by multimodal synchrotron X-ray characterization in aqueous Zn/MnO2 batteries
Presenter: Varun Kankanallu
Abstract: Aqueous Zn/MnO2 batteries with their environmental sustainability and competitive cost, are becoming a promising, safe alternative for grid-scale electrochemical energy storage. Presented as a promising design principle to deliver higher theoretical capacity, this presentation will discuss the fundamental understanding of the dissolution-deposition mechanism of Zn/β-MnO2. A multimodal synchrotron characterization approach including three operando X-ray techniques (powder diffraction, absorption spectroscopy, and fluorescence microscopy) is coupled with elementally resolved synchrotron X-ray nano-tomography. Together they provide a direct correlation between structural evolution, reaction chemistry, and 3D morphological changes. Operando synchrotron X-ray diffraction and spectroscopy show a crystalline-to-amorphous phase transition. Quantitative modeling of the operando data by Rietveld refinement for X-ray diffraction and multivariate curve resolution (MCR) for X-ray absorption spectroscopy are used in a complimentary fashion to track the structural and chemical transitions of both the long-range (crystalline phases) and short-range (including amorphous phases) ordering upon cycling. Scanning X-ray microscopy and full-field nano-tomography visualizes the morphology of electrodes at different electrochemical states with elemental sensitivity to spatially resolve the formation of the Zn- and Mn-containing phases. Overall, these results critically indicate that for Zn/MnO2 aqueous batteries, the reaction pathways involving Zn-Mn complex formation upon cycling become independent of the polymorphs of the initial electrode and shed light on the interplay among structural, chemical, and morphological evolution for electrochemically driven phase transitions. 1 This poster will highlight these key findings and ongoing progress, paving the way for advancing Zn/MnO₂ aqueous batteries as a promising grid-scale energy storage system.
Decoding the morphological information of metal nano-catalysts from X-ray absorption spectroscopy using deep learning and evolutional approach
Presenter: Kaifeng Zheng
Abstract: The morphology of metal nanocatalysts significantly influences their working performance in catalytic reaction. However, as an ensemble-averaged property across the entire sample, this information is challenging to capture with traditional experimental techniques. For example, TEM and SEM provide only 2D images of a small portion of the sample, while diffraction-related techniques, such as XRD and PDF, are limited in their applicability to small particles. In contrast, X-ray absorption spectroscopy (XAS) is sensitive to the local structure of materials in any form. Specifically, X-ray absorption near-edge structure (XANES) is highly effective in the low energy region and is sensitive to the shape of nanoparticles. In this study, we trained a neural network to analyze the shapes of Pt nanoparticles based on features derived from XANES. We then applied this trained model to the experimental spectra. Two key shape descriptors were discussed: oblateness, which measures the degree of flattening in nanoparticles, and surface-volume-ratio (SVR), both of which are essential parameters for characterizing nanoparticle morphology. Additionally, we developed a nanoparticle reconstruction process using a Genetic Algorithm (GA). This process allows us to infer unknown descriptors of reconstructed nanoparticles from known descriptors, and it can also be used to generate a particle database with specific characteristics, such as nanoparticles composed of 300 atoms.
Have Questions?
Contact us at nicholas.marcella@stonybrook.edu