High-resolution transmission electron microscopy (HRTEM) is crucial for observing material’s structural and morphological evolution at Angstrom scales, but the electron beam can alter these processes. Devices such as CMOS-based direct-electron detectors operating in electron-counting mode can be utilized to substantially reduce the electron dosage. However, the resulting images often lead to a low signal-to-noise ratio, which requires frame integration that sacrifices temporal resolution. Several machine learning (ML) models have been recently developed to successfully denoise HRTEM images. Yet, these models are often computationally expensive, and their inference speeds on GPUs are outpaced by the imaging speed of advanced detectors, precluding in situ analysis. Furthermore, the performance of these denoising models on datasets with imaging conditions that deviate from the training datasets has not been evaluated. To mitigate these gaps, we propose a new self-supervised ML denoising pipeline specifically designed for time-series HRTEM images. This pipeline integrates a blind-spot convolution neural network with pre-processing and post-processing steps, including drift correction and low-pass filtering. Results demonstrate that our model outperforms various other ML and non-ML denoising methods in noise reduction and contrast enhancement, leading to improved visual clarity of atomic features. Additionally, the model is drastically faster than U-Net-based ML models and demonstrates excellent out-of-distribution generalization. The model’s computational inference speed is in the order of milliseconds per image, rendering it suitable for application in in-situ HRTEM experiments.

Vapor-phase infiltration (VPI) of inorganic materials in polymers is increasingly becoming popular for synthesizing various functional hybrid materials. While AlO x infiltration using trimethylaluminum (TMA) has been extensively studied, the mechanism of diethylzinc (DEZ)-based ZnO x infiltration, especially one that is initiated by AlO x priming, has not received much attention because highly reactive hydroxyl groups generated by AlO x -priming are expected to dominate the initial binding of DEZ, thus enabling the overall ZnO x VPI. Here, we interrogate the ZnO x infiltration mechanism in AlO x -primed poly(methyl methacrylate) (PMMA) in comparison to the control AlO x -only infiltration by utilizing a suite of complementary characterizations, including quartz crystal microbalance mass gain measurement, transmission electron microscopy, infrared reflection-absorption spectroscopy (IRRAS), and synchrotron X-ray absorption spectroscopy (XAS). The multivalent TMA precursor and associated hyperbranched AlO x network can quickly saturate the AlO x infiltration by clogging the polymer-free volume near the top. On the contrary, the ZnO x infiltration using divalent DEZ precursor, once activated via AlO x -priming, can lead to accelerated ZnO x infiltration. With the help of IRRAS, XAS, and density functional theory (DFT) simulations, we uncover that the AlO x -priming enhances the reactivity of neighboring carbonyl groups toward DEZ and opens up simultaneous reaction pathways, leading to accelerated high-fidelity infiltration of ZnO x .

We use in situ transmission electron microscopy with automated tracking to study supported gold nanoparticles (NPs) during high-temperature vacuum annealing. The average mass loss per NP is governed by a flat, nearly size-independent substrate-mediated evaporation profile. On top of this mean shrinkage, individual NPs show significant fluctuations in apparent growth or shrinkage, and NP volume follows a random-walk-like trajectory. To rationalize both the ensemble-mean behavior and the particle-resolved variability, we develop a self-consistent theory that couples substrate-mediated evaporation to collective 2D Ostwald-type mass exchange through a shared adatom field, described in terms of a renormalized screening length and background concentration. In the experimentally relevant regime, the theory predicts an approximately size-independent mean shrinkage rate and clarifies how net mass loss suppresses classical coarsening. Superimposed on this deterministic drift, we quantify stochastic volume trajectories and capture their fluctuation spectrum with a minimal Langevin description consistent with intermittent adatom attachment and detachment events. In addition, we characterize the lateral diffusive motion of NPs, which is responsible for their coalescence. Altogether, our results highlight that stochasticity is intrinsic at the nanoscale and that predicting the evolution of supported NPs at early and intermediate times requires a unified framework combining substrate-mediated evaporation, collective mass exchange, and stochastic fluctuations.

Autoencoder-transformer model applicable to both simulated and experimental XAS spectra is developed to predict the oxidation state of copper.

X-ray absorption spectroscopy (XAS) and electron energy-loss spectroscopy (EELS) produce detailed information about oxidation state, bonding, and coordination, making them essential for quantitative studies of redox and structure in functional materials. However, high-throughput quantitative analysis of these spectra, especially for mixed valence materials, remains challenging as diverse experimental conditions introduce noise, misalignment, broadening of the spectral features. We address this challenge by training a machine learning model consisting of an autoencoder to standardize the spectra and a transformer model to predict both Cu oxidation state and Bader charge directly from L-edge spectra. The model is trained on a large dataset of FEFF-simulated spectra and evaluates model performance on both simulated and experimental data. The results of the machine learning model exhibit highly accurate prediction across the domains of simulated and experimental XAS as well as experimental EELS. These advances enable future quantitative analysis of Cu redox processes under in situ and operando conditions.

Rare earth elements (REEs) pose challenges for chemical separation due to similarities in charge and ionic radii. Developing materials with high metal specificity, such as molecularly imprinted polymers, could enhance selectivity, but there are limited details available to support the rational design of selective REE binding motifs. In the current study, structural characterization of REE-DOTAM (1,4,7,10-tetrakis(carbamoylmethyl)-1,4,7,10-tetraazacyclododecane) complexes was combined with density functional theory (DFT) and machine learning (ML) to provide specific structural descriptors that can be used to develop selective REE sites. Nineteen REE-DOTAM complexes were synthesized from mixed solvent systems (H2O/DMF, H2O/DMSO, and H2O/DMA) and characterized using single-crystal X-ray diffraction. The resulting structural data were used to construct perfect templates (fixed ligand geometries without further relaxation) and relaxed templates (fully geometry-optimized structures). DFT-derived substitution energies revealed that selectivity between heavy and light REEs could be achieved, with larger energetic penalties observed for the perfect templates. Machine learning analysis attributed selectivity primarily to hydration energies in the perfect templates, with twist angles as the dominant structural control in the relaxed systems. Additional analysis of the ML output suggested specific conditions that could be targeted for rationally designed DOTAM-based materials for improved REE selectivity.

Resolving the local structure motifs that characterize phase evolution as a function of composition is a key challenge in structure characterization of complex materials. In this study, we combine first-principles simulations and x-ray absorption near-edge structures (XANES) analysis to gain insights into the structure evolution revealed by measurements across a combinatorial zinc titanate thin film, which was grown with smoothly varying composition over a wide range of the Ti:Zn ratio. Specifically, we propose a cluster blind-signal-separation (cBSS) method for XANES spectral analysis based on a library of the structures and spectra of representative local motifs. In addition to motifs from zinc titanate crystals, two types of Ti-defect models constructed in this study are key to the understanding of the structure characteristics in the Zn-rich region. The cBSS method makes use of both spectral clustering of the simulated site-XANES spectra library and the BSS procedure to construct high-fidelity spectral basis functions from an experimental spectral sequence. The method provides a rigorous measure of the spectral sensitivity and basis completeness. The results of the XANES analysis are corroborated with other experimental modalities, including x-ray diffraction and spectroscopic ellipsometry, to validate the cBSS method. The calculated motif weights resulting from fitting the XANES spectra with the cBSS basis probe the atomic structure characteristics of both crystalline and amorphous phases as a function of the Ti/Zn composition. The insights of the local structure motif evolution are pivotal to the understanding of the nonmonotonic trend in the optical gap, which may lead to potential applications through tuning the optical properties of zinc titanate. The workflow of the XANES spectral analysis developed in this work can be generalized to construct the structure-property relationship in a broad material space.

Autonomous synthesis platforms promise rapid exploration of vast parameter spaces; yet, integrating in situ structural characterization in closed-loop synthesis optimization remains challenging. We demonstrate a realization of such a closed-loop platform coupled with a droplet-flow microreactor, in situ X-ray scattering methods (SAXS/WAXS), and Gaussian process optimization to synthesize citrate-reduced Au nanoparticles with targeted characteristics. The system efficiently explored ∼19,000 synthesis recipes through 365 experiments, achieving precise control over size (4-60 nm) and polydispersity (σ < 0.11) across large citrate/gold ratios, exceeding traditional synthesis boundaries (1-10). Beyond confirming classical Turkevich-Frens trends, partial-dependence analysis revealed strong nonlinear coupling among precursor, citrate, and pH effects. Combining quantitative SAXS/WAXS analysis with electron microscopy characterization, we uncovered that crystallite size (dc) and particle size (d) follow dc = 0.18d + β, where synthesis chemistry controls the intercept β while maintaining a universal slope. This parallel-band structure enables independent tuning of crystallite domain size at fixed particle diameter through a combination of chloride, gold precursor, citrate, and pH contributions (cross-validated Spearman ρ = 0.7 ± 0.1). High-resolution electron microscopy shows multiple lattice-fringe orientations within single particles, directly confirming polycrystalline domains and the ability to tune dc at the fixed d. The platform's validation includes indistinguishable static versus flowing measurements, stable droplet transport at 100 °C, and <5% run-to-run variation, establishing a robust framework for mapping and controlling multiscale nanoparticle structure across expansive chemical spaces. The developed closed-loop platform can be applied to a borad range of nanosyntheis processes.

In recent years, rapid progress has been made in developing artificial intelligence (AI) and machine learning (ML) methods for x-ray absorption spectroscopy (XAS) analysis. Compared to traditional XAS analysis methods, AI/ML approaches offer dramatic improvements in efficiency and help eliminate human bias. To advance this field, we advocate an AI-driven XAS analysis pipeline that features several inter-connected key building blocks: benchmarks, workflows, databases, and AI/ML models. Specifically, we present two case studies for XAS ML. In the first study, we demonstrate the importance of reconciling the discrepancies between simulation and experiment using spectral domain mapping (SDM). Our ML model, which is trained solely on simulated spectra, predicts an incorrect oxidation state trend for Ti atoms in a combinatorial zinc titanate film. After transforming the experimental spectra into a simulation-like representation using SDM, the same model successfully recovers the correct oxidation state trend. In the second study, we explore the development of universal XAS ML models that are trained on the entire periodic table, which enables them to leverage common trends across elements. Looking ahead, we envision that an AI-driven pipeline can unlock the potential of real-time XAS analysis to accelerate scientific discovery.

X-ray absorption spectroscopy (XAS) is a powerful characterization technique for probing the local chemical environment of absorbing atoms. However, analyzing XAS data presents significant challenges, often requiring extensive, computationally intensive simulations, as well as significant domain expertise. These limitations hinder the development of fast, robust XAS analysis pipelines that are essential in high-throughput studies and for autonomous experimentation. We address these challenges with OmniXAS, a framework that contains a suite of transfer learning approaches for XAS prediction, each uniquely contributing to improved accuracy and efficiency, as demonstrated on the K-edge spectra database covering eight 3d transition metals (Ti-Cu). The OmniXAS framework is built upon three distinct strategies. First, we use M3GNet [Nat. Comput. Sci. 2, 718 (2022)] to derive latent representations of the local chemical environment of absorption sites as input for XAS prediction, achieving significant improvements over conventional featurization techniques. Second, we employ a hierarchical transfer learning strategy, training a universal multitask model across elements before fine-tuning for element-specific predictions. Models based on this cascaded approach after elementwise fine-tuning outperform element-specific models by up to 69%. Third, we implement cross-fidelity transfer learning, adapting a universal model to predict spectra generated by simulation of a different fidelity with a much higher computational cost. This approach improves prediction accuracy by up to 11% over models trained on the target fidelity alone. Our approach significantly boosts the throughput of XAS modeling by orders of magnitude as compared to first-principles simulations and is extendable to XAS prediction for a broader range of elements. The proposed transfer learning framework is generalizable to enhance deep-learning models that target other properties in materials research.

We demonstrate a data-driven approach to interpreting surface reactions by combining time-resolved gas pulsing infrared spectroscopy with chemical reaction neural networks (CRNNs). Using CO adsorption and desorption on Pd(111) at 460-490 K as a model system, we show how transient kinetic data can reveal detailed reaction mechanisms. Starting with a simple one-species model, we systematically evaluate increasingly complex mechanisms involving hollow and bridge site adsorption. Despite the similar goodness of fit to the same experimental absorbance data, our models predict distinct coverage dynamics for different adsorption sites. Through analysis of spectral peak stability and predicted dynamics, we identify a mechanism in which CO primarily adsorbs on bridge sites followed by rapid conversion to hollow sites as being the most physically consistent with experimental observations. This work provides a framework for extracting mechanistic insights from limited experimental data, demonstrating how machine learning can bridge the gap between transient kinetic measurements and a molecular-level understanding of surface reactions.

Accurate classification of molecular chemical motifs from experimental measurement is an important problem in molecular physics, chemistry, and biology. In this work, we present neural network ensemble classifiers for predicting the presence (or lack thereof) of 41 different chemical motifs on small molecules from simulated C, N, and O K-edge X-ray absorption near-edge structure (XANES) spectra. Our classifiers not only achieve class-balanced accuracies of more than 0.95 but also accurately quantify uncertainty. We also show that including multiple XANES modalities improves predictions notably on average, demonstrating a "multimodal advantage" over any single modality. In addition to structure refinement, our approach can be generalized to broad applications with molecular design pipelines.

X-ray absorption spectroscopy (XAS) is an element-specific materials characterization technique that is sensitive to structural and electronic properties. First-principles simulated XAS has been widely used as a powerful tool to interpret experimental spectra and draw physical insights. Recently, there has also been growing interest in building computational XAS databases to enable data analytics and machine learning applications. However, there are non-trivial differences among commonly used XAS simulation codes, both in underlying theoretical formalism and in technical implementation. Reliable and reproducible computational XAS databases require systematic benchmark studies. In this work, we benchmarked Ti K-edge XAS simulations of ten representative Ti-O binary compounds, which we refer to as the Ti-O-10 dataset, using three state-of-the-art codes: xspectra, ocean and exciting. We systematically studied the convergence behavior with respect to the input parameters and developed a workflow to automate and standardize the calculations to ensure converged spectra. Our benchmark comparison shows: (1) the two Bethe-Salpeter equation (BSE) codes (ocean and exciting) have excellent agreement in the energy range studied (up to 35 eV above the onset) with an average Spearman's rank correlation score of 0.998; (2) good agreement is obtained between the core-hole potential code (xspectra) and BSE codes (ocean and exciting) with an average Spearman's rank correlation score of 0.990. Our benchmark study provides important standards for first-principles XAS simulations with broad impact in data-driven XAS analysis.

Calcination is a solid-state synthesis process widely deployed in battery cathode manufacturing. However, its inherent complexity associated with elusive intermediates hinders the predictive synthesis of high-performance cathode materials. Here, correlative in situ X-ray absorption/scattering spectroscopy is used to investigate the calcination of nickel-based cathodes, focusing specifically on the archetypal LiNiO2 from Ni(OH)2. Combining in situ observation with data-driven analysis reveals concurrent lithiation and dehydration of Ni(OH)2 and consequently, the low-temperature crystallization of layered LiNiO2 alongside lithiated rocksalts. Following early nucleation, LiNiO2 undergoes sluggish crystallization and structural ordering while depleting rocksalts; ultimately, it turns into a structurally-ordered layered phase upon full lithiation but remains small in size. Subsequent high-temperature sintering induces rapid crystal growth, accompanied by undesired delithiation and structural degradation. These observations are further corroborated by mesoscale modeling, emphasizing that, even though calcination is thermally driven and favors transformation towards thermodynamically equilibrium phases, the actual phase propagation and crystallization can be kinetically tuned via lithiation, providing freedom for structural and morphological control during cathode calcination.

Development of solar energy converters with earth-abundant and environmentally friendly materials is one of the key routes explored towards a sustainable future. In this work, crystalline delafossite-phase CuAlO2 and CuFeO2 thin film solar water splitting photocathodes were fabricated using pulsed laser deposition. It was found that the desired delafossite phase was formed only after high temperature annealing in an oxygen-free atmosphere. The homogeneous delafossite bulk structure of the films was determined by correlating simulation results from first-principles calculations with synchrotron-based X-ray absorption near edge structure (XANES) spectroscopy. Both CuAlO2 and CuFeO2 photocathodes are active for solar water splitting, with the latter more efficient due to its narrower band gap and improved light absorption.

A new semi-supervised machine learning method for the discovery of structure-spectrum relationships is developed and demonstrated using the specific example of interpreting X-ray absorption near-edge structure (XANES) spectra. This method constructs a one-to-one mapping between individual structure descriptors and spectral trends. Specifically, an adversarial autoencoder is augmented with a novel rank constraint (RankAAE). The RankAAE methodology produces a continuous and interpretable latent space, where each dimension can track an individual structure descriptor. As a part of this process, the model provides a robust and quantitative measure of the structure-spectrum relationship by decoupling intertwined spectral contributions from multiple structural characteristics. This makes it ideal for spectral interpretation and the discovery of new descriptors. The capability of this procedure is showcased by considering five local structure descriptors and a database of over fifty thousand simulated XANES spectra across eight first-row transition metal oxide families. The resulting structure-spectrum relationships not only reproduce known trends in the literature, but also reveal unintuitive ones that are visually indiscernible in large data sets. The results suggest that the RankAAE methodology has great potential to assist researchers to interpret complex scientific data, test physical hypotheses, and reveal new patterns that extend scientific insight.