Automated Image Analysis for Single-Atom Detection in Catalytic Materials by Transmission Electron Microscopy

Improving Nanoparticle Size Estimation from Scanning Transmission Electron Micrographs with a Multislice Surrogate Model

The computational cost of simulating scanning transmission electron microscopy (STEM) images limits the curation of large enough data sets to train accurate and robust machine learning networks for deep feature extraction from atomically resolved STEM images. For nanoparticle size estimation in particular, a diverse data set is essential due to the large variations in size, shape, crystallinity, orientation, and dynamical diffraction effects in experimental data. To address this, we train a 3D convolutional neural network to predict STEM images from voxelized atomic models, achieving a 100x speed-up compared to traditional multislice simulations while maintaining high image quality. We then generate a data set of 100.000 synthetic multislice images and investigate the performance of different size-estimator architectures as a function of training set size. A ResNet18-based model trained on 4000 real and 100.000 synthetic images is found to perform the best, reducing the median size-estimation error from 9.89% without synthetic data to 5.26%.

TiN Boosting the Oxygen Reduction Performance of Fe-N-C through the Relay-Catalyzing Mechanism for Metal-Air Batteries

Metal-air batteries desire highly active, durable, and low-cost oxygen reduction catalysts to replace expensive platinum (Pt). The Fe-N-C catalyst is recognized as the most promising candidate for Pt; however, its durability is hindered by carbon corrosion, while activity is restricted due to limited oxygen for the reaction. Herein, TiN is creatively designed to be hybridized with Fe-N-C (TiN/Fe-N-C) to relieve carbon corrosion and absorb more oxygen when catalyzing oxygen reduction. The half-wave potential of TiN/Fe-N-C is 0.915 V vs reverse hydrogen electrode with 15 mV lost after 30,000 cycles accelerated durability test, higher than 0.893 V and 26 mV of Pt/C. The solid zinc-air battery of TiN/Fe-N-C achieves a peak power density of 238 mW/cm2, 2100 cycle stability at 30 °C, and long-term durability of 1100 h under -20 °C, superior to 150 mW/cm2 and 500 h (-20 °C) of Pt/C. Both calculations and experiments indicate that TiN has dual functions which not only relay abundant oxygen for the reaction but also strengthen the adsorption force for intermediates of carbon corrosion reaction, thus, enhancing the activity and durability of Fe-N-C. Therefore, the proposed relay catalytic strategy by TiN offers an efficient Fe-N-C catalyst for energy conversion devices.

Deep learning-assisted single-atom detection of copper ions by combining click chemistry and fast scan voltammetry

Cell ion channels, cell proliferation and metastasis, and many other life activities are inseparable from the regulation of trace or even single copper ion (Cu+ and/or Cu2+). In this work, an electrochemical sensor for sensitive quantitative detection of 0.4-4 amol L-1 copper ions is developed by adopting: (1) copper ions catalyzing the click-chemistry reaction to capture numerous signal units; (2) special adsorption assembly method of signal units to ensure signal generation efficiency; and (3) fast scan voltammetry at 400 V s-1 to enhance signal intensity. And then, the single-atom detection of copper ions is realized by constructing a multi-layer deep convolutional neural network model FSVNet to extract hidden features and signal information of fast scan voltammograms for 0.2 amol L-1 of copper ions. Here, we show a multiple signal amplification strategy based on functionalized nanomaterials and fast scan voltammetry, together with a deep learning method, which realizes the sensitive detection and even single-atom detection of copper ions.

Design Principles of Catalytic Materials for CO2 Hydrogenation to Methanol

Heterogeneous catalysts are essential for thermocatalytic CO2 hydrogenation to methanol, a key route for sustainable production of this vital platform chemical and energy carrier. The primary catalyst families studied include copper-based, indium oxide-based, and mixed zinc-zirconium oxides-based materials. Despite significant progress in their design, research is often compartmentalized, lacking a holistic overview needed to surpass current performance limits. This perspective introduces generalized design principles for catalytic materials in CO2-to-methanol conversion, illustrating how complex architectures with improved functionality can be assembled from simple components (e.g., active phases, supports, and promoters). After reviewing basic concepts in CO2-based methanol synthesis, engineering principles are explored, building in complexity from single to binary and ternary systems. As active nanostructures are complex and strongly depend on their reaction environment, recent progress in operando characterization techniques and machine learning approaches is examined. Finally, common design rules centered around symbiotic interfaces integrating acid-base and redox functions and their role in performance optimization are identified, pinpointing important future directions in catalyst design for CO2 hydrogenation to methanol.

Hidden Impurities Generate False Positives in Single Atom Catalyst Imaging

Single-atom catalysts (SACs) are an emerging class of materials, leveraging maximum atom utilization and distinctive structural and electronic properties to bridge heterogeneous and homogeneous catalysis. Direct imaging methods, such as aberration-corrected high-angle annular dark-field scanning transmission electron microscopy, are commonly applied to confirm the atomic dispersion of active sites. However, interpretations of data from these techniques can be challenging due to simultaneous contributions to intensity from impurities introduced during synthesis processes, as well as any variation in position relative to the focal plane of the electron beam. To address this matter, this paper presents a comprehensive study on two representative SACs containing isolated nickel or copper atoms. Spectroscopic techniques, including X-ray absorption spectroscopy, were employed to prove the high metal dispersion of the catalytic atoms. Employing scanning transmission electron microscopy imaging combined with single-atom-sensitive electron energy loss spectroscopy, we scrutinized thin specimens of the catalysts to provide an unambiguous chemical identification of the observed single-atom species and thereby distinguish impurities from active sites at the single-atom level. Overall, the study underscores the complexity of SACs characterization and establishes the importance of the use of spectroscopy in tandem with imaging at atomic resolution to fully and reliably characterize single-atom catalysts.

Rapid Identification of Drug Mechanisms with Deep Learning-Based Multichannel Surface-Enhanced Raman Spectroscopy

Rapid identification of drug mechanisms is vital to the development and effective use of chemotherapeutics. Herein, we develop a multichannel surface-enhanced Raman scattering (SERS) sensor array and apply deep learning approaches to realize the rapid identification of the mechanisms of various chemotherapeutic drugs. By implementing a series of self-assembled monolayers (SAMs) with varied molecular characteristics to promote heterogeneous physicochemical interactions at the interfaces, the sensor can generate diversified SERS signatures for directly high-dimensionality fingerprinting drug-induced molecular changes in cells. We further train the convolutional neural network model on the multidimensional SAM-modulated SERS data set and achieve a discriminatory accuracy toward 99%. We expect that such a platform will contribute to expanding the toolbox for drug screening and characterization and facilitate the drug development process.

Quantitative Description of Metal Center Organization and Interactions in Single-Atom Catalysts

Ultra-high-density single-atom catalysts (UHD-SACs) present unique opportunities for harnessing cooperative effects between neighboring metal centers. However, the lack of tools to establish correlations between the density, types, and arrangements of isolated metal atoms and the support surface properties hinders efforts to engineer advanced material architectures. Here, this work precisely describes the metal center organization in various mono- and multimetallic UHD-SACs based on nitrogen-doped carbon (NC) supports by coupling transmission electron microscopy with tailored machine-learning methods (released as a user-friendly web app) and density functional theory simulations. This approach quantifies the non-negligible presence of multimers with increasing atom density, characterizes the size and shape of these low-nuclearity clusters, and identifies surface atom density criteria to ensure isolation. Further, it provides previously inaccessible experimental insights into coordination site arrangements in the NC host, uncovering a repulsive interaction that influences the disordered distribution of metal centers in UHD-SACs. This observation holds in multimetallic systems, where chemically-specific analysis quantifies the degree of intermixing. These fundamental insights into the materials chemistry of single-atom catalysts are crucial for designing catalytic systems with superior reactivity.

Carbon Nitrides from Supramolecular Crystals: From Single Atoms to Heterojunctions and Advanced Photoelectrodes

Carbon nitride materials (CN) have become one of the most studied photocatalysts within the last 15 years. While CN absorbs visible light, its low porosity and fast electron-hole recombination hinder its photoelectric performance and have motivated the research in the modification of its physical and chemical properties (such as energy band structure, porosity, or chemical composition) by different means. In this Concept we review the utilization of supramolecular crystals as CN precursors to tailor its properties. We elaborate on the features needed in a supramolecular crystal to serve as CN precursor, we delve on the influence of metal-free crystals in the morphology and porosity of the resulting materials and then discuss the formation of single atoms and heterojunctions when employing a metal-organic crystal. We finally discuss the performance of CN photoanodes derived from crystals and highlight the current standing challenges in the field.

Atomic metal coordinated to nitrogen-doped carbon electrocatalysts for proton exchange membrane fuel cells: a perspective on progress, pitfalls and prospectives

Proton exchange membrane fuel cells require reduced construction costs to improve commercial viability, which can be fueled by elimination of platinum as the O2 reduction electrocatalyst. The past 10 years has seen significant developments in synthesis, characterisation, and electrocatalytic performance of the most promising alternative electrocatalyst; single metal atoms coordinated to nitrogen-doped carbon (M-N-C). In this Perspective we recap some of the important achievements of M-N-Cs in the last decade, as well as discussing current knowledge gaps and future research directions for the community. We provide a new outlook on M-N-C stability and atomistic understanding with a set of original density functional theory simulations.

Advances in heterogeneous single-cluster catalysis

Heterogeneous single-cluster catalysts (SCCs) comprising atomically precise and isolated metal clusters stabilized on appropriately chosen supports offer exciting prospects for enabling novel chemical reactions owing to their broad structural diversity with unparalled opportunities for engineering their properties. Although the pioneering work revealed intriguing performance trends of size-selected metal clusters deposited on supports, synthetic and analytical challenges hindered a thorough understanding of surface chemistry under realistic conditions. This Review underscores the importance of considering the cluster environment in SCCs, encompassing the development of robust metal-support interactions, precise control over the ligand sphere, the influence of reaction media and dynamic behaviour, to uncover new reactivities. Through examples, we illustrate the criticality of tailoring the entire catalytic ensemble in SCCs to achieve stable and selective performance with practically relevant metal coverages. This expansion in application scope transcends from model reactions to complex and technically relevant reactions. Furthermore, we provide a perspective on the opportunities and future directions for SCC design within this rapidly evolving field.

Single-Atom Heterogeneous Catalysts: Human- and AI-Driven Platform for Augmented Designs, Analytics and Reality-Enabled Manufacturing

Heterogeneous catalysts with targeted functionality can be designed with atomic precision, but it is challenging to retain the structure and performance upon the scaled-up manufacturing. Particularly challenging is to ensure the "atomic economy", where every catalytic site is most gainfully utilized. Given the emerging synergistic integration of human- and artificial intelligence (AI)-driven augmented designs (AD), augmented analytics (AA), and augmented reality manufacturing (AM) platforms, this minireview focuses on single-atom heterogeneous catalysts (SAHCs) and examines the current status, challenges, and future perspectives of translating atomic-level structural precision and data-driven discovery to next-generation industrial manufacturing. We critically examine the atomistic insights into structure-driven SAHCs functionality and discuss the opportunities and challenges on the way towards the synergistic human-AI collaborative data-driven platform capable of monitoring, analyzing, manufacturing, and retaining the atomic-scale structure and functions. Enhanced by the atomic-level AD, AA, and AM, evolving from the current high-throughput capabilities and digital materials manufacturing acceleration, this synergistic human-AI platform is promising to enable atom-efficient and atomically precise heterogeneous catalyst production.

Deep-Learning Aided Atomic-Scale Phase Segmentation toward Diagnosing Complex Oxide Cathodes for Lithium-Ion Batteries

Phase transformation─a universal phenomenon in materials─plays a key role in determining their properties. Resolving complex phase domains in materials is critical to fostering a new fundamental understanding that facilitates new material development. So far, although conventional classification strategies such as order-parameter methods have been developed to distinguish remarkably disparate phases, highly accurate and efficient phase segmentation for material systems composed of multiphases remains unavailable. Here, by coupling hard-attention-enhanced U-Net network and geometry simulation with atomic-resolution transmission electron microscopy, we successfully developed a deep-learning tool enabling automated atom-by-atom phase segmentation of intertwined phase domains in technologically important cathode materials for lithium-ion batteries. The new strategy outperforms traditional methods and quantitatively elucidates the correlation between the multiple phases formed during battery operation. Our work demonstrates how deep learning can be employed to foster an in-depth understanding of phase transformation-related key issues in complex materials.

Quantifying Atomically Dispersed Catalysts Using Deep Learning Assisted Microscopy

The catalytic performance of atomically dispersed catalysts (ADCs) is greatly influenced by their atomic configurations, such as atom-atom distances, clustering of atoms into dimers and trimers, and their distributions. Scanning transmission electron microscopy (STEM) is a powerful technique for imaging ADCs at the atomic scale; however, most STEM analyses of ADCs thus far have relied on human labeling, making it difficult to analyze large data sets. Here, we introduce a convolutional neural network (CNN)-based algorithm capable of quantifying the spatial arrangement of different adatom configurations. The algorithm was tested on different ADCs with varying support crystallinity and homogeneity. Results show that our algorithm can accurately identify atom positions and effectively analyze large data sets. This work provides a robust method to overcome a major bottleneck in STEM analysis for ADC catalyst research. We highlight the potential of this method to serve as an on-the-fly analysis tool for catalysts in future in situ microscopy experiments.

Bioinspired and Bioderived Aqueous Electrocatalysis

The development of efficient and sustainable electrochemical systems able to provide clean-energy fuels and chemicals is one of the main current challenges of materials science and engineering. Over the last decades, significant advances have been made in the development of robust electrocatalysts for different reactions, with fundamental insights from both computational and experimental work. Some of the most promising systems in the literature are based on expensive and scarce platinum-group metals; however, natural enzymes show the highest per-site catalytic activities, while their active sites are based exclusively on earth-abundant metals. Additionally, natural biomass provides a valuable feedstock for producing advanced carbonaceous materials with porous hierarchical structures. Utilizing resources and design inspiration from nature can help create more sustainable and cost-effective strategies for manufacturing cost-effective, sustainable, and robust electrochemical materials and devices. This review spans from materials to device engineering; we initially discuss the design of carbon-based materials with bioinspired features (such as enzyme active sites), the utilization of biomass resources to construct tailored carbon materials, and their activity in aqueous electrocatalysis for water splitting, oxygen reduction, and CO2 reduction. We then delve in the applicability of bioinspired features in electrochemical devices, such as the engineering of bioinspired mass transport and electrode interfaces. Finally, we address remaining challenges, such as the stability of bioinspired active sites or the activity of metal-free carbon materials, and discuss new potential research directions that can open the gates to the implementation of bioinspired sustainable materials in electrochemical devices.

What do we talk about, when we talk about single-crystal termination-dependent selectivity of Cu electrocatalysts for CO2 reduction? A data-driven retrospective

We mine from the literature experimental data on the CO₂ electrochemical reduction selectivity of Cu single crystal surfaces. We then probe the accuracy of a machine learning model trained to predict faradaic efficiencies for 11 CO₂ reduction reaction products, as a function of the applied voltage at which the reaction takes place, and the relative amounts of non equivalent surface sites, distinguished according to their nominal coordination. A satisfactory model accuracy is found only when discriminating data according to their provenance. On one hand, this result points at a qualitative agreement across reported experimental CO₂ reduction reactions trends for single-crystal surfaces with well-defined terminations. On the other, this finding hints at the presence of differences in nominally identical catalysts and/or CO₂ reduction reaction measurements, which result in quantitative disagreement between experiments.

Elucidation of Metal Local Environments in Single-Atom Catalysts Based on Carbon Nitrides

The ability to tailor the properties of metal centers in single-atom heterogeneous catalysts depends on the availability of advanced approaches for characterization of their structure. Except for specific host materials with well-defined metal adsorption sites, determining the local atomic environment remains a crucial challenge, often relying heavily on simulations. This article reports an advanced analysis of platinum atoms stabilized on poly(triazine imide), a nanocrystalline form of carbon nitride. The approach discriminates the distribution of surface coordination sites in the host, the evolution of metal coordination at different stages during the synthesis of the material, and the potential locations of metal atoms within the lattice. Consistent with density functional theory predictions, simultaneous high-resolution imaging in high-angle annular dark field and bright field modes experimentally confirms the preferred localization of platinum in-plane in the corners of the triangular cavities. X-ray absorption spectroscopy (XAS), X-ray photoelectron spectroscopy (XPS), and dynamic nuclear polarization enhanced ¹⁵ N nuclear magnetic resonance (DNP-NMR) spectroscopies coupled with density functional theory (DFT) simulations reveal that the predominant metal species comprise Pt(II) bound to three nitrogen atoms and one chlorine atom inside the coordination sites. The findings, which narrow the gap between experimental and theoretical elucidation, contribute to the improved structural understanding and provide a benchmark for exploring the speciation of single-atom catalysts based on carbon nitrides.