Deep learning-enhanced characterization of bubble dynamics in proton exchange membrane water electrolyzers.

The ever-increasing utility of imaging technology in proton exchange membrane water electrolyzer research raises the demand for rapid and precise image analysis. In particular, for optical video recordings, the challenge primarily lies in the large number of frames that impede the delineation of bubble dynamics with standard methods. In order to address this problem, the present study supports the automation of data analysis to facilitate swift, comprehensive, and measurable insights from captured imagery. We present a deep learning-based framework to perform high-throughput analyses of bubble dynamics using optical images of proton exchange membrane water electrolyzers. Leveraging a relatively small annotated imaging dataset of just 35 images, various configurations of the U-Net architecture were trained to perform bubble segmentation tasks. The best model achieved a precision of 95%, a recall of 78%, and an F1-score of 86% on the validation set. Subsequent to segmentation, the methodology enabled the rapid extraction of parameters such as time-resolved bubble area, size distributions, bubble position probability density, and individual bubble shape analytics. The findings underscore the potential of deep learning to enhance the analysis of polymer electrolyte membrane water electrolyzer imaging, offering a path toward more efficient and informative evaluations in electrochemical research.

UTILE-Gen: Automated Image Analysis in Nanoscience Using Synthetic Dataset Generator and Deep Learning.

This work presents the development and implementation of a deep learning-based workflow for autonomous image analysis in nanoscience. A versatile, agnostic, and configurable tool was developed to generate instance-segmented imaging datasets of nanoparticles. The synthetic generator tool employs domain randomization to expand the image/mask pairs dataset for training supervised deep learning models. The approach eliminates tedious manual annotation and allows training of high-performance models for microscopy image analysis based on convolutional neural networks. We demonstrate how the expanded training set can significantly improve the performance of the classification and instance segmentation models for a variety of nanoparticle shapes, ranging from spherical-, cubic-, to rod-shaped nanoparticles. Finally, the trained models were deployed in a cloud-based analytics platform for the autonomous particle analysis of microscopy images.

Stay Hydrated! Impact of Solvation Phenomena on the CO2 Reduction Reaction at Pb(100) and Ag(100) surfaces.

Herein, a comprehensive computational study of the impact of solvation on the reduction reaction of CO2 to formic acid (HCOOH) and carbon monoxide on Pb(100) and Ag(100) surfaces is presented. Results further the understanding of how solvation phenomena influence the adsorption energies of reaction intermediates. We applied an explicit solvation scheme harnessing a combined density functional theory (DFT)/microkinetic modeling approach for the CO2 reduction reaction. This approach reveals high selectivities for CO formation at Ag and HCOOH formation on Pb, resolving the prior disparity between ab initio calculations and experimental observations. Furthermore, the detailed analysis of adsorption energies of relevant reaction intermediates shows that the total number of hydrogen bonds formed by HCOO plays a primary role for the adsorption strength of intermediates and the electrocatalytic activity. Results emphasize the importance of explicit solvation for adsorption and electrochemical reaction phenomena on metal surfaces.

Low-spin state of Fe in Fe-doped NiOOH electrocatalysts.

Doping with Fe boosts the electrocatalytic performance of NiOOH for the oxygen evolution reaction (OER). To understand this effect, we have employed state-of-the-art electronic structure calculations and thermodynamic modeling. Our study reveals that at low concentrations Fe exists in a low-spin state. Only this spin state explains the large solubility limit of Fe and similarity of Fe-O and Ni-O bond lengths measured in the Fe-doped NiOOH phase. The low-spin state renders the surface Fe sites highly active for the OER. The low-to-high spin transition at the Fe concentration of ~ 25% is consistent with the experimentally determined solubility limit of Fe in NiOOH. The thermodynamic overpotentials computed for doped and pure materials, η = 0.42 V and 0.77 V, agree well with the measured values. Our results indicate a key role of the low-spin state of Fe for the OER activity of Fe-doped NiOOH electrocatalysts.

Deep learning for the automation of particle analysis in catalyst layers for polymer electrolyte fuel cells.

The rapidly growing use of imaging infrastructure in the energy materials domain drives significant data accumulation in terms of their amount and complexity. The applications of routine techniques for image processing in materials research are often ad hoc, indiscriminate, and empirical, which renders the crucial task of obtaining reliable metrics for quantifications obscure. Moreover, these techniques are expensive, slow, and often involve several preprocessing steps. This paper presents a novel deep learning-based approach for the high-throughput analysis of the particle size distributions from transmission electron microscopy (TEM) images of carbon-supported catalysts for polymer electrolyte fuel cells. A dataset of 40 high-resolution TEM images at different magnification levels, from 10 to 100 nm scales, was annotated manually. This dataset was used to train the U-Net model, with the StarDist formulation for the loss function, for the nanoparticle segmentation task. StarDist reached a precision of 86%, recall of 85%, and an F1-score of 85% by training on datasets as small as thirty images. The segmentation maps outperform models reported in the literature for a similar problem, and the results on particle size analyses agree well with manual particle size measurements, albeit at a significantly lower cost.

Properties of the Pt(111)/electrolyte electrochemical interface studied with a hybrid DFT-solvation approach.

Self-consistent modeling of the interface between solid metal electrode and liquid electrolyte is a crucial challenge in computational electrochemistry. In this contribution, we adopt the effective screening medium reference interaction site method (ESM-RISM) to study the charged interface between a Pt(111) surface that is partially covered with chemisorbed oxygen and an aqueous acidic electrolyte. This method proves to be well suited to describe the chemisorption and charging state of the interface at controlled electrode potential. We present an in-depth assessment of the ESM-RISM parameterization and of the importance of computing near-surface water molecules explicitly at the quantum mechanical level. We found that ESM-RISM is able to reproduce some key interface properties, including the peculiar, non-monotonic charging relation of the Pt(111)/electrolyte interface. The comparison with independent theoretical models and explicit simulations of the interface reveals strengths and limitations of ESM-RISM for modeling electrochemical interfaces.

Convolutional neural networks for high throughput screening of catalyst layer inks for polymer electrolyte fuel cells.

The performance of polymer electrolyte fuel cells decisively depends on the structure and processes in membrane electrode assemblies and their components, particularly the catalyst layers. The structural building blocks of catalyst layers are formed during the processing and application of catalyst inks. Accelerating the structural characterization at the ink stage is thus crucial to expedite further advances in catalyst layer design and fabrication. In this context, deep learning algorithms based on deep convolutional neural networks (ConvNets) can automate the processing of the complex and multi-scale structural features of ink imaging data. This article presents the first application of ConvNets for the high throughput screening of transmission electron microscopy images at the ink stage. Results indicate the importance of model pre-training and data augmentation that works on multiple scales in training robust and accurate classification pipelines.

Electronic Structure and Conformational Properties of Polybenzimidazole-Based Ionenes-A Density Functional Theory Investigation.

Polybenzimidazole-based ionenes are explored for use in both alkaline anion-exchange membrane fuel cells and alkaline polymer electrolyzers. Poly-(hexamethyl-p-terphenylbenzimidazolium) (HMT-PMBI), the material of interest in this article, is exceptionally hydroxide-stable and water-insoluble. The impact of the degree of methylation on conformations and electronic structure properties of HMT-PMBI oligomers, from the monomer to the pentamer, is studied with density functional theory calculations. Optimization studies are presented for both the gas phase and in the presence of implicit water. In addition, time-dependent density functional theory is employed to generate the UV-vis absorption spectra of the studied systems. Results are insightful for experimentalists and theorists investigating the impact of synthetic and environmental conditions on the conformation and electronic properties of polybenzimidazole-based membranes.

Interface Properties of the Partially Oxidized Pt(111) Surface Using Hybrid DFT-Solvation Models.

This article reports a theoretical-computational effort to model the interface between an oxidized platinum surface and aqueous electrolyte. It strives to account for the impact of the electrode potential, formation of surface-bound oxygen species, orientational ordering of near-surface solvent molecules, and metal surface charging on the potential profile along the normal direction. The computational scheme is based on the DFT/ESM-RISM method to simulate the charged Pt(111) surface with varying number of oxygen adatoms in acidic solution. This hybrid solvation method is known to qualitatively reproduce bulk metal properties like the work function. However, the presented calculations reveal that vital interface properties such as the electrostatic potential at the outer Helmholtz plane are highly sensitive to the position of the metal surface slab relative to the DFT-RISM boundary region. Shifting the relative position of the slab also affects the free energy of the system. It follows that there is an optimal distance for the first solvent layer within the ESM-RISM framework, which could be found by optimizing the position of the frozen Pt(111) slab. As it stands, manual sampling of the position of the slab is impractical and betrays the self-consistency of the method. Based on this understanding, we propose the implementation of a free energy optimization scheme of the relative position of the slab in the DFT-RISM boundary region. This optimization scheme could considerably increase the applicability of the hybrid method.

Virtual Materials Intelligence for Design and Discovery of Advanced Electrocatalysts.

Similar to advancements gained from big data in genomics, security, internet of things, and e-commerce, the materials workflow could be made more efficient and prolific through advances in streamlining data sources, autonomous materials synthesis, rapid characterization, big data analytics, and self-learning algorithms. In electrochemical materials science, data sets are large, unstructured/heterogeneous, and difficult to process and analyze from a single data channel or platform. Computer-aided materials design together with advances in data mining, machine learning, and predictive analytics are expected to provide inexpensive and accelerated pathways towards tailor-made functionally optimized energy materials. Fundamental research in the field of electrochemical energy materials focuses primarily on complex interfacial phenomena and kinetic electrocatalytic processes. This perspective article critically assesses AI-driven modeling and computational approaches that are currently applied to those objects. An application-driven materials intelligence platform is introduced, and its functionalities are scrutinized considering the development of electrocatalyst materials for CO2 conversion as a use case.

Modelling oxide formation and growth on platinum.

We present a mathematical model of oxide formation and growth on platinum. The motivation stems from the necessity to understand platinum dissolution in the cathode catalyst layer of polymer electrolyte fuel cells. As is known, platinum oxide formation and reduction are strongly linked to platinum dissolution processes. However, a consistent model of the oxidation processes on platinum does not exist. Our oxide growth model links interfacial exchange processes between platinum and oxygen ions with the transport of oxygen ion vacancies via diffusion and migration. A parametric analysis is performed to rationalize vital trends in oxide growth kinetics. The rate determining step of oxide formation and growth is identified as the extraction of platinum atoms at the metal-oxide interface. A kinetic effect is observed while adjusting the potential when growing the oxide layer, and the solution indicates that a structural change occurs at high potentials, around 1.5 VRHE. The model compares well to experimental data for various materials from various sources.

Conformational Properties of Comb-Like Polyelectrolytes: A Coarse-Grained MD Study.

This article presents a coarse-grained molecular dynamics study of single comb-like polyelectrolyte or ionomer chains in aqueous solution. The model polymer is comprised of a hydrophobic backbone chain with grafted side chains that terminate in anionic headgroups. The comb-polymer is modeled at a coarse-grained level with implicit treatment of the solvent. The computational study rationalizes conformational properties of the backbone chain and localization of counterions as functions of side chain length, grafting density of side chains, backbone stiffness, and counterion valence. The main interplay that determines the ionomer properties unfolds between electrostatic interactions among charged groups, hydrophobic backbone interactions, and steric effects induced by the pendant side chains. Depending on the density of branching sites, we have found two opposing effects of side chain length on the backbone gyration radius and local persistence length. Variation in comb-polyelectrolyte architecture is shown to have nontrivial effects on the localization of mobile counterions. Changes in Bjerrum length and counterion valence are also shown to alter the strength of Coulomb interactions and emphasize the role of excluded-volume effects on controlling the backbone conformational behavior. The results of simulations are in qualitative agreement with existing experimental and theoretical studies. The comprehensive conformational picture provides a framework for future studies of comb-polyelectrolyte systems.

Modeling the local potential at Pt nanoparticles in polymer electrolyte membranes.

We present a physical-analytical model for the potential distribution at Pt nanodeposits in a polymer electrolyte membrane (PEM). Experimental studies have shown that solid deposits of Pt in PEM play a dual role in radical-initiated membrane degradation. Surface reactions at Pt particles could facilitate the formation as well as the scavenging of ionomer-attacking radical species. The net radical balance depends on local equilibrium conditions at Pt nanodeposits in the PEM, specifically, their equivalent local electrode potential. Our approach utilizes a continuum description of crossover fluxes of reactant gases, coupled with the kinetics of electrochemical surface reactions at Pt nanodeposits to calculate the potential distribution. The local potential is a function of the PEM structure and composition, which is determined by PEM thickness, concentrations of H2 and O2, as well as the size and density distribution of Pt particles. Model results compare well with experimental data for the potential distribution in PEMs.

Titanium oxynitride interlayer to influence oxygen reduction reaction activity and corrosion stability of Pt and Pt-Ni alloy.

A key advancement target for oxygen reduction reaction catalysts is to simultaneously improve both the electrochemical activity and durability. To this end, the efficacy of a new highly conductive support that comprises of a 0.5 nm titanium oxynitride film coated by atomic layer deposition onto an array of carbon nanotubes has been investigated. Support effects for pure platinum and for a platinum (50 at %)/nickel alloy have been considered. Oxynitride induces a downshift in the d-band center for pure platinum and fundamentally changes the platinum particle size and spatial distribution. This results in major enhancements in activity and corrosion stability relative to an identically synthesized catalyst without the interlayer. Conversely, oxynitride has a minimal effect on the electronic structure and microstructure, and therefore, on the catalytic performance of platinum-nickel. Calculations based on density functional theory add insight with regard to compositional segregation that occurs at the alloy catalyst-support interface.

Radically coarse-grained approach to the modeling of chemical degradation in fuel cell ionomers.

We present a kinetic model of chemical degradation in perfluorosulfonic acid ionomer membranes. It accounts for pathways of radical formation along with mechanisms of ionomer degradation through radical attack. Simplifications in the set of model equations leads to analytical expressions for the concentration of hydroxyl radicals as a function of initial concentrations of iron ions and hydrogen peroxide. The coarse-grained ionomer degradation model distinguishes units that correspond to ionomer head groups, trunk segments of ionomer side chains, and backbone segments between two side chains. A set of differential equations is formulated to describe changes in concentrations of these units. The model is used to study the impact of different degradation mechanisms and ionomer chemistries on fluorine loss and change in ion exchange capacity. Comparison of the model with experimental degradation data for Nafion and Aquivion membranes allows rate constants of degradation processes to be determined. Results of these analyses are discussed in view of strategies to mitigate chemical degradation of ionomer membranes.

Body centered cubic magnesium niobium hydride with facile room temperature absorption and four weight percent reversible capacity.

We have synthesized a new metastable metal hydride with promising hydrogen storage properties. Body centered cubic (bcc) magnesium niobium hydride (Mg(0.75)Nb(0.25))H(2) possesses 4.5 wt% hydrogen gravimetric density, with 4 wt% being reversible. Volumetric hydrogen absorption measurements yield an enthalpy of hydride formation of -53 kJ mol(-1) H(2), which indicates a significant thermodynamic destabilization relative to the baseline -77 kJ mol(-1) H(2) for rutile MgH(2). The hydrogenation cycling kinetics are remarkable. At room temperature and 1 bar hydrogen it takes 30 minutes to absorb a 1.5 µm thick film at sorption cycle 1, and 1 minute at cycle 5. Reversible desorption is achieved in about 60 minutes at 175 °C. Using ab initio calculations we have examined the thermodynamic stability of metallic alloys with hexagonal close packed (hcp) versus bcc crystal structure. Moreover we have analyzed the formation energies of the alloy hydrides that are bcc, rutile or fluorite.