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.

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.

Cation Overcrowding Effect on the Oxygen Evolution Reaction.

The influence of electrolyte ions on the catalytic activity of electrode/electrolyte interfaces is a controversial topic for many electrocatalytic reactions. Herein, we focus on an effect that is usually neglected, namely, how the local reaction conditions are shaped by nonspecifically adsorbed cations. We scrutinize the oxygen evolution reaction (OER) at nickel (oxy)hydroxide catalysts, using a physicochemical model that integrates density functional theory calculations, a microkinetic submodel, and a mean-field submodel of the electric double layer. The aptness of the model is verified by comparison with experiments. The robustness of model-based insights against uncertainties and variations in model parameters is examined, with a sensitivity analysis using Monto Carlo simulations. We interpret the decrease in OER activity with the increasing effective size of electrolyte cations as a consequence of cation overcrowding near the negatively charged electrode surface. The same reasoning could explain why the OER activity increases with solution pH on the RHE scale and why the OER activity decreases in the presence of bivalent cations. Overall, this work stresses the importance of correctly accounting for local reaction conditions in electrocatalytic reactions to obtain an accurate picture of factors that determine the electrode activity.

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.

The effect of side-chain length on the solid-state structure and optoelectronic properties of fluorene-alt-benzothiadiazole based conjugated polymers--a DFT study.

Using the dispersion corrected density functional theory (DFT-D/B97D) approach, we have performed bulk solid-state calculations to investigate the influence of side-chain length on the molecular packing and optoelectronic properties of poly (9,9-di-n-alkylfluorene-alt-benzothiadiazole) or FnBT's where n is the number of CH(2) units in the alkyl side-chains. Our results indicate that the FnBT's with longer side-chains in their most stable configurations, due to the significant intermolecular interactions between the side-chains, form lamellar crystal structures. On the other hand, for the FnBT's with shorter side-chains, two nearly degenerate stable crystal structures with nearly hexagonal symmetries have been found. These different packing structures can be attributed to the microphase separations between the flexible side-chains and the rigid backbones whose existence has been discussed in previous investigations for other hairy rod polymers. As a result of the efficient interchain interactions for the lamellar structures, the dihedral angle between the F and BT units is reduced by about 30°, providing a more planar configuration for the backbone. In turn, a more planar backbone leads to a decrease, about 0.2 and 0.3 eV, of the band gaps of the lamellar structures relative to the gap values for the gas and the nearly hexagonal phases, respectively. Time-dependent DFT (TD-DFT) was used to study the excited states of the monomers of FnBT's with various lengths of side-chains. TD-DFT study suggests that the absorption spectrum of the polymers with longer side-chains is red-shifted relative to the polymers with shorter side-chains and the gas phase.