Data-driven simulations for training AI-based segmentation of neutron images.

Neutron interferometry uniquely combines neutron imaging and scattering methods to enable characterization of multiple length scales from 1 nm to 10 µm. However, building, operating, and using such neutron imaging instruments poses constraints on the acquisition time and on the number of measured images per sample. Experiment time-constraints yield small quantities of measured images that are insufficient for automating image analyses using supervised artificial intelligence (AI) models. One approach alleviates this problem by supplementing annotated measured images with synthetic images. To this end, we create a data-driven simulation framework that supplements training data beyond typical data-driven augmentations by leveraging statistical intensity models, such as the Johnson family of probability density functions (PDFs). We follow the simulation framework steps for an image segmentation task including Estimate PDFs → Validate PDFs → Design Image Masks → Generate Intensities → Train AI Model for Segmentation. Our goal is to minimize the manual labor needed to execute the steps and maximize our confidence in simulations and segmentation accuracy. We report results for a set of nine known materials (calibration phantoms) that were imaged using a neutron interferometer acquiring four-dimensional images and segmented by AI models trained with synthetic and measured images and their masks.

Simultaneous multimaterial operando tomography of electrochemical devices.

The performance of electrochemical energy devices, such as fuel cells and batteries, is dictated by intricate physiochemical processes within. To better understand and rationally engineer these processes, we need robust operando characterization tools that detect and distinguish multiple interacting components/interfaces in high contrast. Here, we uniquely combine dual-modality tomography (simultaneous neutron and x-ray tomography) and advanced image processing (iterative reconstruction and metal artifact reduction) for high-contrast multimaterial imaging, with signal and contrast enhancements of up to 10 and 48 times, respectively, compared to conventional single-modality imaging. Targeted development and application of these methods to electrochemical devices allow us to resolve operando distributions of six interacting fuel cell components (including void space) with the highest reported pairwise contrast for simultaneous yet decoupled spatiotemporal characterization of component morphology and hydration. Such high-contrast tomography ushers in key gold standards for operando electrochemical characterization, with broader applicability to numerous multimaterial systems.

Data from cryo-neutron phase change experiments with LH2 and LCH4.

Cryogenic Propellant management is a critical roadblock to enable long term space missions. Commonly used propellants (liquid hydrogen and methane) undergo constant vaporization but there is limited knowledge on the phase change rate and its implications on long term storage stability. This is, in part, due to the inability to image the liquid-vapor mixture inside opaque metallic containers at cryogenic temperatures. Here, neutron imaging is used as a visualization technique to track the liquid-vapor interface inside Al 6061 and SS 316 test cells. The data contains first known images of steady evaporation/condensation in cryogenic propellants. The experiments were conducted at the NIST Center for Neutron Research using the BT-2 Neutron Imaging facility. The test cells were instrumented with temperature sensors and inserted into a 70-mm liquid helium cryostat before being placed into the neutron beam. Temperatures and pressures were altered to achieve condensation/evaporation and Neutron images were captured during the entire phase change process. Phase change rates were obtained through image processing. The data contains raw images and processed phase change rates along with experimental temperature and pressure. The one-of-a-kind data could be used for model validation, correlation development or serve as a benchmark for future experiments.

Phase retrieval based on deep learning in grating interferometer.

Grating interferometry is a promising technique to obtain differential phase contrast images with illumination source of low intrinsic transverse coherence. However, retrieving the phase contrast image from the differential phase contrast image is difficult due to the accumulated noise and artifacts from the differential phase contrast image (DPCI) reconstruction. In this paper, we implemented a deep learning-based phase retrieval method to suppress these artifacts. Conventional deep learning based denoising requires noise/clean image pair, but it is not feasible to obtain sufficient number of clean images for grating interferometry. In this paper, we apply a recently developed neural network called Noise2Noise (N2N) that uses noise/noise image pairs for training. We obtained many DPCIs through combination of phase stepping images, and these were used as input/target pairs for N2N training. The application of the N2N network to simulated and measured DPCI showed that the phase contrast images were retrieved with strongly suppressed phase retrieval artifacts. These results can be used in grating interferometer applications which uses phase stepping method.

Analysis of a silicon comb structure using an inverse Talbot-Lau neutron grating interferometer.

We describe an inverse Talbot-Lau neutron grating interferometer that provides an extended autocorrelation length range for quantitative dark-field imaging. To our knowledge, this is the first report of a Talbot-Lau neutron grating interferometer (nTLI) with inverse geometry. We demonstrate a range of autocorrelation lengths (ACL) starting at low tens of nanometers, which is significantly extended compared to the ranges of conventional and symmetric setups. ACLs from a minimum of 44 nm to the maximum of 3.5 µm were presented for the designed wavelength of 4.4 Å in experiments. Additionally, the inverse nTLI has neutron-absorbing gratings with an optically thick gadolinium oxysulfide (Gadox) structure, allowing it to provide a visibility of up to 52% while maintaining a large field of view of approximately 100 mm × 100 mm. We demonstrate the application of our interferometer to quantitative dark-field imaging by using diluted polystyrene particles in an aqueous solution and silicon comb structures. We obtain quantitative structural information of the sphere size and concentration of diluted polystyrene particles and the period, height, and duty cycle of the silicon comb structures. The optically thick Gadox structure of the analyzer grating also provides improved characteristics for the correction of incoherent neutron scattering in an aqueous solution compared to the symmetric nTLI.

Effect of Fluid Properties on Contact Angles in the Eagle Ford Shale Measured with Spontaneous Imbibition.

Models of fluid flow are used to improve the efficiency of oil and gas extraction and to estimate the storage and leakage of carbon dioxide in geologic reservoirs. Therefore, a quantitative understanding of key parameters of rock-fluid interactions, such as contact angles, wetting, and the rate of spontaneous imbibition, is necessary if these models are to predict reservoir behavior accurately. In this study, aqueous fluid imbibition rates were measured in fractures in samples of the Eagle Ford Shale using neutron imaging. Several liquids, including pure water and aqueous solutions containing sodium bicarbonate and sodium chloride, were used to determine the impact of solution chemistry on uptake rates. Uptake rate analysis provided dynamic contact angles for the Eagle Ford Shale that ranged from 51 to 90° using the Schwiebert-Leong equation, suggesting moderately hydrophilic mineralogy. When corrected for hydrostatic pressure, the average contact angle was calculated as 76 ± 7°, with higher values at the fracture inlet. Differences in imbibition arising from differing fracture widths, physical liquid properties, and wetting front height were investigated. For example, bicarbonate-contacted samples had average contact angles that varied between 62 ± 10° and ∼84 ± 6° as the fluid rose in the column, likely reflecting a convergence-divergence structure within the fracture. Secondary imbibitions into the same samples showed a much more rapid uptake for water and sodium chloride solutions that suggested alteration of the clay in contact with the solution producing a water-wet environment. The same effect was not observed for sodium bicarbonate, which suggested that the bicarbonate ion prevented shale hydration. This study demonstrates how the imbibition rate measured by neutron imaging can be used to determine contact angles for solutions in contact with shale or other materials and that wetting properties can vary on a relatively fine scale during imbibition, requiring detailed descriptions of wetting for accurate reservoir modeling.

Effect of Membrane Properties on the Carbonation of Anion Exchange Membrane Fuel Cells.

Anion exchange membrane fuel cells (AEMFC) are potentially very low-cost replacements for proton exchange membrane fuel cells. However, AEMFCs suffer from one very serious drawback: significant performance loss when CO2 is present in the reacting oxidant gas (e.g., air) due to carbonation. Although the chemical mechanisms for how carbonation leads to voltage loss in operating AEMFCs are known, the way those mechanisms are affected by the properties of the anion exchange membrane (AEM) has not been elucidated. Therefore, this work studies AEMFC carbonation using numerous high-functioning AEMs from the literature and it was found that the ionic conductivity of the AEM plays the most critical role in the CO2-related voltage loss from carbonation, with the degree of AEM crystallinity playing a minor role. In short, higher conductivity-resulting either from a reduction in the membrane thickness or a change in the polymer chemistry-results in faster CO2 migration and emission from the anode side. Although this does lead to a lower overall degree of carbonation in the polymer, it also increases CO2-related voltage loss. Additionally, an operando neutron imaging cell is used to show that as AEMFCs become increasingly carbonated their water content is reduced, which further drives down cell performance.

Boosting Membrane Hydration for High Current Densities in Membrane Electrode Assembly CO2 Electrolysis.

Despite the advantages of CO2 electrolyzers, efficiency losses due to mass and ionic transport across the membrane electrode assembly (MEA) are critical bottlenecks for commercial-scale implementation. In this study, more efficient electrolysis of CO2 was achieved by increasing cation exchange membrane (CEM) hydration via the humidification of the CO2 reactant inlet stream. A high current density of 755 mA/cm2 was reached by humidifying the reactant CO2 in a MEA electrolyzer cell featuring a CEM. The power density was reduced by up to 30% when the fully humidified reactant CO2 was introduced while operating at a current density of 575 mA/cm2. We reduced the ohmic losses of the electrolyzer by fourfold at 575 mA/cm2 by fully humidifying the reactant CO2. A semiempirical CEM water uptake model was developed and used to attribute the improved performance to 11% increases in membrane water uptake and ionic conductivity. Our CEM water uptake model showed that the increase in ohmic losses and the limitation of ionic transport were the result of significant dehydration at the central region of the CEM and the anode gas diffusion electrode-CEM interface region, which exhibited a 2.5% drop in water uptake.

Electric Field Imaging Using Polarized Neutrons.

We experimentally demonstrate that electrically neutral particles, neutrons, can be used to directly visualize the electrostatic field inside a target volume that can be physically isolated or occupied. Electric field images are obtained using a spin-polarized neutron beam with a recently developed polarimetry method for polychromatic beams that permits detection of a small angular change in spin orientation. This Letter may enable a new diagnostic technique sensitive to the structure of electric potential, electric polarization, charge distribution, and dielectric constant by imaging spatially dependent electric fields in objects that cannot be accessed by other probes.

Neutron grating interferometer with an analyzer grating based on a light blocker.

We study an analyzer grating based on a scintillation light blocker for a Talbot-Lau grating interferometer. This is an alternative way to analyze the Talbot self-image without the need for an often difficult to fabricate absorption grating for the incident radiation. The feasibility of this approach using a neutron beam has been evaluated and experiments have been conducted at the cold neutron imaging facility of the NIST center for Neutron Research. The neutron grating interferometer with the proposed analyzer grating successfully produced attenuation, differential phase, and dark-field contrast images. In addition, numerical simulations were performed to simulate the Talbot pattern and visibility using scintillation screens of different thicknesses and there is good agreement with the experimental measurements. The results show potential for reducing the difficulty of fabricating analyzer grating, and a possibility for the so-called shadow effect to be eliminated and large-area gratings to be produced, especially when applied to X-rays. We report the performance of the analyzer grating based on a light blocker and evaluate its feasibility for the grating interferometer.

Using operando techniques to understand and design high performance and stable alkaline membrane fuel cells.

There is a need to understand the water dynamics of alkaline membrane fuel cells under various operating conditions to create electrodes that enable high performance and stable, long-term operation. Here we show, via operando neutron imaging and operando micro X-ray computed tomography, visualizations of the spatial and temporal distribution of liquid water in operating cells. We provide direct evidence for liquid water accumulation at the anode, which causes severe ionomer swelling and performance loss, as well as cell dryout from undesirably low water content in the cathode. We observe that the operating conditions leading to the highest power density during polarization are not generally the conditions that allow for long-term stable operation. This observation leads to new catalyst layer designs and gas diffusion layers. This study reports alkaline membrane fuel cells that can be operated continuously for over 1000 h at 600 mA cm-2 with voltage decay rate of only 32-µV h-1 - the best-reported durability to date.

Deep learning for high-resolution and high-sensitivity interferometric phase contrast imaging.

In Talbot-Lau interferometry, the sample position yielding the highest phase sensitivity suffers from strong geometric blur. This trade-off between phase-sensitivity and spatial resolution is a fundamental challenge in such interferometric imaging applications with either neutron or conventional x-ray sources due to their relatively large beam-defining apertures or focal spots. In this study, a deep learning method is introduced to estimate a high phase-sensitive and high spatial resolution image from a trained neural network to attempt to avoid the trade-off for both high phase-sensitivity and high resolution. To realize this, the training data sets of the differential phase contrast images at a pair of sample positions, one of which is close to the phase grating and the other close to the detector, are numerically generated and are used as the inputs for the training data set of a generative adversarial network. The trained network has been applied to the real experimental data sets from a neutron grating interferometer and we have obtained improved images both in phase-sensitivity and spatial resolution.

Design of Neutron Microscopes Equipped with Wolter Mirror Condenser and Objective Optics for High-Fidelity Imaging and Beam Transport.

We present and compare the designs of three types of neutron microscopes for high-resolution neutron imaging. Like optical microscopes, and unlike standard neutron imaging instruments, these microscopes have both condenser and image-forming objective optics. The optics are glancing-incidence axisymmetric mirrors and therefore suitable for polychromatic neutron beams. The mirrors are designed to provide a magnification of 10 to achieve a spatial resolution of better than 10 µm. The resolution of the microscopes is determined by the mirrors rather than by the L/Dratio as in conventional pinhole imaging, leading to possible dramatic improvements in the signal rate. We predict the increase in the signal rate by at least two orders of magnitude for very high-resolution imaging, which is always flux limited. Furthermore, in contrast to pinhole imaging, in the microscope, the samples are placed far from the detector to allow for a bulky sample environment without sacrificing spatial resolution.

Understanding water management in platinum group metal-free electrodes using neutron imaging.

Platinum group metal-free (PGM-free) catalysts are a low-cost alternative to expensive PGM catalysts for polymer electrolyte fuel cells. However, due to the low volumetric activity of PGM-free catalysts, the catalyst layer thickness of the PGM-free catalyst electrode is an order of magnitude higher than PGM based electrodes. The thick PGM-free electrodes suffer from increased transport resistance and poor water management, which ultimately limits the fuel cell performance. This manuscript presents the study of water management in the PGM-free electrodes to understand the transport limitations and improve fuel cell performance. In-operando neutron imaging is performed to estimate the water content in different components across the fuel cell thickness. Water saturation in thick PGM electrodes, with similar catalyst layer thickness to PGM-free electrodes, is lower than in the PGM-free electrodes irrespective of the operating conditions, due to high water retention by PGM-free catalysts. Improvements in fuel cell performance are accomplished by enhancing water removal from the flooded PGM-free electrode in three ways: (i) enhanced water removal with a novel microporous layer with hydrophilic pathways incorporated through hydrophilic additives, (ii) water removal through anode via novel GDL in the anode, and (iii) lower water saturation in PGM-free electrode structures with increased catalyst porosity.

Reconciling temperature-dependent factors affecting mass transport losses in polymer electrolyte membrane electrolyzers.

In this work, we investigated the impact of temperature on two-phase transport in low temperature (LT)-polymer electrolyte membrane (PEM) electrolyzer anode flow channels via in operando neutron imaging and observed a decrease in mass transport overpotential with increasing temperature. We observed an increase in anode oxygen gas content with increasing temperature, which was counter-intu.itive to the trends in mass transport overpotential. We attributed this counterintuitive decrease in mass transport overpotential to the enhanced reactant distribution in the flow channels as a result of the temperature increase, determined via a one-dimensional analytical model. We further determined that gas accumulation and fluid property changes are competing, temperature-dependent contributors to mass transport overpotential; however, liquid water viscosity changes led to the dominate enhancement of reactant water distributions in the anode. We present this temperature-dependent mass transport overpotential as a great opportunity for further increasing the voltage efficiency of PEM electrolyzers.

Thick Sintered Electrode Lithium-Ion Battery Discharge Simulations: Incorporating Lithiation-Dependent Electronic Conductivity and Lithiation Gradient Due to Charge Cycle.

In efforts to increase the energy density of lithium-ion batteries, researchers have attempted to both increase the thickness of battery electrodes and increase the relative fractions of active material. One system that has both of these attributes are sintered thick electrodes comprised of only active material. Such electrodes have high areal capacities, however, detailed understanding is needed of their transport properties, both electronic and ionic, to better quantify their limitations to cycling at higher current densities. In this report, efforts to improve models of the electrochemical cycling of sintered electrodes are described, in particular incorporation of matrix electronic conductivity which is dependent on the extent of lithiation of the active material and accounting for initial gradients in lithiation of active material in the electrode that develop as a consequence of transport limitations during charging cycles. Adding in these additional considerations to a model of sintered electrode discharge resulted in improved matching of experimental cell measurements.

Probing transport limitations in thick sintered battery electrodes with neutron imaging.

Lithium-ion batteries have received significant research interest due to their advantages in energy and power density, which are important to enabling many devices. One route to further increase energy density is to fabricate thicker electrodes in the battery cell; however, careful consideration must be taken when designing electrodes as to how increasing the thickness impacts the multiscale and multiphase molecular transport processes, which can limit the overall battery operating power. Design of these electrodes necessitates probing the molecular processes when the battery cell undergoes electrochemical charge/discharge. One tool for in situ insights into the cell is neutron imaging, because neutron imaging can provide information of where electrochemical processes occur within the electrodes. In this manuscript, neutron imaging is applied to track the lithiation/delithiation processes within electrodes at different current densities for a full cell with a thick sintered Li4Ti5O12 anode and LiCoO2 cathode. The neutron imaging reveals that the molecular distribution of Li+ during discharge within the electrode is sensitive to the current density, or equivalently discharge rate. An electrochemical model provides additional insights into the limiting processes occurring within the electrodes. In particular, the impact of tortuosity and molecular transport in the liquid phase within the interstitial regions in the electrodes are considered, and the influence of tortuosity was shown to be highly sensitive to the current density. Qualitatively, the experimental results suggest that the electrodes behave consistent with the packed hard sphere approximation of Bruggeman tortuosity scaling, which indicates that the electrodes are largely mechanically intact but also that a design that incorporates tunable tortuosity could improve the performance of these types of electrodes.

Characterization of the phase sensitivity, visibility, and resolution in a symmetric neutron grating interferometer.

A symmetric Talbot-Lau neutron grating interferometer has been developed for achieving high phase sensitivity. The gratings for the interferometer have been designed by maximizing the intergrating distances available at the experimental facility and optimizing the period of the gratings. The phase sensitivity in a Talbot-Lau grating interferometer has been mathematically modeled and analyzed and compared with experimental data. Evaluation experiments have been performed at the cold neutron imaging facility of the NIST Center for Neutron Research. The symmetric Talbot-Lau neutron grating interferometer exhibits distinct advantages in terms of high sensitivity and wide range for dark-field contrast imaging. Also, the fabrication of gratings is made easier, simpler, and more economical by the Gadox powder filling method, which gives them excellent phase contrast compared with other geometric Talbot-Lau neutron grating interferometers.

Multi-scale analyses of constituent phases in a trip-assisted duplex stainless steel by electron backscatter diffraction, in situ neutron diffraction, and energy selective neutron imaging.

Micrometer to centimeter scale analyses of the crystalline phase volume fractions in a trip-assisted duplex stainless steel were performed under loading using electron backscatter diffraction (EBSD), in situ neutron diffraction, and energy selective neutron imaging (ESNI) methods. In contrast to the localized investigations of EBSD, ESNI provides macroscopic spatial distributions in a volume-averaged manner over the entire specimen with a spatial resolution of about 65 µm. The ESNI shows that the martensite is concentrated on the necking region and estimates its volume fraction of 14% at a strain of 0.2, which is comparable to the neutron diffraction result.