Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “catalyst layer”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 829 records · Page 46
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Atomic layer deposition for surface area determination of solid oxide electrodes

Surface area measurements are important for characterizing porous solids, but commonly used methods such as Brunauer–Emmett–Teller (BET) gas adsorption and computed nanotomography are lacking for certain applications such as the development of solid oxide fuel cell electrodes, which often have absolute surface areas <1 m 2 and contain nanoscale features. Presented is a novel method for the surface area determination of samples with total measurable areas of 1–1000 cm 2 with a standard deviation ± 1 cm 2 , utilizing the atomic layer deposition (ALD) of Al 2 O 3 over microstructurally complex internal porosity. The volume of alumina is then quantified using plasma spectroscopy and converted to an area using the known ALD layer thickness. Under the modest ALD reactor soak times used (8.5 s), the precursor penetration depth is found to be ~50 μm, exceeding the requirement for uniformly coating SOFC functional layers. A model system, (La 0.8 Sr 0.2 ) 0.98 MnO 3–δ /Ce 0.9 Gd 0.1 O 1.95 (LSM/GDC) scaffolds of ~2.7 m 2 g –1 , was measured using the technique and compared against the BET method, and comparable results were obtained but with 1000 times less material needed. The technique is demonstrated for measurements in two example areas of active SOFC research: exsolved (Ni, Fe) nanoparticles on an Sr(Ti, Fe, Ni)O 3–δ electrode and PrO x catalyst nanoparticles infiltrated onto LSM/GDC scaffolds. Finally, the technique may be broadly useful wherever the accurate surface area determination of small absolute quantities of porous ceramic structures on the order of 0.1–50 m 2 g –1 is sought.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Charging Reactions Promoted by Geometrically Necessary Dislocations in Battery Materials Revealed by In Situ Single-Particle Synchrotron Measurements

Crystallographic defects exist in many redox active energy materials, e.g., battery and catalyst materials, which significantly alter their chemical properties for energy storage and conversion. However, there is lack of quantitative understanding of the interrelationship between crystallographic defects and redox reactions. Herein, crystallographic defects, such as geometrically necessary dislocations, are reported to influence the redox reactions in battery particles through single-particle, multimodal, and in situ synchrotron measurements. Through Laue X-ray microdiffraction, many crystallographic defects are spatially identified and statistically quantified from a large quantity of diffraction patterns in many layered oxide particles, including geometrically necessary dislocations, tilt boundaries, and mixed defects. The in situ and ex situ measurements, combining microdiffraction and X-ray spectroscopy imaging, reveal that LiCoO 2 particles with a higher concentration of geometrically necessary dislocations provide deeper charging reactions, indicating that dislocations may facilitate redox reactions in layered oxides during initial charging. The present study illustrates that a precise control of crystallographic defects and their distribution can potentially promote and homogenize redox reactions in battery materials.

25 ENERGY STORAGE↗
DOE OSTI2021

Surface enhanced performance of La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3-δ cathodes by infiltration Pr-Ni-Mn-O progress

Herein the present study reports the enhancement of electrochemical oxygen reduction activity of porous La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3–δ (LSCF) cathodes by coating a thin film of Pr-Ni-Mn oxide (PNM5) using a multi-step infiltration process. XRD examination reveals that PNM5 mainly contains a multiphase mixture of Pr 6 O 11 , PrNiO 3 , MnO and NiO. SEM morphology shows a thin PNM5 film and small particles are formed on the surface of LSCF backbone particles. Impedance spectrum analysis indicates that PNM5 infiltrated LSCF exhibits dramatically reduced polarization resistance (R p ), reaching R p of 0.244 Ω cm 2 at 973 K, which is one-half of the baseline LSCF cathode. The activation energy of LSCF cathodes infiltrated with PNM5 is 1.45 eV, slightly lower than the baseline LSCF cathode (1.77 eV). Distribution of relaxation time (DRT) function analysis shows PNM5 infiltration layer significantly promotes the oxygen reduction reaction (ORR) of cathode surface. With the increase in firing temperature, the total resistance increases and R p changes from ion transport to oxygen reduction reaction. Degradation rate of the PNM5-infiltrated LSCF is also lower, 0.02168% vs 0.07093% for the baseline LSCF over a 200-h period. A single cell testing indicated that the peak power density of the PNM5-infiltrated cell was increased by 140.66%. Overall, PNM5 could be a potential catalyst for boosting the performance of a commercial LSCF cathode for solid oxide fuel cells (SOFC).

36 MATERIALS SCIENCE↗
DOE OSTI2026

Vapor-phase pillarization of MXenes for engineering hierarchical interlayer porosity

MXenes, a family of two-dimensional (2D) multilamellar materials, possess excellent thermal and electronic properties for a range of applications. Their use in heterogeneous catalysis, however, is limited by the low surface area resulting from stacked layers. Pillarization with inorganic oxides can create more open, mesoporous MXene structures, improving accessibility for guest species to diffuse, reside or react in the space between 2D layers. A previous liquid-phase pillarization method, however, involves excessive use of solvent-based precursors and multiple processing steps. Here, we report a vapor-phase pillarization (VPP) strategy to introduce pillars, exemplified by silica pillars, with high pillar precursor usage efficiency and a simplified processing workflow. The resulting silica-pillared mesoporous MXene exhibits significantly increased surface area and porosity. These textural properties can be easily tuned by the VPP synthesis conditions. When applied as a ruthenium (Ru) catalyst support for the hydrogenolysis of low-density polyethylene (LDPE), the silica-pillared MXene enabled high Ru dispersion and catalytic activity. This study highlights the potential of the VPP method for engineering mesoporous, 2D MXene materials and demonstrates the effectiveness of mesoporous MXene as a catalyst support in overcoming mass transport and active-site accessibility challenges in heterogeneous catalysis involving bulky substances, such as plastics upcycling.

Luo, Song [University of Delaware, Newark, DE (Uni↗
DOE OSTI2020

Enhanced stability of silicon for photoelectrochemical water oxidation through self-healing enabled by an alkaline protective electrolyte

Alkaline electrolytes impede the corrosion of Si photoanodes under positive potentials and/or illumination, due to the formation of a SiO x layer that etches 2–3 orders of magnitude more slowly than Si. Hence during water oxidation under illumination, pinholes in protection layers on Si photoanodes result in the local formation of a protective, stabilizing passive oxide on the Si surface. However, operation under natural diurnal insolation cycles additionally requires protection strategies that minimize the dark corrosive etching rate of Si at pinholes. We show herein that addition of [Fe(CN) 6 ] 3- to 1.0 M KOH(aq) results in a self-healing process that extends the lifetime to >280 h of an np + -Si(100) photoanode patterned with an array of Ni catalyst islands operated under simulated day/night cycles. The self-healing [Fe(CN) 6 ] 3- additive caused the exposed Si(100) surface to etch >180 times slower than the Si etch rate in 1.0 M KOH(aq) alone. No appreciable difference in etch rate or facet preference was observed between Si(100) and Si(111) surfaces in 1.0 M KOH(aq) with [Fe(CN) 6 ] 3- , indicating that the surface conformally oxidized before Si dissolved. The presence of [Fe(CN) 6 ] 3- minimally impacted the faradaic efficiency or overpotential of p + -Si/Ni electrodes for the oxygen-evolution reaction.

25 ENERGY STORAGE↗
DOE OSTI2023

A High-Entropy Oxide as High-Activity Electrocatalyst for Water Oxidation

High-entropy materials are an emerging pathway in the development of high-activity (electro)catalysts because of the inherent tunability and coexistence of multiple potential active sites, which may lead to earth-abundant catalyst materials for energy-efficient electrochemical energy storage. In this report, we identify how the multication composition in high-entropy perovskite oxides (HEO) contributes to high catalytic activity for the oxygen evolution reaction (OER), i.e., the key kinetically limiting half-reaction in several electrochemical energy conversion technologies, including green hydrogen generation. We compare the activity of the (001) facet of LaCr 0.2 Mn 0.2 Fe 0.2 Co 0.2 Ni 0.2 O 3-δ with the parent compounds (single B-site in the ABO 3 perovskite). While the single B-site perovskites roughly follow the expected volcano-type activity trends, the HEO clearly outperforms all of its parent compounds with 17 to 680 times higher currents at a fixed overpotential. As all samples were grown as an epitaxial layer, our results indicate an intrinsic composition–function relationship, avoiding the effects of complex geometries or unknown surface composition. In-depth X-ray photoemission studies reveal a synergistic effect of simultaneous oxidation and reduction of different transition metal cations during the adsorption of reaction intermediates. The surprisingly high OER activity demonstrates that HEOs are a highly attractive, earth-abundant material class for high-activity OER electrocatalysts, possibly allowing the activity to be fine-tuned beyond the scaling limits of mono- or bimetallic oxides.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Methane pyrolysis by Joule heating for graphitic carbon and hydrogen production

The global energy transition toward sustainability requires technologies that can decarbonize energy carriers and fuels while producing valuable materials. Methane, a primary component of natural gas, is both a high-energy-density fuel and a significant greenhouse gas. This study reports an approach for methane pyrolysis utilizing Joule heating within the deposition substrate to drive the endothermic reaction. With electric current passing through a resistive porous carbon cloth, heat is generated to break C-H bonds of methane molecules. Here, the decomposition of methane as it flows through the cloth results in hydrogen production and the formation of conformally layered graphite around the carbon fibers. The effects of input power, chamber pressure, feedstock flow rate, and process duration on hydrogen and graphite production are characterized via in situ mass spectrometry and laser absorption spectroscopy, resulting in methane conversion rates up to 88%, with hydrogen and carbon yields of 82% and 72%, respectively. Material characterization verifies uniform high-quality graphite deposition, with a Raman I D /I G ratio of 0.1 and 3.38 Å d-spacing. This Joule heating method for catalyst-free methane pyrolysis offers the potential for advancing hydrogen production technology by simultaneously producing valuable materials such as solid graphite, thus enhancing the economic viability of the fuel decarbonization process.

Energy Resources↗
DOE OSTI2021

Direct methane activation by atomically thin platinum nanolayers on two-dimensional metal carbides

We report efficient and direct conversion of methane to value-added products has been a long-term challenge in shale gas applications. Here, we show that atomically thin nanolayers of Pt with a single or double atomic layer thickness, supported on a two-dimensional molybdenum titanium carbide (MXene), catalyse non-oxidative coupling of methane to ethane/ethylene (C 2 ). Kinetic and theoretical studies, combined with in-situ spectroscopic and microscopic characterizations, demonstrate that Pt nanolayers anchored at the hexagonal close-packed sites of the MXene support can activate the first C–H bond of methane to form methyl radicals that favour desorption over further dehydrogenation and thus suppress coke deposition. At 750 °C and 7% methane conversion, the catalyst runs for 72 hours of continuous operation without deactivation and exhibits >98% selectivity towards C 2 products, with a turnover frequency of 0.2–0.6 s -1 . Our findings provide insights into the design of highly active and stable catalysts for methane activation and create a platform for developing atomically thin supported metal catalysts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Tailoring the Mass Transport to Achieve High-Performing Unitized Regenerative Fuel Cells Considering Practical Operating Conditions

Unitized regenerative fuel cells (URFCs) convert electrical energy to chemical bonds in hydrogen during charge and convert chemical energy to output electricity during discharge, offering a promising solution to long-term energy storage. Recent studies indicate that the round-trip-voltaic efficiency (RTE) and longevity of URFCs are limited by complex mass transport during charging and discharging. Here, we first investigate how different porous transport layer (PTL) structures can impact URFC performance. The preferred PTL has a low tortuosity and high porosity, leading to a high RTE above 50% at 1 A cm −2 using Nafion 212. Moreover, thicker membranes, such as Solvay 90, are required to ensure mechanical stability and minimize H 2 crossover when operating under high differential pressure. Although this assembly inevitably leads to a higher ohmic loss, the RTE can be improved by further tailoring the electrode structures to facilitate mass transport by using supported catalyst, which still achieves over 50% RTEs at 1 A cm −2 . Optimization of porous structure to mitigate mass transport resistance with appropriate materials down selection considering practical application requirements can be a key design principle for achieving high-performing URFCs.

25 ENERGY STORAGE↗
DOE OSTI2023

Interfacial Connections between Organic Perovskite/n + Silicon/Catalyst that Allow Integration of Solar Cell and Catalyst for Hydrogen Evolution from Water

Abstract The rapidly increasing solar conversion efficiency (PCE) of hybrid organic–inorganic perovskite (HOIP) thin‐film semiconductors has triggered interest in their use for direct solar‐driven water splitting to produce hydrogen. However, application of these low‐cost, electronic‐structure‐tunable HOIP tandem photoabsorbers has been hindered by the instability of the photovoltaic‐catalyst‐electrolyte (PV+E) interfaces. Here, photolytic water splitting is demonstrated using an integrated configuration consisting of an HOIP/n + silicon single junction photoabsorber and a platinum (Pt) thin film catalyst. An extended electrochemical (EC) lifetime in alkaline media is achieved using titanium nitride on both sides of the Si support to eliminate formation of insulating silicon oxide, and as an effective diffusion barrier to allow high‐temperature annealing of the catalyst/TiO 2 ‐protected‐n + silicon interface necessary to retard electrolytic corrosion. Halide composition is examined in the (FA 1‐x Cs x )PbI 3 system with a bandgap suitable for tandem operation. A fill factor of 72.5% is achieved using a Spiro‐OMeTAD‐hole‐transport‐layer (HTL)‐based HOIP/n + Si solar cell, and a high photocurrent density of −15.9 mA cm −2 (at 0 V vs reversible hydrogen electrode) is attained for the HOIP/n + Si/Pt photocathode in 1 m NaOH under simulated 1‐sun illumination. While this thin‐film design creates stable interfaces, the intrinsic photo‐ and electro‐degradation of the HOIP photoabsorber remains the main obstacle for future HOIP/Si tandem PEC devices.

08 HYDROGEN↗
DOE OSTI2022

An economic analysis of the role of materials, system engineering, and performance in electrochemical carbon dioxide conversion to formate

The development of technologies that utilize carbon dioxide is important to mitigating climate change. The electrochemical reduction of carbon dioxide is one technology that can utilize greenhouse gasses that would be otherwise be emitted to the atmosphere by producing chemicals and fuels from carbon dioxide and electricity. Significant progress has been made in the experimental performance of carbon dioxide reduction systems with novel catalyst designs, new materials, and systems engineering; however, no work has linked such changes in stack design and materials to capital costs for the stack itself. In this study, we present an analysis that accounts for and analyzes the impacts of alternative materials and system architectures on manufactured costs of carbon dioxide reduction stacks, thus providing a framework to understand exactly how these advances impact the at-scale capital costs of these systems. Specifically, we consider the impact that the addition of a catholyte buffer layer has on an electrolyzer reducing carbon dioxide to formate, finding that the cost of manufacturing this part only increases stack costs by about $30/m 2 at high manufacturing rates, while previous work finds that this part improves system performance. This work shows that the links between system performance, materials, and costs are nonlinear, and that achieving low-cost scalability requires optimization of not just performance parameters but also the use of low-cost and highly scalable materials. These results bridge experimental and techno-economic analysis of processes for carbon dioxide reduction, informing researchers by providing a quantifiable estimate of the impact of advances in electrochemical carbon dioxide reduction technology on manufactured stack capital costs.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
NASA NTRS1985

A model for the origin of photosynthesis--III. The ultraviolet photochemistry of uroporphyrinogen

The photochemical ramifications of the high ultraviolet flux on the primordial earth prior to the formation of the ozone layer have been considered in a study of the ultraviolet photochemistry of uroporphyrinogen (urohexahydroporphyrin), a colorless compound which absorbs strongly at wavelengths less than 220 nanometers. Urohexahydroporphyrin was investigated since it is the first macrocycle formed on the biosynthetic pathway of chlorophyll and can be used to test the hypothesis that the biosynthetic pathway to chlorophyll recapitulates the evolutionary history of photosynthesis. When urohexahydroporphyrin is illuminated in aqueous anaerobic solution, hydrogen gas is produced. More hydrogen gas is produced in the presence of a colloidal platinum catalyst. The products of the photooxidation of urohexahydroporphyrin are urotetrahydroporphyrin (uroporphomethene) and uroporphyrin. This research shows how the oxidation of uroporphyrinogen to uroporphyrin, the first biogenetic porphyrin, could have occurred anaerobically and abiotically on the primordial earth.

Evolution, Chemical↗
NASA NTRS2010

Fabrication of Single, Vertically Aligned Carbon Nanotubes in 3D Nanoscale Architectures

Plasma-enhanced chemical vapor deposition (PECVD) and high-throughput manufacturing techniques for integrating single, aligned carbon nanotubes (CNTs) into novel 3D nanoscale architectures have been developed. First, the PECVD growth technique ensures excellent alignment of the tubes, since the tubes align in the direction of the electric field in the plasma as they are growing. Second, the tubes generated with this technique are all metallic, so their chirality is predetermined, which is important for electronic applications. Third, a wafer-scale manufacturing process was developed that is high-throughput and low-cost, and yet enables the integration of just single, aligned tubes with nanoscale 3D architectures with unprecedented placement accuracy and does not rely on e-beam lithography. Such techniques should lend themselves to the integration of PECVD grown tubes for applications ranging from interconnects, nanoelectromechanical systems (NEMS), sensors, bioprobes, or other 3D electronic devices. Chemically amplified polyhydroxystyrene-resin-based deep UV resists were used in conjunction with excimer laser-based (lambda = 248 nm) step-and-repeat lithography to form Ni catalyst dots = 300 nm in diameter that nucleated single, vertically aligned tubes with high yield using dc PECVD growth. This is the first time such chemically amplified resists have been used, resulting in the nucleation of single, vertically aligned tubes. In addition, novel 3D nanoscale architectures have been created using topdown techniques that integrate single, vertically aligned tubes. These were enabled by implementing techniques that use deep-UV chemically amplified resists for small-feature-size resolution; optical lithography units that allow unprecedented control over layer-to-layer registration; and ICP (inductively coupled plasma) etching techniques that result in near-vertical, high-aspect-ratio, 3D nanoscale architectures, in conjunction with the use of materials that are structurally and chemically compatible with the high-temperature synthesis of the PECVD-grown tubes. The techniques offer a wafer-scale process solution for integrating single PECVD-grown nanotubes into novel architectures that should accelerate their integration in 3D electronics in general. NASA can directly benefit from this technology for its extreme-environment planetary missions. Current Si transistors are inherently more susceptible to high radiation, and do not tolerate extremes in temperature. These novel 3D nanoscale architectures can form the basis for NEMS switches that are inherently less susceptible to radiation or to thermal extremes.

Kaul, Anupama B.↗
DOE OSTI2020

Hot-Carrier Transfer at Photocatalytic Silicon/Platinum Interfaces

Interfacial charge transfer from silicon to heterogeneous catalysts plays a key role in silicon-based photoelectrochemical systems. In general, prior to interfacial charge transfer, carriers that are generated by photons with energies above the bandgap dissipate the excess kinetic energy via hot-carrier cooling, and such energy loss limits the maximum power conversion efficiency. The excess energy of hot-carriers, however, could be utilized through hot-carrier transfer from silicon to the catalysts, but such hot-carrier extraction has not yet been demonstrated. In this work, we exploit transient reflection spectroscopy to interrogate charge transfer at the interface between silicon and platinum. Additionally, quantitative modeling of the surface carrier kinetics indicates that the velocity of charge transfer from silicon to platinum exceeds 2.6 x 10 7 cm s -1 , corresponding to an average carrier temperature of extracted carriers of ~600 K, two times higher than the lattice temperature. The charge transfer velocity can be controllably reduced by inserting silica spacing layers between silicon and platinum.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Prospects for the expansion of standing wave ambient pressure photoemission spectroscopy to reactions at elevated temperatures

Standing wave ambient pressure photoemission spectroscopy (SWAPPS) is a promising method to investigate chemical and potential gradients across solid-vapor and solid-liquid interfaces under close-to-realistic environmental conditions, far away from high vacuum. Until now, these investigations have been performed only near room temperature, but for a wide range of interfacial processes, chief among them being heterogeneous catalysis, measurements at elevated temperatures are required. One concern in these investigations is the temperature stability of the multilayer mirrors, which generate the standing wave field. At elevated temperatures, degradation of the multilayer mirror due to, for example, interdiffusion between the adjacent layers, decreases the modulation of the standing wave field, thus rendering SWAPPS experiments much harder to perform. Here, we show that multilayer mirrors consisting of alternate B4C and W layers are stable at temperatures exceeding 600 °C and are, thus, promising candidates for future studies of surface and subsurface species in heterogeneous catalytic reactions using SWAPPS.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Critical Practices in Rigorously Assessing the Inherent Activity of Nanoparticle Electrocatalysts

Accurate activity measurements for electrocatalytic materials are the backbone of impactful electrocatalyst research. Inherent to accurate measurements is the correct accounting of the active catalyst sites and the design of experiments to ensure that these sites are participating in the measured reaction. Improvements in electrocatalyst activity for fuel cell reactions (hydrogen oxidation and oxygen reduction) are essential for the widespread adoption of this carbon-neutral energy source. Activities for these half-reactions can be reported in a variety of ways including specific activities, mass activities, volumetric activities, and half-wave potentials. These values are traditionally measured with a rotating disk electrode (RDE) in which the electrocatalyst is supported on a porous active layer film attached to the RDE. To accurately measure electrocatalyst activity with an RDE, two major sources of errors must be addressed: (i) accurate electrochemical surface area (ECSA) quantification and (ii) fast diffusion of reactants through the active layer film. In this contribution, we first aim to detail through RDE experiments and mass transport modeling the potential errors that can be observed when these factors are not properly addressed. We then present recommendations and discuss techniques for properly constructing electrocatalyst thin films and characterizing the ECSA on monometallic and novel alloy electrocatalysts. These practices, when adopted, ensure greater confidence and reliability in reports of electrocatalyst activity.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Impact of Varying the Photoanode/Catalyst Interfacial Composition on Solar Water Oxidation: The Case of BiVO 4 (010)/FeOOH Photoanodes

Photoanodes used in a water-splitting photoelectrochemical cell are almost always paired with an oxygen evolution catalyst (OEC) to efficiently utilize photon-generated holes for water oxidation because the surfaces of photoanodes are typically not catalytic for the water oxidation reaction. Suppressing electron–hole recombination at the photoanode/OEC interface is critical for the OEC to maximally utilize the holes reaching the interface for water oxidation. Here, in order to explicitly demonstrate and investigate how the detailed features of the photoanode/OEC interface affect interfacial charge transfer and photocurrent generation for water oxidation, we prepared two BiVO 4 (010)/FeOOH photoanodes with different Bi:V ratios at the outermost layer of the BiVO 4 interface (close to stoichiometric vs Bi-rich) while keeping all other factors in the bulk BiVO 4 and FeOOH layers identical. The resulting two photoanodes show striking differences in the photocurrent onset potential and photocurrent density for water oxidation. The ambient pressure X-ray photoelectron spectroscopy results show that these two BiVO 4 (010)/FeOOH photoanodes show drastically different Fe 2+ :Fe 3+ ratios in FeOOH both in the dark and under illumination with water, demonstrating the immense impact of the interfacial composition and structure on interfacial charge transfer. Using computational studies, we reveal the effect of the surface Bi:V ratio on the hydration of the BiVO 4 surface and bonding with the FeOOH layer, which in turn affect the band alignments between BiVO 4 and FeOOH. These results explain the atomic origin of the experimentally observed differences in electron and hole transfer and solar water oxidation performance of the two photoanodes having different interfacial compositions.

14 SOLAR ENERGY↗
DOE OSTI2021

Latent Representation Learning for Structural Characterization of Catalysts

Supervised machine learning-enabled mapping of the X-ray absorption near edge structure (XANES) spectra to local structural descriptors offers new methods for understanding the structure and function of working nanocatalysts. We briefly summarize a status of XANES analysis approaches by supervised machine learning methods. We present an example of an autoencoder-based, unsupervised machine learning approach for latent representation learning of XANES spectra. This new approach produces a lower-dimensional latent representation, which retains a spectrum–structure relationship that can be eventually mapped to physicochemical properties. Furthermore, the latent space of the autoencoder also provides a pathway to interpret the information content “hidden” in the X-ray absorption coefficient. Our approach (that we named latent space analysis of spectra, or LSAS) is demonstrated for the supported Pd nanoparticle catalyst studied during the formation of Pd hydride. By employing the low-dimensional representation of Pd K-edge XANES, the LSAS method was able to isolate the key factors responsible for the observed spectral changes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗