Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “high solid loading”

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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Understanding the Influence of Li 7 La 3 Zr 2 O 12 Nanofibers on Critical Current Density and Coulombic Efficiency in Composite Polymer Electrolytes

Composite polymer electrolytes (CPEs)are attractive materials for solid-state lithium metal batteries, owing to their high ionic conductivity from ceramic ionic conductors and flexibility from polymer components. As with all lithium metal batteries, however, CPEs face the challenge of dendrite formation and propagation. Not only does this lower the critical current density (CCD) before cell shorting, but the uncontrolled growth of lithium deposits may limit Coulombic efficiency (CE) by creating dead lithium. Here, we present a fundamental study on how the ceramic components of CPEs influence these characteristics. CPE membranes based on poly-(ethylene oxide) and lithium bis-(trifluoro methane sulfonyl)-imide(PEO-LiTFSI) with Li 7 La 3 Zr 2 O 12 (LLZO) nanofibers were fabricated with industrially relevant roll-to-roll manufacturing techniques. Galvanostatic cycling with lithium symmetric cells shows that the CCD can be tripled by including 50 wt % LLZO, but half-cell cycling reveals that this comes at the cost of CE. Varying the LLZO loading shows that even a small amount of LLZO drastically lowers the CE, from 88% at 0 wt % LLZO to 77% at just 2 wt % LLZO. Mesoscale modeling reveals that the increase in CCD cannot be explained by an increase in the macroscopic or microscopic stiffness of the electrolyte; only the microstructure of the LLZO nanofibers in the PEO-LiTFSI matrix slows dendrite growth by presenting physical barriers that the dendrites must push or grow around. This tortuous lithium growth mechanism around the LLZO is corroborated with mass spectrometry imaging. In conclusion, this work highlights important elements to consider in the design of CPEs for high-efficiency lithium metal batteries.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Palladium/Zeolite Low Temperature Passive NOx Adsorbers (PNA): Structure-Adsorption Property Relationships for Hydrothermally Aged PNA Materials

Zeolites with different framework structures (SSZ-13, ZSM-5, BEA) but similar Si/Al ratios and Pd loading (~1 wr%) were synthesized and evaluated as low temperature passive NOx adsorbers (PNA). These materials exhibit high NOx adsorption efficiency with atomically dispersed Pd the active adsorption site. Hydrothermal aging at 750 ºC for 16 hours in the presence of 10% water vapor in air resulted in the formation of PdO nanoparticles in all three samples as evidenced by high energy XRD. Hydrothermal aging of the small-pore Pd/SSZ-13 (Si/Al = 6), which contain ~100-90% atomically dispersed palladium ions, decreases its PNA performance only by ~10-20%. High-field solid state 27Al NMR studies on the fresh and aged samples substantiate dealumination and significant changes in the distribution of Al (and thus, Bronsted acid) sites after hydrothermal aging. FTIR measurements with NO probe molecule and titration of Bronsted acid sites with nitrosyl (NO+) ions further corroborate the 27Al NMR data. Because framework aluminum atoms are the anchoring sites for atomically dispersed Pd ions, their elution from the framework causes the loss of active atomically dispersed Pd species. Furthermore, with the aid of HAADF-STEM imaging and synchrotron XRD studies, we further confirm and visualize the fate of these Pd species – they agglomerate into PdO nanoparticles on the external surface of zeolite. Consequently, these changes lead to the decrease in PNA performance of these materials after hydrothermal aging. The thus formed agglomerates cannot be re-dispersed back to their ionic state due to the loss of framework Al T-sites and/or inherent stability of such large PdO particles. Our study clearly demonstrates that, unlike in previous studies that found increased PNA performance upon HTA , high temperatures hydrothermal aging of PNA materials that contain atomically dispersed Pd initially, results in a decrease in NOx storage efficiency due mostly to the formation of PdO agglomerates. However, we also highlight the high hydrothermal stability of predominantly atomically dispersed 1-3 wt% Pd/SSZ-13 (Si/Al = 6), whose performance decreases only marginally after prolonged hydrothermal aging at 750 ºC. Overall, this study shows that hydrothermally stable passive NOx materials can be prepared using small-pore SSZ-13 zeolite.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

High-performance all-solid-state Li–S batteries enabled by an all-electrochem-active prelithiated Si anode

The development of all-solid-state Li–S batteries has been greatly impeded by dendrite growth and dendrite penetration, which are both related to the Li metal anode. As a compromised alternative, we report lithium silicide (Li x Si) as a dendrite-free and high-capacity anode with Li source. Here, the as-synthesized Li x Si is relatively soft, highly electronically conductive, and with a high Li diffusivity. These distinctive properties make Li x Si anode viable as an “all-electrochem-active” electrode (consisted of 100 wt.% Li x Si). Compared with the typical composite electrode, the all-electrochem-active electrode not merely maximizes the electrode-level energy density but also minimizes the electrolyte-related interfacial degradation. Li x Si symmetric cell demonstrates a reversible cycling at 4 mA cm –2 for over 320 h. Stress change and morphological evolution of the Li x Si electrode are investigated upon dealloying/alloying. When paired with a S cathode (active mass loading of 3 mg cm –2 ), Li x Si–S full cell shows a good cycling behavior over 500 cycles and rate performance (69% capacity retained at 1.2C) even at 25 °C.

25 ENERGY STORAGE↗
DOE OSTI2024

A miniature jet pump for slurries

Here, this work reports on the design and performance of a jet pump system to pump slurries. The jet pump uses a high-velocity water jet to entrain and pump the slurry. This pumping action requires no valves and does not use moving solid parts. Hence, the system can be readily sealed, and the pumping action creates no observable mechanical damage to slurry particles. We tested carbon black beads suspended in water at concentrations of 0.0, 0.2, and 2.0 wt% at primary jet flow rates of 50 and 100 mL/min. Our measurements show an approximately linear decrease in entrainment ratio (slurry-to-jet flow rate) and absolute slurry flow rate with increasing adverse (load) pressure difference. The entrainment ratio is a strong function of the absolute jet flow rate. Pump performance was insensitive to slurry composition. Micron-resolution particle image velocimetry (micro-PIV) revealed qualitative changes in the pump’s internal flow due to changes in absolute jet flow rate, including evidence of the suppression of a vortex at higher flow rates. These findings establish key operational parameters for the jet pump and highlight its potential in a variety of slurry flow pumping applications. The pump is particularly suited for flow electrode slurries (e.g. as used for energy storage and water treatment) as it maintains particle integrity and can be used with standard tubing, tube fittings, and sealing methods.

42 ENGINEERING↗
DOE OSTI2020

Impact of Carbon Support Structure on the Durability of PtCo Electrocatalysts

High performing, low-Pt content fuel cell membrane electrode assemblies (MEAs) are critical to the economic viability of proton exchange membrane fuel cells (PEMFCs) for the transportation industry. Considerable research has been conducted to reduce the Pt content in fuel cells, leading to the development of transition metal alloys, such as Platinum-Cobalt (PtCo). The degree of degradation of PtCo catalysts can be impacted by the catalyst metal itself and its interactions with the carbon support. Several low-loaded PtCo MEAs were fabricated, with various combinations of porous and solid carbon cathode catalyst supports. The MEAs were subjected to an accelerated stress test (AST), and the catalyst degradation characterized using electrochemical, X-ray scattering, and electron microscopy techniques. Porous supports retain more of their electrochemically-active surface area (ECSA) and demonstrate higher performance after the AST. Overall, this is believed to be due to the ability of the porous supports to trap the metal particles within the pores, slowing their dissolution/precipitation, and agglomeration.

25 ENERGY STORAGE↗
DOE OSTI2023

Suppressing Universal Cathode Crossover in High‐Energy Lithium Metal Batteries via a Versatile Interlayer Design**

Abstract The universal cathode crossover such as chemical and oxygen has been significantly overlooked in lithium metal batteries using high‐energy cathodes which leads to severe capacity degradation and raises serious safety concerns. Herein, a versatile and thin (≈25 μm) interlayer composed of multifunctional active sites was developed to simultaneously regulate the Li deposition process and suppress the cathode crossover. The as‐induced dual‐gradient solid‐electrolyte interphase combined with abundant lithiophilic sites enable stable Li stripping/plating process even under high current density of 10 mA cm −2 . Moreover, X‐ray photoelectron spectroscopy and synchrotron X‐ray experiments revealed that N‐rich framework and CoZn dual active sites can effectively mitigate the undesired cathode crossover, hence significantly minimizing Li corrosion. Therefore, assembled lithium metal cells using various high‐energy cathode materials including LiNi 0.7 Mn 0.2 Co 0.1 O 2 , Li 1.2 Co 0.1 Mn 0.55 Ni 0.15 O 2 , and sulfur demonstrate significantly improved cycling stability with high cathode loading.

Xie, Chuyi↗
DOE OSTI2023

Suppressing Universal Cathode Crossover in High‐Energy Lithium Metal Batteries via a Versatile Interlayer Design**

Abstract The universal cathode crossover such as chemical and oxygen has been significantly overlooked in lithium metal batteries using high‐energy cathodes which leads to severe capacity degradation and raises serious safety concerns. Herein, a versatile and thin (≈25 μm) interlayer composed of multifunctional active sites was developed to simultaneously regulate the Li deposition process and suppress the cathode crossover. The as‐induced dual‐gradient solid‐electrolyte interphase combined with abundant lithiophilic sites enable stable Li stripping/plating process even under high current density of 10 mA cm −2 . Moreover, X‐ray photoelectron spectroscopy and synchrotron X‐ray experiments revealed that N‐rich framework and CoZn dual active sites can effectively mitigate the undesired cathode crossover, hence significantly minimizing Li corrosion. Therefore, assembled lithium metal cells using various high‐energy cathode materials including LiNi 0.7 Mn 0.2 Co 0.1 O 2 , Li 1.2 Co 0.1 Mn 0.55 Ni 0.15 O 2 , and sulfur demonstrate significantly improved cycling stability with high cathode loading.

25 ENERGY STORAGE↗
DOE OSTI2020

Shocking Cerium into a New Phase

A material's equation of state (EOS) - which indicates its state under various conditions of pressure, temperature, volume, and energy, and relates them to various phases - is a vital part of understanding the material’s behavior and response in different environments and applications. However, determining the EOS and complete phase diagrams of some materials can pose daunting challenges. One of these materials is cerium, for which some regions of the phase diagram have proven elusive. Investigators from the U.S. Department of Energy’s (DOE’s) Los Alamos National Laboratory (LANL) employed the U.S. DOE’s Advanced Photon Source (APS) to probe the high-pressure solid phase of cerium through shock-wave experiments that provided a detailed look at the cerium's transition from the α-ϵ phases. Their work, which provides the first evidence that an α-ϵ phase transition can be shock-induced in cerium, was published in the Journal of Applied Physics.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Engineering of Self‐Aggregation‐Resistant MnO 2 Heterostructure with A Built‐in Field for Enhanced High‐Mass‐Loading Energy Storage

Abstract Although MnO 2 has been intensively investigated for energy storage, further applications are limited by van der Waals force‐triggered self‐aggregation that always leads to poorly exposed active sites and compromised reaction dynamics, especially under high‐mass‐loading conditions. Herein, by synergistically coupling interfacial modulation with the Kirkendall effect, this work achieves in situ topological structure reorganization of MnOOH toward the high‐aspect‐ratio MnO 2 heterostructure (Heter‐MnO 2 ) with fully exposed active sites, which is ready to assemble into self‐supporting high‐mass‐loading film (30 mg cm −2 ) with restrained self‐aggregation. Theoretical calculation and dynamics analysis results demonstrate the generation of the built‐in field within the heterostructure, thus enhancing the electronic‐transfer and ionic‐adsorption/transport rates. As such, the 30 mg cm −2 Heter‐MnO 2 electrode achieves a superior areal capacitance of 4762 mF cm −2 at 1 mA cm −2 and a sound rate performance (79% at 100 mA cm −2 ) comparable to those of low‐mass‐loading/thin‐film electrodes. As a proof of concept, the fabricated planar interdigital quasi‐solid‐state symmetric micro‐supercapacitor (MSC) based on the Heter‐MnO 2 electrode can deliver a remarkable areal capacitance of 181 mF cm −2 and a considerable volumetric energy density of 10.3 mWh cm −3 . This methodology highlights the promise of surface/interface chemistry modulation for the configuration of easy‐to‐integrate hierarchical nanostructures to better meet practical energy applications.

Chemistry↗
DOE OSTI2020

Enhanced activity and stability of MgO-promoted Ni/Al 2 O 3 catalyst for dry reforming of methane: Role of MgO

Highly dispersed nickel nanoparticles (NPs) were deposited on Al 2 O 3 NPs by atomic layer deposition (ALD). Various amounts of MgO were loaded on Ni/Al 2 O 3 catalysts by the incipient wetness (IW) method for dry reforming of methane. Fresh and used catalysts were systematically characterized to reveal the effects of MgO on the catalytic performance. MgO was found to increase the basic amount and basic strength of catalyst surfaces, which provided additional surface oxygen species and assisted the adsorption and activation of CO 2 . Although the formation of NiO-MgO solid solution, during the calcination of incipient wetness, did not improve the overall reducibility, the Ni NPs from NiO-MgO solid solution after reduction formed an intimate interaction with MgO, which could inhibit Ni NPs from sintering and provide sufficient metal-support interface for CO 2 activation. The MgO-promoted Ni/Al 2 O 3 reached a methane reforming rate of 1780 L CH4 g Ni -1 h -1 at 850 °C, which is 26% more than that of the pristine Ni/Al 2 O 3 . Finally, the higher CO 2 activity enhanced the oxidation rate of the surface carbon generated from side-reactions, thereby resulting in a higher reforming rate and inhibiting coke formation, especially the detrimental graphitic encapsulating carbon on the active nickel surface.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Experimental shock Hugoniot and melt boundary temperatures of aluminum

Aluminum is a ubiquitous component in dynamic compression, pulsed power, and other high energy density physics studies. Its high-pressure behavior and phase diagram are extensively studied standards in shockwave physics. While theoretical calculations and multiphase equations of state have been benchmarked to velocity measurements of loading and unloading waves, pressure and density under shock, and other mechanical data, experimental temperature data under these conditions have not been reported. We conducted a series of experiments shocking and releasing aluminum 6061 and 1100 samples into lithium fluoride windows. We measured temperature at the sample–window interface under steady compression and subsequent isentropic release. These results allow us to constrain the temperature of the solid Hugoniot and the boundary between the liquid and face-centered cubic solid phases.

Hartsfield, Thomas Murray [Sandia National Laborat↗
DOE OSTI2022

Material Plasma Exposure eXperiment (MPEX) High Heat Flux Bellows for Component Alignment in High Microwave Environment

The Material Plasma Exposure eXperiment (MPEX) device is a linear plasma device developed to perform plasma material interaction experiments under the conditions prototypic of a fusion reactor divertor. MPEX has multiple systems that must be precisely aligned to the plasma axis, including an electron cyclotron heating system that emits up to 400 kW of microwave power into the vacuum vessel. Five distinct systems require precise alignment on the MPEX device, thus requiring four bellows, all of which are adjacent to the plasma at a relatively high heat flux of approximately 47 kW/m2 and microwave power regions. The MPEX high heat flux bellows (HHFB) is designed to deliver 6 degrees of freedom positioning. The HHFB includes titanium–zirconium–molybdenum (TZM) inserts that are brazed into a Glidcop AL-15 body using a high-temperature braze alloy, thus blocking direct line of sight to an edge-welded bellows from the plasma and microwave screen to block microwaves from the bellows. Further, a custom ConFlat knife edge is machined into the Glidcop AL-15, so the vacuum flanges do not need a braze or weld joint on the vacuum interface. Fingerstock or copper mesh is used to restrict microwave power from entering the interstitial space between the water-cooled Glidcop AL-15 body and edge-welded bellows. Glidcop AL-15 was selected as the material of choice for the water-cooled body because it can maintain mechanical integrity at elevated temperatures, and it also allows for a high-temperature braze. TZM was selected for its machinability and compatibility with vacuum and plasma requirements. A test article is also being considered for a similarly shaped component, the MPEX limiter, that will demonstrate the integrity of the braze joint under high thermal load. Results from this testing will be extrapolated to deduce the lifetime and integrity of the HHFB design.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Fabrication, oxidation, and combustion of nanoscale magnesium diboride and tetraboride

The difficult ignition and low combustion efficiency of boron particles decrease the performance of boron-loaded, fuel-rich propellants for solid fuel ramjets and ducted rockets. One approach to solving this problem involves the use of magnesium diboride (MgB 2 ), which ignites easier than boron. Magnesium tetraboride (MgB 4 ) offers greater energy density owing to its higher boron content. However, the effect of B/Mg ratio on the ignition and combustion is unknown. Additionally, while nanoscale MgB₂ particles and quasi-2D structures are promising energetic additives, the oxidation and combustion properties of nanoscale MgB₄ have not been explored. To address these knowledge gaps, the present work included synthesis and high-energy ball milling of MgB 2 and MgB 4 powders, thermogravimetric analysis (TGA) of their oxidation, and combustion experiments with thin layers of the obtained powders. Comparison of two synthesis routes (a solid-state reaction in a tube furnace and combustion synthesis) has shown that the former is the superior method for producing magnesium borides. TGA has revealed that oxidation of both MgB 2 and MgB 4 results in a high conversion into the oxides (88–91 %), far exceeding the low conversion of boron (62.5 %). MgB 4 begins to oxidize rapidly at a much lower temperature (∼900 °C) than MgB 2 (∼1200 °C). The burning rates of milled MgB 2 and MgB 4 are about eight and five times, respectively, faster than that of submicron boron. Magnesium borides exhibit a stable, sustained boron flame, needed for high combustion efficiency, whereas physical Mg/B mixtures undergo Mg-driven "flash" combustion.

Boron↗
DOE OSTI2022

Evaluation of the energy balance of chemical looping combustion of solid fuels using CuO-based oxygen carriers

Chemical looping combustion (CLC) can inherently capture the CO2 generated during thermal energy production. When using CuO-based oxygen carriers, the reactions in both the fuel reactor (FR) and air reactor (AR) are exothermic, facilitating autothermal operation. To prevent the oxygen carrier from becoming agglomerated while ensuring a sufficient oxygen release rate, it is necessary to predict the FR temperature and remove an appropriate amount of heat from the AR when burning high-rank fuels. Here in this work, the energy balance of a CuO-based CLC system firing various solid fuels (biomass, coal, and petroleum coke) was investigated by establishing a thermodynamic model using FactSage. The energy distribution characteristics of the FR were quantified based on a virtual five-step thermodynamic sequence. The effects of fuel properties, oxygen carrier properties, and operating conditions on the energy balance were evaluated systematically. The results indicate that the temperature difference between the FR and AR is not very sensitive to the heating value of the fuel, while the CuO loading of the oxygen carrier, the heat capacity of the support material, and the flow rate of the fluidizing gas have significant impact. The insights obtained in this work will help improve the design and heat management of CLC.

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

Modes of strain accommodation in Cu-Nb multilayered thin film on indentation and cyclic shear

Two-phase layered thin films with a high density of semi-coherent interfaces exhibit excellent mechanical properties and thermal stability. Here, in this study, a magnetron-sputtered Cu-Nb dual-layered thin film (~500 nm for Cu and ~150 nm for Nb) having an amorphous interface between Cu and Nb with a high density of aligned growth twins in Cu is subjected to severe surface deformation. The material is loaded using indentation and cyclic shear under tribological testing. The strain accommodation in the subsurface microstructure after deformation varies based on the local structure and deformation mode. Grain refinement and crack formations in the stressed region of the Nb layer and localized crystallization of the amorphous interface are observed after indentation and scratch testing. Pronounced detwinning of growth twins in the Cu layer under the cyclic shear strain leaves large dislocations sites and loops which are observed both by high-resolution transmission electron microscopy and experiment-guided molecular dynamic (MD) simulations. Our simulations provided insights into understanding the pathway for the detwinning process under cyclic shear loading.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Linking Perfluorosulfonic Acid Ionomer Chemistry and High-Current Density Performance in Fuel-Cell Electrodes

Transport phenomena are key in controlling the performance of electrochemical energy-conversion technologies and can be highly complex, involving multiple length scales and materials/phases. Material designs optimized for one reactant species transport however may inhibit other transport processes. We explore such trade-offs in the context of polymer-electrolyte fuel-cell electrodes, where ionomer thin films provide the necessary proton conductivity but retard oxygen transport to the Pt reaction site and cause interfacial resistance due to sulfonate/Pt interactions. We examine the electrode overall gas-transport resistance and its components as a function of ionomer content and chemistry. Low-equivalent-weight ionomers allow better dissolved-gas and proton transport due to greater water uptake and low crystallinity but also cause significant interfacial resistance due to the high density of sulfonic acid groups. These effects of equivalent weight are also observed via in situ ionic conductivity and CO displacement measurements. Of critical importance, the results are supported by ex situ ellipsometry and X-ray scattering of model thin-film systems, thereby providing direct linkages and applicability of model studies to probe complex heterogeneous structures. Structural and resultant performance changes in the electrode are shown to occur above a threshold sulfonic-group loading, highlighting the significance of ink-based interactions. Furthermore, our findings and methodologies are applicable to a variety of solid-state energy-conversion devices and material designs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

The impact of residual solvent on catholyte performance in solid-state batteries

All-solid-state batteries (ASSBs) are attractive due to their safety, use of the Li metal anode, high energy density, and innovative processing routes. However, high interfacial resistance, especially on the cathode side, is one of the major challenges for commercialization of ASSBs. Catholyte, either a liquid or solid, is added to lower the cathode/electrolyte interfacial resistance. In this study, we find that residual N-methylpyrrolidone solvent remaining in a PVDF/LiTFSI solid polymer catholyte after incomplete drying can dramatically lower the interfacial resistance between the Li 6.25 Al 0.25 La 3 Zr 2 O 12 electrolyte and LiNi 1/3 Mn 1/3 Co 1/3 O 2 cathode-active material. Cells with varying amounts of residual solvent are compared to optimize the residual solvent loading. With moderate residual solvent, the discharge capacity reaches 142 mA h g -1 when cycled at 25 °C and 0.5 C and the capacity retention is 60.5% after 125 cycles. Discharge capacity retention is improved at -10 °C. The conductivity of the free-standing PVDF/LiTFSI film, mimicking the PVDF/LiTFSI catholyte, verifies the role of residual NMP in the cathode. Finally, this study demonstrates the possibility of widely differing results for ASSBs when the cathode is not completely dried. Also, it provides a hint for a potential method to lower the cathode/electrolyte interfacial resistance.

25 ENERGY STORAGE↗
DOE OSTI2023

Modular machine learning-based elastoplasticity: Generalization in the context of limited data

The development of highly accurate constitutive models for materials that undergo path-dependent processes continues to be a complex challenge in computational solid mechanics. Challenges arise both in considering the appropriate model assumptions and from the viewpoint of data availability, verification, and validation. Recently, data-driven modeling approaches have been proposed that aim to establish stress-evolution laws that avoid user-chosen functional forms by relying on machine learning representations and algorithms. However, these approaches not only require a significant amount of data but also need data that probes the full stress space with a variety of complex loading paths. Furthermore, they rarely enforce all necessary thermodynamic principles as hard constraints. Hence, they are in particular not suitable for low-data or limited-data regimes, where the first arises from the cost of obtaining the data and the latter from the experimental limitations of obtaining labeled data, which is commonly the case in engineering applications. In this work, we discuss a hybrid framework that can work on a variable amount of data by relying on the modularity of the elastoplasticity formulation where each component of the model can be chosen to be either a classical phenomenological or a data-driven model depending on the amount of available information and the complexity of the response. The method is tested on synthetic uniaxial data coming from simulations as well as cyclic experimental data for structural materials. The discovered material models are found to not only interpolate well but also allow for accurate extrapolation in a thermodynamically consistent manner far outside the domain of the training data. This ability to extrapolate from limited data was the main reason for the early and continued success of phenomenological models and the main shortcoming in machine learning-enabled constitutive modeling approaches. Training aspects and details of the implementation of these models into Finite Element simulations are discussed and analyzed.

42 ENGINEERING↗