Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Lu”

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 1,189 records · Page 66
All sourcesNASA NTRSDOE OSTI
NASA NTRS2000

STS-106 Crew Activity Report/Flight Day 8 Highlights

On this eighth day of the STS-106 Atlantis mission, the flight crew, Commander Terrence W. Wilcutt, Pilot Scott T. Altman, and Mission Specialists Daniel C. Burbank, Edward T. Lu, Richard A. Mastracchio, Yuri Ivanovich Malenchenko, and Boris V. Morukov move into the second half of preparing the International Space Station (ISS) for its first resident crew. Lu and Malenchenko are seen installing the power converters in the Zvezda module and components of the primary oxygen generation system. Mastracchio and Wilcutt moves supplies and logistics from the payload of Atlantis to the ISS. Wilcutt and Altman participate in several interviews and the crew wishes the Olympiads in Sydney good luck in their endeavors. Scenes also include external views of the ISS and images of Earth, including Sydney, Australia.

Source record↗
NASA NTRS2001

STS-106 Mission Highlights Resource Tape

An overview of the STS-106 mission is given through footage of the activities of the first flight days of the mission, starting with flight day one and ending with flight day eight. The crewmembers of Atlantis, Commander Terrence Wilcutt, Pilot Scott Altman, and Mission Specialists Daniel Burbank, Edward Lu, Richard Mastracchio, Yuri Malenchenko, and Boris Morukov, are seen during various prelaunch activities, such as during the ceremonial breakfast, suit up, and boarding Atlantis. The launch is seen, as are the rendezvous and docking of the Orbiter to the International Space Station (ISS) and the spacewalks performed on flight day three by Lu and Malenchenko. Activities for flight days 9-12 can be seen on 'STS-106 Mission Highlights Resource Tape, Part 2 of 2' (internal ID 2001072040).

Source record↗
NASA NTRS2008

Direct Solve of Electrically Large Integral Equations for Problem Sizes to 1M Unknowns

Matrix methods for solving integral equations via direct solve LU factorization are presently limited to weeks to months of very expensive supercomputer time for problems sizes of several hundred thousand unknowns. This report presents matrix LU factor solutions for electromagnetic scattering problems for problem sizes to one million unknowns with thousands of right hand sides that run in mere days on PC level hardware. This EM solution is accomplished by utilizing the numerical low rank nature of spatially blocked unknowns using the Adaptive Cross Approximation for compressing the rank deficient blocks of the system Z matrix, the L and U factors, the right hand side forcing function and the final current solution. This compressed matrix solution is applied to a frequency domain EM solution of Maxwell's equations using standard Method of Moments approach. Compressed matrix storage and operations count leads to orders of magnitude reduction in memory and run time.

Shaeffer, John↗
NASA NTRS2009

Anaerobic Psychrophiles from Lake Zub and Lake Untersee, Antarctica

The study of samples from Antarctica 2008 and 2009 expeditions organized and successfully conducted by Richard Hoover led to the isolation of diverse anaerobic strains with psychrotolerant and psychrophilic physiology. Due to the fact that Lake Untersee has never been subject to microbiological study, this work with the samples has significant and pioneering impact to the knowledge about the biology of this unique ecosystem. Also, the astrobiological significance for the study of these ecosystems is based on new findings of ice covered water systems on other bodies of our solar system. Anaerobic psychrotolerant strain LZ-22 was isolated from a frozen sample of green moss with soils around the rhizosphere collected near Lake Zub in Antarctica. Morphology of strain LZ-22 was observed to be motile, rod shaped and spore-forming cells with sizes 1 x 5-10 micron. This new isolate is a mesophile with the maximum temperature of growth at 40C. Strain LZ-22 is able to live on media without NaCl and in media with up to 7% (w/v) NaCl. It is catalase negative and grows only on sugars with the best growth rate being on lactose. The strain is a neutrophile and grows between pH 5 and 9.0 with the optimum at 7.8. Another two strains UL7-96mG and LU-96m7P were isolated from deep water samples of Lake Untersee. Proteolytic strain LU-96m7P had a truly psychrophilic nature and refused to grow at room temperature. Sugarlytic strain UL7-96mG was found to be psychrotolerant, but its rate of growth at 3C was very high compared with other mesophiles. Two homoacetogenic psychrophilic strains A7AC-96m and AC-DS7 were isolated and purified from samples of Lake Untersee; both of them are able to grow chemolithotrophically on H2+CO2. In the presence of lactate, these strains are able to grow only at 0-18C, and growth at 22C was observed only with yeast extract stimulation. In this paper, physiological and morphological characteristics of novel psychrophilic and psychrotolerant isolates from Antarctica 2008 and 2009 expeditions will be discussed.

Townsend, Alisa↗
NASA NTRS2002

Higher Order, Hybrid BEM/FEM Methods Applied to Antenna Modeling

In this presentation, the authors address topics relevant to higher order modeling using hybrid BEM/FEM formulations. The first of these is the limitation on convergence rates imposed by geometric modeling errors in the analysis of scattering by a dielectric sphere. The second topic is the application of an Incomplete LU Threshold (ILUT) preconditioner to solve the linear system resulting from the BEM/FEM formulation. The final tOpic is the application of the higher order BEM/FEM formulation to antenna modeling problems. The authors have previously presented work on the benefits of higher order modeling. To achieve these benefits, special attention is required in the integration of singular and near-singular terms arising in the surface integral equation. Several methods for handling these terms have been presented. It is also well known that achieving ~he high rates of convergence afforded by higher order bases may als'o require the employment of higher order geometry models. A number of publications have described the use of quadratic elements to model curved surfaces. The authors have shown in an EFIE formulation, applied to scattering by a PEC .sphere, that quadratic order elements may be insufficient to prevent the domination of modeling errors. In fact, on a PEC sphere with radius r = 0.58 Lambda(sub 0), a quartic order geometry representation was required to obtain a convergence benefi.t from quadratic bases when compared to the convergence rate achieved with linear bases. Initial trials indicate that, for a dielectric sphere of the same radius, - requirements on the geometry model are not as severe as for the PEC sphere. The authors will present convergence results for higher order bases as a function of the geometry model order in the hybrid BEM/FEM formulation applied to dielectric spheres. It is well known that the system matrix resulting from the hybrid BEM/FEM formulation is ill -conditioned. For many real applications, a good preconditioner is required to obtain usable convergence from an iterative solver. The authors have examined the use of an Incomplete LU Threshold (ILUT) preconditioner . to solver linear systems stemming from higher order BEM/FEM formulations in 2D scattering problems. Although the resulting preconditioner provided aD excellent approximation to the system inverse, its size in terms of non-zero entries represented only a modest improvement when compared with the fill-in associated with a sparse direct solver. Furthermore, the fill-in of the preconditioner could not be substantially reduced without the occurrence of instabilities. In addition to the results for these 2D problems, the authors will present iterative solution data from the application of the ILUT preconditioner to 3D problems.

Fink, P. W.↗
NASA NTRS2014

Sm-Nd and Rb-Sr Isotopic Systematics of a Heavily Shocked Martian Meteorite Tissint and Petrogenesis of Depleted Shergottites

Tissint is a very fresh Martian meteorite that fell near the town of Tissint in Morocco on July 18, 2011. It contains abundant olivine megacrysts (~23%) in a fine-grained matrix of pyroxene (~55%), maskelynitized plagioclase (~15%), opaques (~4%) and melt pockets (~3%) and is petrographically similar to lithologies A and C of picritic shergottite EETA 79001 [1,2]. The presence of 2 types of shock-induced glasses and all 7 high-pressure mineral phases that were ever found in melt pockets of Martian meteorites suggests it underwent an intensive shock metamorphism of ~25 GPa and ~2000 C localized in melt pockets [2]. Mineral textures suggest that olivines, pyroxenes and plagioclases probably did not experience such hightemperature. Earlier determinations of its age yielded 596+/-23 Ma [3] and 616+/-67 Ma [4], respectively, for the Sm-Nd system and 583+/-86 Ma for the Lu-Hf system [4], in agreement with the 575+/-18 Ma age of the oldest olivine-phyric depleted shergottite Dho 019 [5]. However, the exposure ages of Tissint (~1 Ma [1, 6, 7]) and Dho 019 (~20 Ma [8]) are very different requiring two separate ejection events. These previously determined Sm-Nd and Lu-Hf ages are older than the Ar-Ar maskelynite plateau age of 524+/-15 Ma [9], reversing the pattern usually observed for Martian meteorites. In order to clarify these age issues and place models for Tissint's petrogenesis on a firm basis, we present new Rb-Sr and Sm- Nd isotopic results for Tissint, and discuss (a) the shock effects on them and the Ar-Ar chronometer, (b) correlation of the determined ages with those of other depleted shergottites, and (c) the petrogenesis of depleted shergottites. Since the meteorite is a recent fall, terrestrial contamination is expected to be minimal, but, the strong shock metamorphism might be expected to compromise the equilibrium of the isotopic systems.

Shih, C.-Y.↗
NASA NTRS2021

Thermodynamics of the Lu2O3 – SiO2 system and comparison to other Rare Earth Silicates

Environmental barrier coatings are necessary to protect SiC based ceramics and composites from water vapor degradation in harsh engine environments. Currently, rare earth (RE) silicates are the most promising systems to protect SiC based ceramics and composites. This protection is largely based on reduced silica activity in these rare earth silicates which results in a lowered reactivity with water vapor. To that end, previous Knudsen effusion mass spectrometry (KEMS) studies have explored RE = Yb, Y silicates to measure the reduced silica activity and subsequent water vapor reactivity. Similarly, this work employs the KEMS technique to measure the SiO(g) vapor pressure in Lu containing RE silicates to calculate the activity of silica within the monosilicate [log(a(SiO2)) = -2351.1*1/T-1.6731] and the disilicate (log(a(SiO2)) = -4884.0*1/T + 2.2208) as a function of temperature. The enthalpies of formation for Lu monosilicate from the oxides and the elements were calculated to be -45±3 kJ/mol at 1550 K and -2831.1±12 kJ/mol at 298 K, respectively. The measured enthalpy of formation and those found in literature are compared to modeled values from density functional theory and those estimated using electronegativity.

Knudesen Effusion Mass Spectrometry↗
NASA NTRSDate not supplied

Advancing Dust Tolerant Mechanisms for a Sustained Exploration of the Moon

Introduction: “I think dust is probably one of our greatest inhibitors to a nominal operation on the Moon. I think we can overcome other physio-logical or physical or mechanical problems except dust.”– Gene Cernan, Apollo 17 Technical Debrief The Apollo missions revealed the impact of lu-nar dust on mechanisms. Lunar dust particles are jagged and electrostatically charged, giving them the ability to bind or damage mechanisms and alter thermal properties. Reports documented clogged equipment and jammed mechanisms in every mission, regardless of surface duration, as well as clogged mechanisms in the Extravehicular Mobility Suit (EMS), including zippers, wrist and hose locks, faceplates, and sunshades [1-2]. Several astronauts remarked they could not have sustained surface activity much longer because clogged joints would have frozen up completely [2]. Effective dust mitigation strategies are need-ed to support longer duration stays on the lunar surface [3-4]. State of the Art: Technology for mechanisms able to operate in dusty enviroments is advancing rapidly due to the needs of both Mars rovers and the Artemis program. Vacuum-tight connectors are essential for spacesuits and habitats, and their performance can be dependent on cleaning technologies, which have proven difficult on the lunar surface. Several TRL 3-5 technologies are undergoing tests with the expectation to reach TRL 6 within 1-2 years. Some mechanisms will be infused and tested on the VIPER (Volatiles Inves-tigating Polar Exploration Rover) mission planned for mid-2020s. NASA Funded Efforts: NASA has recognized the need for dust tolerant mechanisms, and has partnered with industry to advance the state-of-the-art. At NASA GRC, KSC, and JSC, the Dust Tolerant Mechanisms Project is working to devel-op advanced actuator seals for rotary joints and rotary bearing technologies for long-term sus-tained operation in lunar dust environments. An-other NASA project at NASA GRC, partnered with GSFC, JPL, and KSC is Motors for Dusty & Ex-treme Cold Environments (MDECE). MDECE is developing an unheated magnetically-geared mo-tor that can operate continuously for a long dura-tion at an ambient temperature of -243 ºC (33 K). NASA GRC has the capability to characterize the effects of dust on seals, mechanisms, and other mating surfaces and components under lunar conditions [5]. Through the SBIR/STTR program, NASA has funded several companies to advance dust toler-ant mechanisms via the Dust Tolerant Mecha-nisms sub-topic with applications in surface mobil-ity, spacesuits, connectors, joints, and more. LSIC and Community Efforts: The Lunar Surface Innovation Consortium (LSIC) Dust Miti-gation focus group has fostered collaborations across NASA, industry, and academia to develop solutions that minimizes the impact of lunar dust on robotic and human systems. Community ef-forts have included topical meetings on dust tol-erant mechanisms, featured technology presenta-tions, and feedback to NASA on potential gaps and needs. Testing: In 2021, NASA released NASA-STD-1008 [6]. This NASA Technical Standard estab-lishes minimum requirements and provides guid-ance for testing systems and hardware to be ex-posed to dust in planetary environments. The standard has specific sections dedicated to Mechanisms Testing (e.g. bearings, gears) as well as Seals and Mating Surfaces Testing (e.g. hatches, docking systems). Gaps and Needs: NASA is tracking dust tol-erant mechanisms as a gap in a cross-directorate analysis of capability areas needed to enable fu-ture human space-flight architectures. Two high-priority gap areas include additional facilities for testing mechanisms in lunar-surface conditions, and a better understanding of vulnerabilities to the smallest, nanometer-scale dust particles. Conclusion: Understanding and mitigating lu-nar dust is critical to successful, sustained opera-tions on the lunar surface – whether autonomous or otherwise. This presentation will discuss both the state-of-the-art and open needs in lunar dust tolerant mechanisms, technology impacts, mitiga-tion approaches, testing, LSIC community efforts, and more. References: [1] Gaier, J. R. (2020). The Im-pact of Dust on Lunar Surface Equipment During Apollo. Lunar Dust 2020. [2] GRC, & Gaier, J. R. (2005). The Effects of Lunar Dust on EVA Sys-tems During the Apollo Missions. [3] Johansen, M. R. (2020). An Update on NASA’s Lunar Dust Mitigation Strategy. Lunar Dust 2020. [4] ASI, CSA, ESA, JAXA, & NASA. (2016). Dust Mitiga-tion Gap Assessment Report. [5] Jimenez, N. et al (2022), LPSC Abstract 2572. [6] NASA-STD-1008, 2021.

J I Nunez↗
NASA NTRSDate not supplied

Evaluation of Human Spaceflight-Related Tissue Weight Relief Using Whole Body Finite Element Model Simulations

Tissue Weight Relief (TWR) is a physiological condition observed in human spaceflight. It not only impacts the injury biomechanics of soft tissue but also the physiological responses of the cardiovascular system both due to fluid redistribution and the effect of tissue-related transmural pressure on the large venous blood vessels. Understanding the effects of tissue weight relief is especially important because of the role it may play in understanding the cause of Space Associated Neuro-Ocular Syndrome (SANS). SANS can be characterized by a number of ocular changes which reduce visual acuity and SANS related symptoms occur in up to 51% of astronauts. A prevailing theory for the causation of SANS is that of headward (cephalad) fluid shift and a prolonged increase of Intracranial Pressure (ICP) similar to intracranial hypertension, which is not fully supported by the experimental data or astronaut symptom reporting. However, it is still believed that SANS is caused by a pressure change in the eye and the surrounding tissues. It has been proposed that TWR plays a substantial role in affecting internal pressures and fluid shifts in microgravity. In this effort, two whole-body Finite Element (FE) models – Elemance and THUMS – are used to ascertain the microgravity-associated TWR of the musculature surrounding the lower body veins. Elemance and THUMS are physics-based computational models that have been validated and verified for several automotive and domestic applications, and as such, can simulate the relief of soft tissue weight due to changes in the gravitation vector. Specifically, the current effort modeled the transition of the gravitational vector from 1G to 0G, applied across the whole-body model in a supine position. For each 1G to 0G transition simulation, the lower body vein’s transmural pressure-time profile was extracted and averaged around the anterior portion of the thigh muscle. The ascertained transmural pressure changes from 1G to 0G transition are given in Figure 1 for the Elemance and the THUMS FE models. The transmural pressure changes of 10 mmHg and 21 mmHg are in the same order of magnitude as Lu’s value of 44 mmHg. It is to be noted that Lu implemented a 0D to 1D lumped parameter model and the Elemance and THUMS are 3D higher order computational models. This proof-of-concept approach demonstrates that TWR pressure can be adequately estimated with in silico techniques however, further in silico investigations need to be conducted to address the unique contributions to the transmural pressure from each of the computational models.

Finite element modeling↗
NASA NTRSDate not supplied

Application of High-Spatial Resolution LA-MC-ICP-MS in Planetary Materials

High-precision stable and radiogenic isotopic measurements of planetary materials can help elucidate processes taking place during early Solar System formation, as well as during subsequent planetary differentiation. Laser Ablation Multi Collector Inductively Coupled Plasma Mass Spectrometry (LA-MC-ICP-MS) adds the valuable ability of preserving the spatial context of these isotopic measurements, by sampling small sample volumes (~10 – 60 um diameter) using a pulsed laser system. In the Center for Isotope Cosmochemistry and Geochronology (CICG) at NASA Johnson Space Center, these in-situ isotopic measurements are primarily made using the Applied SpectraTM iX-fs-Tandem LA-LIBS Instrument coupled to the Nu InstrumentsTM SP1700 MC-ICP-MS. The instrumental set-up also provides simultaneous Laser Induced Breakdown Spectroscopy (LIBS) measurements and the potential use of a low energy path with a collision/reaction cell for additional interference removal, though we focus here on the typical high-energy LA-MC-ICP-MS operation. We will provide examples of stable (e.g., Mg) and radiogenic (e.g., Lu-Hf) isotopic systems measured in planetary materials and their terrestrial analogs. The measurement of mass dependent Mg isotopic variations is an important tool for reconstructing processes such as evaporation and condensation of solids in the solar nebula. For early Solar System samples, Mg isotope measurements are also important for the short-lived (t1/2 = 0.7 Ma) Al-Mg chronometer. During laser ablation analyses, the measurement of the Al-Mg isotopic system is made in medium mass resolution (RP ~9500 at ~20% transmission) to limit interferences from 48Ti2+ and 48Ca2+ on 24Mg+, in addition to 52Cr2+ and 12C14N+ on 26Mg+. Measured in the mineral zircon, the 176Lu-176Hf decay system is often used for understanding the growth of continents on Earth, and is also a vital aid in deciphering early large-scale planetary differentiation processes, such as the crystallization of the lunar magma ocean, which is thought to be responsible for much of the Moon’s crust. Lu-Hf measurements are made in low resolution mode (RP <2000), with 10 ion masses (from 171Yb to 180Hf) measured simultaneously to facilitate corrections for isobaric interferences from 176Yb and 176Lu on 176Hf.

Jacob B Setera↗
DOE OSTI2023

Studying the hadron structure with PANDA and CLAS using machine learning techniques

The hadron spectroscopy and structure are currently very active fields of research to study the non-perturbative regime of quantum chronodynamics. The first one studies the complex structure of excited hadrons by looking at their decay products, while the latter uses lepton scattering on nucleons. Both methods require reconstruction algorithms with great efficiency and good particle identification and background rejection rates. This work aims to provide these by either improving the existing methods or developing new ones. The first part of this document presents a feasibility study of a predicted hybrid charmonium state for the PANDA experiment. Lattice QCD calculations predict the ground state hybrid charmonium to be a spin exotic with quantum numbers of JP C = 1?+ at a mass of around 4.3 GeV with a width to be around 20 MeV. A machine learning based data analysis scheme is proposed to further improve the signal efficiency and the background reduction, alongside with improvements of the analysis software (PandaRoot), that are vital for this study. These improvements include a reworked clustering algorithm for the electromagnetic calorimeter (EMC) and an optimized monte carlo matching for neutral particles. The second part of this document is about studying the proton structure. A multidimensional study of the structure function ratio Fsin(?)LU /FUU has been performed for K±, based on the measurement of beam-spin asymmetries. It uses the high statistics data recorded with the CLAS12 spectrometer at Jefferson Laboratory. Fsin(?)LU is a twist-3 quantity that provides information about the quark gluon correlations in the proton. This document will present for the first time a simultaneous analysis of two kaon channels over a large kinematic range of z, xB , PT and Q2 with virtualities Q2 ranging from 1 GeV2 up to 8 GeV2 using machine learning techniques for improved particle identification.

Kripko, Aron↗
DOE OSTI2023

Measuring the Beam-Spin Asymmetry of Hard Exclusive ¿0 Production off the Neutron with CLAS12

Over the course of the last couple of decades, Generalised Parton Distributions (GPDs) have become a compelling area of focus in both theoretical and experimental research in the ?eld of hadronic structure. GPDs can be indirectly accessed via observables in hard exclusive reactions such as Deeply Virtual Compton Scattering (DVCS) and Deeply Virtual Meson Production (DVMP). In these reactions, an incident lepton scatters o? an individual quark within the target nucleon via the exchange of a virtual photon, and a real photon (DVCS) or a meson (DVMP) are emitted as a result. This thesis presents an analysis of Deeply Virtual ?0 Meson Production o? the neutron in the deuteron (nDV?0P). The data were taken with the CLAS12 detector, which is housed in the experimental Hall B at the Je?erson National Laboratory (JLab) in Virginia, USA. A longitudinally polarised electron beam delivered by JLab?s Continuous Electron Beam accelerator Facility (CEBAF), at an energy of 10.2 and 10.6 GeV, was incident on a liquid deuterium target housed within the CLAS12 detector assembly. The nDV?0P channel is sensitive to the lesser-studied chiral-odd transversity GPDs. GPDs are accessed indirectly via observables related to the cross-sections of hard exclusive processes. One such example is the beam-spin asymmetry (?LU) as a function of Trento-?, the angle between the leptonic plane and the hadronic plane which are de?ned with the beam and scattered electron, and with the recoiling neutron and reconstructed ?0, respectively. Despite the limited statistics, the observable ?LU was successfully extracted for eight kinematic bins (four in Mandelstam ?, and two in ??). This is a ?rst measurement for nDV?0P in the phase-space which is now available with CLAS12 in the 12 GeV era of JLab.

Naidoo, Paul↗
DOE OSTI2022

A comprehensive and fair comparison of two neural operators (with practical extensions) based on $\mathrm{FAIR}$ data

Neural operators can learn nonlinear mappings between function spaces and offer a new simulation paradigm for real-time prediction of complex dynamics for realistic diverse applications as well as for system identification in science and engineering. Herein, we investigate the performance of two neural operators, which have shown promising results so far, and we develop new practical extensions that will make them more accurate and robust and importantly more suitable for industrial-complexity applications. The first neural operator, DeepONet, was published in 2019 (Lu et al., 2019), and its original architecture was based on the universal approximation theorem of Chen & Chen (1995). The second one, named Fourier Neural Operator or FNO, was published in 2020, and it is based on parameterizing the integral kernel in the Fourier space. DeepONet is represented by a summation of products of neural networks (NNs), corresponding to the branch NN for the input function and the trunk NN for the output function; both NNs are general architectures, e.g., the branch NN can be replaced with a CNN or a ResNet. According to Kovachki et al. (2021), FNO in its continuous form can be viewed conceptually as a DeepONet with a specific architecture of the branch NN and a trunk NN represented by a trigonometric basis. In order to compare FNO with DeepONet computationally for realistic setups, we develop several extensions of FNO that can deal with complex geometric domains as well as mappings where the input and output function spaces are of different dimensions. We also develop an extended DeepONet with special features that provide inductive bias and accelerate training, and we present a faster implementation of DeepONet with cost comparable to the computational cost of FNO, which is based on the Fast Fourier Transform. Here we consider 16 different benchmarks to demonstrate the relative performance of the two neural operators, including instability wave analysis in hypersonic boundary layers, prediction of the vorticity field of a flapping airfoil, porous media simulations in complex-geometry domains, etc. We follow the guiding principles of FAIR (Findability, Accessibility, Interoperability, and Reusability) for scientific data management and stewardship. The performance of DeepONet and FNO is comparable for relatively simple settings, but for complex geometries the performance of FNO deteriorates greatly. We also compare theoretically the two neural operators and obtain similar error estimates for DeepONet and FNO under the same regularity assumptions.

42 ENGINEERING↗
DOE OSTI2022

Approximation rates of DeepONets for learning operators arising from advection–diffusion equations

Here we present the analysis of approximation rates of operator learning in Chen and Chen (1995) and Lu et al. (2021), where continuous operators are approximated by a sum of products of branch and trunk networks. In this work, we consider the rates of learning solution operators from both linear and nonlinear advection–diffusion equations with or without reaction. We find that the approximation rates depend on the architecture of branch networks as well as the smoothness of inputs and outputs of solution operators.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

($\mathrm{INVITED}$)Counter-ion effect on the diffusion behavior of $\mathrm{Y}$b, $\mathrm{L}$u, and $\mathrm{N}$d ions in $\mathrm{YAG}$ transparent ceramics

The ability to fabricate additively manufactured laser waveguides with sharp dopant concentration interfaces is limited by diffusion of the dopants at the temperatures required to fully densify the material. Compositional analysis of bilayer samples, where each layer was either undoped YAG or YAG doped with Yb, Lu, or Nd, were fabricated such that all combinations were available for testing. Samples were fabricated at both 1750°C and 1850°C to determine the diffusion behavior of each dopant alone and also in the presence of a second dopant. It was found that the experimental concentration profiles exhibited both intragranular (bulk) and grain boundary contributions, and thus fitting to a complementary error function equation required the use of two diffusion coefficients. Nd always diffused further along grain boundaries than the other dopants, due in part to its small segregation coefficient in YAG. It is shown that the presence of Nd as a counter dopant inhibits intragranular diffusion of other dopants while enhancing their grain boundary diffusion. All the observed trends were attributed to a combination of intragranular lattice strain due to: dopant ions replacing yttrium substitutionally, the relative driving forces for the segregation of dopant ions to grain boundaries, and the ability of one dopant to affect the diffusion of a different dopant in the other direction to maintain charge neutrality.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Coordination Sphere of Lanthanide Aqua Ions Resolved with Ab Initio Molecular Dynamics and X-ray Absorption Spectroscopy

To resolve the fleeting structures of lanthanide Ln 3+ aqua ions in solution, we (i) performed the first ab initio molecular dynamics (AIMD) simulations of the entire series of Ln 3+ aqua ions in explicit water solvent using pseudopotentials and basis sets recently optimized for lanthanides and (ii) measured the symmetry of the hydrating waters about Ln 3+ ions (Nd 3+ , Dy 3+ , Er 3+ , Lu 3+ ) for the first time with extended X-ray absorption fine structure (EXAFS). EXAFS spectra were measured experimentally and generated from AIMD trajectories to directly compare simulation, which concurrently considers the electronic structure and the atomic dynamics in solution, with experiment. In this study, we performed a comprehensive evaluation of EXAFS multiple-scattering analysis (up to 6.5 Å) to measure Ln–O distances and angular correlations (i.e., symmetry) and elucidate the molecular geometry of the first hydration shell. This evaluation, in combination with symmetry-dependent L 3 - and L 1 -edge spectral analysis, shows that the AIMD simulations remarkably reproduces the experimental EXAFS data. The error in the predicted Ln–O distances is less than 0.07 Å for the later lanthanides, while we observed excellent agreement with predicted distances within experimental uncertainty for the early lanthanides. Our analysis revealed a dynamic, symmetrically disordered first coordination shell, which does not conform to a single molecular geometry for most lanthanides. This work sheds critical light on the highly elusive coordination geometry of the Ln 3+ aqua ions.

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

Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems

Double-precision floating-point arithmetic (FP64) has been the de facto standard for engineering and scientific simulations for several decades. Problem complexity and the sheer volume of data coming from various instruments and sensors motivate researchers to mix and match various approaches to optimize compute resources, including different levels of floating-point precision. In recent years, machine learning has motivated hardware support for half-precision floating-point arithmetic. A primary challenge in high-performance computing is to leverage reduced-precision and mixed-precision hardware. We show how the FP16/FP32 Tensor Cores on NVIDIA GPUs can be exploited to accelerate the solution of linear systems of equations Ax = b without sacrificing numerical stability. The techniques we employ include multiprecision LU factorization, the preconditioned generalized minimal residual algorithm (GMRES), and scaling and auto-adaptive rounding to avoid overflow. We also show how to efficiently handle systems with multiple right-hand sides. On the NVIDIA Quadro GV100 (Volta) GPU, we achieve a 4×-5× performance increase and 5× better energy efficiency versus the standard FP64 implementation while maintaining an FP64 level of numerical stability.

GMRES↗