Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Deep”

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 757 records · Page 42
All sourcesNASA NTRSDOE OSTI
DOE OSTI2014

Spectral light separator based on deep-subwavelength resonant apertures in a metallic film

We propose to funnel, select, and collect light spectrally by exploiting the unique properties of deep-subwavelength resonant apertures in a metallic film. In our approach, each aperture has an electromagnetic cross section that is much larger than its physical size while the frequency of the collected light is controlled by its height through the Fabry-Pérot resonance mechanism. The electromagnetic crosstalk between apertures remains low despite physical separations in the deep-subwavelength range. The resulting device enables an extremely efficient, subwavelength way to decompose light into its spectral components without the loss of photons and spatial coregistration errors. As a specific example, we show a subwavelength-size structure with three deep-subwavelength slits in a metallic film designed to operate in the mid-wave infrared range between 3 and 5.5 μm.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2014

High-performance deep ultraviolet photodetectors based on ZnO quantum dot assemblies

A high-performance ZnO quantum dots (QDs)-based ultraviolet (UV) photodetector has been successfully fabricated via the self-assembly of QDs on the Au interdigital electrode. The broadened band gap in ZnO QDs makes the device has the highly selective response for the deep UV detection. The unique QD-QD junction barriers similar to back-to-back Schottky barriers dominate the conductance of the QD network and the UV light-induced barrier-height modulation plays a crucial role in enhancing the photoresponsivity and the response speed. Typically, the as-fabricated device exhibits the fast response and recovery times of within 1 s, the deep UV selectivity of less than 340 nm, and the stable repeatability with on/off current ratio over 10³, photoresponsivity of 5.04×10²A/W, and photocurrent gain of 1.9×10³, demonstrating that the ZnO QD network is a superior building block for deep UV photodetectors.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2015

Proton irradiation effects on deep level states in Mg-doped p-type GaN grown by ammonia-based molecular beam epitaxy

The impact of proton irradiation on the deep level states throughout the Mg-doped p-type GaN bandgap is investigated using deep level transient and optical spectroscopies. Exposure to 1.8 MeV protons of 1 × 10{sup 13 }cm{sup −2} and 3 × 10{sup 13 }cm{sup −2} fluences not only introduces a trap with an E{sub V} + 1.02 eV activation energy but also brings monotonic increases in concentration for as-grown deep states at E{sub V} + 0.48 eV, E{sub V} + 2.42 eV, E{sub V} + 3.00 eV, and E{sub V} + 3.28 eV. The non-uniform sensitivities for individual states suggest different physical sources and/or defect generation mechanisms. Comparing with prior theoretical calculations reveals that several traps are consistent with associations to nitrogen vacancy, nitrogen interstitial, and gallium vacancy origins, and thus are likely generated through displacing nitrogen and gallium atoms from the crystal lattice in proton irradiation environment.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2015

Density relaxation and particle motion characteristics in a non-ionic deep eutectic solvent (acetamide + urea): Time-resolved fluorescence measurements and all-atom molecular dynamics simulations

Temperature dependent relaxation dynamics, particle motion characteristics, and heterogeneity aspects of deep eutectic solvents (DESs) made of acetamide (CH{sub 3}CONH{sub 2}) and urea (NH{sub 2}CONH{sub 2}) have been investigated by employing time-resolved fluorescence measurements and all-atom molecular dynamics simulations. Three different compositions (f) for the mixture [fCH{sub 3}CONH{sub 2} + (1 − f)NH{sub 2}CONH{sub 2}] have been studied in a temperature range of 328-353 K which is ∼120-145 K above the measured glass transition temperatures (∼207 K) of these DESs but much lower than the individual melting temperature of either of the constituents. Steady state fluorescence emission measurements using probe solutes with sharply different lifetimes do not indicate any dependence on excitation wavelength in these metastable molten systems. Time-resolved fluorescence anisotropy measurements reveal near-hydrodynamic coupling between medium viscosity and rotation of a dissolved dipolar solute. Stokes shift dynamics have been found to be too fast to be detected by the time-resolution (∼70 ps) employed, suggesting extremely rapid medium polarization relaxation. All-atom simulations reveal Gaussian distribution for particle displacements and van Hove correlations, and significant overlap between non-Gaussian (α{sub 2}) and new non-Gaussian (γ) heterogeneity parameters. In addition, no stretched exponential relaxations have been detected in the simulated wavenumber dependent acetamide dynamic structure factors. All these results are in sharp contrast to earlier observations for ionic deep eutectics with acetamide [Guchhait et al., J. Chem. Phys. 140, 104514 (2014)] and suggest a fundamental difference in interaction and dynamics between ionic and non-ionic deep eutectic solvent systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2015

Deep proton tunneling in the electronically adiabatic and non-adiabatic limits: Comparison of the quantum and classical treatment of donor-acceptor motion in a protein environment

Analytical models describing the temperature dependence of the deep tunneling rate, useful for proton, hydrogen, or hydride transfer in proteins, are developed and compared. Electronically adiabatic and non-adiabatic expressions are presented where the donor-acceptor (D-A) motion is treated either as a quantized vibration or as a classical “gating” distribution. We stress the importance of fitting experimental data on an absolute scale in the electronically adiabatic limit, which normally applies to these reactions, and find that vibrationally enhanced deep tunneling takes place on sub-ns timescales at room temperature for typical H-bonding distances. As noted previously, a small room temperature kinetic isotope effect (KIE) does not eliminate deep tunneling as a major transport channel. The quantum approach focuses on the vibrational sub-space composed of the D-A and hydrogen atom motions, where hydrogen bonding and protein restoring forces quantize the D-A vibration. A Duschinsky rotation is mandated between the normal modes of the reactant and product states and the rotation angle depends on the tunneling particle mass. This tunnel-mass dependent rotation contributes substantially to the KIE and its temperature dependence. The effect of the Duschinsky rotation is solved exactly to find the rate in the electronically non-adiabatic limit and compared to the Born-Oppenheimer (B-O) approximation approach. The B-O approximation is employed to find the rate in the electronically adiabatic limit, where we explore both harmonic and quartic double-well potentials for the hydrogen atom bound states. Both the electronically adiabatic and non-adiabatic rates are found to diverge at high temperature unless the proton coupling includes the often neglected quadratic term in the D-A displacement from equilibrium. A new expression is presented for the electronically adiabatic tunnel rate in the classical limit for D-A motion that should be useful to experimentalists working near room temperature. This expression also holds when a broad protein conformational distribution of D-A equilibrium distances dominates the spread of the D-A vibrational wavefunction.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2015

Effect of precursor solutions stirring on deep level defects concentration and spatial distribution in low temperature aqueous chemical synthesis of zinc oxide nanorods

Hexagonal c-axis oriented zinc oxide (ZnO) nanorods (NRs) with 120-300 nm diameters are synthesized via the low temperature aqueous chemical route at 80 °C on silver-coated glass substrates. The influence of varying the precursor solutions stirring durations on the concentration and spatial distributions of deep level defects in ZnO NRs is investigated. Room temperature micro-photoluminesnce (μ-PL) spectra were collected for all samples. Cathodoluminescence (CL) spectra of the as-synthesized NRs reveal a significant change in the intensity ratio of the near band edge emission (NBE) to the deep-level emission (DLE) peaks with increasing stirring durations. This is attributed to the variation in the concentration of the oxygen-deficiency with increasing stirring durations as suggested from the X-ray photoelectron spectroscopy analysis. Spatially resolved CL spectra taken along individual NRs revealed that stirring the precursor solutions for relatively short duration (1-3 h), which likely induced high super saturation under thermodynamic equilibrium during the synthesis process, is observed to favor the formation of point defects moving towards the tip of the NRs. In contrary, stirring for longer duration (5-15 h) will induce low super saturation favoring the formation of point defects located at the bottom of the NRs. These findings demonstrate that it is possible to control the concentration and spatial distribution of deep level defects in ZnO NRs by varying the stirring durations of the precursor solutions.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Determination of deep trapping lifetime in organic semiconductors using impedance spectroscopy

A method for determining deep trapping lifetime in semiconductors using an impedance spectroscopy is proposed. A unique feature of the method is the simultaneous determination of the drift mobility and deep trapping lifetime in thin-film electronic devices. The validity of the proposed method is examined by numerical calculation. Simultaneous determinations of the drift mobility and deep trapping lifetime using this method are demonstrated in prototypical hole transporting organic semiconductors.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

Revisiting radiative deep-level transitions in CuGaSe{sub 2} by photoluminescence

Recent defect calculations suggest that the open circuit voltage of CuGaSe{sub 2} solar cells can be limited by deep intrinsic electron traps by Ga{sub Cu} antisites and their complexes with Cu-vacancies. To gain experimental evidence, two radiative defect transitions at 1.10 eV and 1.24 eV are characterized by steady-state photoluminescence on epitaxial-grown CuGaSe{sub 2} thin films. Cu-rich samples are studied, since they show highest crystal quality, exciton luminescence, and no potential fluctuations. Variations of the laser intensity and temperature dependent measurements suggest that emission occurs from two deep donor-like levels into the same shallow acceptor. At 10 K, power-law exponents of 1 (low excitation regime) and 1/2 (high excitation regime) are observed identically for both transitions. The theory and a fitting function for the double power law is derived. It is concluded that the acceptor becomes saturated by excess carriers which changes the exponent of all transitions. Activation energies determined from the temperature quenching depend on the excitation level and show unexpected values of 600 meV and higher. The thermal activation of non-radiative processes can explain the distortion of the ionization energies. Both the deep levels play a major role as radiative and non-radiative recombination centers for electrons and can be detrimental for photovoltaic applications.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2018

Deep learning: A guide for practitioners in the physical sciences

Machine learning is finding increasingly broad applications in the physical sciences. This most often involves building a model relationship between a dependent, measurable output, and an associated set of controllable, but complicated, independent inputs. We present a tutorial on current techniques in machine learning—a jumping-off point for interested researchers to advance their work. We focus on deep neural networks with an emphasis on demystifying deep learning. We begin with background ideas in machine learning and some example applications from current research in plasma physics. We discuss supervised learning techniques for modeling complicated functions, beginning with familiar regression schemes, and then advancing to more sophisticated deep learning methods. We also address unsupervised learning and techniques for reducing the dimensionality of input spaces. Along the way, we describe methods for practitioners to help ensure that their models generalize from their training data to as-yet-unseen test data. We describe classes of tasks—predicting scalars, handling images, and fitting time-series—and prepare the reader to choose an appropriate technique. We finally point out some limitations to modern machine learning and speculate on some ways that practitioners from the physical sciences may be particularly suited to help.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Deep convolutional neural networks for multi-scale time-series classification and application to tokamak disruption prediction using raw, high temporal resolution diagnostic data

In this paper we discuss recent advances in deep convolutional neural networks (CNN) for sequence learning, which allow identifying long-range, multi-scale phenomena in long sequences, such as those found in fusion plasmas. We point out several benefits of these deep CNN architectures, such as not requiring experts such as physicists to hand-craft input data features, the ability to capture longer range dependencies compared to the more common sequence neural networks (recurrent neural networks like long short-term memory (LSTM) networks), and the comparative computational efficiency. We apply this neural network architecture to the popular problem of disruption prediction in fusion energy tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. Initial results trained on a large ECEi dataset show promise, achieving an F 1 -score of ~91% on individual time-slices using only the ECEi data. This indicates the ECEi diagnostic by itself can be sensitive to a number of pre-disruption markers useful for predicting disruptions on timescales not only for mitigation but also avoidance. Future opportunities for utilizing these deep CNN architectures with fusion data are outlined, including impact of recent upgrades to the ECEi diagnostic.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Deep learning surrogate model for kinetic Landau-fluid closure with collision

In this work, the kinetic Landau-fluid (LF) closure with collision and periodic boundary condition is used in the development of the deep learning (DL) surrogate model. A classical neural network, namely, feedforward neural network or sometimes termed multilayer perceptron, is constructed and trained to learn the kinetic LF closure in the static limit and arbitrary mean free path in configuration space. The preliminary relation between best hyperparameters and critical parameters for data generation is found. Compared with the numerical approach (non-Fourier method) of the LF closure, the deep learning surrogate model shows an order of magnitude of improvement in terms of accuracy. Perhaps most importantly, the surrogate model closure has been integrated for the first time with fluid simulations. Our DL-enabled fluid simulations, for the first time, give the correct Landau damping rate for a wide range of wave vectors, while the Hammett–Perkins closure cannot produce the correct damping rate. We correctly connect the collisionless Hammett–Perkins closure and collisional Braginskii closure to reproduce the intrinsic nonlocal feature of the heat flux with DL techniques. We address the most concerning error accumulation problem and find that simulations with the deep learning surrogate model are as good as, if not better than, simulations with the analytic closure in terms of long-term numerical stability in the linear Landau damping test.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Distributed deep learning training using silicon photonic switched architectures

The scaling trends of deep learning models and distributed training workloads are challenging network capacities in today’s datacenters and high-performance computing (HPC) systems. We propose a system architecture that leverages silicon photonic (SiP) switch-enabled server regrouping using bandwidth steering to tackle the challenges and accelerate distributed deep learning training. In addition, our proposed system architecture utilizes a highly integrated operating system-based SiP switch control scheme to reduce implementation complexity. To demonstrate the feasibility of our proposal, we built an experimental testbed with a SiP switch-enabled reconfigurable fat tree topology and evaluated the network performance of distributed ring all-reduce and parameter server workloads. The experimental results show up to 3.6× improvements over the static non-reconfigurable fat tree. Our large-scale simulation results show that server regrouping can deliver up to 2.3× flow throughput improvement for a 2× tapered fat tree and a further 11% improvement when higher-layer bandwidth steering is employed. The collective results show the potential of integrating SiP switches into datacenters and HPC systems to accelerate distributed deep learning training.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Deep learning-based quasi-continuum theory for structure of confined fluids

Predicting the structural properties of water and simple fluids confined in nanometer scale pores and channels is essential in, for example, energy storage and biomolecular systems. Classical continuum theories fail to accurately capture the interfacial structure of fluids. In this work, we develop a deep learning-based quasi-continuum theory (DL-QT) to predict the concentration and potential profiles of a Lennard-Jones (LJ) fluid and water confined in a nanochannel. The deep learning model is built based on a convolutional encoder–decoder network (CED) and is applied for high-dimensional surrogate modeling to relate the fluid properties to the fluid–fluid potential. The CED model is then combined with the interatomic potential-based continuum theory to determine the concentration profiles of a confined LJ fluid and confined water. Further, we show that the DL-QT model exhibits robust predictive performance for a confined LJ fluid under various thermodynamic states and for water confined in a nanochannel of different widths. The DL-QT model seamlessly connects molecular physics at the nanoscale with continuum theory by using a deep learning model.

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

Fast uncertainty estimates in deep learning interatomic potentials

Deep learning has emerged as a promising paradigm to give access to highly accurate predictions of molecular and material properties. A common short-coming shared by current approaches, however, is that neural networks only give point estimates of their predictions and do not come with predictive uncertainties associated with these estimates. Existing uncertainty quantification efforts have primarily leveraged the standard deviation of predictions across an ensemble of independently trained neural networks. This incurs a large computational overhead in both training and prediction, resulting in order-of-magnitude more expensive predictions. Here, we propose a method to estimate the predictive uncertainty based on a single neural network without the need for an ensemble. This allows us to obtain uncertainty estimates with virtually no additional computational overhead over standard training and inference. We demonstrate that the quality of the uncertainty estimates matches those obtained from deep ensembles. We further examine the uncertainty estimates of our methods and deep ensembles across the configuration space of our test system and compare the uncertainties to the potential energy surface. Finally, we study the efficacy of the method in an active learning setting and find the results to match an ensemble-based strategy at order-of-magnitude reduced computational cost.

Chemistry↗
DOE OSTI2023

Identification and characterization of deep nitrogen acceptors in β-Ga 2 O 3 using defect spectroscopies

The ability to achieve highly resistive beta-phase gallium oxide (β-Ga 2 O 3 ) layers and substrates is critical for β-Ga 2 O 3 high voltage and RF devices. To date, the most common approach involves doping with iron (Fe), which generates a moderately deep acceptor-like defect state located at E C -0.8 eV in the β-Ga 2 O 3 bandgap. Recently, there has been growing interest in alternative acceptors, such as magnesium (Mg) and nitrogen (N), due to their predicted deeper energy levels, which could avoid inadvertent charge modulation during device operation. In this work, a systematic study that makes direct correlations between the introduction of N using ion implantation and the observation of a newly observed deep level at E C -2.9 eV detected by deep-level optical spectroscopy (DLOS) is presented. The concentration of this state displayed a monotonic dependence with N concentration over a range of implant conditions, as confirmed by secondary ion mass spectrometry (SIMS). With a near 1:1 match in absolute N and E C -2.9 eV trap concentrations from SIMS and DLOS, respectively, which also matched the measured removal of free electrons from capacitance-voltage studies, this indicates that N contributes a very efficiently incorporated compensating defect. Density functional theory calculations confirm the assignment of this state to be an N (0/-1) acceptor with a configuration of N occupying the oxygen site III [N O(III) ]. The near ideal efficiency for this state to compensate free electrons and its location toward the midgap region of the β-Ga 2 O 3 bandgap demonstrates the potential of N doping as a promising approach for producing semi-insulating β-Ga 2 O 3 .

36 MATERIALS SCIENCE↗
DOE OSTI2020

High-pressure elastic properties of dolomite melt supporting carbonate-induced melting in deep upper mantle

Deeply subducted carbonates likely cause low-degree melting of the upper mantle and thus play an important role in the deep carbon cycle. However, direct seismic detection of carbonate-induced partial melts in the Earth’s interior is hindered by our poor knowledge on the elastic properties of carbonate melts. Here we report the first experimentally determined sound velocity and density data on dolomite melt up to 5.9 GPa and 2046 K by in-situ ultrasonic and sink-float techniques, respectively, as well as first-principles molecular dynamics simulations of dolomite melt up to 16 GPa and 3000 K. Using our new elasticity data, the calculated V P /V S ratio of the deep upper mantle (~180–330 km) with a small amount of carbonate-rich melt provides a natural explanation for the elevated V P /V S ratio of the upper mantle from global seismic observations, supporting the pervasive presence of a low-degree carbonate-rich partial melt (~0.05%) that is consistent with the volatile-induced or redox-regulated initial melting in the upper mantle as argued by petrologic studies. This carbonate-rich partial melt region implies a global average carbon (C) concentration of 80–140 ppm. by weight in the deep upper mantle source region, consistent with the mantle carbon content determined from geochemical studies.

58 GEOSCIENCES↗
DOE OSTI2024

Improved deep learning prediction of antigen–antibody interactions

Identifying antibodies that neutralize specific antigens is crucial for developing effective immunotherapies, but this task remains challenging for many target antigens. The rise of deep learning–based computational approaches presents a promising avenue to address this challenge. Here, we assess the performance of a deep learning approach through two benchmark tests aimed at predicting antibodies for the receptor-binding domain of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein. Three different strategies for constructing input sequence alignments are employed for predicting structural models of antigen–antibody complexes. In our initial testing set, which comprises known experimental structures, these strategies collectively yield a significant top-ranked prediction for 61% of cases and a success rate of 47%. Notably, one strategy that utilizes the sequences of known antigen binders outperforms the other two, achieving a precision of 90% in a subsequent test set of ~1,000 antibodies, balanced between true and control antibodies for the antigen, albeit with a lower recall of 25%. Our results underscore the potential of integrating deep learning methods with single B cell sequencing techniques to enhance the prediction accuracy of antigen–antibody interactions.

Science & Technology - Other Topics↗
DOE OSTI2015

THE ALFA ZOA DEEP SURVEY: FIRST RESULTS

The Arecibo L-band Feed Array Zone of Avoidance (ALFA ZOA) Deep Survey is the deepest and most sensitive blind H i survey undertaken in the ZOA. ALFA ZOA Deep will cover about 300 square degrees of sky behind the Galactic Plane in both the inner (30{sup ∘}⩽l⩽75{sup ∘};b⩽|2{sup ∘}|) and outer (175{sup ∘}⩽l⩽207{sup ∘};−2{sup ∘}⩽b⩽+1{sup ∘}) Galaxy, using the Arecibo Radio Telescope. First results from the survey have found 61 galaxies within a 15 square degree area centered on l=192{sup ∘} and b = −2°. The survey reached its expected sensitivity of rms = 1 mJy at 9 km s{sup −1} channel resolution, and is shown to be complete above integrated flux, F{sub Hi} = 0.5 Jy km s{sup −1}. The positional accuracy of the survey is 28″ and detections are found out to a recessional velocity of nearly 19,000 km s{sup −1}. The survey confirms the extent of the Orion and Abell 539 clusters behind the plane of the Milky Way and discovers expansive voids, at 10,000 and 18,000 km s{sup −1}. Twenty-six detections (43%) have a counterpart in the literature, but only two of these have known redshifts. Counterparts are 20% less common beyond v{sub hel} = 10,000 km s{sup −1} and 33% less common at extinctions higher than A{sub B} = 3.5 mag. The ALFA ZOA Deep survey is able to probe large scale structure beyond redshifts that even the most modern wide-angle surveys have been able to detect in the ZOA at any wavelength.

79 ASTRONOMY AND ASTROPHYSICS↗