Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Diffusion model”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Phase field modeling of hydrogen release in nitrile-butadiene rubber composites after high-pressure hydrogen exposure

Experimental data showed that a simple diffusion model doesn’t suffice to explain the hydrogen (H 2 ) release kinetics in nitrile butadiene rubber (NBR) composites (one with no filler or plasticizer, one with plasticizer only, one with fillers and plasticizer, and one with fillers only) after exposed to hydrogen gas at a pressure of 27.6 MPa. In this work, we developed a phase field model that considers the effect of H 2 gas bubble evolution on H 2 release kinetics. With the model, we simulated the effect of thermodynamic and kinetics properties on the nucleation, growth, and shrinkage of gas bubbles and H 2 release kinetics during and after decompression. Here, the results demonstrated that 1) the model built upon certain thermodynamic and kinetics properties of a given material can well describe its H 2 release kinetics measured in experiments, and 2) the model can predict the gas bubble evolution, which is associated with the material property degradation and failures, with accurate thermodynamic and kinetics properties or H 2 release data from experiments.

08 HYDROGEN↗
DOE OSTI2025

Exploring the energy landscape of RBMs: reciprocal space insights into bosons, hierarchical learning and symmetry breaking

Deep generative models have become ubiquitous due to their ability to learn and sample from complex distributions. Despite the proliferation of various frameworks, the relationships among these models remain largely unexplored, a gap that hinders the development of a unified theory of AI learning. In this work, we address two central challenges: clarifying the connections between different deep generative models and deepening our understanding of their learning mechanisms. We focus on Restricted Boltzmann Machines (RBMs), a class of generative models known for their universal approximation capabilities for discrete distributions. By introducing a reciprocal space formulation for RBMs, we reveal a connection between these models, diffusion processes, and systems of coupled bosons. Our analysis shows that at initialization, the RBM operates at a saddle point, where the local curvature is determined by the singular values of the weight matrix, whose distribution follows the Marc̆enko-Pastur law and exhibits rotational symmetry. During training, this rotational symmetry is broken due to hierarchical learning, where different degrees of freedom progressively capture features at multiple levels of abstraction. This leads to a symmetry breaking in the energy landscape, reminiscent of Landau’s theory. This symmetry breaking in the energy landscape is characterized by the singular values and the weight matrix eigenvector matrix. We derive the corresponding free energy in a mean-field approximation. We show that in the limit of infinite size RBM, the reciprocal variables are Gaussian distributed. Our findings indicate that in this regime, there will be some modes for which the diffusion process will not converge to the Boltzmann distribution. To illustrate our results, we trained replicas of RBMs with different hidden layer sizes using the MNIST dataset. Our findings not only bridge the gap between disparate generative frameworks but also shed light on the fundamental processes underpinning learning in deep generative models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Submodels of model of nonlinear diffusion in the inhomogeneous medium involving absorption

We study the five-parameter model, describing the process of nonlinear diffusion in an inhomogeneous medium in the presence of absorption, for which the differential equation of the model admits a continuous Lie group of transformations, acting on the set of its solutions. We found six submodels of the original model of nonlinear diffusion, with different symmetry properties. Of these six submodels, the five submodels with transient absorption, for which the absorption coefficient depends on time according to a power law, represent the greatest interest with a mathematical point of view and with the point of view of physical applications. For each of these nonlinear submodels, we obtained formulas for producing new solutions that contain arbitrary constants, and we found all invariant submodels. All essentially distinct invariant solutions describing these invariant submodels are found in an explicit form or are reduced to finding the solution of nonlinear integral equations. The presence of the arbitrary constants in the integral equations that determine these solutions provide new opportunities for analytical and numerical study of boundary value problems for the received submodels and, thus, for the original model of nonlinear diffusion. For the received invariant submodels, we studied diffusion processes for which at the initial moment of the time at a fixed point is specified as a concentration and its gradient or as a concentration and its velocity. Solving of boundary value problems describing these processes is reduced to the solving of nonlinear integral equations. We established the existence and uniqueness of solutions of these boundary value problems under some additional conditions. The obtained results can be used to study the diffusion of substances, diffusion of conduction electrons and other particles, diffusion of physical fields and propagation of heat in inhomogeneous medium, and also to study a turbulence (Leith model, differential approximation to wave turbulence)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Mathematical model and calculation algorithm of micro and meso levels of separation process of gaseous mixtures in molecular sieves

In the given article, the main thermodynamic aspects of the issue of modeling diffusion transfer in molecular sieves have been formulated. Dissipation function is used as a basic notion. The differential equation, connecting volume flow with the change of the concentration of catchable component has been derived. As a result, the expression for changing the concentration of the catchable component and the coefficient of membrane detecting has been received. As well, the system approach to describing the process of gases separation in ultra porous membranes has been realized and micro and meso-levels of mathematical modeling have been distinguished. The non-ideality of the shared system is primarily taken into consideration at the micro-level and the departure from the diffusion law of Fick has been taken into account. The calculation method of selectivity considering fractal structure of membranes has been developed at the meso level. The calculation algorithm and its software implementation have been suggested.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2014

Lateral carrier diffusion and current gain in terahertz InGaAs/InP double-heterojunction bipolar transistors

The DC current gain in In{sub 0.53}Ga{sub 0.47}As/InP double-heterojunction bipolar transistors is computed based on a drift-diffusion model, and is compared with experimental data. Even in the absence of other scaling effects, lateral diffusion of electrons to the base Ohmic contacts causes a rapid reduction in DC current gain as the emitter junction width and emitter-base contact spacing are reduced. The simulation and experimental data are compared in order to examine the effect of carrier lateral diffusion on current gain. The impact on current gain due to device scaling and approaches to increase current gain are discussed.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2016

Temperature dependence of magneto-transport properties in Co{sub 2}Fe(Ga{sub 0.5}Ge{sub 0.5})/Cu lateral spin valves

The non-local spin signals of Co{sub 2}Fe(Ga{sub 0.5}Ge{sub 0.5})/Cu lateral spin valves with sub-micron size dimensions were measured with varying temperatures. The non-local spin signal reaches 54 mΩ at 4 K, while it degrades down to 13 mΩ at room temperature. Analysis based on the one-dimensional spin diffusion model clarifies the dominant source for degrading of the spin signal is suppression of the spin diffusion length in Cu, not the spin polarization, indicating Co{sub 2}Fe(Ga{sub 0.5}Ge{sub 0.5}) keeps half-metallic nature even at room temperature. The temperature dependence of non-local spin signal was found to exhibit a downturn at 36 K. The presence of magnetic impurities, detrimental effect of which becomes more pronounced for diffusive transport in long Cu wires, is suggested to cause the observed downturn in non-local spin signals.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Phosphorus Doping of Polycrystalline CdTe by Diffusion

Phosphorus diffusion in single crystal and polycrystalline CdTe material is explored using various methods. Dynamic secondary ion mass spectroscopy (SIMS) is used to determine 1D P diffusion profiles. A 2D diffusion model is used to determine the expected cross-sectional distribution of P in CdTe after diffusion anneals. Time of flight SIMS and cross-sectional cathodoluminescence corroborates expected P distributions. Devices fabricated with diffused P exhibit hole concentrations up to low 1015 cm-3, however a subsequent activation anneal enabled hole concentrations greater than 1016 cm-3. CdCl2 treatments and Cu based contacts were also explored in conjunction with the P doping process.

14 SOLAR ENERGY↗
DOE OSTI2016

Optimizing ZnS/6LiF scintillators for wavelength-shifting-fiber neutron detectors

In this paper we compare the performance of grooved and flat ZnS/6LiF scintillators in a wavelength shifting-fiber (WLSF) detector. Flat ZnS/6LiF scintillators with the thickness L=0.2-0.8 mm were characterized using photon counting and pulse-height analysis and compared to a grooved scintillator of approximately 0.8 mm thick. While a grooved scintillator considerably increases the apparent thickness of the scintillator to neutrons for a given coating thickness, we find that the flat scintillators perform better than the grooved scintillators in terms of both light yield and neutron detection efficiency. The flat 0.8-mm-thick scintillator has the highest light output, and it is 52% higher compared with a grooved scintillator of same thickness. The lower light output of the grooved scintillator as compared to the flat scintillator is consistent with the greater scintillator-WLSF separation and the much larger average emission angle of the grooved scintillator. We also find that the average light cone width, or photon travel-length as measured using time-of-flight powder diffraction of diamond and vanadium, decreases with increasing L in the range of L=0.6-0.8 mm. This result contrasts with the traditional Swank diffusion model for micro-composite scintillators, and could be explained by a decrease in photon diffusion-coefficient or an increase in micro-particle content in the flat scintillator matrix for the thicker scintillators.

Crow, Lowell↗
DOE OSTI2021

Defect characterization and charge transport measurements in high-resolution Ni/n-4H-SiC Schottky barrier radiation detectors fabricated on 250 μ m epitaxial layers

Advances in the growth processes of 4H-SiC epitaxial layers have led to the continued expansion of epilayer thickness, allowing for the detection of more penetrative radioactive particles. In this work, we report the fabrication and characterization of high-resolution Schottky barrier radiation detectors on 250 μm thick n-type 4H-SiC epitaxial layers, the highest reported thickness to date. Several 8 × 8 mm 2 detectors were fabricated from a diced 100 mm diameter 4H-SiC epitaxial wafer grown on a conductive 4H-SiC substrate with a mean micropipe density of 0.11 cm -2 From the Mott–Schottky plots, the effective doping concentration was found to be in the range (0.95–1.85) × 10 14 cm -3 , implying that full depletion could be achieved at ~5.7 kV (0.5 MV/cm at the interface). The current-voltage characteristics demonstrated consistently low leakage current densities of 1–3 nA/cm 2 at a reverse bias of -800 V. This resulted in the pulse-height spectra generated using a 241Am alpha source (5486 keV) manifesting an energy resolution of less than 0.5% full width at half maximum (FWHM) for all the detectors at -200 V. The charge collection efficiencies (CCEs) were measured to be 98–99% with no discernable correlation to the energy resolution. A drift-diffusion model fit to the variation of CCE as a function of bias voltage, revealed a minority carrier diffusion length of ~10 μm. Deep level transient spectroscopy measurements on the best resolution detector revealed that the excellent performance was the result of having ultralow concentrations of the order of 1011 cm-3 lifetime limiting defects—Z 1/2 and EH 6/7 .

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

MOLNs: A Cloud Platform for Interactive, Reproducible, and Scalable Spatial Stochastic Computational Experiments in Systems Biology Using PyURDME

Computational experiments using spatial stochastic simulations have led to important new biological insights, but they require specialized tools and a complex software stack, as well as large and scalable compute and data analysis resources due to the large computational cost associated with Monte Carlo computational workflows. The complexity of setting up and managing a large-scale distributed computation environment to support productive and reproducible modeling can be prohibitive for practitioners in systems biology. This results in a barrier to the adoption of spatial stochastic simulation tools, effectively limiting the type of biological questions addressed by quantitative modeling. In this paper, we present PyURDME, a new, user-friendly spatial modeling and simulation package, and MOLNs, a cloud computing appliance for distributed simulation of stochastic reaction-diffusion models. In conclusion, MOLNs is based on IPython and provides an interactive programming platform for development of sharable and reproducible distributed parallel computational experiments.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2012

Theoretical Investigation of Microstructure Evolution and Deformation of Zirconium Under Cascade Damage Conditions

This work is based on our reaction-diffusion model of radiation growth of Zr-based materials proposed recently in "On the Origin of Radiation Growth of HCP Crystals". In "On the Origin of Radiation Growth of HCP Crystals", the equations for the strain rates in unloaded pure crystal under cascade damage conditions of, e.g., neutron or heavy-ion irradiation were derived as functions of dislocation densities, which include contributions from dislocation loops, and spatial distribution of their Burgers vectors. The model takes into account the intra-cascade clustering of self-interstitial atoms and their one-dimensional diffusion; explains the growth stages, including the break-away growth of pre-annealed samples; and accounts for some striking observations, such as of negative strain in prismatic direction, and co-existence of vacancy- and interstitial-type prismatic loops. In this report, the change of dislocation densities due to accumulation of sessile dislocation loops is taken into account explicitly to investigate the dose dependence of radiation growth. The dose dependence of climb rates of dislocations is calculated, which is important for the climb-induced glide model of radiation creep. The results of fitting the model to available experimental data and some numerical calculations of the strain behavior of Zr for different initial dislocation structures are presented and discussed. The computer code RIMD-ZR.V1 (Radiation Induced Microstructure and Deformation of Zr) developed is described and attached to this report.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2014

Explanation of the local galactic cosmic ray energy spectra measured by Voyager 1. I. Protons

Almost exactly 100 yr after the original discovery of cosmic rays, the V1 spacecraft has observed, for the first time, the local interstellar medium energy spectra of cosmic ray H, He, C/O nuclei at nonrelativistic kinetic energies, after leaving the heliosphere modulation region on 2012 August 25. We explain these observations by modeling the propagation of these particles in the local Galactic environment with an updated steady-state spatial diffusion model including all particle momentum losses with the local interstellar gas (Coulomb/ionization, pion production, adiabatic deceleration, and fragmentation interactions). Excellent agreement with the V1 cosmic ray H observations is obtained if the solar system resides within a spatially homogeneous layer of distributed cosmic ray sources injecting the same momentum power law ∝p {sup –s} with s = 2.24 ± 0.12. The best fit to the V1 H observations also provides an estimate of the characteristic break kinetic energy T{sub C} = 116 ± 27 MeV, representing the transition from ionization/Coulomb energy losses at low energies to pion production and adiabatic deceleration losses in a Galactic wind at high energies. As the determined value is substantially smaller than 217 MeV in the absence of adiabatic deceleration, our results prove the existence of a Galactic wind in the local Galactic environment.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Reversible Tuning of the Surface Plasmon Resonance of Indium Tin Oxide Nanocrystals by Gas-Phase Oxidation and Reduction

Heavily doped oxide nanocrystals exhibit a tunable localized surface plasmon resonance (LSPR) in the infrared, a property that is promising for applications in photonics, spectroscopy, and photochemistry. Nanocrystal carrier density and, thus, spectral response are adjustable via chemical reaction; yet, the fundamental processes that govern this behavior are poorly understood. In this work, we study the time dependence of the LSPR supported by indium tin oxide (ITO) nanocrystals during O 2 and N 2 annealing with in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS). We show that the LSPR red-shifts upon oxidation in O 2 and blue-shifts to its original position upon reduction in N 2 . A reaction–diffusion model allows us to rationalize the underlying physicochemical processes and quantitatively connect nanocrystal redox chemistry, solid-state diffusion, carrier density, and the LSPR.

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

Insights Into Seismicity Associated With Flexibly Operating Enhanced Geothermal System From Real‐Time Distributed Acoustic Sensing

Enhanced Geothermal Systems (EGS) have the capacity to broaden the accessible resource pool for geothermal power generation. Traditionally viewed as a “baseload” resource, their flexible operation might also enable dispatchable load‐following generation and long‐term energy storage, aligning them with the evolving landscape of decarbonized electricity systems. However, increasing permeability and extracting energy during EGS operations can induce microseismic events; for many prior EGS efforts, some associated seismicity has been observed. While energetically beneficial, the flexibility of EGS operations prompts our inquiry into whether new types of operations will yield previously unseen seismicity patterns. We demonstrate the use of distributed acoustic sensing (DAS) with real‐time edge computing to monitor seismicity during a pilot test of a cyclically operated EGS facility at the Blue Mountain geothermal field. Our focus lies in uncovering seismicity insights from the real‐time microseismic catalog, particularly during load‐following dispatchability tests simulating flexible EGS operation. Here, we find that variations in pore pressure consistently correlate with seismicity, and that controlling pressure cycles during flexible operations appears to constrain microseismic activity during subsequent cycles. The spatio‐temporal evolution of microseismic clouds recorded during cyclic injection cycles fits diffusive models over our available observation period. Additionally, seismicity elevation lags behind pore pressure increases, likely due to pressure diffusion to the fracture system boundary. Through real‐time monitoring, we offer novel insights into seismicity associated with flexibly operating EGS. Our findings suggest that leveraging DAS and edge computing can inform EGS operations and help mitigate induced seismicity.

Chamarczuk, Michal [Rice Univ., Houston, TX (Unite↗
DOE OSTI2022

Model Package Report: Composite Analysis Solid Waste Release Model (CASWR Model)

This document describes the implementation of a solid waste form release model in GoldSim for the Hanford Site Composite Analysis (CA) Update. This Composite Analysis Solid Waste Release model (CASWR model) was designed to generate deterministic radionuclide release rates for Hanford’s Central Plateau solid waste disposal sites using single realizations of release model coefficients. Five generalized waste form types are used for the conceptual model of waste release: surplus reactor block, cement, soil-debris, grouted residual waste, and ancillary equipment. The surplus reactor block waste form consists of radionuclide leaching from surplus production graphite reactor core blocks via unspecified processes (White et al., 1984 as cited in PNNL-15965). The cement waste form represents solidified wastes whose permeability is much lower than that of the surrounding soil. The soil-debris waste form type is defined as unconsolidated waste mixed with soil material. Tanks and canyon complexes comprise the grouted residual waste form such that their solid waste will be grouted and capped with a surface barrier at the completion of their cleanup. Finally, the ancillary equipment waste form constitutes contaminant releases from ancillary and auxiliary waste form residues associated with tank farms at closure. Individual sub-models are implemented to numerically represent a respective waste form within the CASWR Model: Surplus Reactor Block Sub-model, Cement Sub-model, SoilDebris Sub-model, Grouted Residual Waste Sub-model, and Ancillary Equipment Submodel. Advection is assumed to be the primary transport process governing the release of radionuclides in the Ancillary Equipment and Soil-Debris Sub-models. Diffusion is assumed to be the primary release process in the Grouted Residual Waste and Cement Sub-models. An unspecified zero-order release process is considered in the Surplus Reactor Block Sub-model due to the lack of information regarding actual processes involved in irradiated graphite leaching. The Surplus Reactor Block, Cement, and SoilDebris Sub-models were compared against analytical solutions (PNNL-11800, Composite Analysis for Low-Level Waste Disposal in the 200 Area Plateau of the Hanford Site). The agreement between results of these analytical solutions and the corresponding waste form models verified their correct implementation in GoldSim. A 1-D modeling abstraction approach for the Grouted Residual Waste and Ancillary Equipment Sub-models was adopted from existing Performance Assessment (PA) models. Despite the simplifications made in these sub-models, they were found to be appropriate representations of the waste forms, similar to what was used in the Waste Management Area C PA model (RPP-ENV-58782, Performance Assessment of Waste Management Area C, Hanford Site, Washington, Rev. 0). A sensitivity analysis was conducted to identify the most influential parameters in each waste form sub-model. Suggestions for considering pH-dependent and redox-dependent release mechanisms are formulated through the development of a conditional constant approach in GoldSim.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2014

Characterizing the hohlraum radiation via one-end driven experiments

A new experiment is designed and performed on the Shenguang III laser facility with the first eight available beams to characterizing the hohlraum radiation, in which the hohlraum with laser entrance holes on both ends is driven through one-end only. The experiment enables us to identify the x-ray radiations originated from the hohlraum reemission wall and high-Z bubble plasmas utilizing their position and spectral characters, which provides a better test on the associated hohlraum models. The total and M-band x-ray radiation fluxes are measured with the flat response x-ray detectors and the filtered M-band x-ray detectors, respectively. Numerical simulations are conducted with the two-dimensional radiation hydrodynamic code LARED-INTEGRATION using the multi-group radiation transfer and/or diffusion models. It is found that the experimentally measured temporal profiles and angular distributions of hohlraum radiation are in good agreement with the predictions of simulation using radiation transfer models, but differ significantly from the results obtained with the multi-group radiation diffusion calculations. We thus note that to accurately represent the hohlraum radiation, a true radiation transfer model is essential.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗