SEARCH · Search NASA
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.
The purpose of this presentation is to describe how the input assumptions for transmission grid supply constraints were developed. These assumptions will be used in a subsequent paper to evaluate the ability of transmission supply chains to meet anticipated transmission development needs.
Understanding the impact of extractant functionalization on metal-binding energetics in liquid-liquid extraction is essential to guide the development of better separation processes. Traditionally, computational extractant design uses electronic structure calculations on metal-ligand clusters to determine the metal-binding energy of the lowest energy state. Although highly accurate, this approach does not account for all of the relevant physics encountered under experimental conditions. Such methodologies often neglect entropic contributions such as temperature effects and ligand flexibility, in addition to approximating solvent-extractant interactions with implicit solvent models. In this study, we use classical molecular dynamics simulations with an advanced sampling method, metadynamics, to map out extractant molecule conformational free energies in the condensed phase. Here we generate the complete conformational landscape in solution for a family of bidentate malonamide-based extractants with different functionalizations of the headgroup and the side chains. In particular, we show how such alkyl functionalization reshapes the free energy landscape, affecting the free energy penalty of organizing the extractant into the cis-like metal-binding conformation from the trans-like conformation of the free extractant in solution. Specifically, functionalizing alkyl tails to the center of the headgroup has a greater influence on increasing molecular rigidity and disfavoring the binding conformation than functionalizing side chains. These findings are consistent with trends in metal-binding energetics based on experimentally reported distribution ratios. We also consider a different bidentate extractant molecule, carbamoylmethylphosphine oxide, and show how the choice of solvent can further reshape the conformational energetic landscape. This study demonstrates the feasibility of using molecular dynamics simulations with advanced sampling techniques to investigate extractant conformational energetics in solution, which, more broadly, will enable extractant design that accounts for entropic effects and explicit solvation.
Molecular and oligomeric amines supported in porous oxide supports are a promising class of CO 2 sorbent materials studied for CO 2 removal from diverse streams such as flue gas and ambient air. Among the various amines investigated, low molecular weight, hyperbranched poly(ethyleneimine) (PEI), and tetraethylenepentamine (TEPA) are among the most extensively studied. While macroscopic structure–performance relationships relating the support structure, amine loading, and other factors affecting CO 2 sorption capacities and kinetics have been developed, structural and dynamic information about the organic amine phase in the porous support is less plentiful. The structure and mobility of amines impregnated in the pores of porous supports directly impact gas sorption, as the accessibility of amine sites in the pores directly relates to amine distribution in the pores and overall pore filling as well as the dynamics of the amine chains. Here, we prepare a family of mesoporous silica SBA-15 materials containing varying loadings of oligomeric (PEI) and molecular (TEPA) amines. 1 H T 1 –T 2 relaxation correlation solid-state NMR experiments are used to characterize the structural and dynamic properties of the confined amines. Both TEPA and PEI are shown to form multiple different domains in the pores, each with distinguishable dynamic properties. TEPA and PEI form more rigid layers around the silica support walls at lower organic loading fractions, characterized by lower mobilities, followed by the formation of more mobile domains less engaged in pore wall interactions at higher loadings. TEPA shows faster mobilities than PEI because of its lower molecular weight. TEPA also appears to more easily transfer between domains within the pores, leading to generally faster CO 2 uptake rates with higher sorption capacities, while PEI located closer to the pore walls remained much less mobile and is thus less engaged in CO 2 capture.
In this paper, we explore the potential of generative machine learning models as an alternative to the computationally expensive Monte Carlo (MC) simulations commonly used by the Large Hadron Collider (LHC) experiments. Our objective is to develop a generative model capable of efficiently simulating detector responses for specific particle observables, focusing on the correlations between detector responses of different particles in the same event and accommodating asymmetric detector responses. Here, we present a conditional normalizing flow model ($\mathcal{CNF}$) based on a chain of Masked Autoregressive Flows, which effectively incorporates conditional variables and models high-dimensional density distributions. We assess the performance of the $\mathcal{CNF}$ model using a simulated sample of Higgs boson decaying to diphoton events at the LHC. We create reconstruction-level observables using a smearing technique. We show that conditional normalizing flows can accurately model complex detector responses and their correlation. This method can potentially reduce the computational burden associated with generating large numbers of simulated events while ensuring that the generated events meet the requirements for data analyses. We make our code available at https://github.com/allixu/normalizing_flow_for_detector_response
An important aspect of how viruses spread and infect is the viral burst size, or the number of new viruses produced by each infected cell. Surprisingly, this value remains poorly characterized for influenza A virus (IAV), commonly known as the flu. In this study, we screened tens of thousands of cells using a microfluidic method called droplet quantitative PCR (dqPCR). The high-throughput capability of dqPCR enabled the measurement of a large population of infected cells producing progeny virus. By measuring the fully assembled and successfully released viruses from these infected cells, we discover that the viral burst sizes for both the seasonal H3N2 and the 2009 pandemic H1N1 strains vary significantly, with H3N2 ranging from 10 1 to 10 4 viruses per cell, and H1N1 ranging from 10 1 to 10 3 viruses per cell. Some infected cells produce average numbers of new viruses, while others generate extensive number of viruses. In fact, we find that only 10% of the single-cell infections are responsible for creating a significant portion of all the viruses. This small fraction produced approximately 60% of new viruses for H3N2 and 40% for H1N1. On average, each infected cell of the H3N2 flu strain produced 709 new viruses, whereas for H1N1, each infected cell produced 358 viruses. This novel method reveals insights into the flu virus and can lead to improved strategies for managing and preventing the spread of viruses.
Diffusion models (DMs) are a class of generative machine learning methods that sample a target distribution by transforming samples of a trivial (often Gaussian) distribution using a learned stochastic differential equation. In standard DMs, this is done by learning a “score function” that reverses the effect of adding diffusive noise to the distribution of interest. Here we consider the generalisation of DMs to lattice systems with discrete degrees of freedom, and where noise is added via Markov chain jump dynamics. We show how to use tensor networks (TNs) to efficiently define and sample such “discrete diffusion models” (DDMs) without explicitly having to solve a stochastic differential equation. We show the following: (i) by parametrising the data and evolution operators as TNs, the denoising dynamics can be represented exactly; (ii) the auto-regressive nature of TNs allows to generate samples efficiently and without bias; (iii) for sampling Boltzmann-like distributions, TNs allow to construct an efficient learning scheme that integrates well with Monte Carlo. We illustrate this approach to study the equilibrium of two models with non-trivial thermodynamics, the d = 1 constrained Fredkin chain and the d = 2 Ising model. Published by the American Physical Society 2025
The cost-effective and sustainable deployment of hydrogen supply and demand networks, especially in large economic regions like California, can be challenging considering the spatial-temporal availability and variability of the different actors across the network such as production processes, distribution modes, and end-users. In this presentation, we will provide an overview and demonstration of a modeling framework used to assess the environmental, economic, and human health impacts of plausible hydrogen production and supply chain networks in California. Scenarios focus on green hydrogen production pathways using water electrolysis and biomass gasification. End-use applications included in the model are transit, medium and heavy-duty trucking, port authorities, and power and aviation companies that currently consume natural gas, diesel, and aviation fuel for their day-to-day operation. Representative locations for hydrogen production and end-use are based on recent projections of the hydrogen economy in California. All mass and energy flows, as well as estimated emissions, are based on H2A process model designs and projections of technology performance, literature review, and LBNL process, economic and life cycle modeling, and not on company data for the sake of this presentation. Human health impacts are included following methodologies developed for the University of California Irvine HyDeal project. Life cycle phases associated with hydrogen production include feedstock preparation (water and biomass), energy production and consumption (renewable, grid, and combination of renewable and grid electricity), maintenance (chemical utilization in electrolysis and natural gas combustion in gasification), carbon sequestration, hydrogen storage (compression and liquefaction), and distribution (truck and pipeline). We apply the framework utilizing California specific emission factors, financial data, and human health damages and explore the impact of network characteristics on results. Example variations include: the inclusion of policy incentives or not, different representations of the electricity grid and source, electrolysis versus gasification versus combinations of both for production, liquefaction versus compression based on producer capacity cutoffs, transportation truck versus pipeline based on existing infrastructure, and ultimate end use. Comparison of these different scenarios can help inform future projects by demonstrating the trade-offs among environmental, economic, and human health impacts. This model, automated in R, is a starting platform upon which new analysis, modeling capabilities, locations, and emission factors can be rapidly tested and integrated.
While Li-ion batteries (LIBs) are a leading energy storage technology, their energy densities are limited by the low capacity of conventional intercalation cathodes, driving interest in high energy-density Li–S batteries that make use of conversion chemistry. Achieving high capacity, reversibility, and cycle stability, and controlling volume changes in conversion batteries during the charge–discharge process, however, remains challenging. Here, in this study, we present a porous, amorphous, sulfide-based Mn x Mo 3 S 13 chalcogel, which concurrently offers high capacity and cycle stability. The solution-processable room temperature synthesized Mn x Mo 3 S 13 (x = 0.25) chalcogel exhibits a local structure that resembles the Mo 3 S 13 cluster with Mn 2+ distributed across the Mo 3 S 13 matrix, as determined by synchrotron X-ray pair distribution function (PDF) and extended X-ray absorption fine structure (EXAFS). Ab initio molecular dynamics (AIMD) simulations reveal that Mn 2+ incorporation shortens the polysulfide chain in the gel matrix compared to the Mo 3 S 13 chalcogel, while forming a coordination environment with disulfide groups, analogous to the experimental findings. A Li/Mn 0.25 Mo 3 S 13 half-cell delivers 897 mAh g –1 capacity during the first discharge and retains 571 mAh g –1 capacity after 100 cycles at a C/3 rate. Distribution of relaxation time (DRT) unveils a stable solid–electrolyte interphase (SEI) formation upon cycling that enables charge–discharge reversibility. Here, the enhanced capacity retention and cycle stability compared to those of the Li/Mo 3 S 13 cell are attributed to the reduced dissolution of active mass into the electrolyte, facilitated by the formation of shorter polysulfide chains within the Mn 0.25 Mo 3 S 13 structure and the strong affinity of Lewis-acidic Mn 2+ for polysulfide anions generated during the charge–discharge process of the Li/Mn 0.25 Mo 3 S 13 cell. Thus, this work illustrates a design principle of material for high-capacity and cycle-stable Li-metal sulfide batteries.
Abstract We report on the variations of the physical parameters of the jet observed in the blazar Mrk 421, and discuss the origin of X-ray flares in the jet, based on analysis of several spectral energy distributions (SEDs). The SEDs are modeled using the one-zone synchrotron self-Compton model, its parameters determined using a Markov chain Monte Carlo method. The lack of data at TeV energies means many of the parameters cannot be uniquely determined and are correlated. These are studied in detail. We find that the optimal solution can be uniquely determined only when we apply a constraint to one of four parameters: the magnetic field (B), the Doppler factor, the size of the emitting region, and the normalization factor of the electron energy distribution. We used 31 sets of SEDs from 2009 to 2014 with optical–UV data observed with UVOT/Swift and the Kanata telescope, X-ray data with XRT/Swift, and γ-ray data with the Fermi Large Area Telescope. The result of our SED analysis suggests that, in the X-ray faint state, the emission occurs in a relatively small area (∼1016 cm) with a relatively strong magnetic field (B ∼ 10−1 G). The X-ray bright state shows a tendency opposite to that of the faint state, that is, a large emitting area (∼1018 cm), probably downstream of the jet, and a weak magnetic field (B ∼ 10−3 G). The high X-ray flux was due to an increase in the maximum energy of electrons. On the other hand, the presence of two kinds of emitting areas implies that the one-zone model is unsuitable for reproducing at least part of the observed SEDs.
We report on the first experimental measurements made at a magnetic confinement fusion device of the tritium(T)-tritium(T) reaction T + T → He 4 + 2 n indicating the presence of the intermediate two-body resonant reaction T + T → He 5 + n . During the second deuterium-tritium campaign (DTE2) at the Joint European Torus, measurements of fusion plasmas with high tritium concentrations, n T / ( n T + n D ) ≈ 0.99 , heated with tritium neutral beam injection, were performed using the neutron time-of-flight (TOF) spectrometer TOFOR. We detect a peak in the neutron emission TOF spectrum consistent with the two-body resonant reaction. The TT neutron emission energy spectrum is modeled using an R -matrix framework where the distributions of the most likely model parameters given our experimental TOF data are determined utilizing a Markov chain Monte Carlo approach. We compare our best estimate of the T + T neutron emission energy spectrum with results obtained at inertial confinement fusion experiments at the OMEGA facility and find a spectral shape that is consistent with the energy dependency in the neutron spectrum observed at OMEGA. Published by the American Physical Society 2024
A fundamental challenge in Bayesian inference is efficient representation of a target distribution. Many nonparametric approaches do so by sampling a large number of points using variants of Markov chain Monte Carlo (MCMC). Here, we propose an MCMC variant that retains only those posterior samples which exceed a kernelized Stein discrepancy (KSD) threshold, which we call KSD thinning. We establish the convergence and complexity trade-offs for several settings of KSD thinning as a function of the KSD threshold parameter, sample size, and other problem parameters. We provide experimental comparisons against other online nonparametric Bayesian methods that generate low-complexity posterior representations. We observe superior consistency/complexity trade-offs across a range of settings including MCMC sampling on two Bayesian inference problems from the biological sciences, and 10 × inference speedup and storage reduction for Bayesian neural networks with no loss of accuracy and no increase in training time. Our code is available at https://github.com/colehawkins/KSD-Thinning.
This report describes a novel fault graph modeling language and a simulation tool for executing models specified in the language. The modeling language has three primary features that distinguish it from similar reliability analysis tools. These are (1) a random variable modeling several distinct outcomes of a single fault; (2) chains of faults in which one fault triggers another; and (3) time to fail sampled from probability distributions including positive normal, exponential, Weibull with a minimum, or immediate. These features are motivated by their use in a historical analysis of centrifuge reliability.
The Monte Carlo method delivers an unbiased estimate of the probability of failure. However, the variance of the estimate depends on the number of evaluated samples. This number must be very large for estimations of a low probability of failure. If the evaluation of each sample is computationally expensive, the crude Monte Carlo simulation strategy is impracticable. Therefore, subset simulations are used to reduce the required number of evaluations. Subset simulations require a Markov Chain Monte Carlo sampler, such as the random walk Metropolis-Hastings algorithm. The algorithm, however, struggles with sampling in low-probability regions, especially if they are narrow. As a consequence, advanced Markov Chain Monte Carlo simulations have been developed. In particular, the Hamiltonian Monte Carlo method explores the target distribution rapidly. Driven by the idea of Hamiltonian dynamics, this sampler provides a non-random walk through the target distribution. The incorporation of subset simulation and Hamiltonian Monte Carlo methods has shown promising results for reliability analysis. One downside of the Hamiltonian Monte Carlo method is that gradient evaluations are computationally expensive, especially when dealing with high-dimensional problems and evaluating long trajectories. We show that integrating Hamiltonian neural networks in Hamiltonian Monte Carlo simulations significantly speeds up the sampling task. Furthermore, the enhancement of adaptive trajectory length within the Hamiltonian Monte Carlo results in the efficient proposal of the following states. Based on this recent enhancement, we provide a fast sampling strategy for subset simulations using Hamiltonian neural networks to replace the evaluation of the gradient and significantly speed up the Hamiltonian Monte Carlo simulation.
This software package offers a novel fault graph modeling language and a simulation tool for executing models specified in the language. The modeling language has three primary features that distinguish it from similar reliability analysis tools. These are (1) a random variable modeling several distinct outcomes of a single fault; (2) chains of faults in which one fault triggers another; and (3) time to fail sampled from probability distributions including positive normal, exponential, Weibull with a minimum, or immediate.
Ongoing progress in synthetic biology, metabolic engineering, and catalysis continues to produce a diverse array of advanced biofuels with complex molecular structure and functional groups. In order to integrate biofuels into existing combustion systems, and to optimize the design of next-generation combustion systems, understanding connections between molecular structure and ignition at low-temperature conditions (< 1000 K) remains a priority that is addressed in part using chemical kinetics modeling. The development of predictive models relies on detailed information, derived from experimental and theoretical studies, on molecular structure and chemical reactivity, both of which influence the balance of chain reactions that occur during combustion – propagation, termination, and branching. In broad context, three main categories of reactions affect ignition behavior: (i) initiation reactions that generate a distribution of organic radicals, ; (ii) competing unimolecular decomposition of and bimolecular reaction of with O2; (iii) decomposition mechanisms of peroxy radical adducts (RO), including isomerization via RO ⇌ OOH. Furthermore, all three categories are influenced by functional groups in different ways, which causes a shift in the balance of chain reactions that unfold over complex temperature- and pressure-dependent mechanisms.
Distributed wind is an electric energy resource segment with strong potential to be deployed in many applications, but special consideration of resilience and cybersecurity is needed to address the unique conditions associated with distributed wind. Distributed wind is a strong candidate to help meet renewable energy and carbon-free energy goals. However, care must be taken as more systems are installed to ensure that the systems are reliable, resilient, and secure. The physical and communications requirements for distributed wind mean that there are unique cybersecurity considerations, but there is little to no existing guidance on best practices for cybersecurity risk management for distributed wind systems specifically. This research develops an architecture for the consideration of cyber risks associated with distributed wind systems. The architecture takes into account the configurations, challenges, and standards for distributed wind to create a risk-focused perspective that considers of threats, vulnerabilities, and consequences, with special emphasis on what sets distributed wind systems apart from other distributed energy resources (DER). We discuss common distributed wind architectures and how they are interconnected to larger power systems. Because cybersecurity cannot exist independently, the cyber-resilience architecture must consider the system holistically. Finally, we discuss the implementation of a risk assessment process that uses the cyber-resilience framework to address challenges specific to distributed wind.
We report the combined experimental and theoretical study of the bulk self-assembly behavior of polystyrene-blockpoly( 2,3,4,5,6-pentafluorostyrene) diblock copolymers. These block copolymers were designed to create highly antagonistic blocks (with a high Flory−Huggins interaction parameter, χ) with minimum disruption to the molecular construct (i.e., only replacing five hydrogen atoms with five fluorine atoms). A large library of diblock copolymers (41 samples) was synthesized by reversible addition− fragmentation chain transfer (RAFT) polymerization to map out a major portion of the phase space. All block copolymers exhibited narrow molecular weight distributions with dispersity (D) values between 1.07 and 1.32, and subsequent thermal annealing revealed phase separation into well-defined nanoscale morphologies depending on their molecular composition, as determined from small-angle X-ray scattering and transmission electron microscopy analyses, with an experimental phase diagram being constructed. The χ value at 25 °C for this block copolymer was estimated to be 0.2 using strong segregation theory, based on trends in phase-separated domain spacing and interfacial width. When applying theoretical approaches, the majority of the domain spacing data trends were captured by a coil−coil diblock copolymer model; however, a better fit to the data for samples with shorter fluorinated blocks was obtained with a rod−coil model, indicating that the chains in these fluorinated blocks likely have a higher inherent stiffness and were thus rod-like. This observation demonstrates that, due to the very high value of χ, a transition from coil−coil to rod−coil behavior can be obtained purely by reducing the length of the stiffer of the two blocks and without varying temperature or the chemical composition of the polymers. Here, this work showcases the presence of strong microphase separation within AB diblock copolymers despite the relatively similar chemical composition of the constituent “A” and “B” units, with a clear transition from rod−coil to coil−coil segregation behavior.
A class of extensions to the Standard Model adds hypothetical long-lived particles (LLPs) that have mass- or kinetic-mixing with neutral mesons, such as pions or rho mesons. The mixing can contribute significantly to the production of LLPs at proton accelerator experiments, and no consistent description of these production modes exists in the literature. In this paper, we develop a framework for studying different LLPs—dark photons, vector mediators coupled to the baryon current, and axionlike particles with different coupling patterns. In particular, we implement the production mechanisms in PYTHIA 8, study how the overall flux and kinematic distributions depend on the LLP’s mass, and compare various subprocesses where the mixing contributes—proton bremsstrahlung, meson decay, and production in the fragmentation chain. We find that our new description of LLP production predicts an integrated flux that differs from current approaches by one to two orders of magnitude, and highlight the unavoidable theoretical uncertainties coming from poor knowledge of the properties of heavy mesons.