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.
Equilibrium is determined by finding chemical composition with minimum free energy. Method of steepest descent is applied to quadratic representation of free-energy surface. Solution is initiated by selecting arbitrary set of mole fractions, from which point on free-energy surface is computed.
Machine-learning function representations such as neural networks have proven to be excellent constructs for constitutive modeling due to their flexibility to represent highly nonlinear data and their ability to incorporate constitutive constraints, which also allows them to generalize well to unseen data. Here, in this work, we extend a polyconvex hyperelastic neural network framework to (isotropic) thermo-hyperelasticity by specifying the thermodynamic and material theoretic requirements for an expansion of the Helmholtz free energy expressed in terms of deformation invariants and temperature. Different formulations which a priori ensure polyconvexity with respect to deformation and concavity with respect to temperature are proposed and discussed. The physics-augmented neural networks are furthermore calibrated with a recently proposed sparsification algorithm that not only aims to fit the training data but also penalizes the number of active parameters, which prevents overfitting in the low data regime and promotes generalization. The performance of the proposed framework is demonstrated on synthetic data, which illustrate the expected thermomechanical phenomena, and existing temperature-dependent uniaxial tension and tension-torsion experimental datasets.
Here we introduce a general, variational scheme for systematic approximation of a given Kohn-Sham free-energy functional by partitioning the density matrix into distinct spectral domains, each of which may be spanned by an independent diagonal representation without requirement of mutual orthogonality. It is shown that by generalizing the entropic contribution to the free energy to allow for independent representations in each spectral domain, the free energy becomes an upper bound to the exact (unpartitioned) Kohn-Sham free energy, attaining this limit as the representations approach Kohn-Sham eigenfunctions. A numerical procedure is devised for calculation of the generalized entropy associated with spectral partitioning of the density matrix. The result is a powerful framework for Kohn-Sham calculations of systems whose occupied subspaces span multiple energy regimes. As a case in point, we apply the proposed framework to warm- and hot-dense matter described by finite-temperature density functional theory, where at high energies the density matrix is represented by that of the free-electron gas, while at low energies it is variationally optimized. We derive expressions for the spectral-partitioned Kohn-Sham Hamiltonian, atomic forces, and macroscopic stresses within the projector-augmented wave (PAW) and the norm-conserving pseudopotential methods. It is demonstrated that at high temperatures, spectral partitioning facilitates accurate calculations at dramatically reduced computational cost. Moreover, as temperature is increased, fewer exact Kohn-Sham states are required for a given accuracy, leading to further reductions in computational cost. Finally, it is shown that standard multiprojector expansions of electronic orbitals within atomic spheres in the PAW method lack sufficient completeness at high temperatures. Spectral partitioning provides a systematic solution for this fundamental problem.
Molecular simulations of biological and physical phenomena generally involve sampling complicated, rough energy landscapes characterized by multiple local minima. In this work, we introduce a new family of methods for advanced sampling that draw inspiration from functional representations used in machine learning and approximation theory. As shown here, such representations are particularly well suited for learning free energies using artificial neural networks. As a system evolves through phase space, the proposed methods gradually build a model for the free energy as a function of one or more collective variables, from both the frequency of visits to distinct states and generalized force estimates corresponding to such states. Implementation of the methods is relatively simple and, more importantly, for the representative examples considered in this work, they provide computational efficiency gains of up to several orders of magnitude over other widely used simulation techniques.
Sharpening the correspondence of Jackiw-Teitelboim (JT) gravity and its dual matrix model description at a finite radial cutoff λ through the $T\overline{T}$ deformation is of interest. To proceed, we simplify the problem by considering the Airy model and deform Airy correlators in the same way as in $T\overline{T}$ -deformed JT gravity. We use those correlators to compute the annealed and quenched free energies for both λ > 0 and λ < 0 from an integral representation of the replica trick. At the leading order in λ and low temperatures, we confirm that the genus-zero quenched free energy monotonically decreases as a function of temperature when perturbation theory is valid. We then study the all-genus quenched free energy at low temperatures, where we discover and discuss subtleties due to non-perturbative effects in the Airy model, as well as the contributions from the non-perturbative branch under the $T\overline{T}$ deformation.
Computing accurate reaction rates is a central challenge in computational chemistry and biology because of the high cost of free energy estimation with unbiased molecular dynamics. In this work, a data-driven machine learning algorithm is devised to learn collective variables with a multitask neural network, where a common upstream part reduces the high dimensionality of atomic configurations to a low dimensional latent space and separate downstream parts map the latent space to predictions of basin class labels and potential energies. Here, the resulting latent space is shown to be an effective low-dimensional representation, capturing the reaction progress and guiding effective umbrella sampling to obtain accurate free energy landscapes. This approach is successfully applied to model systems including a 5D Müller Brown model, a 5D three-well model, the alanine dipeptide in vacuum, and an Au(110) surface reconstruction unit reaction. It enables automated dimensionality reduction for energy controlled reactions in complex systems, offers a unified and data-efficient framework that can be trained with limited data, and outperforms single-task learning approaches, including autoencoders.
In this report the feasibility of a site-selective hydration strategy that enables site-selective atomic layer deposition (ALD) is investigated among four rutile TiO 2 facets [(110), (100), (101) and (001)] and their most prevalent step edges. First-principles simulations of asymmetric slab models were utilized to create accurate representations of pristine terrace and step edge sites. The adsorption free energies for molecular and dissociative adsorption of H 2 O were calculated to evaluate this strategy as a viable route to step edge selectivity. We predict that selective hydroxylation is possible on the 110 and 001 step edges and further computationally evaluate three metalorganic ALD precursors for their compatibility with the selective hydration strategy. Experimental evidence for delayed nucleation of ALD on rutile (001), (110), and (100) TiO2 single crystals corroborates predictions of the dehydration of the surface and suggests the possibility of site-selective ALD.
The development of an understanding of the chemistry of mixtures of metallic hydrogen and abundant, higher-z material such as oxygen, carbon, etc., is important for understanding of fundamental processes of energy release, differentiation, and development of atmospheric abundances in the Jovian planets. It provides a significant theoretical base for the interpretation of atmospheric elemental abundances to be provided by atmospheric entry probes in coming years. Significant differences are found when non-perturbative approaches such as Thomas-Fermi-Dirac (TFD) theory are used. Mapping of the phase diagrams of such binary mixtures in the pressure range from approx. 10 Mbar to approx. 1000 Mbar, using results from three-dimensional TFD calculations is undertaken. Derivation of a general and flexible thermodynamic model for such binary mixtures in the relevant pressure range was facilitated by the following breakthrough: there exists an accurate nd fairly simple thermodynamic representation of a liquid two-component plasma (TCP) in which the Helmholtz free energy is represented as a suitable linear combination of terms dependent only on density and terms which depend only on the ion coupling parameter. It is found that the crystal energies of mixtures of H-He, H-C, and H-O can be satisfactorily reproduced by the same type of model, except that an effective, density-dependent ionic charge must be used in place of the actual total ionic charge.
Type IIB AdS6 solutions with orientifold 7-planes are constructed. The geometry is a warped product of AdS 6 and S 2 over a Riemann surface Σ and the O7-planes correspond to a particular type of puncture on Σ. The solutions are identified as near-horizon limits of (p,q) 5-brane webs with O7-planes. The dual 5d SCFTs have relevant deformations to linear quiver gauge theories which have SO(•) or USp(•) nodes or SU(•) nodes with hypermultiplets in symmetric or antisymmetric representations, in addition to SU(•) nodes with fundamental hypermultiplets. The S 5 free energies are obtained holographically and matched to field theory computations using supersymmetric localization to support the proposed dualities.
As a continuation of the study (in arXiv:2102.07696 and arXiv:2104.12625 ) of strong-coupling expansion of non-planar corrections in \( \mathcal{N} \) = 2 4d superconformal models we consider two special theories with gauge groups SU( N ) and Sp(2 N ). They contain N -independent numbers of hypermultiplets in rank 2 antisymmetric and fundamental representations and are planar-equivalent to the corresponding \( \mathcal{N} \) = 4 SYM theories. These \( \mathcal{N} \) = 2 theories can be realised on a system of N D3-branes with a finite number of D7-branes and O7-plane; the dual string theories should be particular orientifolds of AdS 5 × S 5 superstring. Starting with the localization matrix model representation for the \( \mathcal{N} \) = 2 partition function on S 4 we find exact differential relations between the 1 /N terms in the corresponding free energy F and the \( \frac{1}{2} \) -BPS Wilson loop expectation value \( \left\langle \mathcal{W}\right\rangle \) and also compute their large ’t Hooft coupling ( λ » 1) expansions. The structure of these expansions is different from the previously studied models without fundamental hypermultiplets. In the more tractable Sp(2 N ) case we find an exact resummed expression for the leading strong coupling terms at each order in the 1 /N expansion. We also determine the exponentially suppressed at large λ contributions to the non-planar corrections to F and \( \left\langle \mathcal{W}\right\rangle \) and comment on their resurgence properties. We discuss dual string theory interpretation of these strong coupling expansions.
The high-quality Fermi LAT observations of gamma-ray pulsars have opened a new window to understanding the generation mechanisms of high-energy emission from these systems, The high statistics allow for careful modeling of the light curve features as well as for phase resolved spectral modeling. We modeled the LAT light curves of the Vela and CTA I pulsars with simulated high-energy light curves generated from geometrical representations of the outer gap and slot gap emission models. within the vacuum retarded dipole and force-free fields. A Markov Chain Monte Carlo maximum likelihood method was used to explore the phase space of the magnetic inclination angle, viewing angle. maximum emission radius, and gap width. We also used the measured spectral cutoff energies to estimate the accelerating parallel electric field dependence on radius. under the assumptions that the high-energy emission is dominated by curvature radiation and the geometry (radius of emission and minimum radius of curvature of the magnetic field lines) is determined by the best fitting light curves for each model. We find that light curves from the vacuum field more closely match the observed light curves and multiwavelength constraints, and that the calculated parallel electric field can place additional constraints on the emission geometry
A new method for computing anharmonic thermophysical properties for adsorbates on metal surfaces is presented. Classical Monte Carlo phase space integration is performed to calculate the partition function for the motion of a hydrogen atom on Cu(111). Here, a minima-preserving neural network potential energy surface is used within the integration routine. Two different sampling schema for generating the training data are presented, and two different density functionals are used. The results are benchmarked against direct state counting results by using discrete variable representation. The phase space integration results are in excellent quantitative agreement with the benchmark results. Additionally, both the discrete variable representation and the phase space integration results confirm that the motion of H on Cu(111) is highly anharmonic. The results were applied to calculate the free energy of dissociative adsorption of H 2 and the resulting Langmuir isotherms at 400, 800, and 1200 K in a partial pressure range of 0–1 bar. It shows that the anharmonic effects lead to significantly higher predicted surface site fractions of hydrogen.
The idea of "hyperspace" is suggested as a possible approach to faster-than-light (FTL) motion. A brief summary of a 1986 study on the Euclidean representation of space-time by the author is presented. Some new calculations on the relativistic momentum and energy of a free particle in Euclidean "hyperspace" are now added and discussed. The superimposed Energy-Momentum curves for subluminal particles, tachyons, and particles in Euclidean "hyperspace" are presented. It is shown that in Euclidean "hyperspace", instead of a relativistic time dilation there is a time "compression" effect. Some fundamental questions are presented,
Traditional methods for identifying “hit” molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind.
Dissolved oxygen (DO), total organic carbon (TOC), total nitrogen (TN), molecular data for organic matter, and biochemical reactions for surface water and porewater (i.e., hyporheic zone) collected from a water recirculating flume located at the University of Texas, Austin. The flume contained real river water from Lower Colorado River(Austin, TX) and clean sand. Hyporheic exchange in the flume was induced through The study aims to understand relationships between aerobic metabolism of organic matter and molecular characteristics of organic matter, such as thermodynamic signature and nitrogen content, through the extent of the hyporheic zone at 10 cm- resolution, and through time. During the experiment, organic matter (dry leaves) was added to the flume and removed after 24 hours. The water samples were collected before the addition of leaves, at the time of removal of leaves, and at hour 72. The water samples were analyzed using ultrahigh resolution Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS) and total organic carbon (TOC) and total nitrogen (TN) analysis. Dissolved oxygen content throughout the surface water and the hyporheic zone of the flume was measured with a large planar optode. This data package is associated with the publication ’ Rethinking Aerobic Respiration in the Hyporheic Zone Under Variation in Carbon and Nitrogen Stoichiometry’ published in Environmental Science and Technology (Turețcaia et al., 2023 https://doi.org/10.1021/acs.est.3c04765). The dataset is comprised of five folders (1) Diss_O2_pic, (2) input_files (3) output_files; (4) python_code; and (5) R_code . Diss_O2_pic contains siximages of dissolved oxygen distribution in a bedform at hours 0, 24, and 72 of the experiment conducted in a large recirculation flume. Images are in separate R and G channels (i.e., RGB). The input_files contains (1) a csv file with FTICR peaks identified within each sample, (2) a csv file with molecular information pertinent to FTICR data with Gibbs free energy calculations adjusted for environmental temperature, (3) a csv file containing concentrations of non-purgeable organic carbon measured throughout the experiment , (4) a csv file containing concentrations of total nitrogen measured throughout the experiment, (5) a csv file containing total biochemical reactions (i.e., transformations) identified in the dataset, (6) a csv containing transformation profiles, and (7) a csv file containing transformations with formulas, and (8) a jpg file with schematic representation of locations for sample collection. The output_files contains (1) and xlsx file containing percent biochemical reactions containing nitrogen identified across all 39 sample, (2) a csv file of merged FTICR data and molecular information files, (3) a csv files containing average Gibbs free energy within sampling domains and at each sampling location, (4) a csv file with average concentrations of dissolved oxygen across sampling locations at hour 0, (5) a csv file with average concentrations of dissolved oxygen across sampling locations at hour 24, (6) a csv file with average concentrations of dissolved oxygen across sampling locations at hour 72, (7) a csv file with percent chemical classes identified across sampling locations at hour 0, (8) a csv file with percent chemical classes identified across sampling locations at hour 24, (9) a csv file with percent chemical classes identified across sampling locations at hour 72, and (10) a csv file containing percent nitrogen containing biochemical reactions identified across sampling locations at hours 0, 24, and 72. The python_code contains seven ipynb files which are Jupyter Notebooks used for data analysis and figures generation. The R_code contains 3 R files with R code used for data analysis and figures generation. This data package contains the processed data used in the associated manuscript. This data has not been previously published.
Kohn–Sham density functional theory calculations using conventional diagonalization based methods become increasingly expensive as temperature increases due to the need to compute increasing numbers of partially occupied states. In this work, we present a density matrix based method for Kohn–Sham calculations at high temperatures that eliminates the need for diagonalization entirely, thus reducing the cost of such calculations significantly. Specifically, we develop real-space expressions for the electron density, electronic free energy, Hellmann–Feynman forces, and Hellmann–Feynman stress tensor in terms of an orthonormal auxiliary orbital basis and its density kernel transform, the density kernel being the matrix representation of the density operator in the auxiliary basis. Using Chebyshev filtering to generate the auxiliary basis, we next develop an approach akin to Clenshaw–Curtis spectral quadrature to calculate the individual columns of the density kernel based on the Fermi operator expansion in Chebyshev polynomials and employ a similar approach to evaluate band structure and entropic energy components. We implement the proposed formulation in the SPARC electronic structure code, using which we show systematic convergence of the aforementioned quantities to exact diagonalization results, and obtain significant speedups relative to conventional diagonalization based methods. Finally, we employ the new method to compute the self-diffusion coefficient and viscosity of aluminum at 116 045 K from Kohn–Sham quantum molecular dynamics, where we find agreement with previous more approximate orbital-free density functional methods.
The description of strongly correlated systems interacting with quantized cavity modes poses significant theoretical challenges due to the combinatorial scaling of electronic and photonic degrees of freedom. Recent advances addressing this complexity include cavity quantum electrodynamics (QED) generalizations of complete active space configuration interaction and density matrix renormalization group methods. In this work, we introduce a QED extension of state-averaged complete active space self-consistent field theory, which incorporates cavity-induced correlations through a second-order orbital optimization framework with robust convergence properties. The method is implemented using both photon number state and coherent state representations, with the latter showing robust origin invariance in the energies regardless of the completeness of the photonic Fock space. The implementation enables symmetry-free orbital relaxations to account for photon-mediated symmetry breaking in polaritonic systems. Numerical validation on lithium hydride, hydroxide anion, and magnesium hydride cation demonstrates that this method achieves significantly improved accuracy in modeling ground-state and polariton potential energy surfaces compared to QED-CASCI in a fixed orbital basis. In these studies, we reach sub-kcal/mol accuracy in potential energy surface in much smaller active spaces than are required for QED-CASCI. This advancement provides a more robust approach for studying cavity-altered chemical landscapes for ground and exited strongly coupled systems.
The solar corona routinely exhibits explosive activity, in particular coronal mass ejections and their accompanying eruptive ares, that have global-scale consequences. These events and their smaller counterparts, coronal jets, originate in narrow, sinuous lament channels. The key processes that form and evolve the channels operate on still smaller spatial scales and much longer time scales, culminating in a vast separation of characteristic lengths and times that govern these explosive phenomena. In this article, we describe implementation and tests of an efficient subgrid-scale model for generating eruptive structures in magnetohydrodynamics (MHD) coronal simulations. STITCH {STatistical InjecTion of Condensed Helicity { is a physics-based, reduced representation of helicity condensation: a process wherein small-scale vortical surface convection forms ubiquitous current sheets, and pervasive reconnection across the sheets mediates an inverse cascade of magnetic helicity and free energy, thereby forming the lament channels. We have developed a formalism, STITCH, that abstracts these complex processes into a single term in Ohm's law and the induction equation that directly injects tangential magnetic flux into the low corona. We show that our approach is in very good agreement with a full helicity-condensation calculation that treats all of the dynamics explicitly, while enabling substantial reductions in temporal duration and spatial resolution. In addition, we illustrate the flexibility of STITCH at forming localized lament channels and at energizing complex surface flux distributions that have sinuous boundaries. STITCH is simple to implement and computationally ecient, making it a powerful technique for physics-based modeling of solar eruptive events.