Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “local density approximations”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
NASA NTRS1995

Epsilon Canis Majoris and the ionization of the local cloud

The Lyman continuum radiation from the brightest extreme ultraviolet (EUV) source, the B2 II star epsilon Canis Majoris (Adara), is so intense that it dominates the local stellar EUV radiation field at wavelengths longer than 450 A and therefore sets a lower limit to the ionization of hydrogen in the Local Cloud. Using the EUV (70-730 A) spectrum of epsilon CMa taken with the Extreme Ultraviolet Explorer Satellite (EUVE) and simple models that extrapolate this spectrum to the Lyman edge at 912 A, we have determined the local interstellar hydrogen photionizatin parameter Gamma solely from epsilon CMa to be 1.1 x 10(exp -15)/s. This fiugre is a factor of 7 greater than previous estimates of Gamma calculated for all nearby stars combined (Bruhweiler & Cheng 1988). Using measured values of the density and temperature of neutral interstellar hydrogen gas in the Local Cloud, we derive a particle density of ionized hydrogen n(H(+)) and electrons n(sub e) of 0.015-0.019/cu cm assuming ionization equilibrium and a helium ionization fraction of less than 20%. These values correspond to a hydrogen ionizatin fraction, chi(sub H) from 19% to 15%, respectively. The range of these derived quantities is due to the uncertainties in the local values of the neutral hydrogen and helium interstellar densities derived from both (1) solar backscatter measurements of Ly alpha lines of hydrogen and helium (1216 and 584 A), and (2) the average neutral densities along the line of sight to nearby stars. The local proton density produced by epsilon CMa is enough to allow the ionization mechanism of Ripken & Fahr (1983) to work at the heliopause and explain the discrepancy between the neutral hydrogen density derived from solar backscatter measurements and line-of-sight averages to nearby stars. A large value of electron density in the Local Cloud of n(sub e) is approximately 0.3-0.7/cu cm (T = 7000 K) has recently been reported by Lallement et al. (1994) using observations of Mg II and Mg I toward Sirius A. We show that if such a high value exists, it cannot result from the EUV stellar radiation field and, therefore, must be due to a strong diffuse source of EUV radiation.

Vallerga, J. V.↗
NASA NTRS1995

A localized excess of diffuse gamma radiation

Using archival Energetic Gamma Ray Experiment Telescope (EGRET) gamma-ray data and atomic hydrogen (H I) column densities derived from 21 cm radio observations, we have found a large irregular region in the northern Galactic hemisphere extending from (l approximately 90 deg, b approximately 52 deg) to (l approximately 45 deg, b approximately 77 deg) with a significant enhancement in the gamma-ray emissivity compared to the surrounding sky. The region contains no previously identified gamma-ray point sources. The emission may arise from a localized enhancement in cosmic-ray density or from the presence of matter other than H I. If the emission is due to unseen matter, a column density enhancement equivalent to approximately 2 x 10(exp 20) H-atoms/sq cm is required.

Chen, A.↗
NASA NTRS2017

Feeding and Feedback in the Powerful Radio Galaxy 3C 120

We present a spectral analysis of a 200-kilosecond observation of the broad-line radio galaxy 3C 120, performed with the high-energy transmission grating spectrometer on board the Chandra X-Ray Observatory. We find (i) a neutral absorption component intrinsic to the source with a column density of log N (sub H) equals 20.67 plus or minus 0.05 square centimeters; (ii) no evidence for a warm absorber (WA) with an upper limit on the column density of just log N (sub H) less than 19.7 square centimeters, assuming the typical ionization parameter log xi approximately equal to 2.5 ergs per second per centimeter; the WA may instead be replaced by (iii) a hot emitting gas with a temperature kT approximately equal to 0.7 kiloelectronvolts observed as soft X-ray emission from ionized Fe L-shell lines, which may originate from a kiloparsec-scale shocked bubble inflated by the active galactic nucleus (AGN) wind or jet with a shock velocity of about 1000 kilometers per second determined by the emission line width; (iv) a neutral Fe K alpha line and accompanying emission lines indicative of a Compton-thick cold reflector with a low reflection fraction R approximately equal to 0.2, suggesting a large opening angle of the torus; (v) a highly ionized Fe XXV emission feature indicative of photoionized gas with an ionization parameter log xi equal to 3.75 (sup plus 0.38) (sub minus 0.27) ergs per second per centimeter and a column density of log N (sub H) greater than 22 square centimeters localized within approximately 2 pc from the X-ray source; and (vi) possible signatures of a highly ionized disk wind. Together with previous evidence for intense molecular line emission, these results indicate that 3C 120 is likely a late-state merger undergoing strong AGN feedback.

black hole physics – galaxies: active – line: ↗
DOE OSTI2020

Phase‐Dependent Band Gap Engineering in Alloys of Metal‐Semiconductor Transition Metal Dichalcogenides

Abstract Bandgap engineering plays a critical role in optimizing the electrical, optical and (photo)‐electrochemical applications of semiconductors. Alloying has been a historically successful way of tuning bandgaps by making solid solutions of two isovalent semiconductors. In this work, a novel form of bandgap engineering involving alloying non‐isovalent cations in a 2D transition metal dichalcogenide (TMDC) is presented. By alloying semiconducting MoSe 2 with metallic NbSe 2 , two structural phases of Mo 0.5 Nb 0.5 Se 2 , the 1T and 2H phases, are produced each with emergent electronic structure. At room temperature, it is observed that the 1T and 2H phases are semiconducting and metallic, respectively. For the 1T structure, scanning tunneling microscopy/spectroscopy (STM/STS) is used to measure band gaps in the range of 0.42–0.58 at 77 K. Electron diffraction patterns of the 1T structure obtained at room temperature show the presence of a nearly commensurate charge density wave (NCCDW) phase with periodic lattice distortions that result in an uncommon 4 × 4 supercell, rotated approximately 4° from the lattice. Density‐functional‐theory calculations confirm that local distortions, such as those in a NCCDW, can open up a band gap in 1T ‐Mo 0.5 Nb 0.5 Se 2 , but not in the 2H phase. This work expands the boundaries of alloy‐based bandgap engineering by introducing a novel technique that facilitates CDW phases through alloying.

Wang, Shuxi↗
DOE OSTI2025

DONKEY: A Flexible and Accurate Algorithm for Clustering

We propose an accurate clustering algorithm suitable for the varied and multidimensional data sets that correspond to temporal snapshots from on-the-fly nonadiabatic trajectory-based simulations of photoexcited dynamics. The algorithm approximates the underlying probability density function using variable kernel density estimation, with local maxima corresponding to cluster centers. Each data point is then assigned to one of the maxima by employing a maximization procedure. Finally, clusters artificially separated by minor fluctuations in the probability density are merged. The algorithm does not require parameter tuning, which ensures flexibility and reduces the risk of bias. It is tested on several synthetic data sets, where it consistently outperforms conventional clustering algorithms. As a final example, the algorithm is applied to the excited dynamics of the norbornadiene ⇌ quadricyclane (C 7 H 8 ) molecular photoswitch, demonstrating how distinct reaction pathways can be identified.

algorithms↗
DOE OSTI2025

Charges on a suspended silicon nitride membrane under a high-energy electron beam

Thin silicon nitride (SiN x ) membranes are widely used in gas and liquid phase transmission electron microscopy (TEM) and as phase plates to enhance imaging contrast. SiN x contains trap sites for both positive and negative charges, which can be manipulated by high-energy electron irradiation, external potential biasing, or light exposure. Charge accumulation on the membrane can significantly affect in situ TEM processes, including chemical and electrochemical reactions, nanoparticle dynamics, and catalytic activity, or introduce unwanted phase shifts when used as a phase plate. Here, in this study, charge accumulation on suspended SiN x membranes was investigated using off-axis electron holography combined with model-free charge analysis, supported by custom finite element analysis (FEA) simulations. An average residual positive charge density of approximately 2.8 × 10 −4 C m −2 was measured. Localized and stable regions of both positive and negative charges were identified on the membrane. The global positive and localized positive/negative charges give rise to strong electric fields and electroosmotic slip velocities at the membrane surface, which are sufficient to induce non-Brownian particle behavior and directional fluid flow, offering a physical explanation for previously observed anomalies in particle dynamics, nucleation, and growth during gas and liquid phase TEM experiments. These results provide a benchmark for understanding charge behavior at SiNx interfaces in gas and liquid phase TEM. Furthermore, the FEA simulations establish a framework for future investigations into charge distribution, electrostatic potentials, and electrical double layers at solid–liquid interfaces, particularly in complex geometries and chemically dynamic environments.

Suspended silicon↗
DOE OSTI2024

Simulation of gas mixture dynamics in a pipeline network using explicit staggered-grid discretization

Here we develop an explicit staggered finite difference discretization scheme for simulating the transport of highly heterogeneous gas mixtures through pipeline networks. This study is motivated by the proposed blending of hydrogen into natural gas pipelines to reduce end use carbon emissions while using existing pipeline systems throughout their planned lifetimes. Our computational method accommodates an arbitrary number of constituent gases with very different physical properties that may be injected into a network with significant spatiotemporal variation. In this setting, the gas flow physics are highly location- and time- dependent, so that local composition and nodal mixing must be accounted for. The resulting conservation laws are formulated in terms of pressure, partial densities and flows, and volumetric and mass fractions of the constituents. We include non-ideal equations of state that employ linear approximations of gas compressibility factors, so that the pressure dynamics propagate locally according to a variable wave speed that depends on mixture composition and density. We derive compatibility relationships for network edge boundary values that are more complex than for a homogeneous gas. The simulation method is evaluated on initial boundary value problems for a single pipe and a small network, is cross-validated with a lumped element simulation, and used to demonstrate a local monitoring and control policy for maintaining allowable concentration levels.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1994

Explosions of infalling comets in Jupiter's atmosphere

In view of the expected collision of comet Shoemaker-Levy 9 (1993e) with Jupiter in 1994 July, we calculate basic properties of the initial interaction for a simplified Jovian atmosphere. The comet is expected to impact Jupiter at 60 km/sec and at an angle of 45 deg to the zenith. The shock wave generated by the bolide should be optically thick once it has penetrated to an atmospheric density approximately 10(exp -6) gr/cm(exp 3), and we calculate the post-shock conditions assuming local thermodynamic equilibrium (LTE) for shock velocities v(sub sh) in the range 10 to 60 km/sec and preshock densities rho(sub a) = 10(exp -6) to 10(exp -2) gr/cm(exp 3). Our shock calculations include molecular hydrogen, atomic hydrogen, ionized hydrogen, neutral helium, and singly ionized helium. Even at the highest shock velocity, the gas is only partially ionized and the postshock temperature rises with preshock density in order to maintain the ionization. The value of the effective shock adiabatic index gamma(sub sh) varies from 1.17 (at low v(sub sh) and rho(sub a) to 1.40 (at high v(sub sh) and rho(sub a). The ablation rate is limited by the radiative flux that reaches the bolide surface. We argue that the ablated gas does not efficiently transfer its kinetic energy to the atmosphere, and it ultimately slows in a similar fashion to the comet material. As the bolide initially falls through the atmosphere, the character of the shock emission changes. At rho(sub a) approximately 10(exp -8) gr/cm (exp 3), the gas is optically thin and we expect line emission; in the optical spectrum, Balmer emission is expected from the shocked atmosphere and low-ionization metal lines from ablated cometary material. At rho(sub a) approximately 10(exp -6) gr/m(exp 3), the shocked gas is optically thick and the shock front near the bolide produces a blackbody spectrum. The temperature is favorable for ultraviolet (1000 to 3000 A) emission and the luminosity may be approximately 5 x 10(exp 23) ergs/sec for approximately 0.6 sec for a bolide 1 km in radius. At rho(sub a) approximately 10(exp -4) gr/cm(exp 3), the bolide has passed below the ultraviolet photosphere. The shock front emits considerable ionizing radiation, but it is absorbed in a narrow preshock region. The bolometric correction for the optical luminosity is large and we expect a 3000 to 8000 A luminosity of approximately 3 x 10(exp 23) ergs/sec for approximately 1 sec. The optical emission is strongly peaked in the vicinity of the bolide. The bolide does have a somewhat less luminous, optically thick trail extending greater than or equal to 10 km, but the radiation is characterized by a temperature of 4000 to 5000 K. From the fragmentation model of Chyba, Thomas, & Zahnle (1993), the bolide deposits most of its kinetic energy at rho(sub a) approximately 10(exp -3) gr/cm(exp 3) and this is the effective explosion site. The shock wave from such an explosion can move up about one density scale height. We examine the breakout of the shock front from the Jovian atmosphere and find that the shock acceleration in the decreasing density region is slow, so that the energy flux in the shock front is small. Higher velocities might be generated by shock acceleration along the channel left by the bolide if the shock motion can occur before the channel closes off as a result of radiative cooling. Hot gas created by the explosion ultimately rises due to buoyancy on a timescale of a minute. The luminosity is highest when the bubble first rises into the optically thin part of the atmosphere and may be approximately 1 x 10(exp 25) ergs/sec in the near-infrared. Roughly 1% of the initial bolide energy may be radiated in this way; the rest of the energy is lost to sound waves from the initial explosion and to work done by the bubble on the surrounding atmosphere.

Chevalier, Roger A.↗
NASA NTRS1976

Ultraviolet observations of local gas

From satellite measurements of ultraviolet spectra of stars, an average density of approximately 1.1/cu cm for hydrogen atoms, in both atomic and molecular form, is estimated for regions of space along the galactic plane within about 1 kpc of the sun. About 20% of the atoms are bound in molecular form although this figure is uncertain since the ultraviolet measurements avoid the very dense interstellar clouds. Discrete values for this percentage are observed to vary markedly; regions with less than average density seem to have fractional abundances of H2 several orders of magnitude lower than average. A ratio of CO/H is observed for regions in front of stars observed by the Copernicus satellite.

Jenkins, E. B.↗
NASA NTRS1977

Ultraviolet observations of local gas

From satellite measurements of ultraviolet spectra of stars, an average density of approximately 1.1 cu m/sec for hydrogen atoms, in both atomic and molecular form, is estimated for regions of space along the galactic plane within about 1 kpc of the sun. About 20 percent of the atoms are bound in molecular form, although this figure is uncertain because the ultraviolet measurements avoid the very dense interstellar clouds. Discrete values for this percentage are observed to vary markedly; regions with less than average density appear to have fractional abundances of H2 several orders of magnitude lower than average. A ratio of CO/H which ranges from 10 to the minus 9th to 10 to the minus 6th power is observed for regions in front of stars observed by the Copernicus satellite.

Jenkins, E. B.↗
DOE OSTI2022

Inference of molecular divertor density from filtered camera analysis of molecularly induced Balmer line emission during detachment in JET L-mode plasmas

Here a previously presented Monte Carlo method for estimating local plasma conditions in 2D based on intensity ratios of deuterium Balmer D α , D γ and D ε lines was amended to consider also the D α and D γ emission contributions arising from molecular processes. The obtained estimates were used to infer the molecular divertor density with the help of the molecular databases of EIRENE. The method was benchmarked against EDGE2D-EIRENE simulations and observed to reproduce the molecularly induced emission fractions and the molecular divertor densities primarily within 25% of the references. Experimental analysis of a JET L-mode density scan suggested molecularly induced D α and D γ contributions of up to 60–70% and 20%, respectively, during the process of detachment. The independent estimates of the molecular divertor density inferred from the obtained molecularly induced D α and D γ intensities agree within uncertainties with each other. Both estimates show the molecular density increasing up to approximately 1.0–2.0 x 10 20 m -3 at the outer strike point in deep detachment with its ratio to the local electron density agreeing with EDGE2D-EIRENE predictions within the scatter of the experimental data.

interferometry↗
DOE OSTI2021

Self-interaction-corrected Kohn–Sham effective potentials using the density-consistent effective potential method

Density functional theory (DFT) and beyond-DFT methods are often used in combination with photoelectron spectroscopy to obtain physical insights into the electronic structure of molecules and solids. The Kohn–Sham eigenvalues are not electron removal energies except for the highest occupied orbital. The eigenvalues of the highest occupied molecular orbitals often underestimate the electron removal or ionization energies due to the self-interaction (SI) errors in approximate density functionals. Here, we adapt and implement the density-consistent effective potential method of Kohut, Ryabinkin, and Staroverov (2014) to obtain SI-corrected local effective potentials from the SI-corrected Fermi–Löwdin orbitals and density in the Fermi–Löwdin orbital self-interaction correction scheme. The implementation is used to obtain the density of states (photoelectron spectra) and HOMO–LUMO gaps for a set of molecules and polyacenes. Good agreement with experimental values is obtained compared to a range of SI uncorrected density functional approximations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

CrI 3 revisited with a many-body ab initio theoretical approach

CrI 3 has recently been shown to exhibit low-dimensional, long-range magnetic ordering from few layers to single layers of CrI 3 . The properties of CrI 3 bulk and few layered systems are uniquely defined by a combination of short-range intralayer and long-range interlayer interactions; including strong correlations, exchange and spin orbit coupling. Unfortunately, both the long-range van der Waals interactions, which are driven by dynamic, many-body electronic correlations and the competing strong intralayer correlations, present a formidable challenging for the local or semi local mean-field approximations employed in workhorse electronic structure approaches like density functional theory. In this work, we employ a sophisticated many-body approach that can simultaneously describe long- and short-range correlations. Herein, we establish that the fixed-node diffusion Monte Carlo (FNDMC) method reproduces the experimental interlayer separation distance of bulk CrI 3 for the high-temperature, monoclinic phase, with a reliable prediction of the interlayer binding energy. We subsequently employed, the FNDMC results to benchmark the accuracy of several density functional theory exchange-correlation approximations.

36 MATERIALS SCIENCE↗
NASA NTRS1982

Far-infrared mapping of the double-lobed H II region S106

Photometry of S106 at 50 and 100 microns, as well as its mapping with 35 arcsec resolution and a search for a nearby H2O maser source at 2 and 10 microns, have yielded data that are consistent with a 900 pc distance and an O9 (ZAMS) spectral type for a single, central exciting star. The correlation of dust temperature with optical emission, and anticorrelation of dust optical depth with optical emission, support models of S106 which envision an exciting star surrounded by a dusty disk having very low optical extinction along its approximately north-south axis. A local maximum in the dust column density was found to the west of the optical nebula, near the position of an H2O maser, although no local near- or far-IR flux maximum was detected.

Harvey, P. M.↗
NASA NTRS1995

Are Unidentified Extreme-Ultraviolet Sources the Closest Neutron Stars?

Unidentified extreme-ultraviolet (EUV) sources, detected in the EUVE and ROSAT WFC all-sky surveys, could be isolated old neutron stars, accreting material from the interstellar medium (ISM). The closest neutron stars, which are located in the local ISM bubble of unusually low density, are faint and cool (L approximately 10(exp 27) erg/s, T approximately < 6 eV). The extreme-UV spectrum of these sources is very sensitive to the H I column density, since a large fraction of the energy is emitted just below the hydrogen Lyman edge. The EUVE sources with large count rates in the long-wavelength bandpass (600 A) seem to be the most promising candidates. These sources should have low H I column density (N(sub H I) approximately < 10(exp 18) sq cm), constraining their distances to a few tens of parsecs. Otherwise, their spectra would be significantly modified by ISM absorption, and inevitably they would appear stronger in the short-wavelength (100 A, 200 A) EUVE bandpasses. If these unidentified objects are familiar EUV sources rather than neutron stars, i.e. white dwarfs, late-type stars or cataclysmic variables, they are expected to be identifiable, and generally brighter than V approximately 14.

Shemi, A.↗
NASA NTRS1994

The local distribution of Na I interstellar gas

We present high-resolution absorption measurements (lambda/Delta lambda approximately 75,000) of the interstellar Na I D lines at 5890 A toward 80 southern hemisphere early-type stars located in the local interstellar medium (LISM). Combining these results with other sodium measurements taken from the literature, we produce galactic maps of the distribution of neutral sodium column density for a total of 293 stars generally lying within approximately 250 pc of the Sun. These maps reveal the approximate shape of the mid-plane contours of the rarefied region of interstellar space termed the Local Bubble. Its shape is seen as highly asymmetric, with a radius ranging from 30 to 300 pc, and with an average radius of 60 pc. Similar plots of the Galactic mid-plane distribution of sources emitting extreme ultraviolet radiation show that they also trace out similar contours of the Local Bubble derived from Na I absorption measurements. We conclude that the Local Bubble absorption interface can be represented by a hydrogen column density, Nu(sub ETA) = 2 x 10(exp 19) cm(exp -2), which explains both the local distribution of Na I absorption and the observed galactic distribution of extreme ultraviolet sources. The derived mid-plane contours of the Bubble generally reproduce the large-scale features carved out in the interstellar medium by several nearby galactic shell structures.

Welsh, B. Y.↗
NASA NTRS2013

Concurrent Supermassive Black Hole and Galazy Growth: Linking Environment and Nuclear Activity in Zeta Equals 2.23 H Alpha Emitters

We present results from an approximately equal 100 ks Chandra observation of the 2QZ Cluster 1004+00 structure at z = 2.23 (hereafter 2QZ Clus). 2QZ Clus was originally identified as an overdensity of four optically-selected QSOs at z = 2.23 within a 15 × 15 arcmin square region. Narrow-band imaging in the near-IR (within the K band) revealed that the structure contains an additional overdensity of 22 z = 2.23 H alpha-emitting galaxies (HAEs), resulting in 23 unique z = 2.23 HAEs/QSOs (22 within the Chandra field of view). Our Chandra observations reveal that three HAEs in addition to the four QSOs harbor powerfully accreting supermassive black holes (SMBHs), with 2-10 keV luminosities of approximately equal (8-60) × 10(exp 43) erg s(exp−1) and X-ray spectral slopes consistent with unobscured active galactic nucleus (AGN). Using a large comparison sample of 210 z = 2.23 HAEs in the Chandra-COSMOS field (C-COSMOS), we find suggestive evidence that the AGN fraction increases with local HAE galaxy density. The 2QZ Clus HAEs reside in a moderately overdense environment (a factor of approximately equal 2 times over the field), and after excluding optically-selected QSOs, we find that the AGN fraction is a factor of approximately equal 3.5(+3.8/ −2.2) times higher than C-COSMOS HAEs in similar environments. Using stacking analyses of the Chandra data and Herschel SPIRE observations at 250micrometers, we respectively estimate mean SMBH accretion rates ( M(BH)) and star formation rates (SFRs) for the 2QZ Clus and C-COSMOS samples. We find that the mean 2QZ Clus HAE stacked X-ray luminosity is QSO-like (L(2−10 keV) approximately equal [6-10] × 10(exp 43) erg s(exp −1)), and the implied M(BH)/SFR approximately equal (1.6-3.2) × 10(exp −3) is broadly consistent with the local M(BH)/Stellar Mass relation and z approximately equal 2 X-ray selected AGN. In contrast, the C-COSMOS HAEs are on average an order of magnitude less X-ray luminous and have M(BH)/SFR approximately equal (0.2-0.4) × 10(exp −3), somewhat lower than the local MBH/M relation, but comparable to that found for z approximately equal 1-2 star-forming galaxies with similar mean X-ray luminosities. We estimate that a periodic QSO phase with duty cycle approximately 2%-8% would be sufficient to bring star-forming galaxies onto the local M(BH)/Stellar Mass relation. This duty cycle is broadly consistent with the observed C-COSMOS HAE AGN fraction (Approximately equal 0.4%-2.3%) for powerful AGN with LX approximately greater than 10(exp 44) erg s(exp −1). Future observations of 2QZ Clus will be needed to identify key factors responsible for driving the mutual growth of the SMBHs and galaxies.

Lehmer, B. D.↗
DOE OSTI2024

Machine Learning a Simple Interpretable Short-Range Potential for Silica

A wide array of models, spanning from computationally expensive ab initio methods to a spectrum of force-field approaches, have been developed and employed to probe silica polymorphs and understand growth processes and atomic-level dynamical transitions in silica. However, the quest for a model capable of making accurate predictions with high computational efficiency for various silica polymorphs is still ongoing. Recent developments in short-range machine-learned models, such as GAP and NNPScan, have shown promise in providing reasonable descriptions of silica, but their computational cost remains high compared to force fields such as BKS which are based on simple interpretable functional forms. Here, in this study, we build on the recent success of our reinforcement learning (RL) workflow to derive a new set of optimal parameters for a promising short-range BKS-based model proposed by Soules. We use RL to navigate the eight-dimensional parameter space of the Soules potential using an experimental training data set that includes both local and global structural features from approximately 21 experimentally realized silica polymorphs, including high density phases and porous zeolites. We compare the performance of our machine-learned ML-Soules model with other high quality models including our recent machine-learned parametrization of BKS (ML-BKS), a machine-learned potential (GAP), as well as predictions of ab initio calculations with the highly fidelity SCAN functional. The ML-Soules accurately captures the relative energetic ordering of various polymorphs as well as their structural features at a significantly reduced computational expense. The ML-Soules model also reasonably captures the structure, density, and elastic constants of quartz, as well as metastable silica polymorphs. We further discuss the limitations of the Soules functional form and propose potential enhancements, including the incorporation of additional three-body terms and/or the utilization of different short-ranged functional forms to achieve greater accuracy for both global and local features in the modeling of silica while retaining low computational cost.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗