Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “MD”

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 775 records · Page 43

Practical low-temperature gasoline combustion for very high efficiency off-road, medium- and heavy-duty engines

Low-temperature gasoline combustion (LTGC) with additive-mixing fuel injection (AMFI) is a new combustion strategy that has been demonstrated to deliver 9%–25% better brake thermal efficiency than similar-sized market-leading diesel engines over the operating map. Moreover, the LTGC-AMFI engine shows near-zero smoke, and NOx emissions are 4–100 times lower than those of a diesel, sufficiently low that no aftertreatment, or only passive NOx aftertreatment, would be sufficient (diesel exhaust fluid is not required). LTGC-AMFI combustion is based on kinetically controlled compression ignition of a dilute charge with a variable amount of low-to-moderate fuel stratification. Fast combustion control is provided by adding minute amounts of an ignition-enhancing additive into the fuel each engine cycle to control its reactivity. This strategy was used to operate a medium-duty (MD) LTGC-AMFI engine at loads from idle to 16.3 bar BMEP and speeds from 600 to 2400 rpm with regular E10 gasoline, which covers nearly the entire operating map of a typical MD engine. Turbine-out temperatures were sufficient for an oxidation catalyst to control hydrocarbon and CO emissions. Autonomie simulations over the GEM ARB Transient and the GEM 55 mph Cruise driving cycles for class-6 trucks using this technology showed fuel economies of 8.1 and 11.4 mpg-gasoline-equivalent, respectively, corresponding to 18.6% and 13.4% improvements over a similar-size diesel engine. Engine-out NOx emissions were 0.024 and 0.01 g/bhp-h, respectively, well below current U.S. emission standards. These results show that switching from diesel to LTGC-AMFI engines would greatly reduce greenhouse gas (GHG) emissions for off-road, MD and HD applications, which will continue to rely on combustion engines because electrification is not practical in the foreseeable future. Finally, with their reduced fuel consumption, the lower cost of gasoline compared to diesel fuel, and much lower aftertreatment costs, LTGC-AMFI engines also offer a significantly lower total cost of ownership.

33 ADVANCED PROPULSION SYSTEMS↗

Crystallographic and Molecular Dynamics Simulation Analysis of Escherichia Coli Dihydroneopterin Aldolase

To understand the structural basis for the biochemical differences and further investigate the catalytic mechanism of DHNA, we have determined the structure of EcDHNA complexed with NP at 1.07-Å resolution [PDB:2O90], built an atomic model of EcDHNA complexed with the substrate DHNP, and performed molecular dynamics (MD) simulation analysis of the substrate complex. EcDHNA has the same fold as SaDHNA and also forms an octamer that consists of two tetramers, but the packing of one tetramer with the other is significantly different between the two enzymes. Furthermore, the structures reveal significant differences in the vicinity of the active site, particularly in the loop that connects strands β3 and β4, mainly due to the substitution of nearby residues. The building of an atomic model of the complex of EcDHNA and the substrate DHNP and the MD simulation of the complex show that some of the hydrogen bonds between the substrate and the enzyme are persistent, whereas others are transient. The substrate binding model and MD simulation provide the molecular basis for the biochemical behaviors of the enzyme, including noncooperative substrate binding, indiscrimination of a pair of epimers as the substrates, proton wire switching during catalysis, and formation of epimerization product.

59 BASIC BIOLOGICAL SCIENCES↗

Molecular Dynamics Simulations of Anion Exclusion in Clay Interlayer Nanopores

© 2016, Clays and Clay Minerals. All rights reserved. The aqueous chemistry of water films confined between clay mineral surfaces remains an important unknown in predictions of radioelement migration from radioactive waste repositories. This issue is particularly important in the case of long-lived anionic radioisotopes ( 129 I - , 99 TcO - 4 , 36 Cl - ) which interact with clay minerals primarily by anion exclusion. For example, models of ion migration in clayey media do not agree as to whether anions are completely or partially excluded from clay interlayer nanopores. In the present study, this key issue was addressed for Cl - using MD simulations for a range of nanopore widths (6 to 15 Å) overlapping the range of average pore widths that exists in engineered clay barriers. The MD simulation results were compared with the predictions of a thermodynamic model (Donnan Equilibrium model) and two pore-scale models based on the Poisson-Boltzmann equation under the assumption that interlayer water behaves as bulk liquid water. The simulations confirmed that anion exclusion from clay interlayers is greater than predicted by the pore-scale models, particularly at the smallest pore size examined. This greater anion exclusion stems from Cl - being more weakly solvated in nano-confined water than it is in bulk liquid water. Anion exclusion predictions based on the Poisson-Boltzmann equation were consistent with the MD simulation results, however, if the predictions included an ion closest approach distance to the clay mineral surface on the order of 2.0 ± 0.8 Å. These findings suggest that clay interlayers approach a state of complete anion exclusion (hence, ideal semi-permeable membrane properties) at a pore width of 4.2 ± 1.5 Å.

Tournassat, Christophe↗

Silica Surface States and their Wetting Characteristics

The wetting characteristics of silica (SiO 2 ) surfaces can be described by molecular dynamics (MD) and ab initio simulations, including comparison of silica surfaces (talc (001), siloxanated quartz, tridymite (001) and quartz (001)), some of which have not been considered previously in the literature. Classic MD and ab initio simulation methods have been used to determine the contact angle, interfacial water structure, hydroxylation reaction and hydration energy, the results of which are compared with experimental results reported in the literature. Additionally, wetting of silica surfaces depends on surface polarity and extent of hydroxylation. The non-polar siloxane surfaces are characterized by a contact angle of about 80°, an MD ‘water exclusion zone’ of about 3 Å, a relaxed interfacial water orientation, inertness to hydroxylation and minimal hydration energy. The polar silica surfaces can be wetted by water and have a more ordered interfacial water structure. Silanol groups form at the polar silica surface during hydroxylation reactions, and the calculated hydration energy of -1·2 and -1·6 eV matches the experimental heat of immersion measurements reported in the literature, which correspond to hydrogen (H) bonding with interfacial water. In conclusion, fundamental understanding of silica surfaces is important for understanding flotation phenomena and fluid flow in silica nanopores.

36 MATERIALS SCIENCE↗

Peridynamics as a rigorous coarse-graining of atomistics for multiscale materials design.

This report summarizes activities undertaken during FY08-FY10 for the LDRD Peridynamics as a Rigorous Coarse-Graining of Atomistics for Multiscale Materials Design. The goal of our project was to develop a coarse-graining of finite temperature molecular dynamics (MD) that successfully transitions from statistical mechanics to continuum mechanics. The goal of our project is to develop a coarse-graining of finite temperature molecular dynamics (MD) that successfully transitions from statistical mechanics to continuum mechanics. Our coarse-graining overcomes the intrinsic limitation of coupling atomistics with classical continuum mechanics via the FEM (finite element method), SPH (smoothed particle hydrodynamics), or MPM (material point method); namely, that classical continuum mechanics assumes a local force interaction that is incompatible with the nonlocal force model of atomistic methods. Therefore FEM, SPH, and MPM inherit this limitation. This seemingly innocuous dichotomy has far reaching consequences; for example, classical continuum mechanics cannot resolve the short wavelength behavior associated with atomistics. Other consequences include spurious forces, invalid phonon dispersion relationships, and irreconcilable descriptions/treatments of temperature. We propose a statistically based coarse-graining of atomistics via peridynamics and so develop a first of a kind mesoscopic capability to enable consistent, thermodynamically sound, atomistic-to-continuum (AtC) multiscale material simulation. Peridynamics (PD) is a microcontinuum theory that assumes nonlocal forces for describing long-range material interaction. The force interactions occurring at finite distances are naturally accounted for in PD. Moreover, PDs nonlocal force model is entirely consistent with those used by atomistics methods, in stark contrast to classical continuum mechanics. Hence, PD can be employed for mesoscopic phenomena that are beyond the realms of classical continuum mechanics and atomistic simulations, e.g., molecular dynamics and density functional theory (DFT). The latter two atomistic techniques are handicapped by the onerous length and time scales associated with simulating mesoscopic materials. Simulating such mesoscopic materials is likely to require, and greatly benefit from multiscale simulations coupling DFT, MD, PD, and explicit transient dynamic finite element methods FEM (e.g., Presto). The proposed work fills the gap needed to enable multiscale materials simulations.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Water recovery using waste heat from coal fired power plants.

The potential to treat non-traditional water sources using power plant waste heat in conjunction with membrane distillation is assessed. Researchers and power plant designers continue to search for ways to use that waste heat from Rankine cycle power plants to recover water thereby reducing water net water consumption. Unfortunately, waste heat from a power plant is of poor quality. Membrane distillation (MD) systems may be a technology that can use the low temperature waste heat (<100 F) to treat water. By their nature, they operate at low temperature and usually low pressure. This study investigates the use of MD to recover water from typical power plants. It looks at recovery from three heat producing locations (boiler blow down, steam diverted from bleed streams, and the cooling water system) within a power plant, providing process sketches, heat and material balances and equipment sizing for recovery schemes using MD for each of these locations. It also provides insight into life cycle cost tradeoffs between power production and incremental capital costs.

01 COAL, LIGNITE, AND PEAT↗

Comparison of binary collision approximation and molecular dynamics for displacement cascades in GaAs.

The predictions of binary collision approximation (BCA) and molecular dynamics (MD) simulations of displacement cascades in GaAs are compared. There are three issues addressed in this work. The first is the optimal choice of the effective displacement threshold to use in the BCA calculations to obtain the best agreement with MD results. Second, the spatial correlations of point defects are compared. This is related to the level of clustering that occurs for different types of radiation. Finally, the size and structure of amorphous zones seen in the MD simulations is summarized. BCA simulations are not able to predict the formation of amorphous material.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Collaborative Lubricating Oil Study on Emissions: November 28, 2006 - March 31, 2011

The Collaborative Lubricating Oil Study on Emissions (CLOSE) project was a pilot investigation of how fuels and crankcase lubricants contribute to the formation of particulate matter (PM) and semi-volatile organic compounds (SVOC) in vehicle exhaust. As limited vehicles were tested, results are not representative of the whole on-road fleet. Long-term effects were not investigated. Pairs ofvehicles (one normal PM emitting, one high-PM emitting) from four categories were selected: light-duty (LD) gasoline cars, medium-duty (MD) diesel trucks, heavy-duty (HD) natural-gas-fueled buses, and HD diesel buses. HD vehicles procured did not exhibit higher PM emissions, and thus were labeled high mileage (HM). Fuels evaluated were non-ethanol gasoline (E0), 10 percent ethanol (E10),conventional low-sulfur TxLED diesel, 20% biodiesel (B20), and natural gas. Temperature effects (20 degrees F, 72 degrees F) were evaluated on LD and MD vehicles. Lubricating oil vintage effects (fresh and aged) were evaluated on all vehicles. LD and MD vehicles were operated on a dynamometer over the California Unified Driving Cycle, while HD vehicles followed the Heavy Duty Urban DynamometerDriving Schedule. Regulated and unregulated emissions were measured. Chemical markers from the unregulated emissions measurements and a tracer were utilized to estimate the lubricant contribution to PM.

03 NATURAL GAS↗

Reservoir Characterization of Bridgeport and Cypress Sandstones in Lawrence Field Illinois to Improve Petroleum Recovery by Alkaline-Surfactant-Polymer Flood

Within the Illinois Basin, most of the oilfields are mature and have been extensively waterflooded with water cuts that range up to 99% in many of the larger fields. In order to maximize production of significant remaining mobile oil from these fields, new recovery techniques need to be researched and applied. The purpose of this project was to conduct reservoir characterization studies supporting Alkaline-Surfactant-Polymer Floods in two distinct sandstone reservoirs in Lawrence Field, Lawrence County, Illinois. A project using alkaline-surfactantpolymer (ASP) has been established in the century old Lawrence Field in southeastern Illinois where original oil in place (OOIP) is estimated at over a billion barrels and 400 million barrels have been recovered leaving more than 600 million barrels as an EOR target. Radial core flood analysis using core from the field demonstrated recoveries greater than 20% of OOIP. While the lab results are likely optimistic to actual field performance, the ASP tests indicate that substantial reserves could be recovered even if the field results are 5 to 10% of OOIP. Reservoir characterization is a key factor in the success of any EOR application. Reservoirs within the Illinois Basin are frequently characterized as being highly compartmentalized resulting in multiple flow unit configurations. The research conducted on Lawrence Field focused on characteristics that define reservoir compartmentalization in order to delineate preferred target areas so that the chemical flood can be designed and implemented for the greatest recovery potential. Along with traditional facies mapping, core analyses and petrographic analyses, conceptual geological models were constructed and used to develop 3D geocellular models, a valuable tool for visualizing reservoir architecture and also a prerequisite for reservoir simulation modeling. Cores were described and potential permeability barriers were correlated using geophysical logs. Petrographic analyses were used to better understand porosity and permeability trends in the region and to characterize barriers and define flow units. Diagenetic alterations that impact porosity and permeability include development of quartz overgrowths, sutured quartz grains, dissolution of feldspar grains, formation of clay mineral coatings on grains, and calcite cementation. Many of these alterations are controlled by facies. Mapping efforts identified distinct flow units in the northern part of the field showing that the Pennsylvanian Bridgeport consists of a series of thick incised channel fill sequences. The sandstones are about 75-150 feet thick and typically consist of medium grained and poorly sorted fluvial to distributary channel fill deposits at the base. The sandstones become indistinctly bedded distributary channel deposits in the main part of the reservoir before fining upwards and becoming more tidally influenced near their top. These channel deposits have core permeabilities ranging from 20 md to well over 1000 md. The tidally influenced deposits are more compartmentalized compared to the thicker and more continuous basal fluvial deposits. Fine grained sandstones that are laterally equivalent to the thicker channel type deposits have permeabilities rarely reaching above 250 md. Most of the unrecovered oil in Lawrence Field is contained in Pennsylvanian Age Bridgeport sandstones and Mississippian Age Cypress sandstones. These reservoirs are highly complex and compartmentalized. Detailed reservoir characterization including the development of 3-D geologic and geocellular models of target areas in the field were completed to identify areas with the best potential to recover remaining reserves including unswept and by-passed oil. This project consisted of tasks designed to compile, interpret, and analyze the data required to conduct reservoir characterization for the Bridgeport and Cypress sandstones in pilot areas in anticipation of expanded implementation of ASP flooding in Lawrence Field. Geologic and geocellular modeling needed for reservoir characterization and reservoir simulation were completed as prerequisites for design of efficient ASP flood patterns. Characterizing the complex reservoir geology that identifies the geologic conditions that will optimize oil recoveries for expansion of the ASP pilots in the Bridgeport and Cypress sandstones to other areas of Lawrence Field is the primary objective of this project. It will permit evaluation of efficiency of oil recovery from Bridgeport and Cypress sandstone reservoirs using ASP technology. Additionally evaluation of similar Pennsylvanian and Chesterian reservoirs shows that it is likely that ASP flood technology can be successfully applied to similar reservoirs in the Illinois Basin as well as to other U.S. reservoirs. Chemical flooding was introduced in stages with the first flood initiated in 2010 and a second offset pilot project initiated during 2011. Rex Energy Corporation is reporting a positive response on its ASP Middagh pilot project in the Pennsylvanian Bridgeport B reservoir, Lawrence Field. Oil response in the 15 acre flood has continued to show an increase in oil cut from 1% to 12%. Total pattern production increased from 16 BOPD and stabilized at a range of 65-75 BOPD in the last three months of 2011. Peak production rose to 100 + BOPD. Oil cut in the pilot increased for 1.0% to ~ 12.0% with an individual well showing oil cuts greater than 20%. A second, 58 acre pilot (Perkins-Smith) adjacent to and likely in communication with the Middagh pilot has been initiated. Preliminary brine injection has been implemented and ASP injection was initiated in mid-2012. Response is expected by mid-2013 with peak recovery expected by late 2013. Rex Energy is projecting full scale expansion with the next step of development being a 351 acre project scheduled to begin in mid-2013. Preliminary development has been initiated in this Delta Unit area located in the south half of section 32, T4N, R12W.

02 PETROLEUM↗

Assessment of Offshore Wind Energy Leasing Areas for the BOEM Maryland Wind Energy Area

The National Renewable Energy Laboratory (NREL), under an interagency agreement with the Bureau of Ocean Energy Management (BOEM), is providing technical assistance to identify and delineate leasing areas for offshore wind energy development within the Atlantic Coast Wind Energy Areas (WEAs) established by BOEM. This report focuses on NREL's evaluation of the delineation proposed by the Maryland Energy Administration (MEA) for the Maryland (MD) WEA and two alternative delineations. The objectives of the NREL evaluation were to assess MEA's proposed delineation of the MD WEA, perform independent analysis, and recommend how the MD WEA should be delineated.

17 WIND ENERGY↗

Reaction Mechanisms for Barite Dissolution and Growth

In Stack and Rustad (2007), the reactive flux method (Rey and Hynes, 1996) and molecular dynamics (MD) were used to simulate the {001} barite-water interface structure and water exchange rate of aqueous barium ions and barium surface species. Atomic-level mineralwater interfacial structure and kinetics are being studied with increasing precision due to advances in spectroscopic methods at synchrotron x-ray sources as well as improved computational capacity. Better characterization of these interfaces in turn is leading to advances in the understanding of many macroscopic geochemical properties. Overall the barite-water interfacial structure was found to compare well to that estimated using X-ray reflectivity (XRR) measurements (Fenter et al., 2001), but there was an important difference: the MD predicted an intricate water structure present at the interface with one major peak and several minor peaks whereas the XRR found only a single layer of water. This discrepancy is thought to result from a limited resolution in the Fenter et al. (2001) study as well as over-coordination of surface sulfates by the MD model.

58 GEOSCIENCES↗

Modeling Rare Events in High Dimensional Stochastics Systems

The main issue addressed in this note is the study of an algorithm to accelerate the computation of kinetic rates in the context of molecular dynamics(MD). It is based on parallel simulations of short-time trajectories and its main components are: a decomposition of phase space into macrostates or cells, a resampling procedure that ensures that the number of parallel replicas (MD simulations) in each macro-state remains constant, the use of multiple populations (colored replicas) to compute multiple rates (e.g., forward and backward rates) in one simulation. The method leads to enhancing the sampling of macro-states associated to the transition states, since in traditional MD these are likely to be depleted even after short-time simulations. By allowing parallel replicas to carry different probabilistic weights, the number of replicas within each macro-state can be maintained constant without introducing any bias. The empirical variance of the estimated reaction rate, defined as a probability flux, is expectedly diminished. This note is a first attempt towards a better mathematical and numerical understanding of this method. It provides first a mathematical formalization of the notion of colors. Then, the link between this algorithm and a set of closely related methods having been proposed in the literature within the past few years is discussed. Lastly, numerical results are provided that illustrate the efficiency of the method.

59 BASIC BIOLOGICAL SCIENCES↗

Milestone report: The simulation of radiation driven gas diffusion in UO 2 at low temperature

Below 1000 K it is thought that fission gas diffusion in nuclear fuel during irradiation occurs through atomic mixing due to radiation damage. This is an important process for nuclear reactor performance as it affects fission gas release, particularly from the periphery of the pellet where such temperatures are normal. Here we present a molecular dynamics study of Xe and Kr diffusion due to irradiation. Thermal spikes and cascades have been used to study the electronic stopping and ballistic phases of damage, respectively. Our results predict that O and Kr exhibit the greatest diffusivity and U the least, while Xe lies in between. It is concluded that the ballistic phase does not sufficiently account for the experimentally observed diffusion. Preliminary thermal spike calculations indicate that the electronic stopping phase generates greater fission gas displacement than the ballistic phase, although further calculation must be carried out to confirm this. A good description of the system by the empirical potentials is important over the very wide temperatures induced during thermal spike and damage cascade simulations. This has motivated the development of a parameter set for gas-actinide and gas-oxygen interactions that is complementary for use with a recent many-body potential set. A comprehensive set of density functional theory (DFT) calculations were used to study Xe and Kr incorporation at a number of sites in CeO 2 , ThO 2 , UO 2 and PuO 2 . These structures were used to fit a potential, which was used to generate molecular dynamics (MD) configurations incorporating Xe and Kr at 300 K, 1500 K, 3000 K and 5000 K. Subsequent matching to the forces predicted by DFT for these MD configurations was used to refine the potential set. This fitting approach ensured weighted fitting to configurations that are thermodynamically significant over a broad temperature range, while avoiding computationally expensive DFT-MD calculations. The resultant gas potentials were validated against DFT binding energies and are suitable for simulating combinations of Xe and Kr in solid solutions of CeO 2 , ThO 2 , UO 2 and PuO 2 , providing a powerful tool for the atomistic simulation of conventional nuclear reactor fuel UO 2 as well as advanced MOX fuels.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Modeling Mesoscale Processes of Scalable Synthesis Final Scientific/Technical Report

This is the final scientific and technical report for this award. Molecular dynamics (MD) is a powerful tool to study biomolecules at the nanoscale. One of its main limitations however is that such simulations typically require a timestep of 1 femtosecond while reactions and conformational changes typically happen on the time scale of 1 microsecond or more. This discrepancy severely restricts the impact of this method. We have developed a series of algorithms to mitigate this problem and allow predicting long-time scale events using much shorter simulations. The method is based on the concurrent adaptive sampling method. It consists in using probabilistic weights for trajectories and biasing the sampling such that rare but important trajectories, e.g., those that go over energy barriers and lead to relevant conformational changes in the molecule, are observed with much higher frequency that in a standard MD simulation. We will demonstrate below that this approach has been very successful. In collaboration with Jay Grate at PNNL we have applied this novel method to the study the triazine polymers, which can be con gured using different types of side chains in order to adopt different 3-dimensional structures. We have found novel stable con- formations and have analyzed the impact of side chain modifcations on the native folded shape of the polymer. Direct MD simulations would have been untractable for this problem. The method of replica exchange was also used successfully to calculate free energy profiles, in particular for cis-trans torsional angle changes with high energy barriers. Studying molecules at the atomistic scale is essential to design new molecules in chemical engineering, such as the triazine polymers developed by Grate, in material science to study the strength of metals for examples (e.g., dislocation dynamics), or in biochemistry to understand how proteins fold and how they interact with ligands (a key step in designing drugs). Although new hardware makes these simulations increasingly faster (e.g., GPU, Anton from DE Shaw Research), new algorithms are key to making these methods more tractable and increase their impact. At this point, because of limitations in modeling, one of the main techniques for advancing science remains to a large extent numerous and costly experiments. For exam- ple, when designing drug, where multiple variants must be screened, the ability to run tests on a computer rather than experimentally testing with high throughput methods thousands of compounds would represent a signi cant advance with broad benefits for the public at large.

36 MATERIALS SCIENCE↗

Fouling-Resistant Membranes for Treating Concentrated Brines for Water Reuse in Advanced Energy Systems- Final Technical Report

The high total dissolved solids (TDS) levels in the wastewater quality generated from unconventional oil and gas development make the current state-of-the art approach to water treatment/disposal untenable. Our proposed membrane technology approach addresses the two major challenges associated with this water: 1) the membrane distillation process removes the high TDS content, which is often 8 times higher than that of seawater, and 2) our novel membrane coating prevents the formation of scale that would otherwise pose a significant operational hurdle. This is accomplished through next-generation electrically conductive membranes that mitigate fouling beyond what is currently possible, and allow for the flexibility to treat to the water to levels desirable for multiple reuse options, thus reducing fresh water withdrawal, all the way to direct disposal into the environment. The overall project objective was to demonstrate the efficacy of membrane distillation (MD) as a cost-savings technology to treat concentrated brines (such as, but not limited to, produced waters generated from fossil fuel extraction) that have high levels of TDS for beneficial water reuse in power production and other industrial operations as well as agricultural and municipal water uses. In addition, a novel fouling-resistant nanocomposite membrane was developed to reduce the need for chemicals to address membrane scaling due to the precipitation of divalent ions in high-TDS waters and improve overall MD performance via an electrically conductive membrane distillation process (ECMD). This anti-fouling membrane technology platform is based on incorporating carbon nanotubes (CNTs) into the surface layer of existing, commercially available MD membranes. The CNTs impart electrical conductivity to the membrane surface to prevent membrane scaling and fouling when an electrical potential is applied.

36 MATERIALS SCIENCE↗

Enhancements to the Munson-Dawson Model for Rock Salt

The Munson-Dawson (MD) constitutive model was originally developed in the 1980's to predict the thermomechanical behavior of rock salt. Since then, it has been used to simulate the evolution of the underground in nuclear waste repositories, mines, and storage caverns for gases and liquids. This report covers three enhancements to the MD model. (1) New transient and steady-state rate terms were added to capture salt's creep behavior at low equivalent stresses (below about 8 MPa). These new terms were calibrated against a series of triaxial compression creep experiments on salt from the Waste Isolation Pilot Plant. (2) The equivalent stress measure was changed from the Tresca stress to the Hosford stress. By varying a single exponent, the Hosford stress can reduce to the Tresca stress, the von Mises stress, or a range of behaviors in-between. This exponent was calibrated against true triaxial compression experiments on salt hollow cylinders. (3) The MD model's numerical implementation was overhauled, adding a line search algorithm to the implicit solution scheme. The new implementation was verified against analytical solutions, and benchmarked against a pre-existing implementation on a room closure simulation. The new implementation pre- dicted virtually identical room closure, yet sped up the simulation by 16x . (The source code of the new implementation is included in an appendix of this report.)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Data driven learning of robust nonlocal models

Nonlocal models use integral operators that embed length-scales in their definition. However, the integrands in these operators are difficult to define from the data that are typically available for a given physical system, such as laboratory mechanical property tests. In contrast, molecular dynamics (MD) does not require these integrands, but it suffers from computational limitations in the length and time scales it can address. To combine the strengths of both methods and to obtain a coarse-grained, homogenized continuum model that efficiently and accurately captures materials' behavior, we propose a learning framework to extract, from MD data, an optimal nonlocal model as a surrogate for MD displacements. Our framework guarantees that the resulting model is mathematically well-posed, physically consistent, and that it generalizes well to settings that are different from the ones used during training. The efficacy of this approach is demonstrated with several numerical tests for single layer graphene both in the case of perfect crystal and in the presence of thermal noise.

36 MATERIALS SCIENCE↗

A data-driven peridynamic continuum model for upscaling molecular dynamics

Nonlocal models, including peridynamics, often use integral operators that embed lengthscales in their definition. However, the integrands in these operators are difficult to define from the data that are typically available for a given physical system, such as laboratory mechanical property tests. In contrast, molecular dynamics (MD) does not require these integrands, but it suffers from computational limitations in the length and time scales it can address. To combine the strengths of both methods and to obtain a coarse-grained, homogenized continuum model that efficiently and accurately captures materials’ behavior, we propose a learning framework to extract, from MD data, an optimal Linear Peridynamic Solid (LPS) model as a surrogate for MD displacements. To maximize the accuracy of the learnt model we allow the peridynamic influence function to be partially negative, while preserving the well-posedness of the resulting model. To achieve this, we provide sufficient well-posedness conditions for discretized LPS models with sign-changing influence functions and develop a constrained optimization algorithm that minimizes the equation residual while enforcing such solvability conditions. This framework guarantees that the resulting model is mathematically well-posed, physically consistent, and that it generalizes well to settings that are different from the ones used during training. We illustrate the efficacy of the proposed approach with several numerical tests for single layer graphene. Our two-dimensional tests show the robustness of the proposed algorithm on validation data sets that include thermal noise, different domain shapes and external loadings, and discretizations substantially different from the ones used for training.

homogenization↗