Publications

Abstract not provided.

Abstract not provided.

Abstract not provided.

This SAND report fulfills the completion requirements for the ASC Physics and Engineering Modeling Level 2 Milestone 7836 during Fiscal Year 2021. The Sandia Simplified potential energy clock (SPEC) non-linear viscoelastic constitutive model was developed to predict a whole host of polymer glass physical behaviors in order to provide a tool to assess the effects of stress on these materials over their lifecycle. Polymer glasses are used extensively in applications such as electronics packaging, where encapsulants and adhesives can be critical to device performance. In this work, the focus is on assessing the performance of the model in predicting material evolution associated with long-term physical aging, an area that the model has not been fully vetted in. These predictions are key to utilizing models to help demonstrate electronics packaging component reliability over decades long service lives, a task that is very costly and time consuming to execute experimentally. The initiating hypothesis for the work was that a model calibration process can be defined that enables confidence in physical aging predictions under ND relevant environments and timescales without sacrificing other predictive capabilities. To test the hypothesis, an extensive suite of calibration and aging data was assembled from a combination of prior work and collaborating projects (Aging and Lifetimes as well as the DoD Joint Munitions Program) for two mission relevant epoxy encapsulants, 828DGEBA/DEA and 828DGEBA/T403. Multiple model calibration processes were developed and evaluated against the entire set of data for each material. A qualitative assessment of each calibration's ability to predict the wide range of aging responses was key to ranking the calibrations against each other. During this evaluation, predictions that were identified as non-physical, i.e., demonstrated something that was qualitatively different than known material behavior, were heavily weighted against the calibration performance. Thus, unphysical predictions for one aspect of aging response could generate a lower overall rating for a calibration process even if that process generated better quantitative predictions for another aspect of aging response. This insurance that all predictions are qualitatively correct is important to the overall aim of utilizing the model to predict residual stress evolution, which will depend on the interplay amongst the different material aging responses. The DSC-focused calibration procedure generated the best all-around aging predictions for both materials, demonstrating material models that can qualitatively predict the whole host of different physical aging responses that have been measured. This step forward in predictive capability comes from an unanticipated source, utilization of calorimetry measurements to specify model parameters. The DSC-focused calibration technique performed better than compression-focused techniques that more heavily weigh measurements more closely related to the structural responses to be predicted. Indeed, the DSC-focused calibration procedure was only possible due to recent incorporation of the enthalpy and heat capacity features into SPEC that was newly verified during this L2 milestone. Fundamentally similar aspects of the two material model calibrations as well as parametric studies to assess sensitives of the aging predictions are discussed within the report. A perspective on the next steps to the overall goal of residual stress evolution predictions under stockpile conditions closes the report.

Abstract not provided.

A simple approach to simulate contact between deformable objects is presented which relies on levelset descriptions of the Lagrangian geometry and an optimization-based solver. Modeling contact between objects remains a significant challenge for computational mechanics simulations. Common approaches are either plagued by lack of robustness or are exceedingly complex and require a significant number of heuristics. In contrast, the levelset contact approach presented herein is essentially heuristic free. Furthermore, the presented algorithm enables resolving and enforcing contact between objects with a significant amount of initial overlap. Examples demonstrating the feasibility of this approach are shown, including the standard Hertz contact problem, the robust removal of overlap between two overlapping blocks, and overlap-removal and pre-load for a bolted configuration.

This report is a preliminary test plan of the seismic shake table test. The final report will be developed when all decisions regarding the test hardware, instrumentation, and shake table inputs are made. A new revision of this report will be issued in spring of 2022. The preliminary test plan documents the free-field ground motions that will be used as inputs to the shake table, the test hardware, and instrumentation. It also describes the facility at which the test will take place in late summer of 2022.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Detonations and flames are characterized by three-dimensional (3D) temperature fields, yet state-of- the-art temperature measurement techniques yield information at a point or along a line. The goal of the research documented here was to combine ultrafast laser spectroscopy and structured illumination to deliver an unprecedented measurement capability—three-dimensional, instantaneous temperature measurements in a gas-phase volume. To achieve this objective, different parts of the proposed technique were developed and tested independently. Structured illumination was used to image particulate matter (soot) in a turbulent flame at multiple planes using a single laser pulse and a single camera. Emission spectroscopy with structured detection was demonstrated for emission- based measurements of explosives with enhance dimensionality. Finally, an instrument for multi- planar laser-based temperature measurement technique was developed. Structured illumination techniques will continue to be developed for multi-dimensional and multi-parameter measurements. These new measurement capabilities will be important for heat transfer and fluid dynamic research areas.

This project focused on providing a fundamental physico-chemical understanding of the coupling mechanisms of corrosion- and radiation-induced degradation at material-salt interfaces in Ni-based alloys operating in emulated Molten Salt Reactor(MSR) environments through the use of a unique suite of aging experiments, in-situ nanoscale characterization experiments on these materials, and multi-physics computational models. The technical basis and capabilities described in this report bring us a step closer to accelerate the deployment of MSRs by closing knowledge gaps related to materials degradation in harsh environments.

Abstract not provided.

Abstract not provided.

This user’s guide documents capabilities in Sierra/SolidMechanics which remain “in-development” and thus are not tested and hardened to the standards of capabilities listed in Sierra/SM 5.2 User’s Guide. Capabilities documented herein are available in Sierra/SM for experimental use only until their official release. These capabilities include, but are not limited to, novel discretization approaches such as the conforming reproducing kernel (CRK) method, numerical fracture and failure modeling aids such as the extended finite element method (XFEM) and J-integral, explicit time step control techniques, dynamic mesh rebalancing, as well as a variety of new material models and finite element formulations.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

This document describes the Power and Energy Storage Systems Toolbox for MATLAB, abbreviated as PSTess. This computing package is a fork of the Power Systems Toolbox (PST). PST was originally developed at Rensselaer Polytechnic Institute (RPI) and later upgraded by Dr. Graham Rogers at Cherry Tree Scientific Software. While PSTess shares a common lineage with PST Version 3.0, it is a substantially different application. This document supplements the main PST manual by describing the features and models that are unique to PSTess. As the name implies, the main distinguishing characteristic of PSTess is its ability to model inverter-based energy storage systems (ESS). The model that enables this is called ess.m , and it serves the dual role of representing ESS operational constraints and the generator/converter interface. As in the WECC REGC_A model, the generator/converter interface is modeled as a controllable current source with the ability to modulate both real and reactive current. The model ess.m permits four-quadrant modulation, which allows it to represent a wide variety of inverter-based resources beyond energy storage when paired with an appropriate supplemental control model. Examples include utility-scale photovoltaic (PV) power plants, Type 4 wind plants, and static synchronous compensators (STATCOM). This capability is especially useful for modeling hybrid plants that combine energy storage with renewable resources or FACTS devices.

Subsurface energy activities such as unconventional resource recovery, enhanced geothermal energy systems, and geologic carbon storage require fast and reliable methods to account for complex, multiphysical processes in heterogeneous fractured and porous media. Although reservoir simulation is considered the industry standard for simulating these subsurface systems with injection and/or extraction operations, reservoir simulation requires spatio-temporal “Big Data” into the simulation model, which is typically a major challenge during model development and computational phase. In this work, we developed and applied various deep neural network-based approaches to (1) process multiscale image segmentation, (2) generate ensemble members of drainage networks, flow channels, and porous media using deep convolutional generative adversarial network, (3) construct multiple hybrid neural networks such as convolutional LSTM and convolutional neural network-LSTM to develop fast and accurate reduced order models for shale gas extraction, and (4) physics-informed neural network and deep Q-learning for flow and energy production. We hypothesized that physicsbased machine learning/deep learning can overcome the shortcomings of traditional machine learning methods where data-driven models have faltered beyond the data and physical conditions used for training and validation. We improved and developed novel approaches to demonstrate that physics-based ML can allow us to incorporate physical constraints (e.g., scientific domain knowledge) into ML framework. Outcomes of this project will be readily applicable for many energy and national security problems that are particularly defined by multiscale features and network systems.

Zero-emissions hydrogen fuel cell electrical vehicles (FCEVs) have become more popular in recent years. However, the limited availability of hydrogen fueling stations is considered a critical barrier to sustainable adoption of hydrogen FCEV. To enable the widespread deployment and commercialization of hydrogen FCEV, the availability of hydrogen refueling stations needs to improve. One of the consequences of the lack of hydrogen refueling infrastructure is that consumers can suffer from “range anxiety”, meaning consumers would get anxious of running out of fuel during long-distance trip [4]. A practical solution is to provide a compact emergency hydrogen refueler that can be used if the consumer runs out of hydrogen before reaching the nearest hydrogen refueling station. A safe, compact, and user-friendly hydrogen refueler would give consumers the flexibility they need to feel comfortable using their hydrogen FCEV when planning a long-distance trip. Offering this product would alleviate range anxiety, and it would make Hydrogen FCEV a more attractive alternative to gasoline vehicles. The emergency hydrogen refueler consists of a lithium hydride bed that reacts with liquid water to produce hydrogen gas and lithium hydroxide.

Abstract not provided.

Downscaling of the silicon metal-oxide-semiconductor field-effect transistor technology is expected to reach a fundamental limit soon. A paradigm shift in computing is occurring. Spin field-effect transistors are considered a candidate architecture for next-generation microelectronics. Being able to leverage the existing infrastructure for silicon, a spin field-effect transistor technology based on group IV heterostructures will have unparalleled technical and economical advantages. For the same material platform reason, germanium hole quantum dots are also considered a competitive architecture for semiconductor-based quantum technology. In this project, we investigated several approaches to creating hole devices in germanium-based materials as well as injecting hole spins in such structures. We also explored the roles of hole injection in wet chemical etching of germanium. Our main results include the demonstration of germanium metal-oxide-semiconductor field-effect transistors operated at cryogenic temperatures, ohmic current-voltage characteristics in germanium/silicon-germanium heterostructures with ferromagnetic contacts at deep cryogenic temperatures and high magnetic fields, evaluation of the effects of surface preparation on carrier mobility in germanium/silicon- germanium heterostructures, and hole spin polarization through integrated permanent magnets. These results serve as essential components for fabricating next-generation germanium-based devices for microelectronics and quantum systems.

Abstract not provided.

Rock Valley, in the southern end of the Nevada National Security Site, hosts a fault system that was responsible for a shallow (< 3 km below surface ) magnitude 3.7 earthquake in May 1993. In order to better understand this system, seismic properties of the shallow subsurface need to be better constrained. In April and May of 2021, accelerated weight drop (AWD) active-source seismic data were recorded in order to measure P- and S-wave travel-times for the area. This report describes the processing and phase picking of the recorded seismic waveforms. In total, we picked 7,982 P-wave arrivals at offsets up to ~2500 m, and 4,369 S-wave arrivals at offsets up to ~2200 m. These travel-time picks can be inverted for shallow P-wave and S-wave velocity structure in future studies.

Abstract not provided.

Abstract not provided.

Abstract not provided.

An active source experiment using an accelerated weight drop was conducted in Rock Valley, Nevada National Security Site, during the spring of 2021 in order to characterize the shallow seismic structure of the region. P-wave first arrival travel times picked from this experiment were used to construct a preliminary 3-D compressional wave speed model over an area that is roughly 4 km wide east-west and 8 km north-south to a depth of about 500-600 m below the surface, but with primary data concentration along the transects of the experimental lines. The preliminary model shows good correlation with basic geology and surface features, but geological interpretation is not the focus of this report. We describe the methods used in the tomographic inversion of the data and show results from this preliminary P-wave model.

Nonlocal models naturally handle a range of physics of interest to SNL, but discretization of their underlying integral operators poses mathematical challenges to realize the accuracy and robustness commonplace in discretization of local counterparts. This project focuses on the concept of asymptotic compatibility, namely preservation of the limit of the discrete nonlocal model to a corresponding well-understood local solution. We address challenges that have traditionally troubled nonlocal mechanics models primarily related to consistency guarantees and boundary conditions. For simple problems such as diffusion and linear elasticity we have developed complete error analysis theory providing consistency guarantees. We then take these foundational tools to develop new state-of-the-art capabilities for: lithiation-induced failure in batteries, ductile failure of problems driven by contact, blast-on-structure induced failure, brittle/ductile failure of thin structures. We also summarize ongoing efforts using these frameworks in data-driven modeling contexts. This report provides a high-level summary of all publications which followed from these efforts.

All energy production systems need efficient energy conversion systems. Current Rankine cycles use water to generate steam at temperatures where efficiency is limited to around 40%. As existing fossil and nuclear power plants are decommissioned due to end of effective life and/or societies’ desire for cleaner generation options, more efficient energy conversion is needed to keep up with increasing electricity demands. Modern energy generation technologies, such as advanced nuclear reactors and concentrated solar, coupled to high efficiency sCO₂ conversion systems provide a solution to efficient, clean energy systems. Leading R&D communities worldwide agree that the successful development of sCO₂ Brayton power cycle technology will eventually bring about large-scale changes to existing multi-billion-dollar global markets and enable power applications not currently possible or economically justifiable. However, all new technologies face challenges in the path to commercialization and the electricity sector is distinctively risk averse. The Sandia sCO₂ Brayton team needs to better understand what the electricity sector needs in terms of new technology risk mitigation, generation efficiency, reliability improvements above current technology, and cost requirements which would make new technology adoption worthwhile. Relying on the R&D community consensus that a sCO₂ power cycle will increase the revenue of the electrical industry, without addressing the electrical industry’s concerns, significantly decreases the potential for adoption at commercial scale. With a clear understanding of the market perspectives on technology adoption, including military, private sector, and utilities customers, the Sandia Brayton Team can resolve industry concerns for smoother development and faster transition to commercialization. An extensive customer discovery process, similar to that executed through the NSF’s I-Corp program, is necessary in order to understand the pain points of the market and articulate the value proposition of Brayton systems in terms that engage decision makers and facilitate commercialization of the technology.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Hole spin qubits confined to lithographically - defined lateral quantum dots in Ge/SiGe heterostructures show great promise. On reason for this is the intrinsic spin - orbit coupling that allows all - electric control of the qubit. That same feature can be exploited as a coupling mechanism to coherently link spin qubits to a photon field in a superconducting resonator, which could, in principle, be used as a quantum bus to distribute quantum information. The work reported here advances the knowledge and technology required for such a demonstration. We discuss the device fabrication and characterization of different quantum dot designs and the demonstration of single hole occupation in multiple devices. Superconductor resonators fabricated using an outside vendor were found to have adequate performance and a path toward flip-chip integration with quantum devices is discussed. The results of an optical study exploring aspects of using implanted Ga as quantum memory in a Ge system are presented.

Deep neural networks (NNs) typically outperform traditional machine learning (ML) approaches for complicated, non-linear tasks. It is expected that deep learning (DL) should offer superior performance for the important non-proliferation task of predicting explosive device configuration based upon observed optical signature, a task which human experts struggle with. However, supervised machine learning is difficult to apply in this mission space because most recorded signatures are not associated with the corresponding device description, or “truth labels.” This is challenging for NNs, which traditionally require many samples for strong performance. Semi-supervised learning (SSL), low-shot learning (LSL), and uncertainty quantification (UQ) for NNs are emerging approaches that could bridge the mission gaps of few labels and rare samples of importance. NN explainability techniques are important in gaining insight into the inferential feature importance of such a complex model. In this work, SSL, LSL, and UQ are merged into a single framework, a significant technical hurdle not previously demonstrated. Exponential Average Adversarial Training (EAAT) and Pairwise Neural Networks (PNNs) are chosen as the SSL and LSL methods of choice. Permutation feature importance (PFI) for functional data is used to provide explainability via the Variable importance Explainable Elastic Shape Analysis (VEESA) pipeline. A variety of uncertainty quantification approaches are explored: Bayesian Neural Networks (BNNs), ensemble methods, concrete dropout, and evidential deep learning. Two final approaches, one utilizing ensemble methods and one utilizing evidential learning, are constructed and compared using a well-quantified synthetic 2D dataset along with the DIRSIG Megascene.

Denoising contaminated seismic signals for later processing is a fundamental problem in seismic signals analysis. The most straightforward denoising approach, using spectral filtering, is not effective when noise and seismic signal occupy the same frequency range. Neural network approaches have shown success denoising local signal when trained on short-time Fourier transform spectrograms (Zhu et al 2018; Tibi et al 2021). Scalograms, a wavelet-based transform, achieved ~15% better reconstruction as measured by dynamic time warping on a seismic waveform test set than spectrograms, suggesting their use as an alternative for denoising. We train a deep neural network on a scalogram dataset derived from waveforms recorded by the University of Utah Seismograph Stations network. We find that initial results are no better than a spectrogram approach, with additional overhead imposed by the significantly larger size of scalograms. A robust exploration of neural network hyperparameters and network architecture was not performed, which could be done in follow on work.

Abstract not provided.

In order to establish a zone of influence (ZOI) due to a high energy arcing fault (HEAF) environment, the fragility of the targets must be determined. The high heat flux/short duration exposure of a HEAF is considerably different than that of a traditional hydrocarbon fire, which previous research has addressed. The previous failure metrics (e.g., internal jacket temperature of a cable exposed to a fire) were based on low heat flux/long duration exposures. Because of this, evaluation of different physics and failure modes was considered to evaluate the fragility of cables exposed to a HEAF. Tests on cable targets were performed at high heat flux/short duration exposures to gain insight on the relevant physics and failure modes. These tests yielded data on several relevant failure modes, including electrical failure and sustained ignition. Additionally, the results indicated a relationship between the total energy of exposure and the damage state of the cable target. This data can be used to inform the fragility of the targets.

Abstract not provided.

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.

Metallic lattice structures are being considered for shock mitigation applications due to their superior mechanical properties, energy absorption capability and lightweight characteristics inherent of the additive manufacturing process. In this study, shock compression experiments coupled to x-ray phase contrast imaging (PCI) were conducted on 316L stainless steel lattices. Meso-scale simulations incorporating the as-built lattice structure characterized by computed tomography were used to simulate PCI radiographs in CTH for direct comparison to experimental data. The methodology presented here offers robust validation for constitutive properties to further our understanding of lattice compaction at application-relevant strain rates.

This report details the results of a three-fold investigation of sensitivity analysis (SA) for machine learning (ML) explainability (MLE): (1) the mathematical assessment of the fidelity of an explanation with respect to a learned ML model, (2) quantifying the trustworthiness of a prediction, and (3) the impact of MLE on the efficiency of end-users through multiple users studies. We focused on the cybersecurity domain as the data is inherently non-intuitive. As ML is being using in an increasing number of domains, including domains where being wrong can elicit high consequences, MLE has been proposed as a means of generating trust in a learned ML models by end users. However, little analysis has been performed to determine if the explanations accurately represent the target model and they themselves should be trusted beyond subjective inspection. Current state-of-the-art MLE techniques only provide a list of important features based on heuristic measures and/or make certain assumptions about the data and the model which are not representative of the real-world data and models. Further, most are designed without considering the usefulness by an end-user in a broader context. To address these issues, we present a notion of explanation fidelity based on Shapley values from cooperative game theory. We find that all of the investigated MLE explainability methods produce explanations that are incongruent with the ML model that is being explained. This is because they make critical assumptions about feature independence and linear feature interactions for computational reasons. We also find that in deployed, explanations are rarely used due to a variety of reason including that there are several other tools which are trusted more than the explanations and there is little incentive to use the explanations. In the cases when the explanations are used, we found that there is the danger that explanations persuade the end users to wrongly accept false positives and false negatives. However, ML model developers and maintainers find the explanations more useful to help ensure that the ML model does not have obvious biases. In light of these findings, we suggest a number of future directions including developing MLE methods that directly model non-linear model interactions and including design principles that take into account the usefulness of explanations to the end user. We also augment explanations with a set of trustworthiness measures that measure geometric aspects of the data to determine if the model output should be trusted.

Efficient modeling of low magnetic Reynolds number (low-Rm) magnetohydrodynamics is often challenging and requires the implementation of innovative techniques to avoid key barriers experienced with prior approaches. We detail a new paradigm for first-principles simulation of the solution to the low-Rm governing equations in complex geometries. As a result of a number of innovative numerical advances, the next-generation GPU (graphics processing unit) accelerated physics code LGR has been successfully applied to the modeling of exploding wire problems.

Abstract not provided.

The main goal of this project was to create a state-of-the-art predictive capability that screens and identifies wellbores that are at the highest risk of catastrophic failure. This capability is critical to a host of subsurface applications, including gas storage, hydrocarbon extraction and storage, geothermal energy development, and waste disposal, which depend on seal integrity to meet U.S. energy demands in a safe and secure manner. In addition to the screening tool, this project also developed several other supporting capabilities to help understand fundamental processes involved in wellbore failure. This included novel experimental methods to characterize permeability and porosity evolution during compressive failure of cement, as well as methods and capabilities for understanding two-phase flow in damaged wellbore systems, and novel fracture-resistant cements made from recycled fibers.

Abstract not provided.