SEARCH · Engineering Papers
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.
This perspective highlights the material properties, cell design decisions, and manufacturing costs with the biggest influence on the energy, power, cost, lifetime, and safety of a battery.
Rolling-element bearing material technology, examining fatigue effect of material properties and processing variables
Status of Graphite Oxidation Research Chronic Oxidation Oxidation of Irradiated Graphite Oxidation Resistant Graphite Material Property Changes with Oxidation
This project aims to establish a framework capable of efficiently predicting the properties of structural materials for service in harsh environments over a wide range of temperatures and over long periods of time. The approach is to develop and integrate high throughput first-principles calculations in combination with machine learning (ML) methods, perform high throughput CALPHAD (calculations of phase diagrams) modeling, and carry out finite element method (FEM) simulations. Relevant to high temperature service in fossil power system, nickel-based superalloys such as Inconel 740 and Haynes 282 as well as the associated (Ni-Cr-Co)-Al-C-Fe-Mn-Mo-Nb-Si-Ti system, were investigated. The present framework was built on the concept of phase-based property data, in which properties of individual phases are modeled as a function of internal and external independent variables. This project established an open-source infrastructure with the following capabilities: (1) High throughput implementation of first-principles calculations at finite temperatures and variable compositions using both accurate phonon calculations and the efficient Debye model for thermodynamic properties, elastic constants, diffusion coefficients, vacancy formation, stacking and twin faults, and dislocation mobility; i.e., using the developed code DFTTK; (2) Machine learning capabilities to predict the above properties so that the number of first-principles calculations can be significantly reduced; e.g., using the developed code SIPFENN; (3) High throughput CALPHAD modeling of the above properties as a function of temperature and composition using our unique capability based on ESPEI and PyCalphad; (4) New capabilities to predict the stress-strain behavior of individual phases; and (5) New models for tensile strength prediction in common FEM software with the crystal plasticity finite element simulations (CPFEM).
Here we introduce a denoising diffusion algorithm to discover microstructures with nonlinear fine-tuned properties. Denoising diffusion probabilistic models are generative models that use diffusion-based dynamics to gradually denoise images and generate realistic synthetic samples. By learning the reverse of a Markov diffusion process, we design an artificial intelligence to efficiently manipulate the topology of microstructures to generate a massive number of prototypes that exhibit constitutive responses sufficiently close to designated nonlinear constitutive behaviors. To identify the subset of microcstructures with sufficiently precise fine-tuned properties, a convolutional neural network surrogate is trained to replace high-fidelity finite element simulations to filter out prototypes outside the admissible range. Results of this study indicate that the denoising diffusion process is capable of creating microstructures of fine-tuned nonlinear material properties within the latent space of the training data. More importantly, this denoising diffusion algorithm can be easily extended to incorporate additional topological and geometric modifications by introducing high-dimensional structures embedded in the latent space. Numerical experiments are conducted on the open-source mechanical MNIST data set (Lejeune, 2020). Consequently, this algorithm is not only capable of performing inverse design of nonlinear effective media, but also learns the nonlinear structure–property map to quantitatively understand the multiscale interplay among the geometry, topology, and their effective macroscopic properties.
Low & high prandtl number approaches for heat transfer in laminar flow of incompressible fluids with constant material properties
This paper describes statistical procedures and their importance in obtaining composite material property values in designing structures for aircraft and military combat systems. The property value is such that the strength exceeds this value with a prescribed probability with 95% confidence in the assertion. The survival probabilities are the 99th percentile and 90th percentile for the A and B basis values respectively. The basis values for strain to failure measurements are defined in a similar manner. The B value is the primary concern of this paper.
Relational Information Management (RIM) was selected as the data base management system for a prototype engineering materials data base. The data base provides a central repository for engineering material properties data, which facilitates their control. Numerous RIM capabilities are exploited to satisfy prototype data base requirements. Numerical, text, tabular, and graphical data and references are being stored for five material types. Data retrieval will be accomplished both interactively and through a FORTRAN interface. The experience gained in creating and exercising the prototype will be used in specifying requirements for a production system.
Tensile and strain cycling fatigue properties of test materials for predicting life of turbine blade specimens
Progress in solar energy technology is reported in the following areas: aging and life prediction methodology and devices for solar cell encapsulation; the function of adhesion chemistry, primers, and a new diagnostic technique for estimations of bond durability; a study of fire retardant formulations for decreasing the potential flammability of solar modules; initial studies of the electrical insulating properties of encapsulation materials and measurement of the intrinsic dielectric strength; antisoiling compounds for the prevention of soil build-up on the outer surface of the module; and low temperature processing encapsulants that permit module fabrication at temperatures less than 100 C. Another area of study has been added to determine the degree to which formulation and processes affect the module quality and manufacturing yield.
The physics of spacecraft charging is reviewed, and criteria for selecting and testing semiinsulating polymers (SIPs) to avoid charging are discussed and illustrated. Chapters are devoted to the required properties of dielectric materials, the charging process, discharge-pulse phenomena, design for minimum pulse size, design to prevent pulses, conduction in polymers, evaluation of SIPs that might prevent spacecraft charging, and the general response of dielectrics to space radiation. SIPs characterized include polyimides, fluorocarbons, thermoplastic polyesters, poly(alkanes), vinyl polymers and acrylates, polymers containing phthalocyanine, polyacene quinones, coordination polymers containing metal ions, conjugated-backbone polymers, and 'metallic' conducting polymers. Tables summarizing the results of SIP radiation tests (such as those performed for the NASA Galileo Project) are included.
With the increased use of data-driven approaches and machine learning-based methods in material science, the importance of reliable uncertainty quantification (UQ) of the predicted variables for informed decision-making cannot be overstated. UQ in material property prediction poses unique challenges, including multi-scale and multi-physics nature of materials, intricate interactions between numerous factors, limited availability of large curated datasets, etc. In this work, we introduce a physics-informed Bayesian Neural Networks (BNNs) approach for UQ, which integrates knowledge from governing laws in materials to guide the models toward physically consistent predictions. To evaluate the approach, we present case studies for predicting the creep rupture life of steel alloys. Experimental validation with three datasets of creep tests demonstrates that this method produces point predictions and uncertainty estimations that are competitive or exceed the performance of conventional UQ methods such as Gaussian Process Regression. Additionally, we evaluate the suitability of employing UQ in an active learning scenario and report competitive performance. The most promising framework for creep life prediction is BNNs based on Markov Chain Monte Carlo approximation of the posterior distribution of network parameters, as it provided more reliable results in comparison to BNNs based on variational inference approximation or related NNs with probabilistic outputs.
In this study, different ratios and combinations of common epoxies, curing agents, and flexibilizers were used to find a material with desirable properties. Analysis of several material properties was conducted including glass transition temperature of each cured material which ensures high flexibility at a wide range of temperatures.
Additive manufacturing, cold rolling, and other thermomechanical treatments on metals, especially HCP, can cause texture and oriented grain structures resulting in anisotropic mechanical properties. This can lead to macroscopic response that significantly deviates from isotropic assumptions which motivates studying mechanical properties in multiple directions. In addition, multiaxial loading and complex stress states overlapping the above mentioned material anisotropy can lead to unexpected outcomes. Idaho National Laboratory developed an advanced mechanical testing system to study a range of uniaxial to multiaxial stress states while measuring the anisotropic response bringing insight into the overlap between material properties and stress states. High-temperature capability and stress- or strain-controlled loading is available to enable a variety of experiment types and conditions to measure elastic, plastic, and viscoplastic properties. This poster presents the design and capabilities of this system with preliminary results highlighting the benefits of multiaxial loading and anisotropic analysis in an integrated system.
The structural complexity and robust intermolecular interactions have challenged the incorporation of technical lignin into value-added polymeric materials for decades. To study the correlation between lignin molecular structure and material properties of lignin-based polyurethanes, we applied co-solvent enhanced lignocellulosic fractionation pretreatment followed by sequential precipitation to produce three distinct lignin preparations with narrowly distributed (molecular weight dispersity <2) and comparatively low molecular weight (<1500 g/mol) from poplar biomass. Structural characterization indicated that these lignin preparations differed in average molecular chain length and stiffness as well as hydroxyl group distribution. Secondary hydroxyl group providers such as aliphatic diols and polyethers were incorporated as building blocks into the lignin-based polyurethanes to provide additional hydrogen capacity to improve the dispersion of lignin in the polyurethane network. The selected aliphatic diols and polyethers interacted with lignin molecules at different levels of strength depending on their molecular structure, and their impacts were ultimately reflected in the mechanical and thermal properties of the resulting lignin-based polyurethanes. The copolymerization of technical lignin with tailored structure and secondary hydroxyl providers could provide new strategies in formulating lignin-based/containing polyurethanes for various functional applications.
The analysis of reflected ultrasonic waves induced by oblique insonification of composite materials is a powerful tool for providing informations about defects and material properties. A device was developed to manipulate a pair of transmitting and receiving transducers at vrious angles of wave incidence and propagation with the fiber orientation.
Recent advances in machine learning have demonstrated an enormous utility of deep learning approaches, particularly Graph Neural Networks (GNNs) for materials science. These methods have emerged as powerful tools for high-throughput prediction of material properties, offering a compelling enhancement and alternative to traditional first-principles calculations. While the community has predominantly focused on developing increasingly complex and universal models to enhance predictive accuracy, such approaches often lack physical interpretability and insights into materials behavior. Here, we introduce a novel computational paradigm—Self-Adaptable Graph Attention Networks integrated with Symbolic Regression (SA-GAT-SR)—that synergistically combines the predictive capability of GNNs with the interpretative power of symbolic regression. Our framework employs a self-adaptable encoding algorithm that automatically identifies and adjust attention weights so as to screen critical features from an expansive 180-dimensional feature space while maintaining O(n) computational scaling. The integrated SR module subsequently distills these features into compact analytical expressions that explicitly reveal quantum-mechanically meaningful relationships, achieving 23 × acceleration compared to conventional SR implementations that heavily rely on first-principle calculations-derived features as input. This work suggests a new framework in computational materials science, bridging the gap between predictive accuracy and physical interpretability, offering valuable physical insights into material behavior.
We report using carbon fibers (<100 μm in diameter) as ultrasonic waveguides to measure spatial changes in the environment and material properties. We connected carbon fibers of different lengths to an ultrasonic transducer and measured changes in the times of flight in a pulse-echo mode in response to elevated temperatures. By simultaneously interrogating multiple fibers of different lengths and collectively analyzing the time of flight in each fiber, we demonstrated dynamic measurements of the temperature distribution along the fiber bundle during their heating.