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.
Liquid encapsulated Czochralski crystal growth techniques for producing undoped, high resistivity, low dislocation material suitable for device applications is described. Technique development resulted in reduction of dislocation densities in 3 inch GaAs crystals. Control over the melt stoichiometry was determined to be of critical importance for the reduction of twinning and polycrystallinity during growth.
It has been noted that the crack propagation exponent p for most metals usually varies between values of 2 and 4, and that the motion of dislocations plays an important part in determining the exponent p. Attention is presently given to the significance of the exponent p in terms of the motion of dislocations, in view of the theory of thermally activated plastic flow and the cumulative plastic strain concept for a failure criterion.
The principle of minimum energy was applied to epitaxial interfaces to show the interrelationship beteen misfit, overgrowth thickness and misfit dislocation spacing. The low energy dislocation configurations were presented for selected interfacial geometries. A review of the interfacial energy calculations was made and a critical assessment of the agreement between theory and experiment was presented. Modes of misfit accommodation were presented with emphasis on the distinction between kinetic effects and equilibrium conditions. Two-dimensional and three-dimensional overgrowths were treated together with interdiffusion-modified interfaces, and several models of interfacial structure were treated including the classical and the current models. The paper is concluded by indicating areas of needed investigation into interfacial structure.
Silicon ribbon grown by the dendritic web process passes through a rapidly changing thermal profile in the growth direction. This rapidly changing profile induces stresses which produce changes in the dislocation density in the ribbon. A viscoplastic material response function (Haasen-Sumino model) is used herein to calculate the stresses and the dislocation density at each point in the silicon ribbon. The residual stresses are also calculated.
Inelastic deformation behavior of the cast Ni-base alloy, B1900 + Hf, was investigated using data from step-temperature tensile tests and thermomechanical cyclic tests in the temperature ranges 538-760 C and 760-982 C. The deformation results were correlated with the dislocation structures of deformed specimens, identified by TEM. It was found that, in the 760-982 C temperature range, there are no thermal history effects in the inelastic deformation behavior of B1900 + Hf. In the 538-760 range, anomalous cyclic hardening and, possibly, thermal history effects were observed in thermomechanically deformed alloy, caused by sessile (010) dislocations in the gamma-prime phase.
The source of the asymmetry in the dislocation morphology exhibited in the epitaxial growth of compound semiconductors on (100) was investigated. A thickness wedge of p- and n-type GaAs(0.95)P(0.05) was grown on GaAs by metalorganic chemical vapor deposition, and the effect of misorientation on the resolved shear stress for each slip system was calculated and eliminated as the source of the asymmetry. Another potential source of asymmetry, the thickness gradient, was also eliminated. Results show that the substrate misorientation and the thickness gradient do not significantly contribute to the asymmetry and that the dominant contributor to the asymmetry of misfit dislocations in the (001) epitaxial interface can be attributed to the differences in the Peierls barriers between the two types of dilocations in GaAsP/GaAs.
Limited ductility and toughness of NiAl and related aluminides near room temperature pose major problems in their potential application as structural materials. An analysis of these problems is presented as part of a review of the flow and fracture behavior of binary NiAl. Following this discussion is a demonstration that conditions of elastic and plastic constraint associated with phase boundaries afforded by surface films, internal lamellae, or precipitates may introduce sufficient densities of mobile dislocations to enhance the ductility of NiAl-based materials by significant amounts. Examples of this behavior are presented for several model materials, including 001- and 123-oriented single crystals of oxide-coated NiAl, directionally solidified beta-gamma-prime (Ni70Al30) and beta-gamma (Ni50Fe30Al20) in situ composites, and several NiAl/precipitate systems. The nature of the resulting dislocation substructures and the effects of several materials variables are described.
An evaluation is undertaken of the implications of the friction and frictionless models of misfit dislocation generation in view of: (1) experimental measurements of the critical thickness above which misfit dislocation generation occurs; and (2) the amount of strain relaxation that occurs as a function of layer thickness, time, and temperature. Some of the frictional force terms that were expected to exhibit a strong temperature dependence are shown to be independent of temperature.
Calculations are made to study the dependence of heteroepitaxial InP solar-cell efficiency on dislocation density. Effects of surface recombination velocity and cell emitter thickness are considered. Calculated results are compared with the available experimental results on representative InP solar cells. It is shown that heteroepitaxial InP cells with over 20 percent AM0 efficiency could be fabricated if dislocations are reduced to less than 100,000/sq cm.
Environments that impose operational constraints on conventional silicon-(Si) based semiconductor devices frequently appear in military- and space-grade applications. These constraints include high temperature, high power, and high radiation environments. Silicon carbide (SiC), an alternative type of semiconductor material, has received abundant research attention in the past few years, owing to its radiation-hardened properties as well as its capability to withstand high temperatures and power levels. However, the growth and manufacture of SiC devices is still comparatively immature, and there are severe limitations in present crystal growth and device fabrication processes. Among these limitations is a variety of crystal imperfections known as defects. These imperfections can be point defects (e.g., vacancies and interstitials), line defects (e.g., edge and screw dislocations), or planar defects (e.g., stacking faults and double-positioning boundaries). All of these defects have been experimentally shown to be detrimental to the performance of electron devices made from SiC. As such, it is imperative that these defects are significantly reduced in order for SiC devices to become a viable entity in the electronics world. The NASA Glenn High Temperature Integrated Electronics & Sensors Team (HTIES) is working to identify and eliminate these defects in SiC by implementing improved epitaxial crystal growth procedures. HTIES takes two-inch SiC wafers and etches patterns, producing thousands of mesas into each wafer. Crystal growth is then carried out on top of these mesas in an effort to produce films of improved quality-resulting in electron devices that demonstrate superior performance-as well as fabrication processes that are cost-effective, reliable, and reproducible. In this work, further steps are taken to automate HTIES' SiC wafer inspection system. National Instruments LabVIEW image processing and pattern recognition routines are developed that are capable of quantifying and mapping defects on both the substrate and mesa surfaces, and of quantifying polymorphic changes in the grown materials. In addition, an optical emission microscopy (OEM) system is developed that will facilitate comprehensive study of recombination-enhanced dislocation motion (REDM).
A report describes an experimental study of dislocation-induced spatial ordering of quantum dots (QDs) comprising nanometer-sized In(x)Ga(1-x)As islands surrounded by GaAs. Metastable heteroepitaxial structures were grown by molecular-beam epitaxy of In(x)Ga(1-x)As onto n+ GaAs and semi-insulating GaAs substrates. Then the structures were relaxed during a post-growth annealing/self-organizing process leading to the formation of surface undulations that acted as preferential sites for the nucleation of QDs. Structural effects of annealing times and temperatures on the strain-relaxed In(x)Ga(1-x)As/GaAs and the subsequent spatial ordering of the QDs were analyzed by atomic-force microscopy and transmission electron microscopy. Continuous-wave spectral and time-resolved photoluminescence (PL) measurements were performed to study the effects, upon optical properties, of increased QD positional ordering, increased QD uniformity, and proximity of QDs to arrays of dislocations. PL spectral peaks of ordered QD structures formed on strain-relaxed In(x)Ga(1-x)As/GaAs layers were found to be narrower than those of structures not so formed and ordered. Rise and decay times of time-resolved PL were found to be lower at lower temperatures -- apparently as a consequence of decreased carrier-transport times within the barriers surrounding the QDs.
A multiscale modeling methodology that combines the predictive capability of discrete dislocation plasticity and the computational efficiency of continuum crystal plasticity is developed. Single crystal configurations of different grain sizes modeled with periodic boundary conditions are analyzed using discrete dislocation plasticity (DD) to obtain grain size-dependent stress-strain predictions. These relationships are mapped into crystal plasticity parameters to develop a multiscale DD/CP model for continuum level simulations. A polycrystal model of a structurally-graded microstructure is developed, analyzed and used as a benchmark for comparison between the multiscale DD/CP model and the DD predictions. The multiscale DD/CP model follows the DD predictions closely up to an initial peak stress and then follows a strain hardening path that is parallel but somewhat offset from the DD predictions. The difference is believed to be from a combination of the strain rate in the DD simulation and the inability of the DD/CP model to represent non-monotonic material response.
A multiscale modeling methodology is developed for structurally-graded material microstructures. Molecular dynamic (MD) simulations are performed at the nanoscale to determine fundamental failure mechanisms and quantify material constitutive parameters. These parameters are used to calibrate material processes at the mesoscale using discrete dislocation dynamics (DD). Different grain boundary interactions with dislocations are analyzed using DD to predict grain-size dependent stress-strain behavior. These relationships are mapped into crystal plasticity (CP) parameters to develop a computationally efficient finite element-based DD/CP model for continuum-level simulations and complete the multiscale analysis by predicting the behavior of macroscopic physical specimens. The present analysis is focused on simulating the behavior of a graded microstructure in which grain sizes are on the order of nanometers in the exterior region and transition to larger, multi-micron size in the interior domain. This microstructural configuration has been shown to offer improved mechanical properties over homogeneous coarse-grained materials by increasing yield stress while maintaining ductility. Various mesoscopic polycrystal models of structurally-graded microstructures are generated, analyzed and used as a benchmark for comparison between multiscale DD/CP model and DD predictions. A final series of simulations utilize the DD/CP analysis method exclusively to study macroscopic models that cannot be analyzed by MD or DD methods alone due to the model size.
Homoepitaxial CVD growth of thin lateral cantilevers emanating from the edges of mesa patterns dry-etched into on-axis commercial 4H-SiC substrates prior to growth is reported. Cantilevers on the order of a micrometer thick extending tens of micrometers from the edge of a mesa have been grown. The termination of vertically propagating screw dislocations, including a micropipe, that are overgrown by the cantilevers has been demonstrated, in large part because the crystal structure of the cantilevers is established laterally from the mesa sidewalls. This technique could help reduce performance-degrading dislocations in SiC electrical devices.
The microstructure of metals is key to their mechanical properties. The types, density, composition and morphology of crystal defects all have pronounced impact on the properties. Changes to the microstructure occurring during processing and use can be very striking. The emerging technology additive manufacturing (AM) has the potential to improve performance by allowing optimized designs, but the process and environments can lead to unusual microscale features whose properties must be understood and characterized to enable higher technological readiness levels and application. Experimentally, an extensive evaluation of mechanical properties of 3D printed metals is a challenge, and anomalous effects related to the AM process add complexity. We present a new machine learning (ML) model predicting mechanical response based on dislocation mediated plasticity simulations. A large set of 3D discrete dislocation dynamics simulations with wide ranges of loading conditions is transformed to preprocessed data ready for training with the ML model. The trained model can predict the mechanical response of Mo30W for a given microstructure evolution, providing key information essential for optimization of AM processing.
The microstructure of metals is key to their mechanical properties. The types, density, composition and morphology of crystal defects all have pronounced impact on the properties. Changes to the microstructure occurring during processing and use can be very striking. The emerging technology additive manufacturing (AM) has the potential to improve performance by allowing optimized designs, but the process and environments can lead to unusual microscale features whose properties must be understood and characterized to enable higher technological readiness levels and application. Experimentally, an extensive evaluation of mechanical properties of 3D printed metals is a challenge, and anomalous effects related to the AM process add complexity. We present a new machine learning (ML) model predicting mechanical response based on dislocation mediated plasticity simulations. A large set of 3D discrete dislocation dynamics simulations with wide ranges of loading conditions is transformed to preprocessed data ready for training with the ML model. The trained model can predict the mechanical response of Mo30W for a given microstructure evolution, providing key information essential for optimization of AM processing.
Continuum-scale material deformation models, such as crystal plasticity (CP), can significantly enhance their predictive accuracy by incorporating input from lower-scale (i.e. mesoscale) models. The procedure to generate and extract the relevant information is however typically complex and ad hoc, involving decision and intervention by domain experts, leading to long development times. In this study, we develop a principled approach for calibration of continuum-scale models using lower scale information by representing a CP flow rule as a Gaussian process model. This representation allows for efficient parameter space exploration, guided by the uncertainty embedded in the model through a process known as Bayesian optimization (BO). We demonstrate a semi-autonomous BO loop which instantiates discrete dislocation dynamics simulations whose initial conditions are automatically chosen to optimize the uncertainty of a model CP flow rule. Our self-guided computational pipeline efficiently generated a dataset and corresponding model whose error, uncertainty, and physical feature sensitivities were validated with comparison to an independent dataset four times larger, demonstrating a valuable and efficient active learning implementation readily transferable to similar material systems.
Effects of second phase size and rigidity on stresses associated with blocked screw and edge dislocation arrays