Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “polycrystal metals”

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.

9 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2009

The Development of Directional Decohesion Finite Elements for Multiscale Failure Analysis of Metallic Polycrystals

Atomistic simulations of intergranular fracture have indicated that grain-scale crack growth in polycrystalline metals can be direction dependent. At these material length scales, the atomic environment greatly influences the nature of intergranular crack propagation, through either brittle or ductile mechanisms, that are a function of adjacent grain orientation and direction of crack propagation. Methods have been developed to obtain cohesive zone models (CZM) directly from molecular dynamics simulations. These CZMs may be incorporated into decohesion finite element formulations to simulate fracture at larger length scales. A new directional decohesion element is presented that calculates the direction of Mode I opening and incorporates a material criterion for dislocation emission based on the local crystallographic environment to automatically select the CZM that best represents crack growth. The simulation of fracture in 2-D and 3-D aluminum polycrystals is used to illustrate the effect of parameterized CZMs and the effectiveness of directional decohesion finite elements.

Saether, Erik↗
NASA NTRSDate not supplied

Geometric Requirements for Modeling Polycrystalline Representative Volume Elements Using the Generalized Method of Cells

To accurately predict the mechanical response of advanced metallic parts, the grain structure of the metal must be considered and coupled to the macroscale in a multiscale model. The size of the microscale representative volume element must be sufficiently large, and the grid used to discretize the microscale must be adequately refined, in order to ensure an accurate and converged solution. The generalized method of cells is used to predict the mechanical properties of polycrystalline representative volume elements of copper containing randomly oriented grains with the properties of a single face-centered-cubic crystal. Moreover, the results of a parametric study are used to conclude the geometric requirements for an objective microscale model, with respect to the elastic stiffness.

Multiscale Modeling↗
NASA NTRS1991

Critical experiments of the self-consistent model for polycrystalline Hastelloy-X

A viscoplastic constitutive model is presented for the estimation of the overall mechanical response of Hastelloy-X polycrystalline metals from a knowledge of single crystal behavior. The behavior of polycrystal is derived from that of single crystals using a self-consistent formulation. The single crystal behavior which has been used was developed by summing postulated slip on crystallographic slip systems. The plasticity and creep are treated coupledly using unified viscoplastic model which includes the interaction effects between rapid and slow deformation at elevated temperature. The validity of the model is directly tested by experiments on Hastelloy-X in both single crystal and polycrystalline versions.

Shi, Shixiang↗
NASA NTRS1995

Investigation of Strain Aging in the Ordered Intermetallic Compound beta-NiAl

The phenomenon of strain aging has been investigated in polycrystalline and single crystal NiAl alloys at temperatures between 300 and 1200 K. Static strain aging studies revealed that after annealing at 1100 K for 7200 s (i.e., 2h) followed by furnace cooling, high purity, nitrogen-doped and titanium-doped polycrystalline alloys exhibited continuous yielding, while conventional-purity and carbon-doped alloys exhibited distinct yield points and Luders strains. Prestraining by hydrostatic pressurization removed the yield points, but they could be reintroduced by further annealing treatments. Yield points could be reintroduced more rapidly if the specimens were prestrained uniaxially rather than hydrostatically, owing to the arrangement of dislocations into cell structures during uniaxial deformation. The time dependence of the strain aging events followed at t(exp 2/3) relationship suggesting that the yield points observed in polycrystalline NiAl were the result of the pinning of mobile dislocations by interstitials, specifically carbon. Between 700 and 800 K, yield stress plateaus, yield stress transients upon a ten-fold increase in strain rate, work hardening peaks, and dips in the strain rate sensitivity (SRS) have been observed in conventional-purity and carbon-doped polycrystals. In single crystals, similar behavior was observed; in conventional-purity single crystals, however, the strain rate sensitivity became negative resulting in serrated yielding, whereas, the strain rate sensitivity stayed positive in high purity and in molybdenum-doped NiAl. These observations are indicative of dynamic strain aging (DSA) and are discussed in terms of conventional strain aging theories. The impact of these phenomena on the composition-structure-property relations are discerned. Finally, a good correlation has been demonstrated between the properties of NiAl alloys and a recently developed model for strain aging in metals and alloys developed by Reed-Hill et al.

Weaver, Mark Lovell↗
NASA NTRSDate not supplied

A Convolutional Neural Network for Enhancement of Multi-Scale Localization in Granular Metallic Representative Unit Cells

A convolutional neural network was used to enhance the localization of strain and stress for a generalized method of cells model of a metallic microstructure. Enhanced shear strains, measured in terms of the linear regression coefficients as a function of ground truth strains, were improved from inaccurate and uncorrelated (slope=0.003, Rsq=0.000) to accurate and well correlated (slope=0.890, Rsq=0.882) relative to ground truth (slope=1.0, Rsq=1.0). In applying the convolutional neural network, a convolutional stride of 1.0 (padding=’same’) was only modestly effective while strides of 2 or 3 were more effective yet at higher cost. Additional convolutional layers were generally more expensive than additional dense layers, often with limited benefit. The accuracy of enhanced localized shear strains and stress is expected to yield benefits for damage progression models, especially in the context of hierarchical multi-scale methods where the generalized method of cells is applied at the intermediate scale.

Machine Learning↗