RESEARCH PROJECTS


Multi-Scale Simulation of Dynamic Strain Aging in Al-Mg Alloy

Sponsors: National Science Foundation and Alcoa, Inc.
Participants:

Summary:

The objective of this project is to develop a model to predict phenomena leading to poor formability in Al-Mg alloys based on the underlying deformation mechanisms. Magnesium is added to aluminum to improve strength properties but is found to also have a detrimental effect on formability. The microstructural mechanism responsible for poor formability is a process called dynamic strain aging. Dynamic strain aging results from fast diffusing solute atoms, such as magnesium, interacting with dislocations. These interactions lead to unsteady, collective motion of dislocations, within grains and across grain boundaries, resulting in unstable flow and hence poor formability. At the macroscopic level, a key defining manifestation of dynamic strain aging is a negative strain rate sensitivity. Predicting this phenomenon and the resulting poor formability, based on microstructural events, is the primary objective of this work. More information.


Collaborative Research: Modeling Microstructure Evolution during Hot Bulk Forming of Al-Mg-Si Alloys

Sponsor: National Science Foundation
Participants:

Summary:

Al-Mg-Si alloys are among the most common materials used in aerospace, construction, and automotive industries. The ability to predict final microstructure, and therefore mechanical properties, in a final aluminum part that results form a controlled deformation process is extremely important for the US aluminum industry and its customers. The objective of this collaborative research project is to develop a simulation tool, validated by experiments, capable of predicting the evolution of key microstructural characteristics in Al-Mg-Si alloys during hot, bulk forming. Specifically, the orientation distribution of grains (texture), their size and shape, and the precipitate size and distribution are of interest. It is microstructural characteristics, such as these, that are responsible for the macroscopic behavior of materials. This work will guide process designers to design processes that are lower in cost and produce aluminum products with improved material characteristics.


Determining Stiffness in Human Tissue from Ultrasonic Measurements

Sponsors: National Science Foundation and National Institutes of Health.
Participants:

Summary:

The objective of this research program is to explore methods to improve and extend the information content and the spatial resolution of elastographic images. The elastic properties of tissue are directly related to the underlying structure of the tissue and are therefore strongly affected by pathological changes in the tissue. To be precise, it is the elastic property of shear stiffness, i.e. the property that characterizes resistance to shape change, that is most affected by pathelogical changes. Therefore, the ability to image, with high resolution, the shear stiffness field, or the related property of shear wave speed, would be an extremely valuable diagnostic tool. The long-term goal of this work is to develop methods to achieve this in real-time for an ultrasound-based system developed by collaborator, Prof. M. Fink of the Laboratoire Ondes et Acoustique, E.S.P.C.I., Universite Paris VII. To be specific, the problem to be solved is to determine the 3-D shear stiffness field from ultrasound measurements of interior displacements over time as a pulse-induced elastic wave travels through the body. In mathematics, this is referred to as an inverse problem of parameter identification. These problems are known to be typically ill-posed, meaning the solutions are unstable, and thus, very sensitive to noise in the data, unless special, "intelligent" algorithms are used. In the problem of interest, the primary source of instability is due to the need to differentiate the data twice, if a direct solution method is used. The algorithms that are being explored in this work, either eliminate the need to directly differentiate the displacement data altogether, or reduce the highest derivative to one. Level set methods based on arrival time mapping, geometric optic methods, and finite element based methods are be investigated. Furthermore, the algorithms need to be fast, so fast algorithms and parallel computational methods are being explored too. Another important aspect of this work is to analyze the errors, due to the model, data, and method. Detailed forward finite element models are being developed that include complex material behavior, such as anisotropy and viscoelasticity, to generate simulated data to isolate and investigate these effects. The forward finite element model will also be used to determine sensitivities and identify rich data subsets.


Optimal Design of Bulk Forming Processes

Sponsor: National Science Foundation
Participants:

Summary:

The objective of this work is to develop a simulation tool, which would determine the optimal process geometry, temperature, and speed for a given forming process that will also satisfy design criteria specified by a designer. These design criteria may include objectives, for example produce a product with certain material characteristics or minimize the production time, and they may also include certain design constraints, such as geometric constraints. The work involves developing an optimization algorithm based on an efficient finite element formulation for modeling large deformation, thermo-elasto-viscoplastic contact problems, typical in metal forming.


X-Ray Microbeam Studies of Electromigration

Sponsor: National Science Foundation
Participants:

Summary:

Both electric and stress fields in electronic interconnects drive diffusional processes that can lead to the structural failure of a device through void formation. The grain structure and grain orientation distribution of an interconnect affect these diffusional processes, and thus affect the device reliability. Furthermore, the interconnect feature scale is on the same order as the grain scale. Thus, in order to predict the structural reliability of an interconnect, it is desireable to understand and model these diffusion processes in realistic, three-dimensional grain structures. In this research, x-ray microbeam diffraction and flourescence and other experimental techniques, together with microstructure dependent, grain scale models, and numerical simulations will be used to study the basic underlying mechanisms associated with electromigration in Cu-based conductor lines and thin films. A three-dimensional finite element formulation for modeling electromigration and stress driven diffusion is under development. The model allows for three-dimensional grain structures, grain anisotropy, and non-uniform current density.


Solving Inverse Problems for Determining Unknown Interface Conditions

Participants:

Summary:

Interface conditions, such as tractions and heat flux conditions on the interface between contacting bodies, have a significant impact on wear and surface quality. It is generally difficult to directly measure these interface conditions, however, in many cases, easier and more accurate measurements of temperatures and strains or displacements can be made at regions outside of the contact zone. Problems of interest involve determining unknown traction and heat flux boundary conditions in the contact zone of coupled thermomechanical contact problems using strain (or displacement) and temperature measurements outside of the contact zone. This is a type of inverse problem because the boundary conditions (typically required for a unique solution in mathematical physics) are unknown on part of the boundary and over-specified on another part of the boundary (i.e. part of what is typically the solution is known).

Back to Maniatty Research Group Page

Back to Maniatty's Home Page