Siddiq M. Qidwai

In situ electron microscopy studies of electromechanical behavior in metals at the nanoscale using a novel microdevice-based system

Electrically assisted deformation (EAD) is an emerging technique to enhance formability of metals by applying an electric current through them. Despite its increasing importance in manufacturing applications, there is still an unresolved debate on the nature of the fundamental deformation mechanisms underlying EAD, mainly between electroplasticity (non-thermal effects) and resistive heating (thermal effects). This status is due to two critical challenges: (1) a lack of experimental techniques to directly observe fundamental mechanisms of material deformation during EAD, and (2) intrinsic coupling between electric current and Joule heating giving rise to unwanted thermally activated mechanisms. To overcome these challenges, we have developed a microdevice-based electromechanical testing system (MEMTS) to characterize nanoscale metal specimens in transmission electron microscopy (TEM). Our studies reveal that MEMTS eliminates the effect of Joule heating on material deformation, a critical advantage over macroscopic experiments, owing to its unique scale. For example, a negligible change in temperature (<0.02 °C) is predicted at ∼3500 A/mm2. Utilizing the attractive features of MEMTS, we have directly investigated potential electron-dislocation interactions in single crystal copper (SCC) specimens that are simultaneously subjected to uniaxial loading and electric current density up to 5000 A/mm2. Our in situ TEM studies indicate that for SCC, electroplasticity does not play a key role as no differences in dislocation activities, such as depinning and movement, are observed.

An Extended Finite Element Model of Crevice and Pitting Corrosion

A sharp interface model formulation is developed for simulating the electrochemical environment in crevices/pits due to galvanic corrosion in aqueous media. The concentration of ionic species and the electrical potential in the crevice is established using the non-dimensionalized Nernst-Planck equations along with the assumption of local electro-neutrality. The crevice/pit interface fluxes are defined in terms of the cathodic and anodic current densities using Butler-Volmer kinetics. The extended finite element method is used to discretize the governing equations and the level set function to describe the interface morphology independent of the underlying finite element mesh. The advantage of this formulation is that it eliminates the need for cumbersome mesh generation and remeshing when the interface morphology changes. Numerical investigations of steady-state intergranular crevice corrosion in idealized Al-Mg alloy microstructures in two-dimensions are conducted to establish the viability of the formulation. Simulation results predict large pH and chloride concentration within the crevice environment, which leads us to the conclusion that chemical reactions and precipitation play a prominent role during crevice corrosion.Copyright

Numerical Modeling of Pit Growth in Microstructure

This work presents the numerical modeling of two-dimensional stable corrosion pit growth by solving the Laplace equation which defines the electric potential within the electrolyte. Microstructural features representative of a 316 stainless steel provides the matrix in which the pit grows. Real microstructural features are incorporated into the computational model. The objective is to determine the influence of the microstructure, specifically crystallographic orientation, on the shape of the pit as it grows over time. The high-resolution definition of the microstructure is obtained by the orientation image microscopy (OIM) technique and is incorporated into the finite element model through a grid-based interpolation functionality. The steel-electrolyte corrosion front movement is simulated with the help of the arbitrary Lagrangian-Eulerian (ALE) meshing technique. The front speed, or the material dissolution rate, is approximated with the use of a Butler-Volmer relationship that relates the dissolution current density to the applied overpotential. The results show that small fluctuations (5–10%) in corrosion potential due to the changing crystal orientation ahead of the corrosion front result in variations in pit shape similar to experimental observations reported in the literature.Copyright

Computational Evaluation of Incomplete Coating Coverage

Corrosion is a major cause of removal from service for many industries. Pitting, which involves localized corrosion of metals, can result in catastrophic failures because of resulting crack initiation and failure. The transition from pit to crack is influenced by the pit shape, which in turn is affected by the microstructure of the corroding material. In this work the authors investigate the importance of understanding the construction of a coating layer that may be present over the pit mouth. The coating layer may be by-products of another activity or a repair coating meant to prevent further damage. Stable pit growth occurs under diffusion control at rate that depends upon the extent of protective coating over the opening of the pit. The two extreme cases are: 1) no cover due to total loss of coating and 2) full cover over an existing pit. The cases in between would represent break in coating cover over an evolving pit. To investigate the effect of coating loss, a parametric study based on coating coverage percentage on the metal is investigated. The coating sample length and gap length are taken to be the same for all cases. Coverage percentages of 0% (no coating), 50%, 75% and 100% (fully coated) are analyzed for a set growth time. Severe numerical complications are discovered in the course of these analyses. The movement of the corrosion front parallel to the spatially fixed coating causes considerable mesh distortion that terminates the simulation prematurely requiring an impractically large number of re-meshing steps. The computational concepts investigated will be discussed in addition to evaluating the influence of pit coverage.Copyright

New Metrics for Pit Shape Characterization

Currently a variety of approaches are used to match stress concentration factors with corrosion pit geometry. The majority of these approaches use standardized stress concentration factors, such as concentration factors for circles or ellipses, to estimate the maximum stress values along the pit front. These factors are based on regular geometric shapes. Pits that form in a microstructure are influenced by the individual grains surrounding the pit. These pits often do not have simple shapes. Use of standardized geometric factors do not capture the geometric complexity of the pit. Rather than a single parameter, such as a the aspect ratio of an ellipse, multiple parameters may be required to define the extent and variation in localized curvature along a pit front within a microstructure. Maximum depth and curvature are just two possible candidate metrics. In addition the authors looked to the medical field for potential metrics to adequately describe the convoluted nature of the pit front. Several methods have been developed to mathematically define the serpentine twists in diseased retinal blood vessels. In this work the authors present a methodology for determining characteristics including tortuosity of computationally predicted pit shapes embedded in microstructures. Ultimately it is hoped that maximum curvature, pit tortuosity and other geometric based metrics can be combined to predict the maximum rise in stress associated with a pit embedded in a microstructure.Copyright

Comparison of Mechanical Variable Identifiers of Brain Injury

Multiple mechanical variables have been used to describe the occurrence of brain injury in impact modeling of the human head [1, 2]. The validity of these variables for this purpose is usually established separately through the following process. First, a loading test is performed on an animal. Location, type and spatial extent of injury on the brain are measured upon or after loading. Subsequently, computational simulation is performed based on a particular constitutive model of the brain. Mechanical variables such as pressure or effective stress are plotted for the region of interest. The magnitude of the mechanical variable that results in a contour of the same size as the observed extent of experimental injury is declared as the critical value for that type of injury. The choice of mechanical variable itself could be based on conventional wisdom, precedence, or experience of the researcher. Another, much simpler variable-injury correlation process, which does not rely upon simulations, uses the ex vivo failure response of brain tissue as the criterion. For example, the uniaxial failure strain of the tissue may be taken as the critical value for injury.Copyright