D.H. Warner

Critical length scale controls adhesive wear mechanisms

The adhesive wear process remains one of the least understood areas of mechanics. While it has long been established that adhesive wear is a direct result of contacting surface asperities, an agreed upon understanding of how contacting asperities lead to wear debris particle has remained elusive. This has restricted adhesive wear prediction to empirical models with limited transferability. Here we show that discrepant observations and predictions of two distinct adhesive wear mechanisms can be reconciled into a unified framework. Using atomistic simulations with model interatomic potentials, we reveal a transition in the asperity wear mechanism when contact junctions fall below a critical length scale. A simple analytic model is formulated to predict the transition in both the simulation results and experiments. This new understanding may help expand use of computer modelling to explore adhesive wear processes and to advance physics-based wear laws without empirical coefficients.

An Atomistic-Based Hierarchical Multiscale Examination of Age Hardening in an Al-Cu Alloy

A large class of modern structural alloys derives its strength from precipitation hardening. Precipitates obstruct the motion of dislocations and thereby increase alloy strength. This paper examines the process using an atomistic-based hierarchical multiscale modeling framework. Atomistic modeling is employed to (1) compute solute-dislocation interaction energies for input into a semi-analytic solute hardening model and (2) evaluate precipitate strengths for use in dislocation line tension simulations. The precipitate microstructure in the dislocation line tension simulations is obtained from simple analytic precipitation kinetics relations. Fitting only the rate constants in the precipitation kinetics model, the macroscopic strength predictions of the hierarchical multiscale model are found to correspond reasonably well with experiments. By analyzing the potential sources of discrepancy between the model’s macroscopic predictions and experiments, this work illuminates the importance of specific atomic-scale processes and highlights important challenges that remain before truly predictive mechanism-based plasticity modeling can be realized.

Extended timescale atomistic modeling of crack tip behavior in aluminum

Traditional molecular dynamics (MD) simulations are limited not only by their spatial domain, but also by the time domain that they can examine. Considering that many of the events associated with plasticity are thermally activated, and thus rare at atomic timescales, the limited time domain of traditional MD simulations can present a significant challenge when trying to realistically model the mechanical behavior of materials. A wide variety of approaches have been developed to address the timescale challenge, each having their own strengths and weaknesses dependent upon the specific application. Here, we have simultaneously applied three distinct approaches to model crack tip behavior in aluminum at timescales well beyond those accessible to traditional MD simulation. Specifically, we combine concurrent multiscale modeling (to reduce the degrees of freedom in the system), parallel replica dynamics (to parallelize the simulations in time) and hyperdynamics (to accelerate the exploration of phase space). Overall, the simulations (1) provide new insight into atomic-scale crack tip behavior at more typical timescales and (2) illuminate the potential of common extended timescale techniques to enable atomic-scale modeling of fracture processes at typical experimental timescales.

Effect of normal loading on grain boundary migration and sliding in copper

Molecular dynamics simulations of bicrystals are used to probe the influence of normal loading on grain boundary (GB) shear response. A Σ11(113) Cu GB is chosen to investigate GB migration, while Σ9(221) and Σ27(552) GBs are used to explore GB sliding. The shear strengths of the Σ11(113) and Σ9(221) GBs are found to be largely independent of the normal stresses acting on the GBs. However, the shear strength of the Σ27(552) GB was found to be nonmonotonically correlated with normal stress. The effect of intersecting partial dislocations at the GB is discussed for both GB sliding and migration.

On the debris-level origins of adhesive wear

Significance Wear causes a huge amount of material and energy losses annually, with serious environmental, economic, and industrial consequences. Despite considerable progress in the 19th century, the scientific understanding of wear remains mainly empirical. This study reveals the long-standing microscopic origins of material detachment from solids surface, at the most fundamental level, i.e., wear particles. It discloses that the detached particle volume can be estimated without any empirical factor, via the frictional work. This study unifies previously disconnected and not understood experimental observations. The results open the possibility for developing new wear models with drastically increased predictive ability, with applications to geophysics, physics, and engineering. Every contacting surface inevitably experiences wear. Predicting the exact amount of material loss due to wear relies on empirical data and cannot be obtained from any physical model. Here, we analyze and quantify wear at the most fundamental level, i.e., wear debris particles. Our simulations show that the asperity junction size dictates the debris volume, revealing the origins of the long-standing hypothesized correlation between the wear volume and the real contact area. No correlation, however, is found between the debris volume and the normal applied force at the debris level. Alternatively, we show that the junction size controls the tangential force and sliding distance such that their product, i.e., the tangential work, is always proportional to the debris volume, with a proportionality constant of 1 over the junction shear strength. This study provides an estimation of the debris volume without any empirical factor, resulting in a wear coefficient of unity at the debris level. Discrepant microscopic and macroscopic wear observations and models are then contextualized on the basis of this understanding. This finding offers a way to characterize the wear volume in atomistic simulations and atomic force microscope wear experiments. It also provides a fundamental basis for predicting the wear coefficient for sliding rough contacts, given the statistics of junction clusters sizes.

A continuously growing web-based interface structure databank

The macroscopic properties of materials can be significantly influenced by the presence of microscopic interfaces. The complexity of these interfaces coupled with the vast configurational space in which they reside has been a long-standing obstacle to the advancement of true bottom-up material behavior predictions. In this vein, atomistic simulations have proven to be a valuable tool for investigating interface behavior. However, before atomistic simulations can be utilized to model interface behavior, meaningful interface atomic structures must be generated. The generation of structures has historically been carried out disjointly by individual research groups, and thus, has constituted an overlap in effort across the broad research community. To address this overlap and to lower the barrier for new researchers to explore interface modeling, we introduce a web-based interface structure databank (www.isdb.cee.cornell.edu) where users can search, download and share interface structures. The databank is intended to grow via two mechanisms: (1) interface structure donations from individual research groups and (2) an automated structure generation algorithm which continuously creates equilibrium interface structures. In this paper, we describe the databank, the automated interface generation algorithm, and compare a subset of the autonomously generated structures to structures currently available in the literature. To date, the automated generation algorithm has been directed toward aluminum grain boundary structures, which can be compared with experimentally measured population densities of aluminum polycrystals.

On the Fatigue Performance of Additively Manufactured Ti-6Al-4V to Enable Rapid Qualification for Aerospace Applications

To realize the potential benefits of additive manufacturing technology in airframe and ground vehicle applications, the fatigue performance of load bearing additively manufactured materials must be understood. Due to the novelty of this rapidly developing technology, a very limited, yet swiftly evolving literature exists on the topic. Motivated by these two points, we have attempted to catalogue and analyze the published fatigue performance data of an additively manufactured alloy of significant technological interest, Ti-6Al-4V. Focusing on uniaxial fatigue performance and crack growth, we compare to traditionally manufactured Ti-6Al-4V, discussing failure mechanisms, defects, microstructure, and processing parameters. We then attempt to identify key knowledge gaps that must be addressed before AM technology can safely and effectively be employed in critical load bearing applications.

An interatomic pair potential with tunable intrinsic ductility

A family of interatomic potentials is constructed for which the intrinsic ductility can be tuned systematically. Specifically, the elastic constants and critical energy release rate for Griffith cleavage, G(Ic), are held constant, while the critical energy release rate for dislocation emission, G(Ie), can be varied. This behavior is achieved by modifying a standard near-neighbor pair potential; the new potential is applicable to either 2D (hexagonal lattice) or 3D (FCC/HCP). Analytical expressions are provided for GIe and GIc, enabling a potential with a desired intrinsic ductility to be easily developed. Direct atomistic simulations are used to demonstrate that the new potentials control the intrinsic material ductility, i.e. crack tip dislocation emission versus brittle cleavage, under quasi-static loading. For the 2D potential, the mode I crack tip behavior can be tuned from brittle to ductile; for the 3D potential, such tuning is only possible for certain crack orientations. More generally, the new potentials are expected to be useful in a wide range of physical problems in which behavior is controlled by the ability of the material to nucleate dislocations, including problems involving crack tips, grain boundaries, contact and friction, and bi-material interfaces.

Kohn–Sham density functional theory prediction of fracture in silicon carbide under mixed mode loading

The utility of silicon carbide (SiC) for high temperature structural application has been limited by its brittleness. To improve its ductility, it is paramount to develop a sound understanding of the mechanisms controlling crack propagation. In this manuscript, we present direct ab initio predictions of fracture in SiC under pure mode I and mixed mode loading, utilizing a Kohn–Sham Density Functional Theory (KSDFT) framework. Our results show that in both loading cases, cleavage occurs at a stress intensity factor (SIF) only slightly higher than the Griffith toughness, focusing on a (1 1 1) crack in the 3C-SiC crystal structure. This lattice trapping effect is shown to decrease with mode mixity, due to the formation of a temporary surface bond that forms during decohesion under shear. Comparing the critical mode I SIF to the value obtained in experiments suggests that some plasticity may occur near a crack tip in SiC even at low temperatures. Ultimately, these findings provide a solid foundation upon which to study the influence of impurities on brittleness, and upon which to develop empirical potentials capable of realistically simulating fracture in SiC.