Namas Chandra

Interfacial mechanics of push-out tests : theory and experiments

Abstract The thermo-mechanical characterization of interfaces in composite systems (PMC/MMC/IMC/CMC) is one of the challenging problems in composite mechanics and engineering. Each system has its own distinguishing features; however, in MMCs and IMCs the study is rendered more complex due to the evolving chemical species (both temporally and spatially), and the multi-axial state of residual stresses. Before MMCs or IMCs can be used in actual applications, the role of interfaces in not only the strengthening but also toughening mechanisms needs to be clearly understood. For evaluating the interfacial mechanical properties of interfaces, thin slice push-out test has emerged as the de-facto standard. Though, conceptually the testing procedure is simple, interpretation of the test results is not. It is essential to conduct very careful experiments, make precise meso- and macroscopic chemical/structural/mechanical observations and perform a thorough theoretical/numerical simulation before the test data can be used in a quantitative manner. In this paper, a comprehensive analysis of the push-out test is presented based on the theoretical/numerical and experimental research work of the authors’ group during the past few years. In this work, thin slice push-out tests were conducted primarily on Titanium Matrix Composites at various test temperatures (room and elevated) with different processing conditions (temperature and time). Different composite systems with Titanium based matrices (Ti–6Al–4V, Timetal 21S, Ti–15Nb–3Al) uniaxially reinforced with Silicon Carbide fibers (SCS-6) were chosen for the study. Effect of the evolution of interfacial chemistry and architecture (in matrix, coating and reaction zone) on both shear strength τ s and frictional strength τ f were studied. A novel finite element analysis based on nonlinear finite element method was implemented, in which not only the initiation but propagation of interfacial cracks are simulated. In the analysis, both shear stress and fracture energy based criteria are used to model the initiation of (closed) cracks. Quantitative values of τ s , τ f , G I and G II are then extracted based on the experimental data and the numerical simulation. A critical review of stress and energy based interface-modeling approaches and their applicability to various boundary value problems are made.

Multiscale Model to Study the Effect of Interfaces in Carbon Nanotube-Based Composites

In order to fully harness the outstanding mechanical properties of carbon nanotubes (CNT) as fiber reinforcements, it is essential to understand the nature of load transfer in the fiber matrix interfacial region of CNT-based composites. With controlled experimentation on nanoscale interfaces far off, molecular dynamics (MD) is evolving as the primary method to model these systems and processes. While MD is capable of simulating atomistic behavior in a deterministic manner, the extremely small length and time scales modeled by MD necessitate multiscale approaches. To study the atomic scale interface effects on composite behavior, we herein develop a hierarchical multiscale methodology linking molecular dynamics and the finite element method through atomically informed cohesive zone model parameters to represent interfaces. Motivated by the successful application of pullout tests in conventional composites, we simulate fiber pullout tests of carbon nanotubes in a given matrix using MD. The results of the pullout simulations are then used to evaluate cohesive zone model parameters. These cohesive zone models (CZM) are then used in a finite element setting to study the macroscopic mechanical response of the composites. Thus, the method suggested explicitly accounts for the behavior of nanoscale interfaces existing between the matrix and CNT. The developed methodology is used to study the effect of interface strength on stiffness of the CNT-based composite.

Atomistic simulation of grain boundary sliding and migration

Interatomic potentials using Embedded Atom Method (EAM) are used in conjunction with molecular statics and dynamics calculations to study the sliding and migration of [1 1 0] symmetric tilt grain boundaries (STGB) in aluminum, under both applied displacement and force conditions. For equilibrium grain boundaries (without applied displacements and forces), three low energy configurations (corresponding to three twin structures) are found in the [1 1 0] STGB structures when grain boundary energies at 0 K are computed as a function of grain misorientation angle. “Pure” grain boundary sliding (GBS) without migration is simulated by applying external displacement. When forces are applied, the energy barriers are reduced consequent to the fact that grain boundary sliding of STGB is always coupled with migration. The propensity for “pure” GBS is evaluated by computing the energy associated with incremental equilibrium configurations during the sliding process and compared to the case when sliding is accompanied by migration. The magnitude of the energy barriers is found to be much higher in “pure” GBS than when migration accompanies sliding. Relations between the applied force, internal stress field, and displacement field are established and the role of grain boundary structure on the deformation process are examined. It is found that the GBS displacement is proportional to applied force, GB energy, and time.

Constitutive behavior of superplastic materials

Superplasticity is an intriguing inelastic process in solid materials with deformation upto several thousand percent. Forming sheet and bulk materials using superplastic forming has become an established manufacturing method in aerospace and lately in other industries. Developing the right constitutive behavior is important not only for modeling the process for manufacturing by engineering mechanicians but for choosing the right composition and processing for material scientists. Such an ideal constitutive equation has been eluding the analysts so far. This paper examines some of the fundamental misgivings about the origin of inelastic process in superplasticity compared to other well known deformation processes. Also an attempt is made to understand the basic characteristics of superplastic inelastic deformation at macroscopic, mesoscopic and atomic levels.

Evaluation of interfacial fracture toughness using cohesive zone model

Abstract Interfaces play a critical role in determining the stiffness, strength and fracture properties of polymeric, metallic, and ceramic matrix composites. In this paper, while comparing the origin of interfaces in the three systems, attention is focused on the metal–(intermetal–) matrix composites. The roles of processing induced residual stresses, and the chemistry evolution during in service on the mechanical properties in general, and fracture properties in particular are delineated. Stress-based and energy-based failure criteria to model interfaces are described with examples drawn from Titanium matrix composites. Finally a detailed discussion on using cohesive zone models (CZMs) to describe fracture and failure of interfaces is presented. While it is contented that CZMs present the best alternative from physics and computational perspectives, it is emphasized that the choice of the specific form and parameters is very important.

Elevated temperature interfacial behaviour of MMCs : a computational study

Abstract Metallic and intermetallic matrix composites (MMCs and IMCs) are potential candidates for future use in the aerospace industry because of their high strength-to-weight ratio even at elevated temperatures. The thermomechanical behaviour of the fibre—matrix interface plays an important role in the successful application of this class of composites. The push-out test is emerging as an important experimental tool for characterizing the interfacial behaviour of MMCs and IMCs. In this study, the single-fibre push-out test is modelled using the finite element method, with the objectives of studying the interface failure process and extracting interfacial properties from the experimental test results. Earlier studies by the authors emphasized the significance of processing-induced residual stresses in titanium-based composites and their effects on push-out test results. In the present work, the developed methodology is used to study the interfacial behaviour during push-out tests at elevated temperatures. An attempt is made to predict interfacial shear strengths at elevated temperatures, by correlating the numerical simulations and the experimental results.

Scalable Time-Parallelization of Molecular Dynamics Simulations in Nano Mechanics

Molecular dynamics (MD) is an important atomistic simulation technique, with widespread use in computational chemistry, biology, and materials. An important limitation of MD is that the time step size is small, requiring a large number of iterations to simulate realistic time spans. Conventional parallelization is not very effective for this. We recently introduced a new approach to parallelization, where data from related prior simulations are used to parallelize a new computation along the time domain. In our prior work, the size of the physical system in the current simulation needed to be identical to that of the prior simulations. The significance of this paper lies in demonstrating a strategy that enables this approach to be used even when the physical systems differ in size. Furthermore, this method scaled up to almost 1000 processors with close to ideal speedup in one case, where conventional methods scale to only 2-3 processors

Defect–Defect Interaction in Carbon Nanotubes under Mechanical Loading

Topological defects are formed in carbon nanotubes (CNTs) during processing or subsequent deformation. The presence of defects is found to reduce the Young moduli of CNTs. When the number of defects is more than one, they may be either of the interacting or noninteracting type and this depends on whether the defects are within a specific interacting distance or not. A model for the reduction in Youngs modulus for noninteracting defects has been proposed and validated. However, nonlinear effects dominate for interacting defects. Deviation from linearity can be explained in terms of a transition region followed by a steady-state region governed by interacting distance and the size of the defect. The interaction phenomenon is explained by invoking the concepts of local stresses and strain measures rather than the conventional energy quantities.

Application of reduce order modeling to time parallelization

We recently proposed a new approach to parallelization, by decomposing the time domain, instead of the conventional space domain. This improves latency tolerance, and we demonstrated its effectiveness in a practical application, where it scaled to much larger numbers of processors than conventional parallelization. This approach is fundamentally based on dynamically predicting the state of a system from data of related simulations. In earlier work, we used knowledge of the science of the problem to perform the prediction. In complicated simulations, this is not feasible. In this work, we show how reduced order modeling can be used for prediction, without requiring much knowledge of the science. We demonstrate its effectiveness in an important nano-materials application. The significance of this work lies in proposing a novel approach, based on established mathematical theory, that permits effective parallelization of time. This has important applications in multi-scale simulations, especially in dealing with long time-scales.

Linking Atomistic and Continuum Mechanics Using Multiscale Models

Spatial and temporal limitations of atomistic simulations necessitate the development of multiscale methodologies that link atomic and continuum scales. In this paper, a hierarchical multiscale model is presented to study the thermomechanical behavior of carbon nanotube (CNT) based composites. The method explicitly accounts for the behavior of nanoscale interfaces existing between the matrix and CNT. Molecular dynamics simulations based on Tersoff‐Brenner potential are employed to study the interfaces in CNT based composites, while finite element method simulates macro behavior with the link provided by cohesive zone model parameters. The multiscale model is used to study the effect of interface strength on stiffness of the composite.