Applicability and evaluation of an implicit self-starting unconditionally stable methodology for the dynamics of structures

The applicability and evaluation of a new self-starting, unconditionally stable, implicit methodology of computation for the dynamics of structures is described. The methodology offers different perspectives and architecture for structural dynamics compared with the tranditional (widely advocated and commonly used) time integration methods. It is based on velocity representations and architecture and uses finite elements as the principal analysis tool for structural dynamic modeling/analysis. In particular, the dynamics of beam-type flexural models are considered, and comparative results validate and support the proposed use of the self-starting methodology of computation for the dynamics of linear/nonlinear structures. The overall effectiveness and elegance strongly support its use in most existing commercial codes.     

Theoretical study on strain-induced variations in electronic properties of monolayer MoS2

Ultrathin MoS₂ sheets and nanostructures are promising materials for electronic and optoelectronic devices as well as chemical catalysts. To expand their potential in applications, a fundamental understanding is needed of the electronic structure and carrier mobility as a function of strain. In this paper, the effect of strain on electronic properties of monolayer MoS₂ is investigated using ab initio simulations based on density functional theory. Our calculations are performed in both infinitely large two-dimensional (2D) sheets and one-dimensional (1D) nanoribbons which are theoretically cut from the sheets with semiconducting \( [\bar{1}100] \) (armchair) edges. The 2D crystal is studied under biaxial strain, uniaxial strain, and uniaxial stress conditions, while the 1D nanoribbon is studied under a uniaxial stress condition. Our results suggest that the electronic bandgap of the 2D sheet experiences a direct-indirect transition under both tensile and compressive strains. Its bandgap energy (E _g) decreases under tensile strain/stress conditions, while for an in-plane compression, E _g is initially raised by a small amount and then decreased as the strain varies from 0 to ?6 %. On the other hand, E _g at the semiconducting edges of monolayer MoS₂ nanoribbons is relatively invariant under uniaxial stretches or compressions. The effective masses of electrons at the conduction band minimum (CBM) and holes at the valence band maximum (VBM) are generally decreased as the in-plane extensions or compressions become stronger, but abrupt changes occur when CBM or VBM shifts between different k-points in the first Brillouin zone.     

Effect of the Strain Rate and Microstructure on Damage Growth in Aluminum

Materials used in soldier protective structures, such as armor, vehicles and civil infrastructures, are being improved for performance in extreme dynamic environments. Nanocrystalline metals show significant promise in the design of these structures with superior strengths attributed to the dislocation-based and grain-boundary-based processes as compared to their polycrystalline counterparts. An optimization of these materials, however, requires a fundamental understanding of damage evolution at the atomic level. Accordingly, atomistic molecular dynamics simulations are performed using an embedded-atom method (EAM) potential on three nano-crystalline aluminum atom systems, one a Voronoi-based nano-crystalline system with an average grain size of 10 nm, and the other two single crystals. These simulations are performed under the condition of uniaxial expansion at several strain rates ranging from 10⁶s^-1 to 10¹⁰s^-1. Results for the effective stress are discussed with the aim of establishing the role of the strain rate and microstructure on the evolution of the plastic strain and void volume fraction and the eventual loss of stress carrying capability of the atom systems.     

Convective Heat Transfer and Fluid Dynamic Characteristics of SiO2 Ethylene Glycol/Water Nanofluid

Nanofluids comprised of silicon dioxide (SiO₂) nanoparticles suspended in a 60:40 (% by weight) ethylene glycol and water (EG/water) mixture were investigated for their heat transfer and fluid dynamic performance. First, the rheological properties of different volume percents of SiO₂ nanofluids were investigated at varying temperatures. The effect of particle diameter (20 nm, 50 nm, 100 nm) on the viscosity of the fluid was investigated. Subsequent experiments were performed to investigate the convective heat transfer enhancement of nanofluids in the turbulent regime by using the viscosity values measured. The experimental system was first tested with EG/water mixture to establish agreement with the Dittus-Boelter equation for Nusselt number and with Blasius equation for friction factor. The increase in heat transfer coefficient due to nanofluids for various volume concentrations has been presented. Pressure loss was observed to increase with nanoparticle volume concentration. It was observed that an increase in particle diameter increased the heat transfer coefficient. Typical percentage increases of heat transfer coefficient and pressure loss at fixed Reynolds number are presented.     

An effective finite element modeling/analysis approach for dynamic thermoelasticity due to second sound effects

Of interest here are dynamic thermoelastic problems influenced by second sound effects. In this regard, the effect of the so called heat waves on solid continua is investigated employing a unified explicit computational architecture which uses the finite element method. The approach is robust and effective for transient interdisciplinary thermal-structural modeling/analysis. The non-classical relaxation model of Green and Lindsay (1972) involving two relaxation times is employed in the present work. Numerical simulations relevant to thermal shock problems in an elastic half-space are described for stainless steel via two different illustrative test cases.     

Numerical study of turbulent flow and heat transfer characteristics of nanofluids considering variable properties

Turbulent flow and heat transfer of three different nanofluids (CuO, Al₂O₃ and SiO₂) in an ethylene glycol and water mixture flowing through a circular tube under constant heat flux condition have been numerically analyzed. New correlations for viscosity up to 10% volume concentration for these nanofluids as a function of volume concentration and temperature are developed from the experiments and are summarized in the present paper. In our numerical study, all the thermophysical properties of nanofluids are temperature dependent. Computed results are validated with existing well established correlations. Nusselt number prediction for nanofluids agrees well with Gnielinski correlation. It is found that nanofluids containing smaller diameter nanoparticles have higher viscosity and Nusselt number. Comparison of convective heat transfer coefficient of CuO, Al₂O₃ and SiO₂ nanofluids have been presented. At a constant Reynolds number, Nusselt number increases by 35% for 6% CuO nanofluids over the base fluid.     

Atomistic Modeling of Spall Response in a Single Crystal Aluminum

Materials used in soldier protective structures, such as armor, vehicles and civil infrastructures, are being improved for performance in extreme dynamic environments. Accordingly, atomistic molecular dynamics simulations were performed to study the spall response in a single crystal aluminum atom system. A planar 9.6 picoseconds (ps) shock pulse was generated through impacts with a shock piston at velocities ranging from 0.6 km/s to 1.5 km/s in three <1,0,0>, <1,1,0>, and <1,1,1> crystal orientations. In addition to characterizing the transient spall region width and duration, the spall response was characterized interms of the traditional axial stress vs. axial strain response for gaining an understanding of the material failure in spall. Using an atom section averaging process, the snapshots, or the time history plots of the stress and strain axial distributions in the shock direction, were obtained from the MD simulations’ outputs of the atom level stresses and displacements. These snapshots guided the analyses to an estimation of the spall widths and spall transients. The results were interpreted to highlight the effects of crystal orientation and impact velocity on the spall width, spall duration, spall stress, strain rate, critical strain values at the void nucleation, and the void volume fraction at the void coalescence. For all the combinations of the crystal orientations and the impact velocities, the void nucleation was observed when the stress reached a peak hydrostatic state and the stress triaxiality factor reached a minimum, i.e. by the simultaneous occurring of these three conditions for the stress state: 1. pressure reaching a negative minimum, 2. axial stress reaching the magnitude value of the peak pressure, and 3. the effective stress reaching a zero value. At these conditions, void nucleation was mainly caused by atom de-bonding. In fact, the void nucleation strains were shown to have been preceded by the strains of the stress triaxiality condition in this study, thus confirming the stress triaxiality condition for the void nucleation in spall. Based on the observation that the axial stress reached a maximum value of ∼6 GPa during the void nucleation phase in spall and stayed approximately at that value for different crystal orientations and impact velocities, the value was proposed as a material spall strength.     

An effective data parallel self-starting explicit methodology for computational structural dynamics on the connection machine CM-5

This paper discusses the implementation aspects and our experiences towards a data parallel explicit self-starting finite element transient methodology with emphasis on the Connection Machine (CM-5) for linear and non-linear computational structural dynamic applications involving structured and unstructured grids. The parallel implementation criteria that influence the efficiency of an algorithm include the amount of communication, communication routing, and load balancing. To provide simplicity, high level of accuracy, and to retain the generality of the finite element implementation for both linear and non-linear transient explicit problems on a data parallel computer which permit optimum amount of communications, we implemented the present self-starting dynamic formulations (in comparison to the traditional approaches) based on nodal displacements, nodal velocities, and elemental stresses on the CM-5. Data parallel language CMFortran is employed with virtual processor constructs and with:SERIAL and:PARALLEL layout directives for arrays. The communications via the present approach involve only one gather operation (extraction of element nodal displacements or velocities from global displacement vector) and one scatter operation (dispersion of element forces onto global force vector) for each time step. These gather and scatter operations are implemented using the Connection Machine Scientific Software Library communication primitives for both structured and unstructured finite element meshes. The implementation aspects of the present self-starting formulations for linear and elastoplastic applications on serial and data parallel machines are discussed. Numerical test models for linear and non-linear one-dimensional applications and a two-dimensional unstructured finite element mesh are then illustrated and their performance studies are discussed.     

A computational approach for multi‐scale analysis of heterogeneous elasto‐plastic media subjected to short duration loads

The asymptotic expansion homogenization (AEH) approach has found wide acceptance for the study of heterogeneous structures due to its ability to account for multi‐scale features. The emphasis of the present study is to develop consistent AEH numerical formulations to address elasto‐plastic material response of structures subjected to short‐duration transient loading. A second‐order accurate velocity‐based explicit time integration method, in conjunction with the AEH approach, is currently developed that accounts for large deformation non‐linear material response. The approach is verified under degenerate homogeneous conditions using existing experimental data in the literature and its ability to account for heterogeneous conditions is demonstrated for a number of test problems. Copyright © 2004 John Wiley & Sons, Ltd.