In this paper, melting of long Al nanowires is studied using a phase field model in which deviatoric transformation strain described by a kinetic equation produces a promoting driving force for both melting and solidification and consequently, a lower melting temperature is resolved. The coupled system of the Ginzburg–Landau equation for solidification/melting transformation, the kinetic equation for the deviatoric transformation strain and elasticity equations are solved using the COMSOL finite element code to obtain the evolution of melt solution. A deviatoric strain kinetic coefficient is used which results in the same pressure as that calculated with the Laplace equation in a solid neglecting elastic stresses. The surface and bulk melting temperatures are calculated for different nanowire diameters without mechanical loading which shows a good agreement with existing MD and analytical results. For radii R > 5 nm, a complete surface solid-melt interface is created which propagates to the center. For smaller radii, premelting occurs everywhere starting from the surface and the nanowire melts without creating the interface. The melting rate shows an inverse power relationship with radius for R < 15 nm. For melting under pressure, the model with constant bulk modulus results in an unphysical parabolic variation versus pressure in contrast to the almost linear increase of the melting temperature versus pressure from known MD simulations. Such drawback is resolved by considering the pressure dependence of the bulk modulus through the Murnaghan’s equation due to which an almost linear increase of the melting temperature versus pressure is obtained. Also, a reduction of the interface width and a significant increase of the melting rate versus pressure are found. The presented model and results allow for a better understanding of the premelting and melting of different metallic nanowires with various loading conditions and structural defects.

The interaction between the crack and the grain boundary has been investigated by molecular dynamics simulation. The focus of this research is to study fracture resistance of grain boundary in a three-dimensional pre-cracked Fe-bicrystal. The fracture resistance of Σ5 < 100> {013} symmetric tilt grain boundary (STGB) has been compared with Σ5 < 100> {012} STGB in terms of crack length-time curve, temperature per time diagram, and the stress–strain curve to understand the detailed mechanism of fracture in Fe-bicrystal. Crack delay time at the grain boundary is proposed as a parameter for comparing the fracture resistance of grain boundaries. The results show that the crack delay time at the grain boundary is inversely related to the grain boundary energy. Hence, crack delay time for Σ5{013} STGB with 986 ${\displaystyle \frac{mj}{m^2}}$ energy is more than Σ5{012} STGB with 1098 ${\displaystyle \frac{mj}{m^2}}$ energy. The findings show that Σ5{013} STGB resists more than Σ5{012} STGB against crack propagation. The required stress, which is needed to overcome the grain boundary resistance and cause the crack penetration to the adjacent grain, has been calculated by using stress–strain curve. Modified BCC defect analysis algorithm and centrosymmetry parameter are also employed to analyze propagated defects and their interaction with the grain boundary.

In this paper, the nonlinear effect of interphase properties on the macroscopic plastic response of nylon 6/clay nanocomposites is investigated by applying a stochastic analysis on a multiscale computational model of nanocomposites. The mechanical behavior of interphase is described with respect to that of the matrix by a weakening coefficient. The interphase thickness and properties are considered as the stochastic inputs and the hardening modulus and hardening exponent describing the plastic hardening characteristics of the nanocomposite are the random outputs. The stochastic analysis consists of three procedures including (i) model selection using Akaike information criterion, (ii) uncertainty propagation using Latin Hypercube sampling in conjunction with chi-square test, and (iii) sensitivity analysis using Sobol indices. The results indicate that the exponential hardening model best describes the flow stress–plastic strain response of the nanocomposite. It is also shown that increasing the clay content generally increases the plastic hardening rate of the nanocomposite up to 4% clay content. Besides, the hardening characteristics of the nanocomposite are more sensitive to the weakening coefficient than the interphase thickness.