Research

In this study, we report the grain boundary driven mechanical behavior of 2 polycrystalline ultra-high-temperature ceramics (UHTCs), zirconium diboride (ZrB2) and zirconium carbide (ZrC) with zirconium diboride (ZrC-ZrB2). These nanocomposites were investigated using large-scale molecular dynamics simulations. First, the atomistic models of the polycrystalline ZrB2 and ZrC-ZrB2 nanocomposites were subjected to tensile loading to determine their elastic constants and tensile strengths. It was found that the presence of nanoparticles imparts an insignificant effect on the mechanical properties of ZrB2. It has also been observed that the failure mechanisms of both the ZrB2 and ZrC-ZrB2 nanocomposite are driven by grain boundary deformation. At any instant during the applied load transfer, local tensile stress distribution data indicate that atomic stress becomes much higher near the grain boundaries compared to other locations. The authors performed additional sets of simulations to obtain tensile and shear properties of grain boundary material. When these properties were compared with the adjacent single crystal and overall polycrystalline material properties, it was found that the shear strength and stiffness of the grain boundary materials are significantly lower than the single crystal or polycrystal ZrB2. It is believed that the overall deformation and failure properties of ZrB2 and its composite are controlled by the properties of grain boundary. Hence, the addition of nanoparticles played an insignificant role on the mechanical properties of ZrB2.


Local stress (Stress/ Atom) plot of the model during tensile loading with strain level = 0, 13.4 and 16.7%. Local stress is much higher at the grain boundary compared to other atoms. The atoms are colored according to the stress tensor (Sigma33 and Tau13). Note that the local direction “3” is different from the global direction “z
  • Multi-scale modeling of Ultra high temperature ceramics (from Atomistic to Continuum scale)

vonmises .avi
Tensile test of ZrB2 polycrystalline model. The animation shows the von Mises stress(MPa) developed on ZrB2 for different strain states.

I present an atomistic computational study of electric field and thermal effects on the mechanical behavior of memristor material HfO2. Memristor materials are used for neuromorphic computation which promises to decrease energy consumption and improve the efficiency of important computational tasks, such as perception and decision making. In our study, first, the atomistic model of HfO2 is built on a monoclinic lattice structure. Then, tensile tests have been carried out to study its mechanical behavior. Since the material has non-symmetric crystal structure, we observe varied tensile properties along the x, y and z directions. In addition, the effects of electrical field on mechanical behavior are studied by varying the electrical field intensity from 0 to 0.3 v/Å gradually. For each case, atomistic snapshots are taken to identify the changes occur in the structure due to the electric field. A significant structural damage on the crystal structure of HfO2 is observed after applying 0.3 v/Å electric field, whereas the structural change is insignificant when the magnitude of the electric field is 0.2 v/Å or less. To understand more about the damage of this material, shear loads are applied in different directions and their responses are studied elaborately in this paper.


The efficiency of high-level nuclear waste immobilization by incorporating a host phase within a hollandite material structure can be increased by carefully synthesizing the dispersion phases inside the hollandite matrix. Also, estimation of the leaching rate from these nuclear waste forms is critical. Hence, conformal finite elemental model has been developed to study the effect of morphology of dispersive phases on diffusive nuclear flux. COMSOL Multi-physics is used as a computational tool to solve a Nernst-Plank Equation to study the diffusion leakage flux. A 2D model is built to identify the effect of volume fractions, surface areas, and different shapes of dispersion phase on the exit flux behavior. The results have indicated that there exists an optimum combination of different parameters such as volume fraction, surface area, position with respect to open boundary, and shape of dispersion phases for immobilization.

This study reports estimation of the amount of electrical power produced by thermoelectric generator (TEG) placed between flue gas duct and fresh air duct of an industrial thermal oil heater. Plate fin heat sink on hot and cold side of the TEG module was inserted into the flue gas and fresh air duct respectively. The effect of various design parameters, flow parameters were investigated in order to maximize the electrical power generation. Then the best suited conditions were applied to new thermoelectric generator module based on recently developed thermoelectric materials. A Bi2Te3 based commercial module (HZ-2) produce 3.7 W, where new module, based on p-type (Bi,Sb)2Te3 and n-type hot forged Bi2Te3 generate 4.4 W, at the same operating condition, which is about 19% improvement in output electrical power compared to commercial module. Estimated annual electrical power generation from this proposed system could be around 181,209 kW h. Thermal efficiency of the TEG modules based on recently developed thermoelectric materials could be enhanced up to 8.18%. The specifications of plate fin heat sinks as well as thermoelectric properties of the p-n materials of the system have substantial impact on the performance of TEG module.

  • Interface-driven elastic & failure properties of CNT reinforced Si nanocomposites.

The main objectives of this project is to measure elastic and failure properties of CNT reinforced Si nanocomposite. Due to the abundance of Si and its high electrical conduction properties, Si has extensive application in semiconductor industry. In recent time researches have been focused on developing multifunctional materials and since silicon is highly brittle in nature its mechanical application was limited. On the other hand Carbon Nanotube (CNT) made of single graphene sheet shows excellent elastic modulus if not the largest. These lead to the idea of CNT reinforced Si nanocomposite where it has been proposed that inclusion of CNT in bulk Si matrix may enhance the mechanical property of Si increasing its ductility. Keeping this goal in front, this group has modeled Si/CNT nanocomposite in tension using molecular dynamic (MD) simulation. For this the whole project was divided in three subwork where at first a single CNT was modeled and found its tensile strength followed by modeling bulk Si matrix and Si/CNT composite varying initial crack length and interfacial bonding strength respectively. Several qualitative observations have been made and the MD simulation results showed quite good prediction to experimental data.