Investigation of mechanical behavior of graphene reinforced aluminum nanocomposites

Graphene is 2D material, which consists of C-atom configured hexagonal lattice in one atom thickness. The specific arrangement of atoms causes to emerge versatile extraordinary properties, i.e. low density, high electrical and heat conductivity, flexibility, transparency and high strength. That is why graphene is the prospective material according to the scientific community. Significant numbers of studies have been conducted to predict mechanical properties of graphene. Physical experiments, molecular dynamic simulations (MDS) and multi-scale Analyses are the principal methods of the related studies. Due to nanoscale size of graphene, physical tests become troublesome and costly. In this point, computational simulations are of great importance. Advancement of computational technologies makes possible modeling physical problem almost each time and dimension scale and solving it. In this thesis, mechanical behavior of graphene reinforced Al nanocomposites are investigated utilizing MDS and multi-scale analyses. In each methodology, analyses are conducted in two-level, i.e. only single layer graphene sheet (SLGS) and representative volume element (RVE) are considered, respectively. In multi-scale analyses, linkage between atomistic and continuum is established using finite element method (FEM). In first level FEM analyses, SLGS is modeled using equivalent Euler-Bernoulli beam elements, which represents covalent bond, and applied uniaxial tensile loading. Young's modulus and Poisson's ratios of SLGS are obtained in armchair and zigzag directions. Moreover, variation of mechanical properties of SLGS with its size is investigated too. In second level FEM analyses, uniaxial tensile behavior of RVE (represents graphene reinforced Al nanocomposite) with volume fraction 0.068 is explored. Al matrix is modeled as continuum using hexahedral solid elements, SLGS modeled with beam element and finally heterogeneous interface modeling method is adopted to transfer load between atomistic and continuum environment. Distance-dependent stiffness properties are assigned utilized beam elements on the interface region. In a similar manner to the first level analyses, uniaxial tensile loading is applied to RVE and obtained Young's modulus. Additionally, for RVEs with different volume fractions and its matrix arbitrary chosen Al alloy, stress-strain curves are acquired utilizing non-linear FE code. Finally, mechanical behaviors of NC subjected to different mechanical loading and vibrational behaviors are investigated using multi-scale FE analyses. Molecular dynamic simulations are conducted using open source code called as LAMMPS (Large-scale Atomic/Molecular Massively Parallel Simulator). Same process sequence is adopted with multi-scale analyses, i.e. SLGS and RVE. 1008 C atoms and totally 9120 Al and C atoms are utilized to model SLGS and RVE, respectively. Adaptive Intermolecular Reactive Empirical Bond-Order (AIREBO) potential is employed for C-atom pairs interaction, however Embedded Atom Method EAM potential is used for Al atom interactions. Lennard-Jones potential is assigned to define non-bonded Al-C atoms interaction also. Firstly, to prevent occurrence residual stresses on molecular structure, energy minimization formulation is executed, so optimum atomic configuration is acquired. Then, uniaxial tensile loading is applied to the RVE. One side of RVE, 2-atom array is fixed and totally 2 Å imposed atom group of the other side, gradually. To prevent rupture or instability on the system 20 loop is used for imposed displacements. At the expense of the analyses, stress- strain curves are obtained both SLGS and RVE along to the armchair and zigzag directions. In conclusion, mechanical behavior of SLGS and RVE are acquired using both multi-scale and MD simulations. Obtained results are compliance with each other utilized method in the present study, Rule of Mixture (ROM) analytical calculations and similar studies exists in the literature also. Embedding the SLGS in the middle of the Al matrix conduce to strength enhancement, tremendously. Moreover, multi-scale analysis and MDS are compared in terms of CPU-times.