| Graphene, a single sheet of graphite, has recently emerged as one of the most exciting material systems to study, propelled by the discovery of unusual fundamental physical phenomena such as unconventional quantum Hall effects and surprisingly high room-temperature electron mobility. The intense interest on graphene arises also because of its structure stability despite being as a single atomic layer. The discovery of these novel physical properties of graphene has inspired an endeavor that may lead to the establishment of an entirely new technological platform based on graphene rather than silicon with superior performance. Such a monumental push for new technologies, however, requires a deep physical understanding of the deformation, failure, and the coupled thermo-mechanical-electrical properties of this emergent material system.;This thesis contributes to the multiscale mechanics of graphene and its curved version, carbon nanotubes (CNT). As a one-atom-thick film with large ratio of the in-plane rigidity to out-of-plane bending rigidity, a monolayer graphene ripples even only under the thermal fluctuation. For layered graphene or multi-walled CNTs (MWCNTs), the interplay between the interlayer interactions and in-plane deformation energies often leads to ordered deformation patterns under external loadings. The deformation morphologies of graphene and CNTs not only change their mechanical properties, but only affect their electronic properties. Understanding the deformation morphologies and the kinetics of the defects in graphene and CNTs is critical to the engineering and fabrication of the graphene/CNT-based devices.;Enormous efforts have been undertaken in computationally modeling the deformation morphologies and defect motion in graphene. Classical single-scale methods such as quantum mechanical, molecular dynamics, and continuum mechanics simulations suffer from either length scale and time scale limitations or simulation accuracy. To achieve both physical accuracy and computational efficiency, this thesis strives to develop novel multiscale modeling approaches to bridge phenomena across different length and time scales. In particular, for pristine graphene and CNTs, the interatomic potentials are seamlessly cast into the constitutive relations of the finite crystal elasticity theories for atomic membranes. For interlayer bridged MWCNTs, the interlayer force-separation relations obtained from atomistic simulations are embedded into the continuum level constitutive relations through hierarchical message passing. To overcome the time scale barriers of molecular dynamics simulations, minimum energy paths and the transition states of various atomic processes are determined using the pathway sampling schemes such as the nudged elastic band (NEB) method.;The multiscale modeling approaches enabled large-scale simulations of CNTs and graphene, and revealed a range of interesting deformation phenomena, typically inaccessible to previous single-scale models. For pristine MWCNTs, a variety of ordered deformation patterns such as periodic rippling were identified under different loading conditions, which are in distinct comparison to the local sharp buckling phenomena observed in SWCNTs. For interlayer bridged MWCNTs, the presence of inter-wall covalent bridges not only enhances the post-buckling rigidities of the MWCNTs, but also modifies the deformation morphologies and morphology pathways of MWCNTs. For defected graphene and CNTs, we found that chemical addictives such as hydrogen and oxygen atom regulate the fracture paths. Further, the migration barrier and direction of defects and adatoms (such as Li) strongly depend on the applied stress and strain. The findings from the thesis research are not only scientifically significant, but also offer guidance to the engineering of MWCNTs and graphene as next-generation electronic nanodevices. |
