Computational study of the thermal conductivity in defective carbon nanostructures

We use nonequilibrium molecular dynamics simulations to study the adverse role of defects including isotopic impurities on the thermal conductivity of carbon nanotubes, graphene, and graphene nanoribbons. We find that even in structurally perfect nanotubes and graphene, isotopic impurities reduce thermal conductivity by up to one half by decreasing the phonon mean-free path. An even larger thermal conductivity reduction, with the same physical origin, occurs in presence of structural defects including vacancies and edges in narrow graphene nanoribbons. Our calculations reconcile results of former studies, which differed by up to an order of magnitude, by identifying limitations of various computational approaches.