We theoretically investigate the thermoelectric properties of zigzag graphene nanoribbons in the presence of extended line defects, substrate impurities and edge roughness along the nanoribbon's length. A nearest-neighbor tight-binding model for the electronic structure and a fourth nearestneighbor force constant model for the phonon bandstructure are used. For transport we employ quantum mechanical non-equilibrium Green's function simulations. Starting from the pristine zigzag nanoribbon structure that exhibits very poor thermoelectric performance, we demonstrate how after a series of engineering design steps the performance can be largely enhanced. Our results could be useful in the design of highly efficient nanostructured graphene nanoribbon based thermoelectric devices.

The thermoelectric properties of graphene-based antidot lattices are theoretically investigated. A third nearest-neighbor tight-binding model and a fourth nearest-neighbor force constant model are employed to study the electronic and phononic band structures of graphene antidot lattices with circular, rectangular, hexagonal, and triangular antidot shapes. Ballistic transport models are used to evaluate transport coefficients. Methods to reduce the thermal conductance and to increase the thermoelectric power factor of such structures are studied. Our results indicate that triangular antidot lattices have the smallest thermal conductance due to longer boundaries and the smallest distance between the neighboring antidots. Among them, iso-triangular antidot lattices have also a large power factor and as a result a large figure of merit. V

No abstract

Following our recent study on the electronic properties of rough nanoribbons [1], in this paper the role of geometrical and roughness parameters on the thermal properties of armchair graphene nanoribbons is studied. Employing a fourth nearest-neighbor force constant model in conjuction with the nonequilibrium Green's function method the effect of lineedge-roughness on the lattice thermal conductivity of rough nanoribbons is investigated. The results show that a reduction of about three orders of magnitude of the thermal conductivity can occur for ribbons narrower than 10 nm. The results indicate that the diffusive thermal conductivity and the effective mean free path are directly proportional to the ribbon's width and the roughness correlation length, but inversely proportional to the roughness amplitude. Based on the numerical results an analytical model for the thermal conductivity of narrow armchair graphene nanoribbons is proposed in this paper. The developed model can be used in the analysis of graphene-based nano transistors and thermoelectric devices, where the appropriate selection of geometrical and roughness parameters are essential for optimizing the thermal properties.

We study the effect of confinement on the phonon properties of ultra-narrow silicon nanowires of side sizes of 1 10nm . We use the modified valence force field method to compute the phononic dispersion, and extract the density of states, the transmission function, the sound velocity, the ballistic thermal conductance and boundary scatteringlimited diffusive thermal conductivity. We find that the phononic dispersion and the ballistic thermal conductance are functions of the geometrical features of the structures, i.e. the transport orientation and confinement dimension. The phonon group velocity and thermal conductance can vary by a factor of two depending on the geometrical features of the channel. The <110> nanowire has the highest group velocity and thermal conductance, whereas the <111> the lowest. The <111> channel is thus the most suitable orientation for thermoelectric devices based on Si nanowires since it also has a large power factor. Our findings could be useful in the thermal transport design of siliconbased devices for thermoelectric and thermal management applications.