Spectrally-resolved thermal transport in graphene nanoribbons

Abstract

Thermal transport properties of graphene nanoribbons (GNRs) are investigated using phonon transport studies. Ribbons of varying widths are considered to investigate the transition of key thermal properties with width. The lattice structure of the ribbons is fully resolved, and phonon transport is modeled by accounting for all three-phonon scattering processes using a solution of the linearized Boltzmann transport equation. A 3× reduction in intrinsic thermal conductivity is observed compared to bulk graphene arising from increased strength of three-phonon scattering due to the additional nondegenerate phonon modes that appear due to the finite edges of confined nanoribbons. Strong dependence of thermal conductivity on ribbon width is also observed. The underlying mechanisms for thermal conductivity reduction and width dependence are presented by analyzing frequency- and polarization-resolved phonon transport. The additional scattering pathways present in 1D GNRs lead to a significant reduction in the thermal conductivity of otherwise highly conducting flexural phonons in bulk graphene. In contrast, confinement-induced changes to the density of states, specific heat or group velocity, and the subsequent impact on lattice thermal conductivity are found to be relatively small.

Unusual Width Dependence of Lattice Thermal Conductivity in Ultranarrow Armchair Graphene Nanoribbons with Unpassivated Edges

“…The TWA phonon branch is unique for 1D materials compared to the 2D materials, which is also observed in other 1D materials. ,, Also, the phonon branches of AGNRs largely soften compared to graphene, especially for the LA and TA phonon branches. These features demonstrate the size confinement effect when the width is below 1 nm, which is different from the previous study of ZGNRs …”

“…With the same ribbon width, the BTE study using a reoptimized empirical potential predicted the ratio κ AGNR /κ graphene to be 1.05, much larger than our result. Such discrepancy may attribute to the unresolved AGNR structures and the usage of empirical potential in the previous BTE calculation. Other studies involving ultranarrow AGNRs mostly focused on the phonon ballistic transport regime (that is, the length of AGNRs is very short), , which is different from the current work.…”

“…In their study, the width W ( W = πd ) of AGNR was fixed to the perimeter of a corresponding single-wall carbon nanotube (SWCNT). Because the lattice structures were not fully relaxed, it also led to a contradiction that all AGNRs have a thermal conductivity above or equal to that of graphene …”

“…Moreover, many theoretical studies based on empirical potentials tried to extract the width dependence of the thermal conductivity of GNRs in various size ranges at room temperature. Using the equilibrium molecular dynamics (MD) simulations with empirical potentials, several studies demonstrated that the thermal conductivity of smooth GNRs decreases as the width decreases in the size range from 40 to 1.7 nm, 27,33,35 agreeing well with the expectation that classical size effect suppresses phonon transport (the finite sample size introduces boundaries or edges as extra phonon scattering sources). However, based on the Boltzmann transport equation (BTE) method with the interatomic force constants (IFCs) calculated from an optimized Tersoff potential, 36 Lindsay et al found that the thermal conductivity of AGNRs increases from κ graphene to 1.2κ graphene as the width decreases from 2.7 to 0.42 nm.…”

Phonon Transport in Patterned Two-Dimensional Materials from First Principles

“…Landon et al [15] and Mei et al [16], solved the spacedependent BTE with the phonon Monte Carlo obtaining good agreements with experiments on graphene nanoribbons (GNRs). Marepalli et al [17] solved the BTE in GNRs with short widths, using Tersoff interatomic potentials, and taking into account phonon confinement. * romanog@mit.edu…”