Book Description

Two-dimensional graphene, a single layer of graphite, has emerged as an excellent candidate for future electronic material due to its unique electronic structure and remarkably high carrier mobility. Even higher carrier mobility has been demonstrated in graphene devices using hexagonal boron nitride as an underlying dielectric support instead of silicon oxide. Interestingly, both graphene and boron nitride exhibit superior thermal properties, therefore may potentially offer a solution to the increasingly severe heat dissipation problem in nanoelectronics caused by increased power density. In this thesis, we focus on the investigation of the thermal properties of graphene and hexagonal boron nitride. First, scanning thermal microscopy based on a sub-micrometer thermocouple at the apex of a microfabricated tip was employed to image the temperature profiles in electrically biased graphene devices with ~ 100 nm scale spatial resolution. Non-uniform temperature distribution in the devices was observed, and the "hot spot" locations were correlated with the charge concentrations in the channel, which could be controlled by both gate and drain-source biases. Hybrid contact and lift mode scanning has enabled us to obtain the quantitative temperature profiles, which were compared with the profiles obtained from Raman-based thermometry. The temperature rise in the channel provided an important insight into the heat dissipation mechanism in Joule-heated graphene devices. Next, thermal conductivity of suspended single and few-layer graphene was measured using a micro-bridge device with built-in resistance thermometers. Polymer-assisted transfer technique was developed to suspend graphene layers on the pre-fabricated device. The room temperature thermal conductivity values of 1-7 layer graphene were measured to be lower than that of bulk graphite, and the value appeared to increase with increasing sample thickness. These observations can be explained by the impact of the phonon scattering by polymer residue remaining on the sample surfaces. Lastly, thermal conductivity of few-layer hexagonal boron nitride sample was measured by using the same device and technique used for suspended graphene. Measurements on samples with different suspended lengths but similar thickness allowed us to extract the intrinsic thermal conductivity of the samples as well as the contribution of contact thermal resistance to the overall thermal measurement. The room temperature thermal conductivity of 11 layer sample approaches the basal-plane value reported in the bulk sample. Lower thermal conductivity was measured in a 5 layer sample than an 11 layer sample, which again supports the polymer effect on the thermal transport in few-layer hexagonal boron nitride.