Suspended graphene has the highest measured thermal conductivity of any material at room temperature. However, when graphene is supported by a substrate or encased between two materials, basal-plane heat transfer is suppressed by phonon interactions at the interfaces. We have used frequency domain thermoreflectance to create thermal conductance maps of graphene contacts, obtaining simultaneous measurements of the basal-plane thermal conductivity and crossplane thermal boundary conductance for 1-7 graphitic layers encased between titanium and silicon dioxide. We find that the basal-plane thermal conductivity is similar to that of graphene supported on silicon dioxide. Our results have implications for heat transfer in two-dimensional material systems, and are relevant for applications such as graphene transistors and other nanoelectronic devices. V

We derive a generally applicable formula to calculate the precision of multi-parameter measurements that apply least squares algorithms. This formula, which accounts for experimental noise and uncertainty in the controlled model parameters, is then used to analyze the uncertainty of thermal property measurements with pump-probe thermoreflectance techniques. We compare the uncertainty of time domain thermoreflectance and frequency domain thermoreflectance (FDTR) when measuring bulk materials and thin films, considering simultaneous measurements of various combinations of thermal properties, including thermal conductivity, heat capacity, and thermal boundary conductance. We validate the uncertainty analysis using Monte Carlo simulations on data from FDTR measurements of an 80 nm gold film on fused silica.

As the size of gallium nitride (GaN) transistors is reduced in order to reach higher operating frequencies, heat dissipation becomes the critical bottleneck in device performance and longevity. Despite the importance of characterizing the physics governing the thermal transport in thin GaN films, the literature is far from conclusive. In this letter, we report measurements of thermal conductivity in a GaN film with thickness ranging from 15–1000 nm grown on 4H-SiC without a transition layer. Additionally, we measure the thermal conductivity in the GaN film when it is 1 μm-thick in the temperature range of 300 < T< 600 K and use a phonon transport model to explain the thermal conductivity in this film.

Silicon carbide is used as a substrate for high-power GaN devices because of its closely matched lattice spacing with GaN and its high thermal conductivity. In these devices, thermal resistance at the GaN–SiC interface is a bottleneck to heat flow, making this property an important factor in device design. In this letter, we report the first measurements of the thermal boundary conductance of epitaxial GaN grown directly on SiC without a transition layer. We find that the thermal boundary conductance increases from approximately 230 MW/m2K at 300 K to 330 MW/m2K at 600 K. Our measured values are in good qualitative agreement with the diffuse mismatch model for thermal boundary conductance and are in good quantitative agreement when we include a correction factor based on the ratio of Debye temperatures of the two materials. We also report the thermal conductivity of the GaN film, the thermal conductivity of 4H-SiC, and the thermal boundary conductance between Ni and GaN.

Optical pump-probe techniques based on thermoreflectance, such as time domain thermoreflectance and frequency domain thermoreflectance (FDTR), have been widely used to characterize the thermal conductivity of thin films and the thermal conductance across interfaces. These techniques typically use a transducer layer to absorb the pump light and improve the thermoreflectance signal. The transducer, however, complicates the interpretation of the measured signal because the approximation that all the energy from the pump beam is deposited at the transducer surface is not always accurate. In this paper, we consider the effect of laser absorption in the top layer of a multilayer sample, and derive an analytical solution for the thermoreflectance signal in the diffusion regime based on volumetric heating. We analyze the measurement sensitivity to the pump absorption depth for transducers with different thermal conductivities, and investigate the additional effect of probe laser penetration depth on the measured signal. We validate our model using FDTR measurements on 490 nm thick amorphous silicon films deposited on fused silica and silicon substrates.