Surface wetting to enhance thermoreflectance characterization of integrated circuits.

Thermoreflectance (TR) imaging enables non-contact thermal imaging of devices and integrated circuits (ICs) with sub-µm spatial resolution. TR coefficient of most metals and semiconductors in visible wavelengths is in the 10-5 to 10-3 K-1 range, which gives a temperature resolution of 0.1-0.5 °C with a few minutes of averaging. Here, we show that surface wetting with various solvents, such as water, methanol, as well as Fluorinert, which is a commonly used coolant for high-power ICs, can enhance the TR coefficient by up to 19 times. Systematic characterizations as a function of the heating power, illumination-wavelength, liquid layer thickness, sample's tilt, and objective lens's numerical aperture are presented. TR images are distorted due to interference in the liquid layer, but this technique could be used for fast detection of small temperature variations and hot spots in ICs.

Characterization of the temperature dependence of the thermoreflectance coefficient for conductive thin films.

We describe a novel approach for calibration of the thermoreflectance coefficient, ideally suited for measurements in a vacuum thermostat, and present the high temperature thermoreflectance coefficients for several metals commonly encountered in electronic devices: gold, platinum, and aluminum. The effect of passivation on these metals is also examined, and we demonstrate the signal to noise ratio of a thermoreflectance measurement can be improved with informed selection of the dielectric layer thickness. Furthermore, the thermo-optic coefficients of the metals are extracted over a wide temperature range. The results presented here can be utilized in the optimization of experimental configurations for high temperature thermoreflectance imaging.

Heat transfer simulation and thermal measurements of microfabricated x-ray transparent heater stages.

A microfabricated amorphous silicon nitride membrane-based nanocalorimeter is proposed to be suitable for an x-ray transparent sample platform with low power heating and built-in temperature sensing. In this work, thermal characterization in both air and vacuum are analyzed experimentally and via simulation. Infrared microscopy and thermoreflectance microscopy are used for thermal imaging of the sample area in air. While a reasonably large isothermal area is found on the sample area, the temperature homogeneity of the entire sample area is low, limiting use of the device as a heater stage in air or other gases. A simulation model that includes conduction, as well as radiation and convection heat loss, is presented with radiation and convection parameters determined experimentally. Simulated temperature distributions show that the homogeneity can be improved by using a thicker thermal conduction layer or reducing the pressure of the gas in the environment but neither are good solutions for the proposed use. A new simple design that has improved temperature homogeneity and a larger isothermal area while maintaining a thin thermal conduction layer is proposed and fabricated. This new design enables applications in transmission x-ray microscopes and spectroscopy setups at atmospheric pressure.

A new REBO potential based atomistic structural model for graphene sheets.

A new atomistic structural model is developed here for graphene sheets based on the stiffnesses from the REBO potential. Using this model, the flexural vibration natural frequencies and buckling loads of rectangular single-layer graphene sheets of different sizes, chiralities and boundary conditions are calculated. The newly developed atomistic structural model is verified by comparing the calculated fundamental natural frequencies for small-sized graphene sheets with those obtained from ab initio density functional theory (DFT) frequency analysis. The vibration and buckling analysis results are also compared with those of an earlier atomistic structural model based on the AMBER potential as well as the equivalent continuum model for graphene sheets. Through this study, it is observed that graphene sheets display very slight anisotropic characteristics in flexural vibration and buckling. Also, it is shown that the atomistic structural model cannot be replaced by a classical equivalent continuum model such as a plate model. Most significantly, we verify that the new atomistic structural model based on the REBO potential predicts more accurate natural frequencies and buckling loads for graphene sheets, which are considerably lower than those predicted by the earlier atomistic structural model based on the AMBER potential.

Application of network identification by deconvolution method to the thermal analysis of the pump-probe transient thermoreflectance signal.

The paper discusses the possibility to apply network identification by deconvolution (NID) method to the analysis of the thermal transient behavior due to a laser delta pulse excitation in a pump-probe transient thermoreflectance experiment. NID is a method based on linear RC network theory using Fourier's law of heat conduction. This approach allows the extraction of the thermal time constant spectrum of the sample under study after excitation by either a step or pulse function. Furthermore, using some mathematical transformations, the method allows analyzing the detail of the heat flux path through the sample, starting from the excited top free surface, by introducing two characteristic functions: the cumulative structure function and the differential structure function. We start by a review of the theoretical background of the NID method in the case of a step function excitation and then show how this method can be adjusted to be used in the case of a delta pulse function excitation. We show how the NID method can be extended to analyze the thermal transients of many optical experiments in which the excitation function is a laser pulse. The effect of the semi-infinite substrate as well as extraction of the interface and thin film thermal resistances will be discussed.

"Nanoparticle-in-alloy" approach to efficient thermoelectrics: silicides in SiGe.

We present a "nanoparticle-in-alloy" material approach with silicide and germanide fillers leading to a potential 5-fold increase in the thermoelectric figure of merit of SiGe alloys at room temperature and 2.5 times increase at 900 K. Strong reductions in computed thermal conductivity are obtained for 17 different types of silicide nanoparticles. We predict the existence of an optimal nanoparticle size that minimizes the nanocomposite's thermal conductivity. This thermal conductivity reduction is much stronger and strikingly less sensitive to nanoparticle size for an alloy matrix than for a single crystal one. At the same time, nanoparticles do not negatively affect the electronic conduction properties of the alloy. The proposed material can be monolithically integrated into Si technology, enabling an unprecedented potential for micro refrigeration on a chip. High figure-of-merit at high temperatures (ZT approximately 1.7 at 900 K) opens up new opportunities for thermoelectric power generation and waste heat recovery at large scale.