Thermal boundary resistance (TBR) is a key property for the thermal management of high power micro- and opto-electronic devices and for the development of high efficiency thermal barrier coatings and thermoelectric materials. Prediction of TBR is important for guiding the discovery of interfaces with very low or very high TBR. In this study, we report the prediction of TBR by the machine learning method. We trained machine learning models using the collected experimental TBR data as training data and materials properties that might affect TBR as descriptors. We found that the machine learning models have much better predictive accuracy than the commonly used acoustic mismatch model and diffuse mismatch model. Among the trained models, the Gaussian process regression and the support vector regression models have better predictive accuracy. Also, by comparing the prediction results using different descriptor sets, we found that the film thickness is an important descriptor in the prediction of TBR. These results indicate that machine learning is an accurate and cost-effective method for the prediction of TBR.

Ideal materials for modern electronics packaging should be highly thermoconductive. This may be achieved through designing multifunctional polymer composites. Such composites may generally be achieved via effective embedment of functional inorganic fillers into desirable polymeric bodies. Herein, two types of high-performance 3D h-BN porous frameworks (3D-BN), namely, h-BN nanorod-assembled networks and nanosheet-interconnected frameworks, are successfully created via an in-situ carbothermal reduction substitution chemical vapor deposition using carbon-based nanorodinterconnected networks as templates. These 3D-BN porous materials with densely-interlinked frameworks, excellent mechanical robustness and integrity, highly-isotropous and multiple heat transfer paths, enable reliable fabrications of diverse 3D-BN/polymer porous composites. The composites exhibit combinatorial multi-functional properties, such as excellent mechanical strength, light weight, ultra-low coefficient of thermal expansion, highly isotropic thermal conductivities (~ 26 -This article is protected by copyright. All rights reserved.3 51 multiples of pristine polymers), relatively-low dielectric constants and super-low dielectric losses, and high resistance to softening at elevated temperatures. In addition, the regarded 3D-BN frameworks are easily recycled from their polymer composites, and may be reliably reutilized for multi-functional reuse. Thus, these materials should be valuable for new-era advanced electronic packaging and related applications.

In this study, we investigated the temperature dependence and size effect of the thermal boundary resistance at Si/Ge interfaces by non-equilibrium molecular dynamics (MD) simulations using the direct method with the Stillinger-Weber potential. The simulations were performed at four temperatures for two simulation cells of different sizes. The resulting thermal boundary resistance decreased with increasing temperature. The thermal boundary resistance was smaller for the large cell than for the small cell. Furthermore, the MD-predicted values were lower than the diffusion mismatch model (DMM)-predicted values. The phonon density of states (DOS) was calculated for all the cases to examine the underlying nature of the temperature dependence and size effect of thermal boundary resistance. We found that the phonon DOS was modified in the interface regions. The phonon DOS better matched between Si and Ge in the interface region than in the bulk region. Furthermore, in interface Si, the population of low-frequency phonons was found to increase with increasing temperature and cell size. We suggest that the increasing population of low-frequency phonons increased the phonon transmission coefficient at the interface, leading to the temperature dependence and size effect on thermal boundary resistance.

No abstract

For harvesting energy from waste heat, the power generation densities and fabrication costs of thermoelectric generators (TEGs) are considered more important than their conversion efficiency because waste heat energy is essentially obtained free of charge. In this study, we propose a miniaturized planar Si-nanowire micro-thermoelectric generator (SiNW-μTEG) architecture, which could be simply fabricated using the complementary metal–oxide–semiconductor–compatible process. Compared with the conventional nanowire μTEGs, this SiNW-μTEG features the use of an exuded thermal field for power generation. Thus, there is no need to etch away the substrate to form suspended SiNWs, which leads to a low fabrication cost and well-protected SiNWs. We experimentally demonstrate that the power generation density of the SiNW-μTEGs was enhanced by four orders of magnitude when the SiNWs were shortened from 280 to 8 μm. Furthermore, we reduced the parasitic thermal resistance, which becomes significant in the shortened SiNW-μTEGs, by optimizing the fabrication process of AlN films as a thermally conductive layer. As a result, the power generation density of the SiNW-μTEGs was enhanced by an order of magnitude for reactive sputtering as compared to non-reactive sputtering process. A power density of 27.9 nW/cm2 has been achieved. By measuring the thermal conductivities of the two AlN films, we found that the reduction in the parasitic thermal resistance was caused by an increase in the thermal conductivity of the AlN film and a decrease in the thermal boundary resistance.