We present experimental measurements of the thermal boundary conductance (TBC) from 78–500 K across isolated heteroepitaxially grown ZnO films on GaN substrates. This data provides an assessment of the underlying assumptions driving phonon gas-based models, such as the diffuse mismatch model (DMM), and atomistic Green’s function (AGF) formalisms used to predict TBC. Our measurements, when compared to previous experimental data, suggest that TBC can be influenced by long wavelength, zone center modes in a material on one side of the interface as opposed to the ‘“vibrational mismatch”’ concept assumed in the DMM; this disagreement is pronounced at high temperatures. At room temperature, we measure the ZnO/GaN TBC as 490­[+150,–110] MW m–2 K–1. The disagreement among the DMM and AGF, and the experimental data at elevated temperatures, suggests a non-negligible contribution from other types of modes that are not accounted for in the fundamental assumptions of these harmonic based formalisms, which may rely on anharmonicity. Given the high quality of these ZnO/GaN interfaces, these results provide an invaluable, critical, and quantitative assessment of the accuracy of assumptions in the current state of the art computational approaches used to predict phonon TBC across interfaces.

Thermophotovoltaics (TPVs) convert predominantly infrared wavelength light to electricity via the photovoltaic effect, and can enable approaches to energy storage1,2 and conversion3–9 that use higher temperature heat sources than the turbines that are ubiquitous in electricity production today. Since the first demonstration of 29% efficient TPVs (Fig. 1a) using an integrated back surface reflector and a tungsten emitter at 2,000 °C (ref. 10), TPV fabrication and performance have improved11,12. However, despite predictions that TPV efficiencies can exceed 50% (refs. 11,13,14), the demonstrated efficiencies are still only as high as 32%, albeit at much lower temperatures below 1,300 °C (refs. 13–15). Here we report the fabrication and measurement of TPV cells with efficiencies of more than 40% and experimentally demonstrate the efficiency of high-bandgap tandem TPV cells. The TPV cells are two-junction devices comprising III–V materials with bandgaps between 1.0 and 1.4 eV that are optimized for emitter temperatures of 1,900–2,400 °C. The cells exploit the concept of band-edge spectral filtering to obtain high efficiency, using highly reflective back surface reflectors to reject unusable sub-bandgap radiation back to the emitter. A 1.4/1.2 eV device reached a maximum efficiency of (41.1 ± 1)% operating at a power density of 2.39 W cm–2 and an emitter temperature of 2,400 °C. A 1.2/1.0 eV device reached a maximum efficiency of (39.3 ± 1)% operating at a power density of 1.8 W cm–2 and an emitter temperature of 2,127 °C. These cells can be integrated into a TPV system for thermal energy grid storage to enable dispatchable renewable energy. This creates a pathway for thermal energy grid storage to reach sufficiently high efficiency and sufficiently low cost to enable decarbonization of the electricity grid.

β-Ga2O3 is a wide-bandgap semiconductor of significant technological importance for electronics, but its low thermal conductivity is an impeding factor for its applications. In this work, an interatomic potential is developed for β-Ga2O3 based on a deep neural network model to predict the thermal conductivity and phonon transport properties. Our potential is trained by the ab initio energy surface and atomic forces, which reproduces phonon dispersion in good agreement with first-principles calculations. We are able to use molecular dynamics (MD) simulations to predict the anisotropic thermal conductivity of β-Ga2O3 with this potential, and the calculated thermal conductivity values agree well with experimental results from 200 to 500 K. Green–Kubo modal analysis is performed to quantify the contributions of different phonon modes to the thermal transport, showing that optical phonon modes play a critical role in the thermal transport. This work provides a high-fidelity machine learning-based potential for MD simulation of β-Ga2O3 and serves as a good example of exploring thermal transport physics of complex semiconductor materials.

Molecular dynamics simulations have been extensively used to study phonons and gain insight, but direct comparisons to experimental data are often difficult, due to a lack of accurate empirical interatomic potentials for different systems. As a result, this issue has become a major barrier to realizing the promise associated with advanced atomistic-level modeling techniques. Here, we present a general method for specifically optimizing empirical interatomic potentials from ab initio inputs for the study of phonon transport properties, thereby resulting in phonon optimized potentials. The method uses a genetic algorithm to directly fit the empirical parameters of the potential to the key properties that determine whether or not the atomic level dynamics and most notably the phonon transport are described properly.

No abstract