Ultrafast electron diffuse scattering as a tool for studying phonon transport: Phonon hydrodynamics and second sound oscillations.

Hydrodynamic phonon transport phenomena, like second sound, have been observed in liquid helium more than 50 years ago. More recently second sound has been observed in graphite at over 200 K using transient thermal grating (TG) techniques. In this work, we explore signatures of phonon hydrodynamic transport and second sound oscillations in ultrafast electron diffuse scattering patterns, which can provide time, momentum, and branch resolved information on the state-of-excitation of the phonon system beyond that available through TG experiments. We use the density functional theory and solve the Boltzmann transport equation to determine time-resolved non-equilibrium phonon populations and model phonon transport in graphite. This model also provides the information necessary to calculate the time evolution of one-phonon structure factors and diffuse scattering patterns during thermal transport covering ballistic, diffusive, and hydrodynamic regimes where the effect of a second sound oscillation on the phonon distribution is observed. Direct measurements of how the phonon distribution varies in time and space in various thermal transport regimes should yield new insights into the fundamental physics of the underlying processes.

Anomalous Defect Dependence of Thermal Conductivity in Epitaxial WO3 Thin Films.

Lattice defects typically reduce lattice thermal conductivity, which has been widely exploited in applications such as thermoelectric energy conversion. Here, an anomalous dependence of the lattice thermal conductivity on point defects is demonstrated in epitaxial WO3 thin films. Depending on the substrate, the lattice of epitaxial WO3 expands or contracts as protons are intercalated by electrolyte gating or oxygen vacancies are introduced by adjusting growth conditions. Surprisingly, the observed lattice volume, instead of the defect concentration, plays the dominant role in determining the thermal conductivity. In particular, the thermal conductivity increases significantly with proton intercalation, which is contrary to the expectation that point defects typically lower the lattice thermal conductivity. The thermal conductivity can be dynamically varied by a factor of ≈1.7 via electrolyte gating, and tuned over a larger range, from 7.8 to 1.1 W m-1 K-1 , by adjusting the oxygen pressure during film growth. The electrolyte-gating-induced changes in thermal conductivity and lattice dimensions are reversible through multiple cycles. These findings not only expand the basic understanding of thermal transport in complex oxides, but also provide a path to dynamically control the thermal conductivity.