Breakdown of the Wiedemann-Franz law at the Lifshitz point of strained Sr$_2$RuO$_4$

Abstract

Strain tuning Sr2RuO4 through the Lifshitz point, where the Van Hove singularity of the electronic spectrum crosses the Fermi energy, is expected to cause a change in the temperature dependence of the electrical resistivity from its Fermi liquid behavior ρ ∼ T 2 to ρ ∼ T 2 log (1/T ), a behavior consistent with experiments by Barber et al. [Phys. Rev. Lett. 120, 076601 (2018)]. This expectation originates from the same multi-band scattering processes with large momentum transfer that were recently shown to account for the linear in T resistivity of the strange metal Sr3Ru2O7. In contrast, the thermal resistivity ρQ ≡ T /κ, where κ is the thermal conductivity, is governed by qualitatively distinct processes that involve a broad continuum of compressive modes, i.e. long wavelength density excitations in Van Hove systems. While these compressive modes do not affect the charge current, they couple to thermal transport and yield ρQ ∝ T 3/2 . As a result, we predict that the Wiedemann-Franz law in strained Sr2RuO4 should be violated with a Lorenz ratio L ∝ T 1/2 log (1/T ). We expect this effect to be observable in the temperature and strain regime where the anomalous charge transport was established.

Tuning the Fermi Liquid Crossover in Sr$_2$RuO$_4$ with Uniaxial Stress

“…We note that an increase of the latter under strain was found in a previous density-functional theory calculation [25]. Another interesting issue is the departure from Fermi liquid theory at the critical strain [25,42].…”

Dynamical mean field theory of moiré bilayer transition metal dichalcogenides: phase diagram, resistivity, and quantum criticality

“…The linearity is theoretically expected for a van Hove singularity of degree three within the DMFT approximation [39,57]. The single-site DMFT is a local approximation; effects on the van Hove contribution to the self energy [17,58,59] and resistivity beyond the local self energy approximation [26,47] are left to future work. Finally, we consider the doping-driven crossover of the metallic resistivity from the T -linear behavior found when the Fermi surface touches the third order van Hove point to the T 2 behavior expected when the van Hove point is not at the Fermi surface.…”