Speaker
Description
The symmetry of the superconducting order parameter of Sr2RuO4 has been debated for three decades. Its structural similarity to the cuprates and an early, now disproven consensus on odd-parity pairing entrenched the assumption that the pairing is governed by spin fluctuations, leaving the role of phonons largely unexplored. In this talk I revisit that assumption from first principles. Anharmonic phonons obtained within the stochastic self-consistent harmonic approximation reproduce the measured inelastic neutron spectra, and the electron-phonon coupling computed from them accounts for the measured evolution of Tc under both in-plane and c-axis uniaxial strain, where spin-fluctuation pairing computed by functional renormalization group predicts the wrong sign under c-axis compression. The resulting gap is strongly anisotropic but sign-preserving, consistent with the absence of a neutron spin resonance. At the density functional theory level the coupling underestimates Tc, as expected for a correlated metal, and several effects compound to close the gap: spin-orbit coupling enhances the density of states, the coupling anisotropy raises Tc in a multiband treatment, and correlations enhance the electron-phonon matrix elements. Together these bring the predicted Tc into the range of the measured 1.4 K. Our results establish the electron-phonon interaction as a key ingredient of the superconductivity in Sr2RuO4. A quantitative description will nevertheless require treating it on equal footing with electronic correlations and spin fluctuations.