Unified Picture for the Colossal Thermopower Compound FeSb2.

We identify the driving mechanism of the gigantic Seebeck coefficient in FeSb2 as the phonon-drag effect associated with an in-gap density of states that we demonstrate to derive from excess iron. We accurately model electronic and thermoelectric transport coefficients and explain the so far ill-understood correlation of maxima and inflection points in different response functions. Our scenario has far-reaching consequences for attempts to harvest the spectacular power factor of FeSb2.

Strain-engineering Mott-insulating La2CuO4.

The transition temperature Tc of unconventional superconductivity is often tunable. For a monolayer of FeSe, for example, the sweet spot is uniquely bound to titanium-oxide substrates. By contrast for La2-xSrxCuO4 thin films, such substrates are sub-optimal and the highest Tc is instead obtained using LaSrAlO4. An outstanding challenge is thus to understand the optimal conditions for superconductivity in thin films: which microscopic parameters drive the change in Tc and how can we tune them? Here we demonstrate, by a combination of x-ray absorption and resonant inelastic x-ray scattering spectroscopy, how the Coulomb and magnetic-exchange interaction of La2CuO4 thin films can be enhanced by compressive strain. Our experiments and theoretical calculations establish that the substrate producing the largest Tc under doping also generates the largest nearest neighbour hopping integral, Coulomb and magnetic-exchange interaction. We hence suggest optimising the parent Mott state as a strategy for enhancing the superconducting transition temperature in cuprates.

Protected Fe valence in quasi-two-dimensional α-FeSi2.

We report the first comprehensive study of the high temperature form (α-phase) of iron disilicide. Measurements of the magnetic susceptibility, magnetization, heat capacity and resistivity were performed on well characterized single crystals. With a nominal iron d(6) configuration and a quasi-two-dimensional crystal structure that strongly resembles that of LiFeAs, α-FeSi2 is a potential candidate for unconventional superconductivity. Akin to LiFeAs, α-FeSi2 does not develop any magnetic order and we confirm its metallic state down to the lowest temperatures (T = 1.8 K). However, our experiments reveal that paramagnetism and electronic correlation effects in α-FeSi2 are considerably weaker than in the pnictides. Band theory calculations yield small Sommerfeld coefficients of the electronic specific heat γ = Ce/T that are in excellent agreement with experiment. Additionally, realistic many-body calculations further corroborate that quasi-particle mass enhancements are only modest in α-FeSi2. Remarkably, we find that the natural tendency to vacancy formation in the iron sublattice has little influence on the iron valence and the density of states at the Fermi level. Moreover, Mn doping does not significantly change the electronic state of the Fe ion. This suggests that the iron valence is protected against hole doping and indeed the substitution of Co for Fe causes a rigid-band like response of the electronic properties. As a key difference from the pnictides, we identify the smaller inter-iron layer spacing, which causes the active orbitals near the Fermi level to be of a different symmetry in α-FeSi2. This change in orbital character might be responsible for the lack of superconductivity in this system, providing constraints on pairing theories in the iron based pnictides and chalcogenides.

Downfolded self-energy of many-electron systems.

Starting from the full many-body Hamiltonian of interacting electrons the effective self-energy acting on electrons residing in a subspace of the full Hilbert space is derived. This subspace may correspond to, for example, partially filled narrow bands, which often characterize strongly correlated materials. The formalism delivers naturally the frequency-dependent effective interaction (the Hubbard U) and provides a general framework for constructing theoretical models based on the Green's function language. It also furnishes a general scheme for first-principles calculations of complex systems in which the main correlation effects are concentrated on a small subspace of the full Hilbert space.