Recent measurements of the Fermi surface with de Haas-van Alphen oscillations in LaFePO showed a shrinking of the Fermi pockets with respect to first-principle calculations, suggesting an energy shift of the hole and electrons bands with respect to local-density approximations. We show that this shift is a natural consequence of the strong particle-hole asymmetry of electronic bands in pnictides, and that it provides an indirect experimental evidence of a dominant interband scattering in these systems.

We study the newly-discovered Pt phosphides APt3P (A=Sr, Ca, La) [T. Takayama, et al., Phys. Rev. Lett. 108, 237001 ] using first-principles calculations and Migdal-Eliashberg theory. Given the remarkable agreement with the experiment, we exclude the charge-density wave scenario proposed by previous first-principles calculations, and give conclusive answers concerning the superconducting state in these materials. The pairing increases from La to Ca and Sr due to changes in the electronphonon matrix elements and low frequency phonons. Although we find that all three compounds are well described by conventional s-wave superconductivity and spin orbit coupling of Pt plays a marginal role, we show that it could be possible to tune the structure from centrosymmetric to non centrosymmetric opening new perspectives towards the understanding of unconventional superconductivity.

We present a method to correct the magnetic properties of itinerant systems in local spin density approximation (LSDA) and we apply it to the ferromagnetic-paramagnetic transition under pressure in a typical itinerant system, Ni3Al. We obtain a scaling of the critical fluctuations as a function of pressure equivalent to the one obtained within Moryia's theory. Moreover we show that in this material the role of the bandstructure is crucial in driving the transition. Finally we calculate the magnetic moment as a function of pressure, and find that it gives a scaling of the Curie temperature that is in good agreement with the experiment. The method can be easily extended to the antiferromagnetic case and applied, for instance, to the Fe-pnictides in order to correct the LSDA magnetic moment. PACS numbers: 71.15.Mb,71.20.Be,,75.50.Cc Density functional theory (DFT), in its most common implementations (local spin density approximation, LSDA, with or without gradient corrections, GGA), is in principle the only way to access the ground state of real materials. 1 And indeed the agreement with experiment concerning the ground state properties, such as crystal and electronic structures, is excellent, especially for itinerant system, where local correlations play a minor role. Nevertheless, a well known problem of LSDA is the overestimation of the tendency to magnetism in itinerant magnets near the quantum critical point (QCP). This problem can be traced down to the fact that LSDA is essentially a mean field theory, which does not take into account a detrimental effect of near-critical fluctuations on the long-range magnetism. Thus, while the itinerant nature of systems like FeAl, 2 Pd 3 or the more recent and better known Fe-pnictides, 4-6 make LDA and GGA reproduce very well the paramagnetic bandstructure, whenever a (magnetic) quantum critical point (QCP) is approached the theory fails miserably. The importance of this problem is demonstrated by the amount of papers dealing with the problem of correcting the magnetic moment of Fe-pnicitides. 7-12 There, the usual argument is that correlations beyond mean field suppress the (LSDA) local ordered moment. It was shown that one can reduce the calculated magnetic moment by using the LDA+U method with a negative U 8 , but there is no physical justification for this procedure. Such many-body approaches as Dynamic Mean Field Theory (DMFT) 9-11 and Gutzwiller 12 were also successfully used; since these methods introduce additional fluctuations into the system, they obviously work in the right direction. However, the concept of substituting long-range critical fluctuations by the on-site ones is rather questionable. 2 Furthermore the effect of non-local fluctuations was recently found to be crucial, also in localized models, whenever the critical behavior is analyzed 13 , and in any event computational load in these methods is incomparably heavier than in LDA calculations.For these reasons we propose a different approach which corrects LSDA within DFT and takes into account the ...

We analyze the outcomes of an extended-Drude-model approach to the optical spectra of pnictides, where the multiband nature of the electronic excitations requires a careful analysis of the role of interband processes in the optical conductivity. Through a direct comparison between model calculations of the intraband optical spectra and experimental data, we show that interband transitions, whose relevance is shown by first-principle calculations, give a non negligible contribution already in the infrared region. This leads to a substantial failure of the extended-Drude-model analysis on the measured optical data without subtraction of interband contributions.Optical studies are a useful experimental probe to analyze the interactions at play in different classes of correlated materials, 1 for which it would be desirable to devise a common theoretical scheme. A typical example is the so-called extended-Drude-model (EDM) analysis, where the experimental optical conductivity is analyzed in terms of a Drude-like model with a frequencydependent inverse lifetime and effective mass. 1This approach is well justified for a single-band system interacting with a continuum of bosonic excitations (such for instance phonons in a metal) where it can provide useful information on the relevant collective modes interacting with the electronic particle-hole excitations. 2The EDM has been widely used also for strongly correlated materials, such as the cuprate superconductors; here, intraband excitations are well separated from the optical interband transitions, which pose a natural cut-off (typically of order of 4000 cm −1 ) to the applicability of the EDM itself in these compounds. 2,3After the discovery of superconductivity in iron-based superconductors an intense experimental and theoretical research has been devoted to the EDM analysis in these materials as well. 4-8A general outcome of this analysis is a relatively large frequency-dependent in-plane inverse lifetime τ −1 (ω) ∼ ω,that, within the EDM, is interpreted in terms of a strong-coupling regime (λ ≃ 3 − 4) for the relevant bosonic excitations, 5,6 located around 20-60 meV. These are usually identified with the spin fluctuations between the hole and electron pockets of the Fermi surface, that from the very beginning have been suggested as the most promising candidates for the superconducting pairing. 9There are however two main open issues concerning these results. On one hand, a strong coupling to these low-energy bosonic modes would imply also an effective mass at low energy much larger than that measured by other probes, , even though some authors suggested that they are present at lower energies. 13-15Low-energy interband transitions may invalidate the conclusions of the EDM analysis, introducing spurious effects.In this paper, we quantify the relative role of interband and intraband transitions in the EDM analysis of the in-plane optical conductivity in pnictides, focusing on LaFePO where no magnetic transition occurs and where the coupling to spin fluctuations with...

No abstract