Over the last few years, Sr 2 IrO 4 , a single-layer member of the Ruddlesden-Popper series iridates, has received much attention as a close analog of cuprate high-temperature superconductors.Although there is not yet firm evidence for superconductivity, a remarkable range of cuprate phenomenology has been reproduced in electron-and hole-doped iridates including pseudogaps, Fermi arcs, and d-wave gaps. Further, a number of symmetry breaking orders reminiscent of those decorating the cuprate phase diagram have been reported using various experimental probes. We discuss how the electronic structures of Sr 2 IrO 4 through strong spin-orbit coupling leads to the low-energy physics that had long been unique to cuprates, what the similarities and differences between cuprates and iridates are, and how these advance the field of high-temperature superconductivity by isolating essential ingredients of superconductivity from a rich array of phenomena that surround it. Finally, we comment on the prospect of finding a new high-temperature superconductor based on the iridate series. b c a Sr O Ir a b c pseudospin-1/2 J a b FIG. 1. Crystal and magnetic structures of Sr 2 IrO 4 . (a) The crystal structure is based on Ref. [26] according to which the space group is I4 1 /acd. However, recent studies indicate that the symmetry is lower and most probably is I4 1 /a [27-29]. (b) Quasi-2D network of Ir and O. (c) The magnetic structure from Ref. [2].
II. MAPPING ONTO CUPRATE PHYSICSSoon after the discovery of HTSC in the cuprates, many complex oxides with a K 2 NiF 4 structure (isostructural to La 2 CuO 4 ) and its variants were searched for signs of superconductivity. Among these were Sr 2 RhO 4 and Sr 2 IrO 4 [30], which are "one-hole" systems, albeit with a t 2g -hole in the low-spin d 5 configuration as opposed to an e g -hole in cuprates. However, it was quickly apparent that 4d and 5d systems tend to be rather weakly correlated. In fact, Sr 2 RhO 4 is a Fermi liquid metal [31, 32] with a Fermi surface understood qualitatively in terms of a band structure calculated using density functional theory within the local density approximation (LDA) [33, 34]. Interestingly, the Fermi surfaces of Sr 2 RhO 4 and Sr 2 IrO 4 , calculated without SOC, are near indistinguishable due to their almost identical crystal structures with less than 1% difference in their lattice parameters [30]. A closer inspection, however, reveals that LDA does not accurately replicate the measured FermiX M X E-μ (eV) E-μ (eV) E-μ (eV) M X Γ Γ Γ Γ Γ Γ U ζ ζ μ wide band metal narrow J eff =1/2 band J eff =1/2 Mott state LDA+SOC LDA LDA+SOC+U t 2g band J eff =1/2 band J eff =3/2 band J eff =3/2 band J eff =1/2 UHB J eff =1/2 LHB FIG. 2. Illustration of the SOC driven Mott transition. Introduction of SOC splits off a narrow band near the Fermi level (orange solid lines), for which a moderate Coulomb interaction U ∼2 eV is sufficient to open a gap.observation of a significant gap reduction across T N , a typical Slater behavior, supports this viewpoint [42,43]. On th...