Using the local density approximation and its combination with dynamical mean-field theory, we show that electronic correlations induce a single-sheet, cuprate-like Fermi surface for hole-doped 1/1 LaNiO3/LaAlO3 heterostructures, even though both eg orbitals contribute to it. The Ni 3d 3z 2 −1 orbital plays the role of the axial Cu 4s-like orbital in the cuprates. These two results indicate that "orbital engineering" by means of heterostructuring should be possible. As we also find strong antiferromagnetic correlations, the low-energy electronic and spin excitations in nickelate heterostructures resemble those of high-temperature cuprate superconductors. PACS numbers: 71.27.+a, 71.10.Fd, 74.78.Fk The discovery of high-temperature superconductivity (HTSC) in hole-doped cuprates [1] initiated the quest for finding related transition-metal oxides with comparable or even higher transition temperatures. In some systems such as ruthenates [2] and cobaltates [3] superconductivity has been found. However, in these t 2g systems superconductivity is very different from that in cuprates and transition temperatures (T c 's) are considerably lower.As it became possible to grow transition-metal oxides in heterostructures, this quest got a new direction: Novel effectively two-dimensional (2D) systems could be engineered. But which oxides, besides cuprates, are most promising for getting high T c 's?The basic band structure of the hole-doped cuprates is that of a single 2D Cu 3d x 2 −y 2 -like band which is less than half-filled (configuration d 9−h ). In this situation, antiferromagnetic fluctuations prevail and are often believed to mediate the superconductivity. The Fermi surface (FS) from this x 2 − y 2 band has been observed in many overdoped cuprates and found to agree with the predictions of density-functional (LDA) band theory.Recently the following idea for arriving at a cupratelike situation in nickelates was presented [4]: Bulk LaNiO 3 d 7 has one electron in two degenerate e g bands, but sandwiching a LaNiO 3 layer between layers of an insulating oxide such as LaAlO 3 will confine the 3z 2 − 1 orbital in the z-direction and may remove this band from the Fermi level, thus leaving the electron in the x 2 − y 2 band. The possibility of finding bulk nickelates with an electronic structure analogous to that of cuprates was discarded a while ago [5], but heterostructures offer new perspectives.Indeed, a major reconstruction of orbital states at oxide interfaces may recently have been observed [6], and this kind of phenomenon could lead to novel phases not present in the bulk. Extensive theoretical studies of mechanisms for orbital selection in correlated systems [7] have revealed the complexity of this problem, where details of the electronic structure and lattice distortions play decisive roles. It is therefore crucial to examine nickelate heterostructures by means of state-of-the-art theoretical methods and find the optimal conditions for x 2 − y 2 orbital selection.In this Letter we present results of electronic-str...

Systems of adatoms on semiconductor surfaces display competing ground states and exotic spectral properties typical of two-dimensional correlated electron materials which are dominated by a complex interplay of spin and charge degrees of freedom. We report a fully ab initio derivation of low energy Hamiltonians for the adatom systems Si(111):X, with X=Sn, Si, C, Pb, that we solve within self-consistent combined GW and dynamical mean field theory ("GW+DMFT"). Calculated photoemission spectra are in agreement with available experimental data. We rationalize experimentally observed tendencies from Mott physics towards charge-ordering along the series as resulting from substantial long-range interactions.PACS numbers: 71.15. Mb, 73.20.At, 71.10.Fd, 71.30.h Understanding the electronic properties of materials with strong electronic Coulomb correlations remains one of the biggest challenges of modern condensed matter physics. The interplay of delocalization and interactions is not only at the origin of exotic ground states, but also determines the excitation spectra of correlated materials. The "standard model" of correlated fermions, the Hubbard model, in principle captures these phenomena. Yet, relating the model to the material on a microscopic footing remains a challenge. Even more importantly, the approximation of purely local Coulomb interactions can become severe in realistic materials, where long-range interactions and charge fluctuation physics cannot be neglected.Systems of adatoms on semiconducting surfaces, such as Si (111):X with X=Sn, C, Si, Pb, have been suggested [1] to be good candidates for observing low-dimensional correlated physics. Commonly considered to be realizations of the one-band Hubbard model and toy systems for investigating many-body physics on the triangular lattice, such surfaces have been explored experimentally [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18] and theoretically [19][20][21][22][23][24][25][26][27][28][29][30][31].These so-called α-phases show a remarkable variety of interesting physics including commensurate charge density wave (CDW) states [5,6,9] and isostructural metal to insulator transitions (MIT) [14]. However, while specific systems and/or phenomena have been investigated also theoretically, a comprehensive understanding including materials trends is still lacking. A central goal of our work is to present a unified picture that relates, within a single framework, different materials (adatom systems), placing them in a common phase diagram.We derive low-energy effective Hamiltonians ab initio from a combined density functional and constrained random phase approximation (cRPA) scheme [32] in the implementation of [33] (see also the extension to surface systems in [34]). While the first surprise are the relatively large values of the onsite interactions which we find to be of the order of the bandwidth (≈ 1 eV), most importantly we show that non-local interactions are large (nearestneighbor interaction of ≈ 0.5 eV) and, hence, an essential part of t...