Table of Contents
Fetching ...

Universal electronic structure of multi-layered nickelates via oxygen-centered planar orbitals

Christine C. Au-Yeung, X. Chen, S. Smit, M. Bluschke, V. Zimmermann, M. Michiardi, P. C. Moen, J. Kraan, C. S. B. Pang, C. T. Suen, S. Zhdanovich, M. Zonno, S. Gorovikov, Y. Liu, G. Levy, I. S. Elfimov, M. Berciu, G. A. Sawatzky, J. F. Mitchell, A. Damascelli

TL;DR

The paper addresses the origin of low-energy electronic states that enable superconductivity in multilayer nickelates by performing a comparative ARPES study of mixed 2222/1313 $La_3Ni_2O_7$ systems and related LNO4310. It demonstrates a universal low-energy structure dominated by oxygen-centered planar orbitals, with a SDW-driven Fermi surface reconstruction that produces a translated $eta$ band $t\beta$ characterized by $Q_{t\beta} \approx (0.61,0)$, and shows that these features arise from Zhang-Rice singlet (ZRS)-like states evolving from 3-spin polaron (3SP) character. A tight-binding model including in-plane O $p$ orbitals reproduces the polarization-dependent ARPES dichroism and supports an orbital picture where ZRS-like states mediate SDW scattering; the gap size and $t\beta$ coherence depend on the dopant level, linking density-wave order to oxygen content. The work suggests that superconductivity in these nickelates is favored by hole doping that enhances ZRS-like states, drawing a direct parallel to cuprates and highlighting oxygen content as a key control parameter for the competition between SDW and superconductivity.

Abstract

Superconductivity has been demonstrated in the family of multi-layered nickelates La$_3$Ni$_2$O$_7$ and La$_4$Ni$_3$O$_{10}$. Key questions remain open regarding the low-energy electronic states that support superconductivity in these compounds. Here we take advantage of the natural polymorphism between bilayer (2222) and alternating monolayer-trilayer (1313) stacking sequences that arises in bulk La$_3$Ni$_2$O$_7$ crystals, and by employing angle-resolved photoemission spectroscopy (ARPES) we identify a universal low-energy electronic structure in this family of materials. We observe the fingerprint of a doping-dependent spin-density wave (SDW) instability -- strong and coherent enough to reconstruct the Fermi surface, both by gapping out regions of the low-energy electronic structure as well as translating the $β$ pocket by a vector $Q_{tβ}$ consistent with the results of previous neutron and x-ray scattering experiments. Using an effective tight-binding model, we simulate the spectral weight distribution observed in our ARPES dichroism experiments and establish that the low-energy electronic phenomenology is dominated by oxygen-centered planar orbitals, which evolve from the $d_{3x^2-r^2}$ and $d_{3y^2-r^2}$ symmetry characteristic of 3-spin polarons (3SP) to the familiar $d_{x^2-y^2}$ Zhang-Rice singlets (ZRS) that support high-temperature superconductivity in cuprates. By inclusion of magnetic moments on plaquettes of oxygen orbitals in our model, we show that ZRS-like states mediate the SDW. Combined with the observation that oxygen annealing is required to induce superconductivity in both thin films and bulk La$_3$Ni$_2$O$_7$, this demonstrates that the ZRS population dictates whether the ground state favors density-wave order or superconductivity -- with hole doping suppressing the former and stabilizing the latter, as in the cuprates.

Universal electronic structure of multi-layered nickelates via oxygen-centered planar orbitals

TL;DR

The paper addresses the origin of low-energy electronic states that enable superconductivity in multilayer nickelates by performing a comparative ARPES study of mixed 2222/1313 systems and related LNO4310. It demonstrates a universal low-energy structure dominated by oxygen-centered planar orbitals, with a SDW-driven Fermi surface reconstruction that produces a translated band characterized by , and shows that these features arise from Zhang-Rice singlet (ZRS)-like states evolving from 3-spin polaron (3SP) character. A tight-binding model including in-plane O orbitals reproduces the polarization-dependent ARPES dichroism and supports an orbital picture where ZRS-like states mediate SDW scattering; the gap size and coherence depend on the dopant level, linking density-wave order to oxygen content. The work suggests that superconductivity in these nickelates is favored by hole doping that enhances ZRS-like states, drawing a direct parallel to cuprates and highlighting oxygen content as a key control parameter for the competition between SDW and superconductivity.

Abstract

Superconductivity has been demonstrated in the family of multi-layered nickelates LaNiO and LaNiO. Key questions remain open regarding the low-energy electronic states that support superconductivity in these compounds. Here we take advantage of the natural polymorphism between bilayer (2222) and alternating monolayer-trilayer (1313) stacking sequences that arises in bulk LaNiO crystals, and by employing angle-resolved photoemission spectroscopy (ARPES) we identify a universal low-energy electronic structure in this family of materials. We observe the fingerprint of a doping-dependent spin-density wave (SDW) instability -- strong and coherent enough to reconstruct the Fermi surface, both by gapping out regions of the low-energy electronic structure as well as translating the pocket by a vector consistent with the results of previous neutron and x-ray scattering experiments. Using an effective tight-binding model, we simulate the spectral weight distribution observed in our ARPES dichroism experiments and establish that the low-energy electronic phenomenology is dominated by oxygen-centered planar orbitals, which evolve from the and symmetry characteristic of 3-spin polarons (3SP) to the familiar Zhang-Rice singlets (ZRS) that support high-temperature superconductivity in cuprates. By inclusion of magnetic moments on plaquettes of oxygen orbitals in our model, we show that ZRS-like states mediate the SDW. Combined with the observation that oxygen annealing is required to induce superconductivity in both thin films and bulk LaNiO, this demonstrates that the ZRS population dictates whether the ground state favors density-wave order or superconductivity -- with hole doping suppressing the former and stabilizing the latter, as in the cuprates.

Paper Structure

This paper contains 7 sections, 4 figures.

Figures (4)

  • Figure 1: Electronic structure of polymorph LNO327.(a,e) FS of 2222-LNO327 and 1313-LNO327 measured with 100 eV photons, integrated over light polarization and symmetrized around $k_x = 0$ of the tetragonal BZ (white diamond indicates the orthorhombic BZ), and (b,d) corresponding lattice structure (blue: oxygen; red: lanthanum; grey: nickel). The number of electrons $n_e$ is calculated as the ratio between the area of the $\alpha$ and the $\beta$ FS as labeled (counting electrons and including spin degeneracy), and the area of the tetragonal BZ (see SI Sect. S5). (c) Valence band ARPES spectra integrated at the tetragonal zone boundary for both the 2222 and 1313 phases. (f,h) Integrated ($h$$k$ 0) zone XRD patterns of extracted crystallites on which the FS in (a,e) were taken, with diffraction spots showing the pure 2222 and pure 1313 structure, respectively. (g) Extraction of crystallites via ion milling: top inset, image of cleaved LNO327 after ARPES experiments, with green circle indicating the region measured by ARPES; bottom inset, extracted section of crystal for XRD of approximately 40 × 40 $\mu\text{m}^2$ wide and 20 $\mu\text{m}$ thick.
  • Figure 2: Linear dichroism and symmetry of electronic wavefunctions.(a-c) Polar plot of the simulated LNO327 ARPES intensity for linear vertical and linear horizontal (LV, LH) polarizations in the first BZ, where the distance from the center defines the magnitude of the signal at a given angle along the FS. The polar plots are obtained from (a) an effective tight-binding model with only Ni $e_g$ orbitals, (b) experiment, and (c) an effective tight-binding model with both Ni $e_g$ and O $p$ orbitals. (d,f) (top) Schematic representation of wavefunctions at the indicated momenta along the FS with relative phases in red and blue (O and Ni sites in red and silver, respectively), along with (bottom) the simulated ARPES dichroism (LV-LH) from the same model as in (a,c). (e) (top) Experimental geometry, showing the polarization vector of the light with respect to the in-plane orbitals, along with (bottom) the experimental dichroism measured by ARPES on 1313-LNO327.
  • Figure 3: Anomalous electron pocket originating from doping-dependent incommensurate band folding.(a) ARPES spectra and corresponding FS from 1313-LNO327, measured with 45 eV photons and symmetrized around $k_y = 0$. Cut 1 is taken at $k_x\!=\!-0.61$, cut 2 at $k_x\!=\!0$, cut 3 at $k_x\!=\!0.61$, and cut 4 at $k_y\!=\!-0.5$ (expressed in reciprocal lattice units of the orthorhombic BZ); orange arrows indicate the momentum transfer $Q_{t\beta}$ connecting the $\beta$ band to the translated $t\beta$ band. Inset: the gray solid line shows the orthorhombic BZ, with its high-symmetry points labeled. The $\beta$ FS is highlighted with continuous yellow lines; it is translated by the orange arrow ($Q_{t\beta}$) to the $t\beta$ FS, with continuous and dashed segments highlighting portions detected and undetected by ARPES, respectively. (b) Left inset: Tight-binding FS translation by $q_1$, which maximizes the overlap of the circular portions of the $\alpha$ and the $\beta$ FS; right inset: translation by $q_2$, which maximizes the overlap of the straight portions of the $\beta$ FS. Main panel: Tight-binding FS autocorrelation along the $k_x$ direction as a function of the Luttinger volume $n_e\!=\![0.84,1.21]$; the dashed orange and green lines follow the $n_e$-dependence of $q_1$ and $q_2$, respectively (expressed in reciprocal lattice units of the orthorhombic BZ). The autocorrelation calculations for the full FS with $n_e\!=\!0.84$ and the ZRS-projected FS are highlighted with light blue gradient, and correspond to the same effective electron counting [see SI Sect. S12 and Fig. S13(b)]. (c) Comparison of simulated ($q_{1,2}$) and experimental ($Q_{t\beta}$) translation vectors versus the number of electrons $n_e$, with ARPES-derived data from this work (2222-LNO327 and 1313-LNO327), as well as from the literature for LNO327 (Yang et al. Yang2023) and LNO4310 (Li et al. li2017; Du et al.du2024).
  • Figure 4: Doping- and orbital-dependent Fermi surface gapping due to SDW.(a) Schematics of a NiO$_2$ bilayer with a (0.5,0) SDW order, with the magnetic moments residing on plaquettes of oxygen sites (blue and red color represent opposite spin orientations), and antiferromagnetic coupling between adjacent planes. (b) Experimental FS measured with LV + LH polarized light at 45 eV (left), and 100 eV (middle and right); the Fermi volumes $n_e$ are obtained from the tight-binding fits shown as white lines. (c) Band dispersion as measured by ARPES along the momentum cuts indicated by the thick orange lines in (b), with the MDC at $E_{\text{F}}$ shown at the top (integrated over a 20 meV window around $E_{\text{F}}$). (d) Simulation of the electronic dispersion along the same cuts as in (c), obtained with using the SDW model described in the main text and S.I. for net magnetic moments per oxygen plaquette $m_p$ = 0.6 (left), 0.4 (middle), and 0.2 (right) $\mu_B$, as correspondingly indicated.