Marc Gabay, Stefano Gariglio and Jean-Marc Triscone
“Epitaxial growth followed by topotactic reduction yields superconducting nickelate phases with the same hole-doping level as that obtained by chemical substitution, without causing structural disorder”
Fig. 1 | Key differences between the structures of the cuprates and the nickelates. Both cuprate (left) and nickelate (right) structures include planar d-ion–O2 square unit cells. Apical oxygens are located above/below the Cu atoms of the CuO2 planes and they (hole-) dope the CuO2 plane. The Ni structure does not feature apical oxygen; the ‘functional’ blocks sandwiching the NiO2 stacks produce holes in the Ni eg bands. Another important difference between the two structures is the location of the holes. In the hole is present. When one hole is added to a CuO2 unit, it hops on and off each of the four O surrounding the Cu site (the plaquette), as suggested by the circular arrow. The energy of this state is lower than that obtained when two holes reside at the Cu site (charge-transfer limit). The hole moving around the plaquette plus the hole at the Cu site form a Zhang–Rice spin singlet8. In the NiO2 case, it is energetically preferable to create a second hole on Ni (Mott limit).
