The single-band Hubbard model of interacting electrons on a lattice is believed to be the minimal model for the famous cuprate high-temperature unconventional superconductors. Additional physical realizations of this model would be highly desirable in the search for new superconductors and to benchmark our theories of strong electron correlations, given the cuprates’ very complicated phase diagram. Owing to the prevalence of high-degeneracy octahedral and tetrahedral coordinations in transition metal oxides, however, additional compounds described by the single-band Hubbard model have yet to be discovered.
Here I will discuss two approaches to design a single-band correlated material via electronic structure calculations coupled with high-throughput computing and data mining. In the first, a specific material (trigonal prismatic vanadium disulfide) is proposed as a candidate based on crystal field theory. Density functional theory calculations, with and without strong correlation corrections based on dynamical mean-field theory, illustrate the one-band nature of the material and suggest it is a promising candidate for strong-correlation physics. In the second, the Open Quantum Materials Database containing electronic structure calculations for more than 500,000 known and hypothetical inorganic compounds is systematically screened for single-band correlated materials. A screening strategy based on chemistry, atomic coordination, nominal electron count and thermodynamic stability correctly captures already-known correlated materials and identifies several new promising candidates.
