AIP Advances (Jul 2023)

Mean-field approximation of the Fermi–Hubbard model expressed in a many-body basis

  • Antoine Honet,
  • Luc Henrard,
  • Vincent Meunier

DOI
https://doi.org/10.1063/5.0153076
Journal volume & issue
Vol. 13, no. 7
pp. 075210 – 075210-13

Abstract

Read online

The effective independent-particle (mean-field) approximation of the Fermi–Hubbard Hamiltonian is described in a many-body basis to develop a formal comparison with the exact diagonalization of the full Fermi–Hubbard model using small atomic chain as test systems. This allows for the development of an intuitive understanding of the shortcomings of the mean-field approximation and how critical correlation effects are missed in this popular approach. The description in the many-body basis highlights a potential ambiguity related to the definition of the density of states. Specifically, satellite peaks are shown to emerge in the mean-field approximation, in departure from the common belief that they characterize correlation effects. The scheme emphasizes the importance of correlation and how different many-body corrections can improve the mean-field description. The pedagogical treatment is expected to make it possible for researchers to acquire an improved understanding of many-body effects as found in various areas related to the electronic properties of molecules and solids.