Related Experiment Video
Updated: Aug 6, 2025

Computation of Atmospheric Concentrations of Molecular Clusters from ab initio Thermochemistry
Published on: April 8, 2020
The cumulant Green's functions method for the Hubbard model
R N Lira1, P S Riseborough2, J Silva-Valencia3
1Instituto de Física, Universidade Federal Fluminense, Niterói, RJ, Brazil.
Abstract:
We use the cumulant Green's functions method (CGFM) to study the single-band Hubbard model. The starting point of the method is to diagonalize a cluster ('seed') containingNcorrelated sites and employ the cumulants calculated from the cluster solution to obtain the full Green's functions for the lattice. All calculations are done directly; no variational or self-consistent process is needed. We benchmark the one-dimensional results for the gap, the double occupancy, and the ground-state energy as functions of the electronic correlation at half-filling and the occupation numbers as functions of the chemical potential obtained from the CGFM against the corresponding results of the thermodynamic Bethe ansatz and the quantum transfer matrix methods. The particle-hole symmetry of the density of states is fulfilled, and the gap, occupation numbers, and ground-state energy tend systematically to the known results as the cluster size increases. We include a straightforward application of the CGFM to simulate the singles occupation of an optical lattice experiment with lithium-6 atoms in an eight-site Fermi-Hubbard chain near half-filling. The method can be applied to any parameter space for one, two, or three-dimensional Hubbard Hamiltonians and extended to other strongly correlated models, like the Anderson Hamiltonian, thet - J, Kondo, and Coqblin-Schrieffer models.
Related Concept Videos
Clausius-Clapeyron Equation
Reynolds Transport Theorem
Gibbs Free Energy
Theorems of Pappus and Guldinus: Problem Solving
Euler's Formula for Pin-Ended Columns
To calculate the critical load,...
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation
On...

