Analysis of the dynamical cluster approximation for the Hubbard model
arXiv:cond-mat/0201186 · doi:10.1103/PhysRevB.65.153102
Abstract
We examine a central approximation of the recently introduced Dynamical Cluster Approximation (DCA) by example of the Hubbard model. By both analytical and numerical means we study non-compact and compact contributions to the thermodynamic potential. We show that approximating non-compact diagrams by their cluster analogs results in a larger systematic error as compared to the compact diagrams. Consequently, only the compact contributions should be taken from the cluster, whereas non-compact graphs should be inferred from the appropriate Dyson equation. The distinction between non-compact and compact diagrams persists even in the limit of infinite dimensions. Non-local corrections beyond the DCA exist for the non-compact diagrams, whereas they vanish for compact diagrams.
4 pages, 6 figures. To appear in Phys. Rev. B as a Brief Report
Cited by in corpus (13)
- Electronic Structure Calculations with Dynamical Mean-Field Theory: A Spectral Density Functional Approach
- Quantum Cluster Theories
- Defects in correlated metals and superconductors
- Tracking the Footprints of Spin Fluctuations: A MultiMethod, MultiMessenger Study of the Two-Dimensional Hubbard Model
- Practical consequences of Luttinger-Ward functional multivaluedness for cluster DMFT methods
- Real-space cluster dynamical mean-field theory: Center focused extrapolation on the one- and two particle level
- Dynamical Cluster Approximation Employing FLEX as a Cluster Solver
- Fermi Arcs From Dynamical Variational Monte Carlo
- A novel FLEX supplemented QMC approach to the Hubbard model
- Multi-channel fluctuating field approach to competing instabilities in interacting electronic systems
- Maximum entropy analytic continuation of anomalous self-energies
- Dynamical Variational Monte Carlo as a quantum impurity solver: Application to Cluster Dynamical Mean-Field Theory
- Towards analytical approaches to the dynamical-cluster approximation