Symmetric single-impurity Kondo model on a tight-binding chain: a comparison of analytical and numerical ground-state approaches
Gergely Barcza, Kevin Bauerbach, Fabian Eickhoff, Frithjof B. Anders,, Florian Gebhard, and \"Ors Legeza

TL;DR
This paper compares analytical and numerical methods for studying the ground state of the symmetric single-impurity Kondo model on a tight-binding chain, highlighting the strengths and limitations of each approach across different coupling regimes.
Contribution
It provides a comprehensive comparison of perturbative, variational, DMRG, and NRG methods for the Kondo model, demonstrating the accuracy and applicability of each in various regimes.
Findings
Gutzwiller approach is exact in strong coupling
DMRG can handle chains with about 1000 sites for susceptibilities
NRG is reliable for small Kondo couplings and general density of states
Abstract
We analyze the ground-state energy, local spin correlation, impurity spin polarization, impurity-induced magnetization, and corresponding zero-field susceptibilities of the symmetric single-impurity Kondo model on a tight-binding chain with bandwidth and coupling strength . We compare perturbative results and variational upper bounds from Yosida, Gutzwiller, and first-order Lanczos wave functions to the numerically exact data obtained from the Density-Matrix Renormalization Group (DMRG) and from the Numerical Renormalization Group (NRG) methods. The Gutzwiller variational approach becomes exact in the strong-coupling limit and reproduces the ground-state properties from DMRG and NRG for large couplings. We calculate the impurity spin polarization and its susceptibility in the presence of magnetic fields that are applied globally/locally to the impurity spin. The…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
