Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2412.13327 · Physical Review A

SU( N ) Fermi-Hubbard model on two sites: Bethe ansatz solution and quantum phase transition of the Lipkin-Meshkov-Glick model in the large- N limit

Pierre Nataf

Published Feb 10, 2025• 3 citations
arXivHTMLPDF

Abstract

The authors find that the SU($N$) Fermi-Hubbard model on two sites can be mapped to the Lipkin-Meshkov-Glick model. It admits a Bethe ansatz solution and exhibits a quantum phase transition in the large-$N$ limit that could be experimentally accessible using current technologies involving SU($N$) ultracold atoms or molecules.

DOI
10.1103/physreva.111.l020201

Related papers

A quasi-exactly solvable Lipkin–Meshkov–Glick model

Feng Pan, Jijie Lin, Xiaogang Xue +1 more

We prove that a special Lipkin–Meshkov–Glick model is quasi-exactly solvable with solutions that can be expressed in the SU (2) coherent state form. Ground-state properties of the model are studied analytically. We also show that the model reduces to the standard two-site Bose–Hubbard model in the large- N limit for finite U / t or large ( N − 1)| U |/ t cases with finite N , which proves that in these cases the ground state of the standard two-site Bose–Hubbard model is an SU (2) coherent state.

Journal of Physics A Mathematical and Theoretical•Apr 15, 2010•0 citations

Solution of One-Dimensional Bose–Hubbard Model in Large-U Limit

Yong Zheng

The one-dimensional Bose-Hubbard model in large-$U$ limit has been studied via reducing and mapping the Hamiltonian to a simpler one. The eigenstates and eigenvalues have been obtained exactly in the subspaces with fixed numbers of single- and double-occupancies but without multiple-occupancies, and the thermodynamic properties of the system have been calculated further. These eigenstates and eigenvalues also enable us to develop a new perturbation treatment of the model, with which the ground-state energy has been calculated exactly to first order in $1/U$.

arXiv:2104.08670•Journal of the Physical Society of Japan•Dec 20, 2021•1 citation

Exactly Solvable Bose and Fermi many-body Hamiltonian with higher order terms based on the S2 symmetry

Feng Pan, L. R. Dai

It is shown that the two component Fermi or Bose many-body Hamiltonian, such as the two-orbit fermion pairing and two-site Bose-Hubbard model with arbitrary finite higher order terms can always be solved exactly by using Bethe ansatz vector construction based on the permutation of two components of bosons or fermions involved. As examples of the solution, the extended one-dimensional dimer Bose -Hubbard model with multi-body interactions and the mean-fifield plus orbit-dependent non-separable pairing model with two non-degenerate j-orbits are demonstrated with the eigenstates and the eigen-energy and the related Bethe ansatz equations. It is shown that the main feature of the solutions lies in the fact that the Bethe ansatz vectors can be expressed in terms of binomials of the boson or fermion operators times the related symmetric functions. As the consequence, two-component quantum many-body systems , such as the extended Lipkin-Meshkov-Glick model with higher-order interactions, can be solved in a similar way.

•Aug 3, 2021•0 citations

Simple $N$-scaling above the superexchange energy of thermodynamic observables in the SU($N$) Fermi Hubbard Model at $1/N$ filling in the two dimensional square lattice

Eduardo Ibarra-García-Padilla, Sohail Dasgupta, Haotian Wei +4 more

Bulletin of the American Physical Society•Jun 2, 2021•0 citations

Large- n limit of the Hubbard-Heisenberg model

John Bradley Marston, Ian K. Affleck

To gain insight into the behavior of the Hubbard model, we define a $\mathrm{SU}(n)$ invariant generalization of the Hubbard-Heisenberg model and, in the large-$n$ limit, solve it in one dimension and in two dimensions on a square lattice. In one dimension the ground state is completely dimerized near half filling. We show that this behavior agrees with a renormalization-group solution of the one-dimensional $\mathrm{SU}(n)$ Hubbard model. In two spatial dimensions we find several different ground states depending on the size of the hopping term $t$, the doping $\ensuremath{\delta}$, and the biquadratic spin interaction $\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{J}$. In particular, the undimerized "flux" or "$s+id$" phase is the ground state at half filling for sufficiently large $t$ or $\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{J}$. We study the electronic and spin excitations of the various phases and comment on the relevance of the large-$n$ problem to the high-${T}_{c}$ superconductors.

Physical review. B, Condensed matter•Jun 1, 1989•494 citations

The Generalized Lipkin-Meshkov-Glick Model and the Modified Algebraic Bethe Ansatz

T. Skrypnyk

We show that the Lipkin-Meshkov-Glick 2N-fermion model is a particular case of one-spin Gaudin-type model in an external magnetic field corresponding to a limiting case of non-skew-symmetric elliptic r-matrix and to an external magnetic field directed along one axis. We propose an exactly-solvable generalization of the Lipkin-Meshkov-Glick fermion model based on the Gaudin-type model corresponding to the same r-matrix but arbitrary external magnetic field. This model coincides with the quantization of the classical Zhukovsky-Volterra gyrostat. We diagonalize the corresponding quantum Hamiltonian by means of the modified algebraic Bethe ansatz. We explicitly solve the corresponding Bethe-type equations for the case of small fermion number N=1,2.

arXiv:2210.04454•Symmetry Integrability and Geometry Methods and Applications•Oct 10, 2022•0 citations
The authors find that the SU(NNN) Fermi-Hubbard model on two sites can be mapped to the Lipkin-Meshkov-Glick model. It admits a Bethe ansatz solution and exhibits a quantum phase transition in the large-NNN limit that could be experimentally accessible using current technologies involving SU(NNN) ultracold atoms or molecules.
We prove that a special Lipkin–Meshkov–Glick model is quasi-exactly solvable with solutions that can be expressed in the SU (2) coherent state form. Ground-state properties of the model are studied analytically. We also show that the model reduces to the standard two-site Bose–Hubbard model in the large- N limit for finite U / t or large ( N − 1)| U |/ t cases with finite N , which proves that in these cases the ground state of the standard two-site Bose–Hubbard model is an SU (2) coherent state.
The one-dimensional Bose-Hubbard model in large-UUU limit has been studied via reducing and mapping the Hamiltonian to a simpler one. The eigenstates and eigenvalues have been obtained exactly in the subspaces with fixed numbers of single- and double-occupancies but without multiple-occupancies, and the thermodynamic properties of the system have been calculated further. These eigenstates and eigenvalues also enable us to develop a new perturbation treatment of the model, with which the ground-state energy has been calculated exactly to first order in 1/U1/U1/U.
It is shown that the two component Fermi or Bose many-body Hamiltonian, such as the two-orbit fermion pairing and two-site Bose-Hubbard model with arbitrary finite higher order terms can always be solved exactly by using Bethe ansatz vector construction based on the permutation of two components of bosons or fermions involved. As examples of the solution, the extended one-dimensional dimer Bose -Hubbard model with multi-body interactions and the mean-fifield plus orbit-dependent non-separable pairing model with two non-degenerate j-orbits are demonstrated with the eigenstates and the eigen-energy and the related Bethe ansatz equations. It is shown that the main feature of the solutions lies in the fact that the Bethe ansatz vectors can be expressed in terms of binomials of the boson or fermion operators times the related symmetric functions. As the consequence, two-component quantum many-body systems , such as the extended Lipkin-Meshkov-Glick model with higher-order interactions, can be solved in a similar way.
Simple NNN-scaling above the superexchange energy of thermodynamic observables in the SU(NNN) Fermi Hubbard Model at 1/N1/N1/N filling in the two dimensional square lattice
To gain insight into the behavior of the Hubbard model, we define a SU(n)\mathrm{SU}(n)SU(n) invariant generalization of the Hubbard-Heisenberg model and, in the large-nnn limit, solve it in one dimension and in two dimensions on a square lattice. In one dimension the ground state is completely dimerized near half filling. We show that this behavior agrees with a renormalization-group solution of the one-dimensional SU(n)\mathrm{SU}(n)SU(n) Hubbard model. In two spatial dimensions we find several different ground states depending on the size of the hopping term ttt, the doping $\ensuremath{\delta}$, and the biquadratic spin interaction $\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{J}$. In particular, the undimerized "flux" or "s+ids+ids+id" phase is the ground state at half filling for sufficiently large ttt or $\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{J}$. We study the electronic and spin excitations of the various phases and comment on the relevance of the large-nnn problem to the high-Tc{T}_{c}Tc​ superconductors.
We show that the Lipkin-Meshkov-Glick 2N-fermion model is a particular case of one-spin Gaudin-type model in an external magnetic field corresponding to a limiting case of non-skew-symmetric elliptic r-matrix and to an external magnetic field directed along one axis. We propose an exactly-solvable generalization of the Lipkin-Meshkov-Glick fermion model based on the Gaudin-type model corresponding to the same r-matrix but arbitrary external magnetic field. This model coincides with the quantization of the classical Zhukovsky-Volterra gyrostat. We diagonalize the corresponding quantum Hamiltonian by means of the modified algebraic Bethe ansatz. We explicitly solve the corresponding Bethe-type equations for the case of small fermion number N=1,2.