| Bose–Hubbard model | |
|---|---|
| Name | Bose–Hubbard model |
| Caption | Lattice schematic of interacting bosons |
| Field | Quantum mechanics; Condensed matter physics |
| Introduced | 1989 |
| Introduced by | Gutzwiller approach and development from Hubbard model concepts |
Bose–Hubbard model
The Bose–Hubbard model is a theoretical lattice model describing interacting bosons on a discrete lattice and is central to studies of strongly correlated systems in Quantum mechanics and Condensed matter physics. It captures the competition between kinetic energy (hopping) and interaction energy (on-site repulsion), and underpins understanding of the superfluid–Mott insulator quantum phase transition realized in modern ultracold atomic gas experiments. The model provides a bridge between fundamental many-body theory and experimental platforms such as optical lattices.
The Bose–Hubbard model formalizes interactions of bosonic particles in a periodic potential first motivated by extensions of the Hubbard model and by early work on quantum lattice gases. It is relevant to many-body physics, quantum phase transitions, and the study of quantum simulation using systems like ultracold atoms in optical lattices. Insights from the model inform research on high-temperature superconductivity analogues, quantum magnetism, and proposals for quantum information processing with lattice-based qubits. Seminal theoretical contributions include work by Jaksch et al. that connected the model to realizable cold-atom experiments.
The standard Bose–Hubbard Hamiltonian on a lattice Λ is written as H = -t ∑_{⟨i,j⟩} (b_i^† b_j + h.c.) + (U/2) ∑_i n_i(n_i - 1) - μ ∑_i n_i, where t is the nearest-neighbor hopping amplitude, U is the on-site interaction strength, μ is the chemical potential, b_i^† and b_i are bosonic creation and annihilation operators at site i, and n_i = b_i^† b_i. The model is defined on lattices such as the square lattice, cubic lattice, or one-dimensional chains, and may include additional terms for longer-range hopping or interactions. The Hamiltonian directly connects to second-quantized descriptions used in quantum field theory and lattice realizations in optical lattice setups. Parameters t and U can be related to microscopic quantities via Wannier function integrals, as shown in calculations employing the tight-binding model.
In limiting regimes the model admits analytic understanding. For U ≪ t (weak interactions) the ground state is a delocalized superfluid described by a macroscopic condensate and Bogoliubov theory captures low-energy excitations. For U ≫ t at integer filling one obtains a Mott insulating state with gapped particle–hole excitations; strong-coupling perturbation theory and the atomic limit give access to excitations and spectral gaps. In one dimension, mappings to the Bose gas and techniques from Tomonaga–Luttinger liquid theory elucidate correlation decay and critical properties. Mean-field approximations such as the Gutzwiller approximation provide qualitative phase boundaries in higher dimensions.
The Bose–Hubbard model exhibits a quantum phase diagram in the t/U versus μ/U plane featuring lobes of Mott insulating phases at commensurate fillings separated by superfluid regions. The transition between Mott insulator and superfluid is a zero-temperature quantum phase transition characterized by universality classes that depend on dimensionality and symmetry; in many cases the transition maps to the XY model universality class or to relativistic O(2) criticality. Renormalization group analyses, quantum Monte Carlo simulations, and series expansions have been used to compute critical points and exponents. Experimental confirmation of the phase diagram has validated the role of quantum fluctuations and collective modes such as the Higgs amplitude mode in lattice boson systems.
Numerous extensions enrich the basic model: adding random on-site potentials yields the disordered Bose–Hubbard model and the possibility of a Bose glass phase; multi-component (spinor) extensions treat several bosonic species or internal states and connect to spinor Bose–Einstein condensate physics and SU(N) symmetric models; long-range interactions (e.g., dipolar or Rydberg-mediated) lead to extended Bose–Hubbard models with nearest-neighbor or power-law interactions and new ordered phases like density waves and supersolids. Other variations include lattice geometries with frustration (e.g., triangular lattice), synthetic gauge fields yielding chiral phases, and driven-dissipative versions relevant to photonic simulators and circuit QED platforms.
The paradigm experimental realization uses ultracold atoms loaded into an optical lattice formed by interfering laser beams; landmark experiments by groups at University of Cambridge, MIT, and Max Planck Institute for Quantum Optics observed the superfluid–Mott transition. Control over lattice depth, interaction via Feshbach resonances, and single-site imaging with a quantum gas microscope enable precise tests of theory. Alternative platforms include polar molecules, magnetic atoms with long-range dipolar interactions, arrays of superconducting qubits and circuit quantum electrodynamics where Bose–Hubbard-like Hamiltonians can be engineered, and photonic lattices using coupled cavities.
A variety of numerical techniques are employed to study the Bose–Hubbard model. Exact diagonalization provides benchmarks for small clusters, while density matrix renormalization group (DMRG) and matrix product state methods yield nearly exact results in one dimension. Quantum Monte Carlo (path-integral and stochastic series expansion) is widely used in higher dimensions absent a sign problem. Dynamical mean-field theory (DMFT) and cluster extensions capture local correlations in large lattices; tensor network states and variational Monte Carlo are active areas for simulating dynamics and entanglement. Computational studies guide experiment design and test analytical approximations, reinforcing the model's central role in contemporary quantum many-body research.
Category:Quantum models Category:Many-body physics