Hubbard1D — 1D Hubbard Chain (Lieb–Wu Bethe Ansatz)
Phase 1 exposes the Lieb–Wu (1968) closed-form integrals at half filling only ($\mu = U/2$). General filling and finite-temperature QTM observables are deferred to Phase 2; off-half-filling fetches raise DomainError.
Hamiltonian
\[H = -t \sum_{i, \sigma} \bigl(c_{i,\sigma}^\dagger c_{i+1,\sigma} + \text{h.c.}\bigr) + U \sum_i n_{i,\uparrow} n_{i,\downarrow} - \mu \sum_i n_i, \qquad t > 0,\ U > 0.\]
Half filling corresponds to $\mu = U/2$ by particle-hole symmetry; in this regime the chain is a Mott insulator for any $U > 0$ (no Mott transition — the celebrated Lieb–Wu result).
Phase 1 coverage
| Quantity | Infinite (half filling) | Off half filling |
|---|---|---|
GroundStateEnergyDensity | Lieb–Wu integral | DomainError |
ChargeGap | Lieb–Wu integral | DomainError |
SpinGap | $0$ exactly | DomainError |
Closed-form integrals (Lieb–Wu 1968)
Ground-state energy per site
\[\frac{E_0}{N} = -4 t^2 \int_0^\infty d\omega \, \frac{J_0(\omega)\, J_1(\omega)}{\omega \, \bigl[1 + \exp(\omega U / 2t)\bigr]}.\]
Mott (charge) gap
\[\Delta_c = \frac{16 t^2}{U} \int_1^\infty d\omega \, \frac{\sqrt{\omega^2 - 1}}{\sinh(2\pi t \omega / U)}.\]
Spin gap
\[\Delta_s = 0 \quad \text{exactly} \quad (\text{rigorous gapless spinons}).\]
Both integrals are evaluated numerically with QuadGK.quadgk at relative tolerance $10^{-12}$, well below the test atol of $10^{-8}$.
Asymptotic limits (test anchors)
| Limit | $E_0/N$ | $\Delta_c$ |
|---|---|---|
| $U/t \to 0$ (free fermion) | $-4t/\pi$ | $0$ (exponentially small) |
| $U/t \to \infty$ (Heisenberg AFM) | $-4 t^2 \log 2 / U$ | $U - 4t + 8 t^2 \log 2 / U$ |
The $U \to \infty$ limit reproduces the Hulthén ground-state energy of the Heisenberg AFM chain with effective coupling $J = 4 t^2 / U$.
Quick-look code
using QAtlas
m = Hubbard1D(; t=1.0, U=4.0, μ=2.0) # half filling
QAtlas.fetch(m, GroundStateEnergyDensity(), Infinite()) # ≈ -0.5737293678984494
QAtlas.fetch(m, ChargeGap(), Infinite()) # ≈ 1.2867270220129354
QAtlas.fetch(m, SpinGap(), Infinite()) # 0.0 exactly
# Off half filling raises DomainError (Phase 2 territory)
QAtlas.fetch(Hubbard1D(; t=1, U=6, μ=2),
GroundStateEnergyDensity(), Infinite()) # DomainErrorReferences
- E. H. Lieb, F. Y. Wu, "Absence of Mott transition in an exact solution of the short-range, one-band model in one dimension", Phys. Rev. Lett. 20, 1445 (1968).
- F. H. L. Essler, H. Frahm, F. Göhmann, A. Klümper, V. E. Korepin, The One-Dimensional Hubbard Model, Cambridge University Press (2005).
- E. H. Lieb, F. Y. Wu, "The one-dimensional Hubbard model: a reminiscence", Physica A 321, 1 (2003).
Related
- Heisenberg1D — the $U/t \to \infty$ effective-spin reduction of the half-filled Hubbard chain (Hulthén 1938).
- XXZ1D — the same Bethe-ansatz machinery applied to the anisotropic spin-½ chain (issue #108 — TBA/NLIE).
- Lieb tight-binding lattice — Lieb's other flat-band lattice result (1989); unrelated to the Lieb–Wu Hubbard solution despite the shared surname.
<!– ATLAS:HUBS:START – auto-generated by docs/atlas/generate.jl. Do not edit by hand; edits between these markers are overwritten on next regen. –>
Verified hubs
In the Verified Atlas, these 2 models register 5 hubs (quantity / BC pair). The badge column shows the R1 assurance level; click a hub link to see the exact verify(...) calls, references, and corroboration mechanism.
| Model | Quantity | BC | Assurance | Cards |
|---|---|---|---|---|
ExtendedHubbard1D | ChargeGap | Infinite | 🔵 coherent | 1 |
Hubbard1D | ChargeGap | Infinite | 🔵 coherent | 2 |
Hubbard1D | GroundStateEnergyDensity | Infinite | 🔵 coherent | 2 |
Hubbard1D | LuttingerParameter | Infinite | 🟢 corroborated-at-p | 1 |
Hubbard1D | SpinGap | Infinite | 🟢 corroborated-at-p | 1 |
<!– ATLAS:HUBS:END –>