The Hubbard-Holstein model (HHM) — itinerant fermions with on-site Hubbard repulsion coupled to optical phonons — captures the competition between Coulomb repulsion and phonon-mediated attraction in unconventional superconductors. Hague, Kornilovitch and MacCormick (Phys. Rev. A 102, 033333 (2020)) proposed a Rydberg-dressed cold-atom quantum simulator for its extended form, with off-site phonon-mediated interactions of tunable range and sign, and predicted experimentally accessible preformed pairs and Berezinskii-Kosterlitz-Thouless (BKT) transitions; however, all their quantitative phase-diagram predictions rest on the Lang-Firsov strong-coupling, antiadiabatic approximation. They explicitly left open the physics of the experimentally relevant intermediate regime (dimensionless electron-phonon coupling λ ≈ 1, phonon frequency ℏω_ph ≈ t, with Hubbard repulsion, hopping, and phonon-mediated interaction all comparable — cuprate-like parameters). In the later literature, the local HHM at half filling is now mapped by sign-problem-free quantum Monte Carlo, and fragmentary results exist for extended couplings and for pseudogap observables, but the doped two-dimensional extended model's phase boundaries, BKT transition, preformed-pair regime, and pseudogap spectroscopy at intermediate coupling remain undetermined, and no Hubbard-Holstein quantum simulator has been realized. The problem — to determine these phases and signatures — remains open.

Background

The Hubbard-Holstein model (HHM) is a minimal model of solids in which strong Coulomb repulsion and electron-phonon coupling compete. It describes spin-1/2 fermions hopping between neighboring sites of a lattice with amplitude t, an on-site repulsion U between two fermions of opposite spin on the same site (the Hubbard interaction), and a linear coupling g of the fermion density to a dispersionless quantum oscillator (a 'Holstein' phonon) of frequency ω_ph that may sit on or between the fermion sites. At energies below the phonon frequency, exchanging these phonons produces a retarded attraction between fermions whose overall strength is measured by a dimensionless coupling λ (of order g²/(ℏω_ph W), with W the bandwidth; in the source paper W = 4t and W_λ denotes the resulting phonon-mediated interaction scale) and whose spatial range follows the coupling kernel Φ between sites: the extended HHM additionally contains off-site phonon-mediated interactions, of either sign, on nearest-neighbor and diagonal bonds. Concretely, in the source paper's simulator Hamiltonian H_QS = H_hop + H_R-ph + H_ph + H_Hub (Eq. 19 of arXiv:2001.11980, 'an analogue of an extended HHM'), the fermion density n_i on site i couples linearly to the displacement of each phonon-site oscillator (j,ν) with amplitude f_{ij,ν}, and the phonon-mediated interaction between fermion sites i and i' follows the kernel Φ_{ii'} = Σ_{j,ν} f_{ij,ν} f_{i'j,ν} (Eq. 20); in the source's geometry (r_c much less than the lattice spacing a) only the on-site, nearest-neighbor, and diagonal (next-nearest-neighbor) values Φ_00, Φ_NN, Φ_NNN are needed, and they are tunable in sign and range. The dimensionless coupling is λ = Φ_00/(2 W M_Rb ω_ph²) (Eq. 21), with M_Rb the mass of the atom playing the phonon role. In the antiadiabatic strong-coupling (Lang-Firsov) limit this retarded model reduces to static interactions — the on-site attraction together with nearest-neighbor and diagonal couplings V_1 = −2Φ_NN Wλ and V_2 = −2Φ_NNN Wλ — while away from that limit the retardation of the full fermion-phonon model (Eq. 19) matters; the full retarded model, not its static Lang-Firsov reduction, is the object targeted by this problem. The model is motivated by unconventional superconductors such as the cuprates, where U is comparable to the bandwidth and the relevant phonon frequencies are comparable to the hopping (ℏω_ph ≈ t), so that pairing emerges from a competition between repulsion and retarded attraction rather than from weak-coupling theory.

Two controlled limits bracket the problem. In the antiadiabatic strong-coupling limit (ℏω_ph much greater than t, λ much greater than 1), the Lang-Firsov transformation — a canonical shift of the phonon coordinates that dresses each fermion with its phonon cloud — maps the model onto fermions with a phonon-renormalized hopping t' = t·exp[−W_λ(1 − Φ_NN/Φ_00)/ℏω_ph] and static mediated interactions, for which two-body pairing thresholds, pair masses, and pairing temperatures can be computed analytically. In the weak-coupling limit (λ much less than 1), Migdal-Eliashberg theory — the standard diagrammatic treatment of phonon-mediated superconductivity, which neglects vertex corrections — applies. Between these limits, at λ ≈ 1 and ℏω_ph ≈ t with U, W_λ, and t all comparable, neither expansion is controlled, and this is precisely the regime relevant to cuprate superconductors. The source delimits this window in two related ways — as λ ∼ 1, ℏω_ph ∼ t (its conclusions) and as the regime W_λ ∼ U ∼ t in which 'theoretical techniques often fail' (its Sec. III); with W = 4t the two characterizations agree only up to a numerical factor of order unity, so the target is the intermediate window in this qualitative sense — coupling of order unity, phonon frequency of order t, and phonon-mediated interaction comparable to U and t — rather than a sharply bounded parameter region. In two dimensions, superconducting order in such models appears through a Berezinskii-Kosterlitz-Thouless (BKT) transition — the topological unbinding of vortex-antivortex pairs that establishes quasi-long-range order at finite temperature — and 'preformed' fermion pairs may bind at a temperature T_pair above the condensation temperature T_BKT; whether such preformed pairs deplete the single-particle spectrum and open a 'pseudogap' (a partial gap in the excitation spectrum above the transition temperature, without global symmetry breaking, as observed in the cuprates), as against pseudogaps driven by competing orders such as spin or charge fluctuations, is a central open question of unconventional superconductivity. Numerically, determinant quantum Monte Carlo (DQMC, an exact sampling of the finite-temperature path integral that is sign-problem-free here only at half filling and for local coupling, via the Karakuzu-Seki-Sorella construction), density-matrix renormalization group / matrix product state (DMRG/MPS) methods, dynamical mean-field theory (DMFT, a self-consistent local approximation), and functional renormalization group (fRG, a scheme that integrates out fluctuations momentum-shell by momentum-shell so as to resum infinite classes of diagrams) methods each cover only part of this parameter space.

Hague, Kornilovitch and MacCormick (Phys. Rev. A 102, 033333 (2020)) proposed to access this regime experimentally with a cold-atom quantum simulator: two Rydberg-dressed atomic species (Rydberg dressing off-resonantly admixes a highly polarizable Rydberg state to a ground-state atom, producing strong tunable ranged interactions) held in a painted optical monolayer (a two-dimensional light-shift lattice potential created by projecting a shaped, programmable light pattern), one species representing the fermions and the other the phonon sites, with the Rydberg-dipole interactions allowing independent tuning of U, the local electron-phonon coupling, and the off-site kernel Φ — including the unusual case of repulsive phonon-mediated off-site interactions. They predicted that preformed pairs, a pairing transition, and BKT condensation at temperatures of order 20 nK are experimentally accessible. However, their quantitative predictions (two-body pairing thresholds U^(C), pairing temperatures T_pair, pair masses, and T_BKT) were all computed in the Lang-Firsov strong-coupling, antiadiabatic limit, and they explicitly delimited what remains unknown: the actual phases and their boundaries at the 'modest' couplings λ ≈ 1 and ℏω_ph ≈ t where, as they note, theoretical techniques often fail.

Problem Statement

As delimited by the source work, the phase-diagram predictions for the extended Hubbard-Holstein model (HHM) — and for its proposed Rydberg-dressed cold-atom quantum simulator — are computed only in the strong-coupling, antiadiabatic (Lang-Firsov) limit, so the behavior in the experimentally relevant intermediate regime, with dimensionless electron-phonon coupling λ ≈ 1, phonon frequency ℏω_ph ≈ t, and Hubbard repulsion, hopping, and phonon-mediated interaction all of comparable size, remains to be determined. The problem is to establish, for the two-dimensional extended Hubbard-Holstein model with local Hubbard interaction U and off-site phonon-mediated interactions of tunable range and sign (the Hamiltonian engineered by the proposed simulator: the full retarded fermion-phonon model H_QS = H_hop + H_R-ph + H_ph + H_Hub of Eq. 19 of arXiv:2001.11980, whose mediated interactions follow the kernel Φ_{ii'} defined in the Background and of which the static Lang-Firsov form is recovered only in the antiadiabatic strong-coupling limit), in this intermediate regime: (a) the equilibrium finite-temperature phase boundaries among the normal state, the phonon-mediated preformed-pair regime, and the Berezinskii-Kosterlitz-Thouless (BKT) condensed superconducting phase, in particular whether and how the strong-coupling two-body pairing thresholds and pairing temperature T_pair shift at moderate coupling and with retardation; (b) the BKT transition temperature T_BKT and the pair mobility (effective mass) as functions of λ, ℏω_ph/t, U, and the off-site interaction kernel; and (c) the single-particle spectral signatures of the preformed-pair regime — the presence, magnitude, and temperature dependence of any pseudogap — and how they change when the off-site phonon-mediated interactions are switched on or off or their range and sign are varied, i.e. whether preformed pairs alone can be consistent with a pseudogap, as against competing-order or two-gap scenarios. These deliverables serve one objective — the intermediate-regime phase diagram and pairing phenomenology of the extended HHM — and an answer confined to the strong-coupling Lang-Firsov limit does not resolve it, that limit being precisely the one in which the existing predictions were computed.

References

  1. J. P. Hague, P. E. Kornilovitch, C. MacCormick, 'A cold atom quantum simulator to explore pairing, condensation, and pseudogaps in extended Hubbard-Holstein models', Phys. Rev. A 102, 033333 (2020). DOI: 10.1103/PhysRevA.102.033333, https://doi.org/10.1103/PhysRevA.102.033333 (also arXiv:2001.11980, https://arxiv.org/abs/2001.11980)
  2. G. D. Adebanjo, J. P. Hague, P. E. Kornilovitch, 'Quantum Monte Carlo Simulation of Bipolaron Superconductivity in Extended Hubbard-Holstein models on Face-Centered-Cubic and Body-Centered-Cubic Lattices', Annalen der Physik 538, 202500343 (2025). DOI: 10.1002/andp.202500343, https://doi.org/10.1002/andp.202500343 (arXiv:2507.17398, https://arxiv.org/abs/2507.17398)
  3. J. P. Mendonça, K. Jachymski, 'Quantum simulation of extended electron-phonon-coupling models in a hybrid Rydberg atom setup', Phys. Rev. A 107, 032808 (2023). DOI: 10.1103/PhysRevA.107.032808, https://doi.org/10.1103/PhysRevA.107.032808 (arXiv:2208.11473, https://arxiv.org/abs/2208.11473)
  4. N. C. Costa, K. Seki, S. Yunoki, S. Sorella, 'Phase diagram of the two-dimensional Hubbard-Holstein model', Communications Physics 3, 80 (2020). DOI: 10.1038/s42005-020-0342-2, https://doi.org/10.1038/s42005-020-0342-2 (arXiv:1910.01146, https://arxiv.org/abs/1910.01146)
  5. S. Karakuzu, K. Seki, S. Sorella, 'Solution of the sign problem for the half-filled Hubbard-Holstein model', Phys. Rev. B 98, 201108 (2018). DOI: 10.1103/PhysRevB.98.201108, https://doi.org/10.1103/PhysRevB.98.201108 (arXiv:1808.07759, https://arxiv.org/abs/1808.07759)
  6. S. Karakuzu, L. F. Tocchio, S. Sorella, F. Becca, 'Superconductivity, charge-density waves, antiferromagnetism, and phase separation in the Hubbard-Holstein model', Phys. Rev. B 96, 205145 (2017). DOI: 10.1103/PhysRevB.96.205145, https://doi.org/10.1103/PhysRevB.96.205145 (arXiv:1709.00278, https://arxiv.org/abs/1709.00278)
  7. A. Al-Eryani, S. Andergassen, M. M. Scherer, 'Intertwined fluctuations and isotope effects in the Hubbard-Holstein model on the square lattice from functional renormalization', Phys. Rev. Research 7, 043052 (2025). DOI: 10.1103/6vyb-h379, https://doi.org/10.1103/6vyb-h379 (arXiv:2504.10863, https://arxiv.org/abs/2504.10863)
  8. J. Coulter, A. J. Millis, 'Electron-phonon coupling in correlated materials: insights from the Hubbard-Holstein model', Phys. Rev. B 113, 165145 (2026). DOI: 10.1103/3qk3-py1n, https://doi.org/10.1103/3qk3-py1n (arXiv:2505.08081, https://arxiv.org/abs/2505.08081)
  9. I. Esterlis, S. A. Kivelson, D. J. Scalapino, 'Pseudogap crossover in the electron-phonon system', Phys. Rev. B 99, 174516 (2019). DOI: 10.1103/PhysRevB.99.174516, https://doi.org/10.1103/PhysRevB.99.174516 (arXiv:1811.10622, https://arxiv.org/abs/1811.10622)
  10. Y. Wang, I. Esterlis, T. Shi, J. I. Cirac, E. Demler, 'Zero-temperature phases of the two-dimensional Hubbard-Holstein model: A non-Gaussian exact diagonalization study', Phys. Rev. Research 2, 043258 (2020). DOI: 10.1103/PhysRevResearch.2.043258, https://doi.org/10.1103/PhysRevResearch.2.043258 (arXiv:1910.01792, https://arxiv.org/abs/1910.01792)
  11. P.-Y. Zhao, K. Ding, S. Yang, 'Chebyshev pseudosite matrix product state approach for the spectral functions of electron-phonon coupling systems', Phys. Rev. Research 5, 023026 (2023). DOI: 10.1103/PhysRevResearch.5.023026, https://doi.org/10.1103/PhysRevResearch.5.023026 (arXiv:2210.09208, https://arxiv.org/abs/2210.09208)
  12. S. Jiang, T. Shi, 'Scalable Simulation of Strongly Correlated Electron-Phonon Systems via Non-Gaussian Matrix Product States', arXiv:2607.08176 (2026), https://arxiv.org/abs/2607.08176
  13. S. Karakuzu, A. Tanjaroon Ly, P. Mai, J. Neuhaus, T. A. Maier, S. Johnston, 'Stripe correlations in the two-dimensional Hubbard-Holstein model', Communications Physics 5, 311 (2022). DOI: 10.1038/s42005-022-01092-x, https://doi.org/10.1038/s42005-022-01092-x (arXiv:2205.15464, https://arxiv.org/abs/2205.15464)
  14. K. Kovač, A. Nocera, A. Damascelli, J. Bonča, M. Berciu, 'Signature of Preformed Pairs in Angle-Resolved Photoemission Spectroscopy', Phys. Rev. Lett. 134, 096502 (2025). DOI: 10.1103/PhysRevLett.134.096502, https://doi.org/10.1103/PhysRevLett.134.096502 (arXiv:2409.14490, https://arxiv.org/abs/2409.14490)
  15. P. Weckesser, K. Srakaew, T. Blatz, D. Wei, D. Adler, S. Agrawal, A. Bohrdt, I. Bloch, J. Zeiher, 'Realization of a Rydberg-dressed extended Bose-Hubbard model', Science 390, 849-853 (2025). DOI: 10.1126/science.adq7082, https://doi.org/10.1126/science.adq7082 (arXiv:2405.20128, https://arxiv.org/abs/2405.20128)
  16. U. Bhattacharya, T. Graß, A. Bachtold, M. Lewenstein, F. Pistolesi, 'Phonon-Induced Pairing in Quantum Dot Quantum Simulator', Nano Letters 21, 9661-9667 (2021). DOI: 10.1021/acs.nanolett.1c03457, https://doi.org/10.1021/acs.nanolett.1c03457 (arXiv:2106.09418, https://arxiv.org/abs/2106.09418)
  17. T. Bode, R. Roma, S. Schmitz, A. Ciani, D. S. Shapiro, D. Bagrets, F. K. Wilhelm, 'Quantum Simulation of Strongly Correlated Fermion-Phonon Models in Circuit QED', arXiv:2607.11516 (2026), https://arxiv.org/abs/2607.11516
  18. S. Giuli, R. J. Campos-Lopes, E. Moghadas, M. Capone, 'Ghost-RISB for Correlated Electron-Phonon Systems: Application to the Hubbard-Holstein Model', arXiv:2608.04816 (2026), https://arxiv.org/abs/2608.04816