Bipolaronic superconductivity out of a Coulomb gas
J. Sous,1, 2, ∗C. Zhang,3, †M. Berciu,4, 5 D. R. Reichman,6B. V. Svistunov,7, 8 N. V. Prokof’ev,7and A. J. Millis9, 10, ‡
1Department of Physics, Stanford University, Stanford, CA 93405, USA
2Stanford Institute for Theoretical Physics, Stanford University, Stanford, CA5, USA
3State Key Laboratory of Precision Spectroscopy,
East China Normal University, Shanghai 200062, China
4Department of Physics and Astronomy, University of British Columbia, Vancouver, British Columbia V6T 1Z1, Canada
5Stewart Blusson Quantum Matter Institute, University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada
6Department of Chemistry, Columbia University, New York, New York 10027, USA
7Department of Physics, University of Massachusetts, Amherst, Massachusetts 01003, USA
8Wilczek Quantum Center, School of Physics and Astronomy and T. D. Lee Institute,
Shanghai Jiao Tong University, Shanghai 200240, China
9Department of Physics, Columbia University, New York, New York 10027, USA
10Center for Computational Quantum Physics, Flatiron Institute, 162 5th Avenue, New York, NY 10010, USA
(Dated: December 5, 2022)
Employing unbiased sign-problem-free quantum Monte Carlo, we investigate the effects of long-
range Coulomb forces on BEC of bipolarons using a model of bond phonon-modulated electron hop-
ping. In absence of long-range repulsion, this model was recently shown to give rise to small-size,
light-mass bipolarons that undergo a superfluid transition at high values of the critical transition
temperature Tc. We find that Tcin our model even with the long-range Coulomb repulsion remains
much larger than that of Holstein bipolarons, and can be on the order of or greater than the typical
upper bounds on phonon-mediated Tcbased on the Migdal-Eliashberg and McMillan approxima-
tions. Our work points to a physically simple mechanism for superconductivity in the low-density
regime that may be relevant to current experiments on dilute superconductors.
Introduction. Understanding the mechanisms of su-
perconductivity in the dilute density regime is an ac-
tive theme of research, relevant to polar materials [1,2],
doped topological insulators [3,4], transition metal
dichalcogenides [5], moir´e materials [6–11] and other ma-
terials [12]. This large and growing list of ultra-low car-
rier density superconductors motivates theoretical exam-
ination of superconductivity in electron-phonon coupled
systems at very small densities where, as a matter of
principle, the Fermi liquid/Migdal-Eliashberg paradigm
must fail. Bose-Einstein condensation (BEC) of pre-
formed pairs (“bipolarons”) in principle offers a robust
route to superconductivity at low densities. But, in
the low-density regime, the Coulomb repulsion is weakly
screened and thus the pairing “glue” required to bind
electron pairs into bound states must be strong enough
to overcome the Coulomb repulsion. A strong pairing
interaction is usually believed to result in heavy bound
states, implying low values of the critical transition tem-
perature Tc. These considerations [13,14] are widely be-
lieved to severely limit the maximum Tcobtained from
phonon-mediated binding of electrons into bipolarons.
We have recently shown [15] that even in the pres-
ence of a short-ranged interaction parameterized by a
large onsite Hubbard repulsion U, electrons coupled to
phonons via bond phonon-modulated electron hopping
form small-size, light-mass bipolarons [16,17] that un-
dergo a superfluid (“BEC”) transition at values of Tcthat
are much larger than those obtained in (Holstein) mod-
els in which the electron density is coupled to phonons or
from Migdal-Eliashberg theory of superconductivity out
of a Fermi liquid. This work did not include the long-
range part of the Coulomb interaction, so is relevant to
two-dimensional (2D) materials in which the Coulomb
repulsion is completely screened by gating or proximity
to a substrate [15]. However, in ungated 2D materials
and in three-dimensional (3D) materials in which the
Coulomb repulsion cannot be screened by an external
gate, the question of the effects of long-range Coulomb re-
pulsion on bipolaronic superconductivity (and other non-
phononic BEC mechanisms [18–22]) remains open.
In this letter, we study BEC of bipolarons occuring
in a dilute, 3D Coulomb gas, showing that Tcof bond-
coupled bipolarons is still higher than that of density-
coupled (Holstein) bipolarons, and in line with the value
of Tctypically found in experiments on 3D materials be-
lieved to be close to or in the low-density regime. In a
Coulomb system, the two-electron bound state retains a
finite size even at the critical interaction strength asso-
ciated with unbinding [23], and thus the maximum Tcis
determined by a combination of binding strength, mass
and size with the constraint that the size cannot be in-
finite. To the best of our knowledge, despite decades
of debate [13,24,25], our theory is the first quantita-
tive effort that takes the presence of long-range Coulomb
interaction into account and (i) demonstrates, using an
unbiased approach, a realistic mechanism for BEC for-
mation at relatively high values of Tcand (ii) unveils the
properties of bipolarons, e.g. their mass and size, in 3D.
Model. We consider the bond-Peierls [26] (also known
as bond-Su-Schrieffer-Heeger [27]) electron-phonon cou-
pling on a 3D cubic lattice. In this model the electronic
arXiv:2210.14236v2 [cond-mat.supr-con] 1 Dec 2022