What pairs electrons in doped SrTiO3 at Fermi energies below phonons
In plain words
Strontium titanate with very few added electrons superconducts below about 0.4 K. Its electrons have less kinetic energy than the lattice vibrations that should pair them, so standard theory does not apply.
Precise statement
Electron-doped SrTiO3 (Nb, La or oxygen vacancies) superconducts from $n\ \text{below}\ 1\mathrm{e}18\,\mathrm{cm}^{-3}\ \text{to about}\ 1\mathrm{e}21\,\mathrm{cm}^{-3}$ with maximal $T_c$ of approximately $0.4\,\mathrm{K}$ near $n \sim 1\mathrm{e}20\,\mathrm{cm}^{-3}$; at the lowest densities $E_F \sim 1\,\mathrm{meV}$ is far below the longitudinal optical phonon energies (up to about $100\,\mathrm{meV}$). Identify the pairing interaction (soft transverse-optical mode via two-phonon or Rashba-type coupling, dynamically screened longitudinal-optical phonons and plasmons, or other) that reproduces $T_c(n)$ quantitatively over the full dome without adjustable couplings. An answer is a mechanism with computed $T_c(n)$ within a factor of about 2.
What would settle it
A first-principles strong-coupling calculation reproducing $T_{c}(n)$, the isotope effect and the gap-to-$T_{c}$ ratio across the dome.
Status in the literature
Unverified note
Strong-coupling theories with ferroelectric fluctuations (e.g. Saha et al., npj Quantum Materials 2025) reproduce trends, but no mechanism has quantitative consensus.
Related problems
- More general than Does ferroelectric quantum criticality raise the $T_{c}$ of SrTiO3