The Stretched Conductor

Carbon nanotubes are not superconductors under normal conditions. They conduct well — the cylindrical geometry and delocalized pi electrons make them excellent one-dimensional conductors — but Cooper pairs do not form at accessible temperatures. The electron-phonon coupling is too weak. The phonons are too stiff. The density of states at the Fermi level is too low to support the instability that produces pairing.

The authors stretch the nanotube. A uniaxial tensile strain of 4.5% — pulling the tube along its axis by less than five percent — raises the predicted critical temperature to 162 kelvin. Not a few kelvin. Not a modest enhancement. A transition from non-superconducting to high-temperature superconducting, above the liquid nitrogen threshold, in a structure made entirely of carbon.

Three mechanisms conspire. First, the strain softens the phonons — the lattice vibrations decrease in frequency because the bonds are weakened by stretching. Softer phonons couple more strongly to electrons because the electron-lattice interaction is enhanced when the lattice is more responsive. Second, the density of electronic states at the Fermi level increases — the strain modifies the band structure, creating a van Hove singularity near the Fermi energy that concentrates the electronic states available for pairing. Third, the two effects multiply: more electrons to pair, and stronger pairing interaction per electron.

Quasi-one-dimensionality helps rather than hinders. The reduced dimensionality concentrates the density of states into sharp features that the strain can push to the Fermi level. In three dimensions, the features would be smeared. In one dimension, they are sharp enough to produce a dramatic enhancement from a modest mechanical perturbation.

Pull the tube. The phonons soften. The electrons pair. Carbon becomes a superconductor.