3.41 Superconductivity

This module calculates the Eliashberg function α2F(ω), the electron-phonon coupling parameter λ, and the superconducting transition temperature Tc using the Allen-Dynes expression. It builds on the electron-phonon coupling / electronic friction infrastructure described in Section 3.40.

A dedicated publication focused on the superconductivity implementation is forthcoming. Until then, please cite and refer to the following reference when using this module:

  • C. L. Box, W. G. Stark, R. J. Maurer, “Ab initio calculation of electron-phonon linewidths and molecular dynamics with electronic friction at metal surfaces with numeric atom-centred orbitals,” Electron. Struct. 5, 035005 (2023), https://doi.org/10.1088/2516-1075/acf3c4.

Scope and Workflow

The present implementation obtains phonon linewidths from the friction/EPC machinery and then:

  1. 1.

    constructs α2F(ω) on a user-defined frequency grid,

  2. 2.

    evaluates λ=2𝑑ωα2F(ω)/ω,

  3. 3.

    evaluates ωlog and Tc via Allen-Dynes.

For periodic systems, the current friction-based EPC workflow corresponds to the Γ-point / 𝐪=0 setup. The superconductivity driver can be combined with scans in friction broadening and electronic temperature (see keywords below).

Theory

The Eliashberg function is evaluated as

α2F(ω)=12πNε𝐅𝐪,νδ(ωω𝐪ν)γ𝐪νω𝐪ν. (3.185)

The electron-phonon coupling parameter is

λ=2𝑑ωα2F(ω)ω. (3.186)

In the present implementation, the linewidths γ𝐪ν passed from the friction module are treated as full widths at half maximum (FWHM).

The superconducting transition temperature is computed with the Allen-Dynes form

Tc=f1f2ωlog1.20exp(1.04(1+λ)λμ(1+0.62λ)), (3.187)

with

f1 =(1+[λ2.46(1+3.8μ)]3/2)1/3, (3.188)
f2 =(1+λ2(ω¯2/ωlog1)λ2+[1.82(1+6.3μ)(ω¯2/ωlog)]2).

Tags for general section of control.in

 

Tag: calculate_superconductivity(control.in)

Usage: calculate_superconductivity type
Purpose: Enables superconductivity analysis (linewidths, α2F, λ, ωlog, Tc).
type: Method for EPC matrix elements:

  • numerical: finite-difference EPC matrix elements (default).

  • DFPT: DFPT EPC matrix elements (experimental in this context).

 

Tag: alpha2f_broadening_width(control.in)

Usage: alpha2f_broadening_width width
Purpose: Gaussian broadening width (in eV) used to represent delta functions when building α2F(ω).
This setting affects α2F, and therefore also λ, ωlog, and Tc.
Default: 0.001 eV.

 

Tag: alpha2f_grid(control.in)

Usage: alpha2f_grid Emin Emax dE
Purpose: Frequency grid (in eV) used for tabulating α2F(ω).
Emin: minimum frequency in eV.
Emax: maximum frequency in eV (must be larger than Emin).
dE: grid spacing in eV (must be positive).
Default: 0.0 0.5 0.0001 (eV).

 

Tag: supercond_mu_star(control.in)

Usage: supercond_mu_star mu_star
Purpose: Effective Coulomb pseudopotential μ used in Allen-Dynes Tc.
Typical range: 0.1–0.2.
Default: 0.13.

Friction Keywords Used by Superconductivity

The superconductivity driver reuses friction settings for linewidth generation:

If both are scans, all combinations are evaluated (Cartesian product of smearing and temperature grids).

Output Files

Two files are written by the superconductivity module:

  • eliashberg_function.out: frequency grid (eV) and α2F columns for each parameter point.

  • superconductivity_tc.out: one row per parameter point with smearing (eV), electronic temperature (K), μ, λ, ωlog (meV), DOS at EF (1/eV/spin), Allen-Dynes prefactors f1, f2, and Tc (K).

Additionally, use empty_states large enough (or calculate_all_eigenstates) to ensure sufficient unoccupied states are available for the EPC/frequency-resolved analysis.