Screened hydrogenic orbitals as starting guess

[jagot]

Convergence can be quite a lot faster if we have better starting guesses than the unscreened hydrogenic orbitals we have now. Either we can implement some heuristic method that counts the number of electrons "below" the orbital to be initialized, or we use the Roothan-Hartree-Fock wavefunctions as tabulated by

• Clementi, E., & Roetti, C. (1974). Roothaan-Hartree-Fock Atomic
  Wavefunctions. Atomic Data and Nuclear Data Tables, 14(3-4),
  177–478. http://dx.doi.org/10.1016/s0092-640x(74)80016-1