Is there a specific integral/formula that you would like implemented?
We currently have the Ehrenfest Force/Hessian implemented using the virial theorem (via the stress tensor). It would be nice to have a direct implementation, using (fancier) integrals. This would have better numerical characteristics, as the virial-theorem-based definition has spurious nodes resulting from the fourth (force) and fifth (Hessian) derivatives of GTOs.
Additional context
James Anderson has notes on this, so the formulas are explicit (and worked out).
Is there a specific integral/formula that you would like implemented?
We currently have the Ehrenfest Force/Hessian implemented using the virial theorem (via the stress tensor). It would be nice to have a direct implementation, using (fancier) integrals. This would have better numerical characteristics, as the virial-theorem-based definition has spurious nodes resulting from the fourth (force) and fifth (Hessian) derivatives of GTOs.
Additional context James Anderson has notes on this, so the formulas are explicit (and worked out).