Ou and Carter¶
Ou and Carter developed an inversion method inspired by the mRKS.
Ou, Qi, and Emily A. Carter. “Potential Functional Embedding Theory with an Improved Kohn–Sham Inversion Algorithm.” Journal of chemical theory and computation 14.11 (2018): 5680-5689.
In this method we are only required to have a KS set of equations. Then we can derive a self-consisent calculation for the exchange correlation potential as:
where
By replacing the Kohn-Sham density everywhere by the accurate input density and the external potential by an effective external potential. The final expression for this method is:
And we can generate our sample calculation as:
Ne = psi4.geometry(
"""
Ne
noreorient
nocom
units bohr
symmetry c1
""" )
psi4.set_options({"reference" : "rhf",
'DFT_SPHERICAL_POINTS': 350, # Usually specify the DFT spherical grid is highly recommended.
'DFT_RADIAL_POINTS': 210, # See [https://psicode.org/psi4manual/master/dft.html] for options.
'CUBIC_BASIS_TOLERANCE': 1e-21,
'DFT_BASIS_TOLERANCE': 1e-21,
}) # Spin-Restricted
wfn = psi4.properties("CCSD/cc-pcvqz", return_wfn=True, molecule=Ne, properties=["dipole"])[1]
ine = n2v.Inverter(wfn)
We will need to introduce a grid to express the potential for visualization
x = np.linspace(-5,10,1501)
y = [0]
z = [0]
grid, shape = ine.generate_grids(x,y,z)
And we can finally invert the density.
v = ine.invert("OC", vxc_grid=grid, guide_potential_components=["hartree"],
opt_max_iter=35, frac_old=0.9, init="SCAN")