Instead of supplying maxiter as an argument to the energy function, treat it as an option. Just include one of the following lines before your energy call:
set maxiter 500 to apply to all methods, or set scf maxiter 500 to apply to just scf.
Might I ask for the coordinates of this troublesome molecule? Improving our SCF code is high on the developer to-do list, and we’d appreciate any examples of molecules we can use to test the SCF changes.
Thank you for your advice, and as you mentioned the coordinates of the molecule cause trouble. I want to calculate torsion energy of the molecule gaining dihedral angle in 10 degrees from 0 to 360.
from rdkit import Chem
from rdkit.Chem import rdMolTransforms
from psikit import Psikit
pk = Psikit()
pk.read_from_smiles("CNS(=O)(=O)CC")
pk.optimize(basis_sets="scf/sto-3g")
conformer = pk.mol.GetConformer(0)
for dihedral_degree in range(0, 360, 10):
rdMolTransforms.SetDihedralDeg(conformer, 0, 1, 2, 3, dihedral_degree) # C-N-S-O dihedral angle
print(dihedral_degree, pk.energy(basis_sets="scf/sto-3g", maxiter=500))
and this is the coordinates of the molecule.
C 4.1849348070947405 -1.7924348429601273 -1.3799272940965122
N 2.84385380774969 0.6704251410262807 -1.2943204629125742
S 0.04279646300828439 0.6222406801042148 0.4580332088292887
O -1.3526430954985227 3.7220850053069308 -0.13212277347509963
O 0.84730520881222 -1.064849617058772 3.346549936962613
C -1.9808929419126597 -1.5621970383882722 -1.317891473501073
C -4.69892399334976 -1.4094671557700051 -0.30289755933589646
H 3.239892498121687 -3.207206103234504 -2.5491088281086873
H 6.0695509123425 -1.463660901576179 -2.139476640284267
H 4.349485483910852 -2.50938090852209 0.5426187252955574
H 2.417734051237826 1.2940252363261038 -3.096551475837313
H -1.225227863560315 -3.4615053542407934 -1.1053693679927319
H -1.910129694750547 -1.0224499650561665 -3.3013290044687893
H -4.7670500895862675 -1.996162869540755 1.6629212744325275
H -5.912865977583351 -2.648060642810516 -1.4006534937099253
H -5.406344934503997 0.5124787385120418 -0.4461539329691137
units angstrom
Since your question seems answered, I marked one of my previous posts as solving the issue. Let me know if you’re still having problems, and I should be able to unmark it.