Hi. Recently, I’m learning about Psi4.
When I optimize the intermidiate structure of nucleophilic acyl substitution (NAS) reaction, a structure in which the intermediate is split into two molecules is obtained.
The code is shown below. How can I solve it.
psi4 version: 1.9
mol = psi4.geometry(‘’’
0 1
C -5.83855 3.77077 0.00028
Cl -4.05191 3.80472 -0.03228
C -6.33538 2.58145 -0.82520
O -6.31848 4.95464 -0.56649
N -6.30435 3.66872 1.38592
H -5.98045 2.66907 -1.87429
H -7.44596 2.55246 -0.82966
H -5.95506 1.62841 -0.39940
H -5.94250 2.79153 1.82648
H -7.35016 3.66232 1.41585
H -5.95905 4.48931 1.93497
‘’')
energy = psi4.optimize(‘hf/sto-3g’, molecule=mol)
In addition, optimization with a protonated intermidiate of oxygen atoms yields a tetrahedral structure.
The STO-3G basis set is very small (poor) and may not capture your target. The first thing I would look at is whether the energy was going down step-by-step in your optimization as it proceeded. If the answer is “yes” then either 1) there is no other minimum there with this basis set and HF; 2) your optimization did not begin close enough to this minimum to be in its local well; or 3) the step size is too large (particularly the very first step) and a smaller step might help.
Thank you very much for your kind response. I have tried the advice you gave me.
First, I tried B3LYP and MP2 as well as HF as calculation methods, and 6-31+G(d,p) and cc-pVDZ as basis set, but they did not work.
Then, I tried setting “interfrag_step_limit” and “intrafrag_step_limit” to 0.2 to reduce the step size, but it also did not work.
Therefore, the cause of this problem is either 1) there is no other minimum there with these basis sets or 2) this optimization did not begin close enough to this minimum to be in its locall well.
To clarify this point, I tried to optimize the same structure in GAMESS and it also did not work, so I think the initial structure is bad. I will do some literature research and try to correct the initial structure in the future. Thank you very much.
If you have any suggestions on how to find the initial structure, I would appreciate it.