Multiwfn forum

saeed_E

Member

Registered: 2019-12-21

Posts: 157

About MP2 density within ESP calculation

Dear Developers,

In the manual of Multiwfn is it evidently indicated that the default value of "cubegendenstype" is "SCF".
In this regards, I have been encountered with a problem when I am going to calculate electrostatic potentials. Please let me explain my problem with a simple example:
Suppose that a given molecular system has been optimized at the MP2/6-31G(d) computational level and I want to calculate value of (V_s,max) on a specific atom using MP2 density. As far as I know, two possibilities exist:
1- In the "settings.ini" file, the value of "cubegendenstype" is changed to MP2 and, then, upload the ".chk" file generated within optimization to the Multiwfn and attempt to calculate V_s,max through main function 12.
2- The value of "cubegendenstype" in "settings.ini" is left unchanged (the default "SCF" is kept). Then, try to generate a "wfn" file for the fully optimized structure using "MP2/6-31G(d) out=wfn density=current" command line. Then, the generated "wfn" file is uploaded to Multiwfn and attempt to compute value of V_s,max through main function 12.

While it is expected that both of two above-mentioned procedures should lead to an identical value of V_s,max I am looking for, procedures 1 and 2 lead to the  two different values for V_s,max. If possible, please kindly let me know what is the problem and which one of  procedures 1 and 2 should be taken into account.

In advance, your kind attention is highly appreciated and I hope you can reply to my message soon as much as possible.
Sincerely,
Saeed

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,628

Website

Re: About MP2 density within ESP calculation

Dear Saeed,

Please note that two factors affect the ESP analysis on vdW surface. (1) The density for generating vdW surface (i.e. electron density isosurface) (2) The ESP on vdW surface.

In your case 1, the density for constructing vdW surface corresponds to HF density (because by default the orbitals recorded in chk file is SCF orbitals), while ESP corresponds to MP2 density (because you request cubegen calculate ESP based on MP2 density matrix)

In your case 2, both the density for constructing vdW surface and calculating ESP is MP2 density (because the orbitals in the input file are MP2 natural orbitals, and the ESP is directly calculated by Multiwfn rather than by invoking cubegen)

Clearly, the two cases give different result.

Best regards,

Tian

saeed_E

Member

Registered: 2019-12-21

Posts: 157

Re: About MP2 density within ESP calculation

Dear Tian,
Thank you very very much.

Sincerely,
Saeed