Multiwfn forum

Multiwfn official website: http://sobereva.com/multiwfn. Multiwfn forum in Chinese: http://bbs.keinsci.com/wfn

You are not logged in.

#1 2023-05-22 16:58:37

amin.alibakhshi@hotmail.c
Member
Registered: 2019-08-25
Posts: 23

some potential bug in units?

Dear All,
for computation of free and effective volume and polarizability of atoms based on Tkatchenko-Scheffler method ( 15-Fuzzy atomic space analysis->option13), the output printed by the software indicates that the computed volumes are in a.u., nevertheless, they seem pretty much to be in Angstrom.^3 unit (for free volumes, computed radii based on spherical shape assumption perfectly matches the vdW radii in Angstrom ). Is it a bug or I got something wrong?
Best,
Amin

Offline

#2 2023-05-22 19:57:12

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,660
Website

Re: some potential bug in units?

Dear Amin,

The printed unit is correct. Please note that the atomic volume defined in Tkatchenko-Scheffler method is not the volume in common sense, it is just used to partition molecular polarizability.

Best,

Tian

Offline

#3 2023-05-23 08:30:21

amin.alibakhshi@hotmail.c
Member
Registered: 2019-08-25
Posts: 23

Re: some potential bug in units?

Dear Tian,
many thanks for the information and quick consideration.
Best regards,
Amin

Offline

#4 2023-05-25 11:49:24

amin.alibakhshi@hotmail.c
Member
Registered: 2019-08-25
Posts: 23

Re: some potential bug in units?

Dear Tian,
I also have another question. Whenever we try to calculate atomic effective volume, the software in the first step asks for wfn for individual atoms. Then, once all wfns are given, then it searches for wfn of the exact same atoms (for which wfn has already been given bu user) in atomwfn and if it cannot find them, ask for the path of Gaussian. I suppose there should be no difference between the two sets of wfn, but since the software asks for them two times separately, I am wondering, shall they be computed differently?
Best,
Amin

Offline

#5 2023-05-25 21:57:05

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,660
Website

Re: some potential bug in units?

Dear Amin,

The atomic .wfn files provided for the first time are those used to calculate atomic free volumes; if you choose to use Hirshfeld weighting function, you will also be asked to provide atomic .wfn files used to calculate the weighting functions. The two sets of atomic wavefunction files can be the same, but it is not mandatory.

Best,

Tian

Offline

#6 2023-05-31 12:26:56

amin.alibakhshi@hotmail.c
Member
Registered: 2019-08-25
Posts: 23

Re: some potential bug in units?

Dear Tian,
many thanks for your response. I am wondering, shouldn't the wavefunctions used to compute free volume and also Hirshfeld weights be (preferably) computed based on the same level of theory to yield consistent results?
The second question is about the equation of q_A in page 78 of manual (first formula in section 3.9.1 Hirshfeld atomic charge). I suppose in that formula the weight only should be multiplied by rho(r) and not rho_diff (i.e. rho^pro should not be weighted). What do you think?
Best regards,
Amin

Offline

#7 2023-05-31 18:17:20

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,660
Website

Re: some potential bug in units?

Dear Amin,

1 The two sets of atomic wavefunction files are not necessarily identical, and Hirshfeld weighting function is relatively less sensitive to the computational level for generating the atomic wavefunction files used in this stage. Of course, you can use the same atomic wavefunction files for the two stages, and this way is somewhat more physically sound.

2 The formula is correct. I just updated Section 3.9.1 to add a few equations to better explain the physical meaning, please re-download the manual.

Best,

Tian

Offline

#8 2023-06-02 09:28:07

amin.alibakhshi@hotmail.c
Member
Registered: 2019-08-25
Posts: 23

Re: some potential bug in units?

Dear Tian,
many thanks, everything is crystal clear. Just one quick thing: I suppose "dr" in these equations should be dr^3. Am I right?
Best,
Amin

Offline

#9 2023-06-02 18:15:24

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,660
Website

Re: some potential bug in units?

Dear Amin,

Please note that r in the equations is bolded to indicate that it is a vector, therefore dr is equivalent to dr^3

Best,

Tian

Offline

Board footer

Powered by FluxBB