Multiwfn forum

saeed_E

Member

Registered: 2019-12-21

Posts: 244

Two questions regarding NBO calculations

Dear Tian,
If you kindly let me, I would like to ask two questions regarding NBO calculations:

1- Please suppose A...B interaction. The filled-empty density delocalizations (associated with E(2) perturbation theory) can simply be calculated for this system including two units i and 2. There are some cases, presumably due to the fact that interaction is very strong and inter-nuclear distances are smaller than threshold values, for which NBO is not able to characterize two different units and takes A..B as whole and one unit. Please let me know is there any solution for such cases so that we can manually introduce atoms corresponded to units 1 and 2 in desired manner? If so, please provide a simple example or refer me to the related section in GenNBO 5.01 manual.

2- In the second question, I am going to describe my personal idea to calculate value of polarization (Pol) and charge transfer (CT) in a given non-covalent interaction such as FCl...NCH halogen-bond using NBO. You know much better than me that Pol is defined as filled-empty electron delocalization in one monomer due to presence other monomer while CT is referred to as filled-empty electron delocalization between two monomers. If we want to calculate Pol FCl due to presence of NCH, the steps should be as follows:
a- After full geometry optimization of FCl, an NBO calculation using "nostar" keyword allows the energy difference due to filled-empty delocalizations ie printed out. Let this value to be E1.
b- After full geometry optimization of FCl...NCH complex, the NCH fragment is removed (so that the FCl structure at complex geometry is obtained) and an NBO calculation using "nostar" keyword is again performed over fragment FCl. Let this value to be E2.
It seems delta_E1=E2-E1 should be related to the polarization energy of FCl due to presence of NCH. Evidently, similar calculations can be employed to calculate polarization energy of NCH due to presence of FCl.

Now, I am going to explain how the value of CT could be determined. The corresponding steps should be as follows:
c- Over the fully optimized geometry of complex, an NBO calculation using "nostar" keyword is performed and the empty orbitals of FCl fragment are all deleted. The value of change in energy printed out at the end of calculation is named E3. It seems delta_E=E3-delta_E1 should be corresponded to the energy of CT from NCH to FCl. I think in this calculation, two different posibilities are avoided through "nostar" keyword: one) the filled-empty delocalization in FCl fragment (polarization of FCl due to presence NCH and two)electron density delocalization between filled orbitals of NCH and empty ones of FCl. Thus, E3-delta_E1 should satisfy the value of CT from NCH toward FCl. Again, a similar calculation allows the value of CT from FCl toward NCH is obtained but, this times, the empty orbitals of FCl are deleted.

Please kindly let me know your very valuable opinion and much informative comments on the validity of mentioned idea.

In advance, your kind attention is highly appreciated.

Sincerely,
Saeed

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,830

Website

Re: Two questions regarding NBO calculations

Dear Saeed,

1 You should use $CHOOSE to manually define the Lewis pattern to avoid detecting a BD orbital between the two moieties. There are examples of using $CHOOSE in NBO manual.

2 I don't think "nostar" keyword can be used to reasonably evaluate Pol and CT energy, there are much better ways, such as Morokuma energy decomposition, which is supported by GAMESS-US, and the ALMO energy decomposition, which can be used in Q-Chem.

Tian

saeed_E

Member

Registered: 2019-12-21

Posts: 244

Re: Two questions regarding NBO calculations

Dear Tian,
Too many thanks for your so nice and much valuable comments.
While, as you nicely suggested, GAMESS and Q-Chem are able to perform EDA, my idea may be useful when these codes are not available. On the other hand, you are quite right and a more careful and closer examination of Gen NBO 5.0 manual shows that there are many keywords which allow us to perform such analysis in a much more simpler manner.

Once again, your kind attention is very appreciated.

Sincerely,
Saeed