On performing NBO analysis (Pop=nboread, $nbo bndidx $end ) using g09, the matrix showing bond orders has some negative values. What does that signify?
Note that the NBO module outputs three kinds of bond order when "bndidx" keyword is used, they are quite different:
(1)Wiberg bond order in NAO basis
(2)Atom-atom overlap-weighted NAO bond order
(3)MO bond order
You only need to consider (1). (2) is not as robust and reliable as (1), and in my viewpoint (3) is completely meaningless.
(1) is always positive, because it is simply calculated as the sum of square of corresponding elements of the density matrix in NAO basis.
(2) defines bond order between A and B as ∑[i∈A]∑[j∈B]P(i,j)*S'(i,j), where P is density matrix in NAO basis and S' is overlap matrix in PNAO basis. To some extent, (2) is similar to Mulliken bond order, and its negative value implies anti-bonding effect between the two atoms.
Dear Angana,
Negative bond orders are meaningless. In the other words, when a bond order becomes negative, the unique logical meaning is that there is no any bond between two corresponding atoms.
It should be noted that using other methods/basis sets may provide positive bond order but very close to zero emphasizing,again, there is no any bond between two atoms.
Best,
Saeedreza Emamian ... gaussian is giving negative bond orders (~ -0.2) for some C-C bonds in graphene and those C-atoms are connected..
Dear Angana Ray,
Usally, Gaussian 09W gives negative values in frequency calculation for some less stable configurations. this is the first time i came to know like this. I think " problem arises due to molecular configuration". Suppose if the configuration is a stable conformer, try the same calculation in another basis set or calculate the bond orders separately.
all the best
T.karthick
Note that the NBO module outputs three kinds of bond order when "bndidx" keyword is used, they are quite different:
(1)Wiberg bond order in NAO basis
(2)Atom-atom overlap-weighted NAO bond order
(3)MO bond order
You only need to consider (1). (2) is not as robust and reliable as (1), and in my viewpoint (3) is completely meaningless.
(1) is always positive, because it is simply calculated as the sum of square of corresponding elements of the density matrix in NAO basis.
(2) defines bond order between A and B as ∑[i∈A]∑[j∈B]P(i,j)*S'(i,j), where P is density matrix in NAO basis and S' is overlap matrix in PNAO basis. To some extent, (2) is similar to Mulliken bond order, and its negative value implies anti-bonding effect between the two atoms.
Dear Angana Ray,
I understood your position, my suggestion is that once you reoptimize your compound with a suitable method and basis set and check again the negative bond order problem is getting resolved or not. It may resolve for a better method and basis set.
Best wishes...
Thank you Tian Lu, the Wiberg bond order is what I missed and shuld have considered. It is working fine now.
Negative bond order is usually occured for very low interaction. This means that the BO is around zero or 0.05 or less
On Wed, 1 Nov 1995, Vernon G. Box wrote:
> Hi All
>
> What is the significance of a "negative" bond order? Is it
simply
> an expression of the presence of an anti-bond or no bond?
>
> VB
>
The question is not trivial. As far as I remember, one cannot produce
negative bond orders in any two electron - two basis orbitals problem.
Moreover, filling twice either a bonding or an antibonding orbital will
give a bond order of one, while filling both we get zero - provided
that the bonding and antibonding orbitals are composed of the same
hybrids, and differ only by their signs (and normalization constant).
Negative bond order between two atoms can, therefore, be obtained if
there are interferences from different bonding and antibonding contributions.
One should also take into account that no negative bond orders are possible
in an orthogonalized basis - then the bond order reduces to the Wiberg index
which represent a sum of squares. So negative mond orders are a subbtle
overlap effect.
Yours sincerely,
Prof. Istvan Mayer
Please use e-mails: mayer - at - cric.chemres.hu
IB13LVIB - at - HUEARN.sztaki.hu
H1376May - at - ella.hu
Central Research Institute for Chemistry
of the Hungarian Academy of Sciences
H-1525 Budapest, P.O.Box 17
Hungary
negative bond order is obtained due to basis sets are used in your calculations are not suitable. Thus if you try better method and basis sets, the problem will be solved
See the following paper for details of a recent and more generally applicable bond order that is non-negative: T. A. Manz, "Introducing DDEC6 atomic population analysis: part 3. Comprehensive method to compute bond orders," RSC Advances, 7 (2017) 45552-45581 (http://dx.doi.org/10.1039/C7RA07400J ). This method is implemented in the free Chargemol program (http://ddec.sourceforge.net).
- Badges
- Science topic