try keyword Pop=Full

This properties keyword controls printing of molecular orbitals and several types of population analysis and atomic charge assignments.

Dear Mer. As I can see you are using Gaussian package. Try to run a NBO calculation. You can obtain the % from the results. Good luck!!!

Dear Francisco,

I did try with NBO calculation, as bellow: %nprocshared=8

"%mem=8GB

%chk=AzBr2op.chk

# b3lyp/6-31g(d) Pop = NBOread geom=connectivity......."

But still not so sure if that is what i should try, and the output file make me confused, don't know how the result was presented. Have you done it before? if yes, could you please share with me some point, like which command should I add, and how to read the data from .out file. Thanks!

So far, with the command POP=REG, i get some thing very close to what I need. I can see the main contribution, but not so sure how to calculate the composition in percentage %. Can you help me?

Molecular Orbital Coefficients:

101 102 103 104 105

O O O O O

Eigenvalues -- -0.23068 -0.22207 -0.21484 -0.20827 -0.18124

1 1 C 1S -0.00006 0.00223 -0.00018 -0.00071 0.00002

2 2S 0.00001 -0.00626 0.00017 0.00096 0.00015

3 3S 0.00196 -0.00707 0.00483 0.00131 -0.00275

4 4PX 0.01770 -0.01791 -0.00093 0.00555 0.00053

5 4PY 0.25054 -0.00052 -0.03617 0.00200 0.00721

6 4PZ -0.00262 0.00371 0.00022 0.00110 -0.00009

7 5PX 0.00129 -0.02391 -0.00547 0.01842 0.00100

8 5PY 0.11136 0.00125 -0.01669 0.00018 0.00446

9 5PZ -0.00228 -0.00359 -0.00174 0.00677 -0.00065

10 2 C 1S 0.00003 -0.00374 0.00022 -0.00145 -0.00008

11 2S 0.00006 0.00607 -0.00007 0.00170 0.00011

12 3S 0.00019 0.04274 -0.00519 0.02200 0.00172

13 4PX 0.00690 0.01692 -0.00501 -0.00006 -0.01832

14 4PY 0.09837 -0.00854 -0.05291 0.00686 -0.25930

15 4PZ -0.00116 -0.01533 0.00132 -0.00833 0.00264

16 5PX -0.00069 -0.00051 -0.00248 0.01043 -0.01287

17 5PY 0.05079 -0.00472 -0.02799 0.00218 -0.15859

18 5PZ 0.00269 0.00014 0.00638 -0.02725 0.00106

.............................

try keyword Pop=Full

This properties keyword controls printing of molecular orbitals and several types of population analysis and atomic charge assignments.

Thank you very much. Let me try it and report back later!

Dear Mer: I send you a script to run in Gaussian09 a NBO calculation:

%chk=C:\calcgaus\cuala2w2.chk

%mem=1GB

# b3lyp/6-311++g(d,p) pop=(nbo,savenbo) output=wfn

cu-ala-2w

1 2

N -0.50033200 -1.37213053 -0.61972658

C -1.82260350 -0.75423737 -0.25557285

C -1.69629620 0.77944092 -0.26553564

O -2.67569586 1.47086329 -0.34873739

H -0.43763984 -1.47808121 -1.63393865

O -0.46234510 1.23069582 -0.10396317

H -0.44634631 -2.31524211 -0.23367284

H -2.57898484 -1.04292633 -0.98870463

C -2.23736917 -1.21155812 1.13819510

H -2.39214046 -2.29394962 1.17029338

H -3.17442863 -0.72277114 1.41004894

H -1.48320603 -0.93506733 1.88442440

Cu 0.90905374 -0.02679417 -0.04106181

O 2.05657426 1.57896049 0.40615712

H 2.57922065 1.72186361 1.20698352

H 1.42327723 2.31614704 0.32558531

O 2.58922228 -1.14271671 -0.07018457

H 3.43900495 -0.74846893 -0.31364303

H 2.68657730 -2.10385492 -0.06719331

cuala2w2.wfn

In this calculation NBO study is carried out and NBO orbitals are saved in the chk file (pop=(nbo,savenbo). At the same time the wavefunctions is exported to be used in a QTAIM analysis (output=wfn + cuala2w2.wfn, the name of the file where wavefunction must be saved, at the end of the script).

You obtain the following NBO analysis results:

.....................................................................................................................................................

(Occupancy) Bond orbital/ Coefficients/ Hybrids

---------------------------------------------------------------------------------

1. (0.99605) BD ( 1) N 1 - C 2

( 63.59%) 0.7974* N 1 s( 31.52%)p 2.17( 68.45%)d 0.00( 0.03%)

0.0000 0.5612 0.0105 -0.0088 0.0001

-0.7206 0.0147 0.0082 0.0015 0.3527

0.0015 -0.0049 0.0004 0.2011 0.0003

-0.0047 0.0007 -0.0081 -0.0045 0.0000

0.0129 -0.0081

( 36.41%) 0.6034* C 2 s( 21.01%)p 3.75( 78.85%)d 0.01( 0.14%)

-0.0001 0.4572 -0.0334 0.0025 -0.0002

0.7738 -0.0211 -0.0017 -0.0030 -0.3694

0.0260 -0.0014 0.0007 -0.2266 0.0287

-0.0026 0.0000 -0.0231 -0.0134 0.0055

0.0203 -0.0153

2. (0.99413) BD ( 1) N 1 - H 5

( 70.69%) 0.8408* N 1 s( 24.05%)p 3.15( 75.88%)d 0.00( 0.07%)

0.0002 -0.4903 -0.0039 -0.0091 0.0000

-0.1112 0.0128 0.0009 -0.0005 0.0607

-0.0043 0.0015 -0.0010 0.8614 -0.0220

0.0096 0.0004 0.0008 -0.0014 -0.0080

0.0025 -0.0254

( 29.31%) 0.5414* H 5 s( 99.94%)p 0.00( 0.06%)

-0.9997 -0.0022 0.0006 0.0002 -0.0002

-0.0030 -0.0241

...................................................................................................................................................

As you can see, the NBO analysis gives the % of each atom to the bond orbital and the atomic orbitals involved in the bond.

You can obtain the NBO tutorials from the NBO home page:

http://www.chem.wisc.edu/~nbo5/

Thank you, Francisco. But your out put should be the NBO, not the MO right? It is more likely to be the composition of a specific bond between the two specific atoms. I'd like MO composition, like the conjugated system of HOMO orbital, LUMO orbital. Is there a way to convert NBO to the conjugated NBO or HOMO?

Thanks!

You can obtain the Natural Localized Molecular Orbitals using the keywords in the NBO tutorials.

By other hand, in the Gaussian manual there are another option to know atomic contribution to molecular orbitals. The URL is

http://www.gaussian.com/g_tech/g_ur/k_population.htm

The example is done with a Pd compound.

I hope this help you.

Multiwfn is the most powerful program that supports MO composition analysis, which can be freely downloaded at http://multiwfn.codeplex.com. Multiwfn supports Mulliken, SCPA, Stout-Politzer, NAO and Hirshfeld methods for decomposing MO into atom contributions. Corresponding tutorials can be found in Section 4.8 of the program manual, related theories are introduced in Section 3.10 of the manual. If you have any further question during using the program please feel free to E-mail the program author.

Thank you all for your helpful suggestion. In one my of my orbital composition calculation, the contribution from a fragment appears to be negative (-0.0x%). Should it be problem to see the negative contribution like that?

Thanks!

There are numerous methods to compute orbital composition (as discussed in http://sioc-journal.cn/Jwk_hxxb/CN/abstract/abstract340458.shtml), if Mulliken analysis is employed, then presence of negative contributions is very common, in particular for virtual orbitals. This is one of primary drawbacks of Mulliken analysis. However, if the negative value is not large, you can simply ignore it.

If you change to better methods, such as SCPA, NAOMO, Hirshfeld/Becke method to derive the orbital compositions, the negative value will never occur.

Ignore small negative values.

As regard to thr orbital contribution, generally do the following:

-look what orbital you want to analyze, visualize it and find its eigenvalue,

- for a particular eigenvalue, check "Molecular Orbital Coefficients" and for the particular column simply sum up coefficients with respect to atoms/orbitals;

So afterwards you can say smth. like: first C atom has ~37% (make sure that you use normalized values) contribution to x-orbital with eigenvalue of -0.23068; the biggest contribution to this MO of 50% comes from 4p AO orbitals of 3rd carbon atom , ...

Thank you Danylo. Yes i used the method you suggested before use small freeware to assist me with that.

@Tian Lu: Thank you! Yes, i read about the negative value in some material also. Your reply helped me to confirm it.

Thank you all for spending time with my problem!

when I perform an NBO calculation, but the only NAO calculation is performed

What we can infer from these contributions of fragments in molecular orbitals???