Multiwfn forum

icamps

Member

Registered: 2018-05-17

Posts: 31

<< Poincare-Hopf relationship is not satisfied >>

Hello,

I found the following warning when searching for CPs:

Poincare-Hopf relationship verification:  164  -  232  +  101  -   14  =  19
Warning: Poincare-Hopf relationship is not satisfied, some CPs may be missing

How to improve the search?

I changed the maximal iterations to 2500 and the scale factor of stepsize to 0.5, but nothing changed.

Regards,

Camps

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,624

Website

Re: << Poincare-Hopf relationship is not satisfied >>

I need more information about your calculation.
Which real space function did your analyzed in topology analysis module?
How did you search the CPs (which commands you inputted?)
How did you generate your input file? What keywords were used in your quantum chemistry program?
What is the chemical formula of your system?

icamps

Member

Registered: 2018-05-17

Posts: 31

Re: << Poincare-Hopf relationship is not satisfied >>

Hello,

Here are the infos:

>> Which real space function did your analyzed in topology analysis module?
Electron density

>> How did you search the CPs (which commands you inputted?)
I am using the following sequence of commands:

2    <- Topological analysis
-1      <- Set CP searching parameters
1       <- Set maximal iterations:
2500
2       <- Set scale factor of stepsize
0.5
0
2    <- Search CPs from nuclear positions
3    <- Search CPs from midpoint of atom pairs
8    <- Generate the path connected (3,-3) and (3,-1)    #9    <- Generate the path connected (3,+1) and (3,+3)
-5    <- Modify or print detail or export paths, or calculate property along a path
1    <- Print summary of paths
4       <- Save points of all paths to paths.txt in current folder
6       <- Export paths as paths.pdb file in current folder
0       <- Return
0    <- Print and visualize all generated CPs, paths and surfaces
-4    <- Modify or export CPs (critical points)
-1    <- Print summary of CPs (in Angstrom)
4    <- Save CPs to CPs.txt in current folder
6    <- Export CPs as CPs.pdb file in current folder
0    <- Return
7       <- Show real space function values at specific CP or all CPs
0       <- If input 0, then properties of all CPs will be outputted to CPprop.txt in current folder
-10    <- Return

>> How did you generate your input file? What keywords were used in your quantum chemistry program?
I generate the input using GAUSSIAN.
The keywords were: #p scf=qc int=ultrafine output=wfn pm7
Here I put a zip with the input and output wavefunction and unformatted check file.

>> What is the chemical formula of your system?
My system is a boron-nitride nanotube with 4 cadmium atoms. The nanotubes ends were competed with hydrogens.

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,624

Website

Re: << Poincare-Hopf relationship is not satisfied >>

I have checked your file, the most severe problem is that you employed semi-empirical method PM7. Multiwfn doesn't support semi-empirical wavefunction. I suggest you performing a single point task using DFT for your optimized structure to yield DFT wavefunction for AIM analysis.

If the system is too large to easily conduct a single point task at DFT level, an alternative choice is using xtb code (https://github.com/grimme-lab/xtb/) to perform the calculation via GFN-xTB theory (a semi-empirical DFT method), and let xtb to generate .molden file. Though the wavefunction quality of GFN-xTB theory is never as satisfactory as DFT, at least the corresponding .molden file can be normally analyzed by Multiwfn.

icamps

Member

Registered: 2018-05-17

Posts: 31

Re: << Poincare-Hopf relationship is not satisfied >>

OK. Thanks for your advice.