[CPMD-list] partial charges on CNTs

[Axel Kohlmeyer]

On Wed, 2 Apr 2008, Alessio Alexiadis wrote:

AA> Hi everybody,
AA>    
AA>   I calculated the wave function of a small nanotube
AA>    
AA>   &INFO
AA>   (5,5) Carbon Nanotube.
AA>   &END

alessio,

AA>   &CPMD
AA>    OPTIMIZE WAVEFUNCTION
AA>    ELECTROSTATIC POTENTIAL 
AA>    CONVERGENCE ORBITALS
AA>     1.0d-7
AA>   &END
AA>    
AA>   &SYSTEM
AA>    SYMMETRY
AA>     8
AA>    ANGSTROM
AA>    CELL
AA>     10.784 0.997 0.913  0.0  0.0  0.0
AA>    CUTOFF
AA>    70.0
AA>   &END 

(n,n) nanotubes are metallic but you are using only a 
gamma point calculation with fixed occupations. i doubt 
that your electron structure is very meaningful.

[...]
    
AA>   According to the CPMD guide the method of Hirshfeld 
AA> Theret.Cim.Acta 44 126 1977 is used to calculated the partial 
AA> charges. I obtained this result


AA>      ATOM                 COORDINATES                   CHARGES
AA>                       X         Y         Z           INT       ESP
AA>       1  C        16.5994   10.1554    1.1622       4.000     3.279
AA>       2  C        16.0457   12.7632    1.1622       4.000    -0.231
AA>       3  C        12.1698   16.2516    1.1622       4.000     6.516
AA>       4  C         9.5185   16.5313    1.1622       4.000    -0.457
AA>       5  C         5.0040   13.9235    1.1622       4.000    -4.526
AA>       6  C         3.9193   11.4876    1.1622       4.000    -3.923
AA>       7  C         5.0040    6.3873    1.1622       4.000     0.872
AA>       8  C         6.9844    4.6034    1.1622       4.000   -14.763
AA>       9  C        12.1698    4.0591    1.1622       4.000     1.313
AA>      10  C        14.4791    5.3914    1.1622       4.000   -14.970
AA>      11  C        15.3748   13.9235    3.4865       4.000    -3.955
AA>      12  C        13.3944   15.7074    3.4865       4.000    -7.806
AA>      13  C         8.2090   16.2516    3.4865       4.000     9.126
AA>      14  C         5.8997   14.9194    3.4865       4.000   -11.176
AA>      15  C         3.7795   10.1554    3.4865       4.000    24.056
AA>   
80 C atoms
AA>    

AA>   The charges are by fare too high; they should be around 0.1e. 

the hirshfeld charges are the INT column and the value of 4.0 
makes perfect sense for a carbon atom with 4 valence electrons.
you get the polarization (which is not the charge!) by subtracting 
the nominal charge (ZV in the pseudopotential) from them.

the ESP derived charges are bogus, of course, i suspect because 
of the "flatness" of the electrostatic potential along the surface 
of the nanotube.

AA> Anybody can help me to understand what I did wrong? (I also tried 
AA> different pseudopotentials, cut-offs or convergence values but the 
AA> problem remains)

PEBCAC


cheers,
   axel.


AA>    
AA>   Thanks in advance
AA>    
AA>   Alessio
AA> 
AA>        

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.