[CPMD-list] (no subject)

[Axel Kohlmeyer]

On Mon, 7 May 2007, davidofkozlowski wrote:

DK> Dear all,

dear 'davidofkozlowski',

DK> We would like to calculate the BDE of phenol in water. To perform  
DK> such calculation with a large numbers of water molecules, we believe  
DK> that CPMD offers the best possibility.

can you please comment on what you call a 'large' number of water 
molecules. the method employed by CPMD does not scale linearly, 
so calculations on larger systems tend to become very costly.

DK> However this is a new technique for us, so we would appreciate some  
DK> help with the following questions.
DK> 
DK> 1st question : Which are the best functionals and pseudopotentials  
DK> for radicals? In standard DFT calculations hybrid B3LYP is known to  
DK> underestimate the BDE.

there is no general answer for that. please note that CPMD usually
employs 'plain' DFT (i.e. BLYP instead of B3LYP). in theory it is
possible to run hybrid functionals, but for 'large' systems the cost
of computing the HF-exchange is prohibitively expensive (easily two 
or three orders of magnitude).

DK> 2nd question : When using MT pseudopotential a large cutoff (70 ry)  

for MT oxygen pseudopotentials a cutoff of 70ry is not large, on the 
contrary. the cutoff generally is determined by the 'hardest' atom, 
which in your case is the oxygen and there about 85-90ry is a 
conservative choice and at about 120ry you are 'fully' converged.

DK> is required. If we run calculations (PBC) of phenol with only the  
DK> first solvation shell (12x12x10) is it possible that the solute  
DK> interacts with its image?

definitely yes. you have two effects: the water molecules 'on the right'
are directly interacting through PBC with the ones 'on the left' and 
thus you have an (unphysical) preference for motions that are
commensurate with this and you have rather long ranging coulomb
interactions (since phenole is polar and you'll probably want to
abstract a proton, i guess).

DK> 3rd question : We know that CPMD calculations can yield the Helmholtz  
DK> Free Energy (DeltaA or DeltaF) ) but we require the enthalpy (DeltaH)  
DK> to calculate the BDE. Is there an easy way to convert one to the  
DK> other or to obtain DeltaH directly ?

i'd actually worry more about the accuracy of the calculation 
altogether. in principle you can do a 'constant pressure' MD
with CPMD as well, but i suspect that the additional 
(computational and human) effort required to perform
that kind of run accurate enough will be justified.

you also didn't say exactly, how you intend to compute the 
BDE itself, that may have some impact on accuracy on its own.

cheers,
   axel.

DK> 
DK> Thank you for any suggestions.
DK> _______________________________________________
DK> CPMD-list mailing list
DK> CPMD-list at cpmd.org
DK> http://cpmd.org/mailman/listinfo/cpmd-list
DK> 

-- 
=======================================================================
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.