[Pw_forum] what's the difference between 'md' in pw.x and cp.x

[Axel Kohlmeyer]

On Tue, 2009-06-23 at 13:34 +0800, vega lew wrote:
> Dear friends,
> 
> Thank you for you  time for reading and answering my questions.
> 
> I realized that the ensemble is much more important than one snap of
> the whole run, resembling the classical MD to some extent.

there is _no_ difference between classical MD and ab initio MD
in this respect. the only difference between the two is how you
get the forces. statistical mechanics doesn't care about the 
forces only about configurations and ensembles. 

[...]

> we put a integrated water molecule at certain site on the surface. and
> then we perform MD calculations of different method, such as CPMD,
> cp.x, pw.x(calcualtion='md'). I wonder whether the dissociative state
> could be acheived on the surface at very low temparature (such as 100K
> or below) via all the MD calculations mentioned above?

you have to look at this from a different perspective. what you will
see, depends on the potential surfaces of the system, that you are
studying. running MD allows your system to sample different local
minima, provided the barriers between them are small enough. if you
run at low temperature, your MD will not be very different from 
a geometry optimization.  

> From your answers, I see cp.x and CPMD of IBM CORP. is using
> 'Car-Parrinello Molecular Dynamics', while the 'md' calculation in
> pw.x performs born-oppenheimer MD. Paolo just told us that the
> born-oppenheimer MD only perform MD at ground state. Does CPMD or cp.x
> perform the MD calculation at ground state or exited state? 

both are methods that look at the ground state wavefunction. 
the CP method uses a "trick" that would allow to propagate
the wavefunction without having to do a full optimization.
CP can be faster, but has some approximations and limitations,
e.g. CP requires a band gap and has not been implemented for
k-points (at least not in the codes you mentioned, and i don't
know of a code that has it, even though it should be technically
possible).

doing excited state calculations with DFT is a different 
problem, as (regular) DFT is a ground state method.


cheers,
    axel.
> 
> thank you very much for you time 
> 
> best, 
> 
> vega
> 
> 
> -- 
> ==================================================================================
> Vega Lew ( weijia liu)
> PH.D Candidate in Chemical Engineering
> State Key Laboratory of Materials-oriented Chemical Engineering
> College of Chemistry and Chemical Engineering
> Nanjing University of Technology, 210009, Nanjing, Jiangsu, China
> ******************************************************************************************************************
> Email: vegalew at gmail.com
> Office: Room A705, Technical Innovation Building, Xinmofan Road 5#,
> Nanjing, Jiangsu, China  
> ****************************************************************************************************************** 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

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