[CPMD-list] VDW energy

[Axel Kohlmeyer]

On Thu, 5 Apr 2007, [GB2312] ÓÚ°×Èã wrote:

dear bairy yu,

BY> Dear all,
BY> Recently, I try to calculated the interaction energy of N2 dimer. First, I
BY> optimized the wavefunction, then I did the md run including &VDW, but I
BY> don't know where to find the interaction energy in the output file.



BY> Here is the output  file:

[...]


BY> 
BY>  (K+E1+L+N+X)           TOTAL ENERGY =          -39.18132072 A.U.
BY>  (K)                  KINETIC ENERGY =           26.50686024 A.U.
BY>  (E1=A-S+R)     ELECTROSTATIC ENERGY =          -38.70761894 A.U.
BY>  (S)                           ESELF =           39.89422804 A.U.
BY>  (R)                             ESR =            0.39814272 A.U.
BY>  (L)    LOCAL PSEUDOPOTENTIAL ENERGY =          -22.86724066 A.U.
BY>  (N)      N-L PSEUDOPOTENTIAL ENERGY =            5.20996446 A.U.
BY>  (X)     EXCHANGE-CORRELATION ENERGY =           -9.32118382 A.U.
BY>          GRADIENT CORRECTION ENERGY =            -0.56423256 A.U.

this looks reasonable.

[...]

BY>  (K+E1+L+N+X)           TOTAL ENERGY =           58.11218987 A.U.
BY>  (K)                  KINETIC ENERGY =           26.50648334 A.U.
BY>  (E1=A-S+R)     ELECTROSTATIC ENERGY =           40.92818932 A.U.

OOOOPS. TOTAL ENERGY >> 0. ouch. what went wrong?


[...]

BY>        NFI    EKINC   TEMPP           EKS      ECLASSIC
BY> EHAM         DIS    TCPU
BY>         18   0.145E+01 0.2E+13   0.58112E+02   0.31288E+08   0.31288E+08

eeeeek! EKINC > 1.0 for a system with so few electrons.
your wavefunction is totally bogus. didn't you see the warning
about the missing wavefunction velocities in the restart file?
please retry that run with QUENCH BO ELECTRONS

[...]

BY> 0.683E+090.46E-01
BY>        282   0.114E+03 0.2E+13   0.34292E+02   0.31286E+08   0.31286E+08
BY> 0.688E+090.46E-01
BY>        283   0.115E+03 0.2E+13   0.28415E+02   0.31286E+08   0.31286E+08
BY> 0.694E+090.46E-01
BY>        284   0.115E+03 0.2E+13   0.19110E+02   0.31286E+08   0.31286E+08
BY> 0.699E+090.46E-01
BY>        285   0.115E+03 0.2E+13   0.31286E+02   0.31286E+08   0.31286E+08
BY> 0.704E+090.46E-01

[...]

<sigh> please, please, PLEASE. don't copy long, (whole) trajectory 
output files to mailing lists.</sigh> better you quote only a few
relevant bits and make the rest available to people in demand.
nobody wants to look at a pointless output like this one.

[...]

BY> And why the final atomic coordinates become so large(eg:28303.103680 a.u.)?
BY> Could anybody help me? Thanks very much.

well, first of all you have to help yourself: pick up a textbook
or a review on car-parrinello MD and try to understand _why_
the fictitious dynamics has to have little kinetic energy and
why you'll get a bogus wavefunction and thus bogus forces and thus
a bogus trajectory if it is not. please have a look at:
http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-tutor/part6.html
for another example of the same phenomenon.


cheers,
   axel.
BY> 
BY> Bairu YU
BY> 

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