[CPMD-list] problem with path integrals restart

[Serge Ivanov]

Dear Marek,

Here are my two cents: DIPOLE DYNAMICS is not implemented in PI branch
I have a local hack for PI-BO, but not for PI-CP. I hope to find time to 
code it as well, but probably not earlier than in summer.

PI simulations with 2 beads are a little bit ill-defined. I would suggest 
to increase the number of beads at least to four. I will need to have a 
look on how those radii are computed, but such a problem can be connected 
to your bead number. In fact, you have to study the bead convergence for 
your system and temperature. The rule of thumb is that if you have 
hydrogens in your system at around 100K - then you need at least 32 beads. 

As Axel pointed out, I can't say anything more specific without having 
more information from your side.

WBR, Serge


On Wed, 2 Apr 2008, Marek Naryniecki wrote:

> Date: Wed, 2 Apr 2008 11:50:01 +0200 (CEST)
> From: Marek Naryniecki <marek at elrond.chem.uni.wroc.pl>
> To: cpmd-list at cpmd.org
> Subject: [CPMD-list] problem with path integrals restart
> 
> 
> Dear All
> 
> I have some problems with calculations using PI. When I start MD or when I
> restart MD i get:
> 
> ***************************************************************
>  *                    CHARACTERISTIC RADII                     *
>  ***************************************************************
>    SPECIES  R OF GYRATION  FREE PARTICLE  (IN ANGSTROM)
>    1   P    0.06799       Infinity
>    2   O    0.05446       Infinity
>    3   K    0.08015       Infinity
>    4   H    0.12800       Infinity
>             R OF SUSCEPTIBILITY
>    1   P        NaN       Infinity
>    2   O        NaN       Infinity
>    3   K        NaN       Infinity
>    4   H        NaN       Infinity
>             R OF CORRELATION
>    1   P    0.13599       Infinity
>    2   O    0.10892       Infinity
>    3   K    0.16030       Infinity
>    4   H    0.25600       Infinity
> 
> I noticed that when I used  DEBROGLIE CENTROID  500.d0 during replics
> generation a then start MD in 120K i get this problem. when i start MD
> 320K it starts ok, but after 5 steps i stop, restart md and the problem
> apear.
> 
> what is the reason of that? After that calculations go on and average
> values for bonds seems to be good.
> I looked into cpmd code but without  succes.
> 
> in my input i put
> &CPMD
> PATH INTEGRALS
> MOLECULAR DYNAMICS CP
>  RESTART COORDINATES WAVEFUNCTION LATEST VELOCITEIES
>  DIPOLE DYNAMICS
>  TIMESTEP
>  3.0
>  MAXSTEP
>  100000
>  STRUCTURE BONDS ANGLES
> &END
> &DFT
>  FUNCTIONAL PBE
> &END
> &SYSTEM
>    TESR
>      3
>    ANGSTROM
>    SYMMETRY
>     TETRAGONAL
>    CELL ABSOLUTE DEGREE
>     7.4521 7.4521 6.9740 90.0 90.0 90.0
>    CUTOFF
>     100.0
>  &END
> ..
> &PIMD
>    TROTTER DIMENSION
>      2
>   NORMAL MODES
>   1.d0
>  FACMASS
>   1.d0
>  PRINT LEVEL
>   1
> &END
> 
> Please help me.
> 
> Best Regards,
> 
> 
> Marek Naryniecki
> Wydzial Chemii
> Uniwersytet Wroclawski
> 
> 
> 
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://cpmd.org/mailman/listinfo/cpmd-list
> 

 _____________________________________________________________________
| Dr. Sergei Ivanov                  | Phone : ++49 (0)234-32 26751   |
| Lehrstuhl fuer Theoretische Chemie | Fax   : ++49 (0)234-32 14045   |
| Ruhr-Universitaet Bochum           | sergei.ivanov at theochem.rub.de  |
| D-44780 Bochum                     | http://www.theochem.rub.de/    |
|_____________________________________________________________________|