[CPMD-list] lattice parameter optimize

[Ari.P.Seitsonen at iki.fi]

Dear Xiong,

> I want to optimize the lattice parameter of Aluminum using cpmd
> source. I dont know whether the input file is right and what I will
> get after running cpmd. what kind of information I can get through
> the results.

> Al input file is attached.
> &CPMD
> OPTIMIZE GEOMETRY
> MAXSTEP
> 20000
> STEEPEST DESCENT CELL
> EMASS
> 400
> MOVIE
> 5
> &END
> &SYSTEM
> SYMMETRY
> 1
> CELL
> 4.0494  1.0  1.0  0.0  0.0  0.0
> CUTOFF
> 12.0
> &END
> &ATOMS
> *Al_SGS_BLYP KLEINMAN-BYLANDER
> LMAX=P
> 4
> 0.0000  0.0000  0.0000  1
> 0.0000  2.0247  2.0247  1
> 2.0247  2.0247  0.0000  1
> 2.0247  0.0000  2.0247  1
> &END

There are a couple of essential issues to be changed here: This is a
bulk system, so you need to use k points and include several cells in
the Ewald summation (please see the variable TESR in the manual). In
CPMD v3.5.3 you have to use diagonalisation instead of DIIS to solve
the wave functions. Second, you need to use fractional occupation
numbers, using FREE ENERGY FUNCTIONAL.

> btw,when I special the AlN structure as Hexgonal and point group as
> value 22, it gave me a error information on operation is wrong. If I
> use point group as value 24,it worked. In fact, the AlN is 22 point
> group and hexgonal structure. even used 24 point group, it noted me
> that the structure is triclinic. I dont know what is happened.

Did you try POINT GROUP AUTO, what does it give...?

    Greetings,

       apsi

-- 
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
 Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/
 Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35      
 Anschrift: Physikalisch Chemisches Institut (PCI), Universität Zürich (UniZh)
 Indirizzo: Winterthurerstraße 190, CH-8057 Zürich
 Address:   Schweiz / Svizzera / Suisse / Svizra / Switzerland