The ABINIT Users Mailing List ( CLOSED )

[Matthieu Verstraete <verstraete@pcpm.ucl.ac.be>]

Hi,

you should send in fuller inputs and outputs. your energy tolerance is not 
very tight, but especially I have no idea what happens when diemix is 0. That 
means your results are probably pure noise. Read its definition and set it to 
0.5 or 1.

If you have a real problem, it could be you need to add ecutsm different from 
0.0, but this seems to be a much bigger problem.

Matthieu



Quoting apirath@hpcc.nectec.or.th:

> Hi all, 
> 
> I have a problem in determining the bond length of a Carbon dimer
> molecule. 
> I have calculated the total-energy curve as a function of the seperation
> between C atoms, and I found that there was a "discontinuity" in the
> total-energy curve. The calculated forces were competely wrong as well.
> 
> 
> THIS IS MY RESULTS ---------------------------- 
> distance(angstrom)  etotal(Ha)    force(Ha/Bohr) 
>     1.20  -1.1651337270E+01 1.3031579214E-01 
>     1.21  -1.1654900731E+01 1.3727504151E-01 
>     1.22  -1.1657978193E+01 1.4365447983E-01 
>     1.23  -1.1660595870E+01 1.4949382326E-01 
>     1.24  -1.1662778804E+01 1.5483024408E-01 
>     1.25  -1.1664550918E+01 1.5969856326E-01 
>     1.26  -1.1665935074E+01 1.6413199426E-01 
>     1.27  -1.1666953111E+01 1.6816076344E-01 
>     1.28  -1.1667625891E+01 1.7181292372E-01 
>     1.29  -1.1667973365E+01 1.7511469508E-01 
>     1.30  -1.1668014577E+01 1.7808921357E-01 
>     1.31  -1.1667767719E+01 1.8075753719E-01 
>     1.32  -1.1667250181E+01 1.8313998960E-01 
>     1.33  -1.1666478571E+01 1.8525566926E-01 
>     1.34  -1.1665468752E+01 1.8712238288E-01 
>     1.35  -1.1664235868E+01 1.8875757765E-01 
>     1.36  -1.1662794390E+01 1.9017574862E-01 
>     1.37  -1.1661158127E+01 1.9139192204E-01 
>     1.38  -1.1659340222E+01 1.9241851915E-01 
>     1.39  -1.1657353280E+01 1.9326725253E-01 
>     1.40  -1.1655209271E+01 1.9394818125E-01 
>     1.41  -1.1652919625E+01 1.9447164545E-01 
>     1.42  -1.1650495230E+01 1.9484767225E-01 
>     1.425 -1.1649235767E+01 1.9498321914E-01 
>     1.43  -1.1647946464E+01 1.9508545971E-01 
>     1.44  -1.1645283198E+01 1.9519402188E-01 
>     1.444 -1.1644187951E+01 1.9520318369E-01 
>     1.448 -1.1630758882E+01 6.8353016987E-02 <<=== !! Discontinous !! 
>     1.45  -1.1630603864E+01 6.9325554195E-02 
>     1.452 -1.1630444414E+01 7.0286281207E-02 
>     1.456 -1.1630112457E+01 7.2172709473E-02 
>     1.46  -1.1629763477E+01 7.4013056829E-02 
>     1.47  -1.1628819530E+01 7.8416984820E-02 
> -------------------------------------------- 
> 
> THIS IS MY INPUT------------------ 
> the pseudopotential file = 6c.pspnc 
> 
> ndtset 15 
>                                                                         
>                                       
> #Definition of the unit cell 
> acell 40 40 40  angstrom 
>                                                                         
>                                       
> #Definition of the atom types 
> ntypat 1 
> znucl 6 
>                                                                         
>                                       
> #Definition of the atoms 
> natom 2 
> typat 1 1 
> xangst: 
>    6.00 6.0 6.0 
>    7.20 6.0 6.0 
>                                                                         
>                                       
> xangst+ 
>    0.00 0.0 0.0 
>    0.02 0.0 0.0 
>                                                                         
>                                       
> #Definition of the planewave basis set 
> ecut 60.0 
>                                                                         
>                                       
> #Definition of the k-point grid 
> nkpt 1 
>                                                                         
>                                       
> #Definition of the SCF procedure 
> nstep 100 
> toldfe 1.0d-6 
> diemac 3.0 
> diemix 0.0 
> ---------------------------------- 
> I used "ABINIT version 4.3.3" compiled by intel compiler "efc" with the
> "MKL library". 
> This run on "Itanium2" and OS is "Linux".   
> 
> Could anyone please check the result and tell me what's wrong with this
> calculation. 
> 
> Thank you, 
> Apirath 
> 


-------------------------------------------------
This mail sent through IMP: atlas.pcpm.ucl.ac.be

----- End forwarded message -----


-------------------------------------------------
This mail sent through IMP: atlas.pcpm.ucl.ac.be