<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
  <title></title>
</head>
<body>
Dear Ari,<br>
<br>
thanks for your suggestions. &nbsp;First of all I gave the atoms a random displacement
of 0.05 by the option randomize in the inputfile and then did again a geometry
optimization. &nbsp;The programs ends normally, but the atoms do not return to
their original position, the energy however is converged. &nbsp;I must conclude
then that the system is indeed unstable. &nbsp;The input file is given below of
this simulations. &nbsp;By the way I did not optimize the cell parameters, is
this necessary? &nbsp;<br>
<br>
Now some simulations run on the larger system. &nbsp;this takes longer so I do
not know the result yet. &nbsp;<br>
<br>
Is it an option to use the nose keyword for the electrons to keep them close
to the BO surface? &nbsp;<br>
Then the question remains how to choose this electronic mass? &nbsp;<br>
<br>
Thanks in advance<br>
veronique<br>
<br>
PS : the thing movie sample is for some conversion programs that I have written.
&nbsp;They transform the MOVIE file into car and arc files which can be read in
by CERIUS.<br>
<br>
<br>
<br>
input file for geometry optimization :<br>
&amp;CPMD<br>
&nbsp;RESTART LATEST WAVEFUNCTION<br>
&nbsp;OPTIMIZE GEOMETRY<br>
&nbsp;TIMESTEP<br>
&nbsp; 5<br>
&nbsp;MAXSTEP<br>
&nbsp; 3000<br>
&nbsp;PRINT<br>
&nbsp; 10<br>
&nbsp;RANDOMIZE COORDINATES<br>
&nbsp; 0.05<br>
&nbsp;STRUCTURE BONDS ANGLES<br>
&nbsp;MOVIE SAMPLE<br>
&nbsp; 1<br>
&amp;END<br>
<br>
&amp;SYSTEM<br>
&nbsp;ANGSTROM<br>
&nbsp;SYMMETRY<br>
&nbsp;1<br>
&nbsp;CELL<br>
&nbsp;6.3 1.0 1.0 0.0 0.0 0.0<br>
&nbsp;CUTOFF<br>
&nbsp;25.<br>
&amp;END<br>
<br>
&amp;ATOMS<br>
*Cl_VDB_U BINARY NEWF<br>
&nbsp;LMAX=P<br>
&nbsp;4<br>
&nbsp;0.000000&nbsp;&nbsp;&nbsp; 0.000000&nbsp;&nbsp;&nbsp; 0.000000<br>
&nbsp;3.150000&nbsp;&nbsp;&nbsp; 0.000000&nbsp;&nbsp;&nbsp; 3.150000<br>
&nbsp;0.000000&nbsp;&nbsp;&nbsp; 3.150000&nbsp;&nbsp;&nbsp; 3.150000<br>
&nbsp;3.150000&nbsp;&nbsp;&nbsp; 3.150000&nbsp;&nbsp;&nbsp; 0.000000<br>
*019-K-ca-sp-vgrp.uspp BINARY NEWF<br>
&nbsp;LMAX=P<br>
&nbsp;4<br>
&nbsp;3.150000&nbsp;&nbsp;&nbsp; 3.150000&nbsp;&nbsp;&nbsp; 3.150000<br>
&nbsp;0.000000&nbsp;&nbsp;&nbsp; 0.000000&nbsp;&nbsp;&nbsp; 3.150000<br>
&nbsp;0.000000&nbsp;&nbsp;&nbsp; 3.150000&nbsp;&nbsp;&nbsp; 0.000000<br>
&nbsp;3.150000&nbsp;&nbsp;&nbsp; 0.000000&nbsp;&nbsp;&nbsp; 0.000000<br>
&amp;END<br>
<br>
&amp;DFT<br>
&nbsp;LDA CORRELATION PZ<br>
&nbsp;FUNCTIONAL BP<br>
&nbsp;GC-CUTOFF<br>
&nbsp;5.D-5<br>
&amp;END<br>
<br>
<br>
<br>
<a class="moz-txt-link-abbreviated" href="mailto:Ari.P.Seitsonen@iki.fi">Ari.P.Seitsonen@iki.fi</a> wrote:<br>
<blockquote type="cite"
 cite="mid200305111432.h4BEWCE27921@magadino.cscs.ch">
  <pre wrap="">Dear Veronique,

  Some suggestions:

 - The system is too small for Gamma point only, it might be that the
   ionic configuration is unstable; you could test this by giving the
   ions a small displacement from the equilibrium positions (by the
   way, for such crystals I find the keyword 'SCALED' very useful) and
   then relax the geometry and see, if the atoms return to the lattice
   positions. The natural solution, if the structure is indeed
   unstable, is to use a bigger cell (2x2x2, i.e. 32 atoms of each
   species)

 - You don't specify the electronic mass, the default might be
   unsuitable for this system (also the electrons heat up slowly in
   your simulation)

  Please tell if this doesn't help, then we can try something else.

    Greetings,

       apsi

PS I guess that you use the option 'MOVIE SAMPLE \n 1' only for the
  debugging, right?

  </pre>
  <blockquote type="cite">
    <pre wrap="">I have some problems simulating the KCL lattice at elevated temperature. 
 First I did an optimizations on the lattice.  Next I tried to heat up 
the crystal with the inputfile as given below.  Normally I would expect 
only small oscillations of the atoms around their equilibrium positions, 
due to the low temperature.  Strangely the atoms undergo very large 
variations and I get a totally distorted system.  The temperature in the 
ENERGIES file raises also drastically.  Can somebody tell me what I do 
wrong?  The energies file is given in attachment.  
When I do a MD simulations microcanonically, everything goes fine and 
the system stays in the minimuum.

Thanks in advance

veronique


&amp;CPMD
 RESTART LATEST WAVEFUNCTIONS
 MOLECULAR DYNAMICS CP
 TIMESTEP
  5
 MAXSTEP
  2000
 PRINT
  10
 TEMPERATURE
  5
 QUENCH BO
 STRUCTURE BONDS ANGLES
 MOVIE SAMPLE
  1
&amp;END

&amp;SYSTEM
 ANGSTROM
 SYMMETRY
 1
 CELL
 6.3 1.0 1.0 0.0 0.0 0.0
 CUTOFF
 25.
&amp;END

&amp;ATOMS
*Cl_VDB_U BINARY NEWF
 LMAX=P
 4
 0.000000    0.000000    0.000000
 3.150000    0.000000    3.150000
 0.000000    3.150000    3.150000
 3.150000    3.150000    0.000000
*019-K-ca-sp-vgrp.uspp BINARY NEWF
 LMAX=P
 4
 3.150000    3.150000    3.150000
 0.000000    0.000000    3.150000
 0.000000    3.150000    0.000000
 3.150000    0.000000    0.000000
&amp;END

&amp;DFT
 LDA CORRELATION PZ
 FUNCTIONAL BP
 GC-CUTOFF
 5.D-5
&amp;END
    </pre>
  </blockquote>
  <pre wrap=""><!---->
  </pre>
</blockquote>
<br>
<pre class="moz-signature" cols="$mailwrapcol">-- 
-----------------------------------------------------------------------
Dr. ir. Van Speybroeck Veronique
Laboratorium voor Theoretische Fysica
Universiteit Gent
Proeftuinstraat 86
9000 Gent
Tel +32-9-264.65.58  GSM : +32/474/259767
Fax +32-9-264.65.60
email : <a class="moz-txt-link-abbreviated" href="mailto:veronique.vanspeybroeck@rug.ac.be">veronique.vanspeybroeck@rug.ac.be</a>
<a class="moz-txt-link-freetext" href="http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html">http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html</a>
-----------------------------------------------------------------------</pre>
<br>
</body>
</html>