<div>first thank your for your help...... </div>
<div> </div>
<div>the relaxed structure  from vasp at 500kba is </div>
<div>Ba                                   <br>   3.12000000000000     <br>     1.0202516248839648    0.0000000000000000    0.0000000000000000<br>    -0.5101258124419824    0.8835632605198342    0.0000000000000000<br>     0.0000000000000000    0.0000000000000000    1.6154107754936728<br>
   2<br>Direct<br>  0.3333333300000021  0.6666659999999993  0.2500000000000000<br>  0.6666666599999971  0.3333330000000032  0.7500000000000000</div>
<div>  external pressure =        1.67 kB  Pullay stress =      500.00 kB<br><br> pwscf input is celldm(1)=6.015350,celldm(3)=1.58, <br><font color="#cc0033">Bohr</font>=5.2917706   angstrom</div>
<div>I have changed the input file like this after accepting your suggestions</div>
<div>&amp;control<br>    calculation=&#39;vc-relax&#39;,<br>    restart_mode=&#39;from_scratch&#39;,<br>    dt=80,<br>    prefix=&#39;ba&#39;,<br>    pseudo_dir = &#39;/home/solid/Work/PW/pseudo/&#39;,<br>    outdir=&#39;/home/solid/Work/PW/temp/&#39;<br>
    tstress=.t.,<br>    tprnfor=.t.<br>    nstep= 100<br>/<br> &amp;system<br>    ibrav= 4,<br>    celldm(1)=6.0,<br>    celldm(3)=1.58,<br>    nat= 2,<br>    ntyp=1,<br>    nbnd=30<br>    ecutwfc=30.0,ecutrho = 300<br>    nosym = .f. ,<br>
    occupations = &#39;smearing&#39; ,<br>    degauss = 0.014D0 ,<br>    smearing = &#39;methfessel-paxton&#39; ,<br>/<br> &amp;electrons<br>    mixing_beta = 0.7<br>    conv_thr =  1.0d-8<br>/<br> &amp;IONS<br>    ion_dynamics = &#39;bfgs&#39; ,<br>
/<br> &amp;CELL<br>    cell_dynamics = &#39;bfgs&#39; ,<br>    press_conv_thr = 0.1 ,<br>    press = 500<br>/<br>ATOMIC_SPECIES<br>Ba  137.327 Ba.pbe-nsp-van.UPF <br>ATOMIC_POSITIONS {crystal}<br>Ba 0.3333333333333286 0.6666666666666714 0.2500000000000000<br>
Ba 0.6666666666666714 0.3333333333333286 0.7500000000000000<br>K_POINTS {automatic}<br>14 14 8  0 0 0<br>then the error still is </div>
<div>task #         0<br>     from scale_h : error #         1<br>     Not enough space allocated for radial FFT: try restarting with a larger cell_factor.<br>if I add the cell_factor =3.0 to the &amp;CELL, then the error became</div>

<div> </div>
<div> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>     task #         2<br>     from electrons : error #         1<br>     charge is wrong<br> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>
<br>the relaxed pressure changes unexpectly as following </div>
<div>  grep kba *<br>ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=  356.22<br>ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=  420.98<br>ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=  105.79<br>
ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=-1505.88<br>ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P= 9429.67<br>ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=-1476.75<br>
ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=-1702.03<br>ba.optim.out_500:          total   stress  (Ry/bohr**3)                   (kbar)     P=-2923.68<br><br></div>
<div class="gmail_quote">2010/3/3 Lorenzo Paulatto <span dir="ltr">&lt;<a href="mailto:paulatto@sissa.it">paulatto@sissa.it</a>&gt;</span><br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div class="im">On Wed, 03 Mar 2010 12:30:32 +0100, Wei Zhou &lt;<a href="mailto:zdw2000@gmail.com">zdw2000@gmail.com</a>&gt; wrote:<br>&gt;&gt; the initial structure is get from the VASP soft,and is consist with the<br>
&gt;&gt; experiment at 500kba,<br><br></div>Dear Wei,<br>I think you may have done a unit of measure mistake while converting the<br>system parameters from VASP to QE. In order to get the huge pressure you<br>are requiring the cell parameter has to shrink by a factor ~ 0.57 (i.e not<br>
too far away from the bohr/Angstrom conversion). Furthermore, you smearing<br>degauss is probably too high (remember it is expressed in Rydberg) and you<br>cutoff energy is probably unnecessarily high (the pseudopotential you are<br>
using is well converged at much lower ecutwfc and ecutrho).<br>
<div class="im"><br>&gt; I wonder if the two Ba pseudo potential of pwscf web cannot  work at<br>&gt; high pressre..........<br><br></div>It is possible: when the atoms get very close to each other you need<br>harder pseudopotentials; however that&#39;s not the cause of your problem, I<br>
would bet on the smearing instead.<br><br>regards<br><br>--<br>Lorenzo Paulatto<br><br>*** Note: my affiliation has changed! please send future<br>correspondence to: &lt;<a href="mailto:Lorenzo.Paulatto@impmc.upmc.fr">Lorenzo.Paulatto@impmc.upmc.fr</a>&gt; ***<br>
<br>post-doc @ IMPMC/UPMC - Université Paris 6<br>phone: +33 (0)1 44 27 74 89<br>www:   <a href="http://www-int.impmc.upmc.fr/~paulatto/" target="_blank">http://www-int.impmc.upmc.fr/~paulatto/</a><br><br>previously:<br>phd student @ SISSA  &amp;  DEMOCRITOS (Trieste)<br>
phone: +39 040 3787 511<br>www:   <a href="http://people.sissa.it/~paulatto/" target="_blank">http://people.sissa.it/~paulatto/</a><br>
<div>
<div></div>
<div class="h5">_______________________________________________<br>Pw_forum mailing list<br><a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br><a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank">http://www.democritos.it/mailman/listinfo/pw_forum</a><br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>ZhouDawei<br>JiLin Universiyt ,ChangChun ,China<br><a href="mailto:zdw2000@gmail.com">zdw2000@gmail.com</a><br>