<p>Hi Dear user,</p>
<p>I have tried to test the k-points and ecutwfc convergence for<br>a Mg(002) slab which contains 7 layers and the middle 5 layers<br>have been fixed.</p>
<p>PROBLEM I:<br>I have tested (1x1x1, 2x2x1, 4x4x1, 5x5x1, 7x7x1, 8x8x1, 10x10x1) M-P<br>k-points grids for one input file, other parameters are the same(see below).</p>
<p>For 2x2x1, 5x5x1 and 8x8x1, I have got<br>&quot;The maximum number of steps has been reached.<br>&nbsp;End of BFGS Geometry Optimization&quot; <br>However, for others, say, (1x1x1, 4x4x1, 7x7x1, 10x10x1)I have got <br>&quot;bfgs converged in xx scf cycles and xx bfgs steps
<br>&nbsp;End of BFGS Geometry Optimization&quot;</p>
<p>For the latter cases, the convergences have completed successfully, <br>but for the former cases, they seemed not finished. I wonder, if just<br>for k-points test, whether it is worth to restart the jobs to make the<br>
optimization converged? Or, choosing new K sampling, such as (3x3x1, <br>6x6x1, 9x9x1, etc.)?</p>
<p>PROBLEM II<br>when testing ecutwfc convergence, I have chosen 10, 14, 18, 22, 26, 30 Ry<br>as the value of the ecutwfc. It works well for ecutwfc=18, 26 and 30. When<br>ecutwfc=14 or 22, the following errors occured:<br>
...<br>&nbsp;%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>&nbsp;&nbsp;&nbsp;&nbsp; from bfgs : error #&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 1<br>&nbsp;&nbsp;&nbsp;&nbsp; bfgs history already reset at previous step<br>&nbsp;%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%</p>
<p>&nbsp;&nbsp;&nbsp;&nbsp; stopping ...</p>
<p>I have checked the mailing list archives, like:<br><a href="http://www.democritos.it/pipermail/pw_forum/2006-November/005331.html">http://www.democritos.it/pipermail/pw_forum/2006-November/005331.html</a><br>suggested that restarting the jobs to make the geometry optimization converged.
<br>However, I wonder why the optimizaition is failed for certain lower values of <br>ecutwfc which seems would be more easily to be completed. </p>
<p>In PROBLEM I, ecutwfc=24Ry; In PROBLEM II, K_POINTS are 5 5 1 0 0 0</p>
<p>the following is the input file:<br>&amp;CONTROL<br>calculation&nbsp; = &#39;relax&#39;,<br>prefix&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = &#39;Mgrelax&#39;,<br>pseudo_dir&nbsp;&nbsp; = &#39;/....../pseudo/&#39;,<br>outdir&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = &#39;/....../tmp/&#39;,<br>tstress=.true.
<br>tprnfor=.true.<br>/<br>&amp;SYSTEM<br>ibrav&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 14,<br>a&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 3.149307D0,<br>b&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 3.149307D0,<br>c&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 32.624496,<br>cosab&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 0.5D0,<br>cosac&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 0.D0,<br>cosbc&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 0.D0,<br>nat&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 7,
<br>ntyp&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; = 1,<br>ecutwfc&nbsp;&nbsp;&nbsp;&nbsp; = xx.D0,<br>occupations = &quot;smearing&quot;,<br>smearing&nbsp;&nbsp;&nbsp; = &quot;methfessel-paxton&quot;,<br>degauss&nbsp;&nbsp;&nbsp;&nbsp; = 0.05D0,<br>/<br>&amp;ELECTRONS<br>/<br>&amp;IONS<br>ion_dynamics=&#39;bfgs&#39;,
<br>pot_extrapolation = &quot;second_order&quot;,<br>wfc_extrapolation = &quot;second_order&quot;,<br>/<br>ATOMIC_SPECIES<br>Mg&nbsp; 24.305&nbsp; Mg.pw91-np-van.UPF<br>ATOMIC_POSITIONS {angstrom}<br>Mg&nbsp;&nbsp; 2.099538&nbsp;&nbsp; 2.099538&nbsp;&nbsp; 0.000000
<br>Mg&nbsp;&nbsp; 1.049769&nbsp;&nbsp; 1.049769&nbsp;&nbsp; 2.718708&nbsp;&nbsp; 0 0 0<br>Mg&nbsp;&nbsp; 2.099538&nbsp;&nbsp; 2.099538&nbsp;&nbsp; 5.437416&nbsp;&nbsp; 0 0 0<br>Mg&nbsp;&nbsp; 1.049769&nbsp;&nbsp; 1.049769&nbsp;&nbsp; 8.156124&nbsp;&nbsp; 0 0 0<br>Mg&nbsp;&nbsp; 2.099538&nbsp;&nbsp; 2.099538&nbsp; 10.874832&nbsp;&nbsp; 0 0 0<br>Mg&nbsp;&nbsp; 1.049769&nbsp;&nbsp; 1.049769&nbsp; 13.59354
&nbsp;&nbsp;&nbsp; 0 0 0<br>Mg&nbsp;&nbsp; 2.099538&nbsp;&nbsp; 2.099538&nbsp; 16.312248<br>K_POINTS {automatic}<br>nk1 nk2 nk3 0 0 0<br></p>