<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <p>Hello Lachlan,</p>
    <p><br>
    </p>
    <p>the automated k-point grid correctly generates points without
      inversion folding in this case, so something like</p>
    <p><br>
    </p>
    <p>    KPointsAndWeights = SupercellFolding {<br>
              6 0 0<br>
              0 6 0<br>
              0 0 6<br>
              0.5 0.5 0.5<br>
          }<br>
    </p>
    <br>
    is fine.<br>
    <br>
    Regards<br>
    <br>
    Ben<br>
    <br>
    <br>
    <div class="moz-cite-prefix">On 08/05/17 15:15, Lachlan Oberg wrote:<br>
    </div>
    <blockquote
cite="mid:MEXPR01MB08390BA6D8B2FD7F10F6065BA6EE0@MEXPR01MB0839.ausprd01.prod.outlook.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      <meta name="Generator" content="Microsoft Exchange Server">
      <!-- converted from text -->
      <style><!-- .EmailQuote { margin-left: 1pt; padding-left: 4pt; border-left: #800000 2px solid; } --></style>
      <meta content="text/html; charset=UTF-8">
      <style type="text/css" style="">
<!--
p
        {margin-top:0;
        margin-bottom:0}
-->
</style>
      <div dir="ltr">
        <div id="x_divtagdefaultwrapper" dir="ltr"
          style="font-size:12pt; color:#000000;
          font-family:Calibri,Arial,Helvetica,sans-serif">
          <p>Hello <font size="2"><span style="font-size:12pt">Bálint,</span></font></p>
          <p><font size="2"><span style="font-size:12pt"><br>
              </span></font></p>
          <font size="2"><span style="font-size:12pt">Thanks for</span></font>
          your quick reply, your solution works! Could I also ask a
          question regarding a paragraph in the SpinOrbit section of the
          manual (section 2.4.3, page 29):<br>
          <p><br>
          </p>
          <p>"For periodic systems, use of this keyword automatically
            prevents the folding by<br>
          </p>
          <div>inversion normally used in SupercellFolding{}, but
            manually specified KPointsAndWeights should<br>
            not be reduced by inversion."<br>
            <br>
            I took this to mean that manual k-point sampling should be
            used in spin orbit calculations, and this is the reason I
            used 2x2x2 sampling as it was quick to write out. Is this
            not a correct interpretation of that paragraph?<br>
            <br>
            Kind regards<br>
            Lachlan<br>
          </div>
        </div>
        <hr tabindex="-1" style="display:inline-block; width:98%">
        <div id="x_divRplyFwdMsg" dir="ltr"><font style="font-size:11pt"
            color="#000000" face="Calibri, sans-serif"><b>From:</b>
            DFTB-Plus-User
            <a class="moz-txt-link-rfc2396E" href="mailto:dftb-plus-user-bounces@mailman.zfn.uni-bremen.de"><dftb-plus-user-bounces@mailman.zfn.uni-bremen.de></a> on
            behalf of Bálint Aradi <a class="moz-txt-link-rfc2396E" href="mailto:aradi@uni-bremen.de"><aradi@uni-bremen.de></a><br>
            <b>Sent:</b> Monday, May 8, 2017 6:58:13 PM<br>
            <b>To:</b> User list for DFTB+ related questions<br>
            <b>Subject:</b> Re: [DFTB-Plus-User] Error: Failure in
            diagonalisation routine ZHEEVR, unable to complete Cholesky
            factorization of B 46</font>
          <div> </div>
        </div>
      </div>
      <font size="2"><span style="font-size:10pt;">
          <div class="PlainText">Dear Lachlan,<br>
            <br>
            > Regardless of the spin coupling constants I enter (with
            the exception of<br>
            > non-zero values for the s orbital), I obtain the
            following error after<br>
            > 10 geometry steps:<br>
            > <br>
            > -> Failure in diagonalisation routine ZHEEVR, unable
            to complete<br>
            > Cholesky factorization of B     46<br>
            > <br>
            > Any ideas on what I can do to fix this? For reference,
            I am using the<br>
            > si-band parameter set.<br>
            <br>
            Note, the siband parameter set is not meant for geomertry
            optimisation<br>
            at all. It only contains the electronic parameterisation and
            a dummy<br>
            spline. So, if you do geometry optimisation, it definitely
            ends up in a<br>
            completely unphysical geometry, where the overlap becomes
            non-positive<br>
            definite and therefore the diagonalization fails.<br>
            <br>
            By the way, a side note: You should use a better k-point set
            as 2x2x2<br>
            for the primitive cell. Actually, if you k-point sampling is
            decent<br>
            enough, you should see no forces on the atom, as those
            cancel out in a<br>
            cubic structure...<br>
            <br>
            > Changing to the pbc parameter set, geometry convergence
            DOES occur.<br>
            > However, after sampling along the k-lines L - G - X and
            using dp_bands,<br>
            > I obtain the error:<br>
            > <br>
            >   File<br>
            >
            "/apps/dftbp/1.3.1/lib/python2.7/site-packages/dptools/bandout.py",
            line<br>
            > 69, in fromfile<br>
            >     nkpt = len(eigvalarrays) / nspin<br>
            > ZeroDivisionError: integer division or modulo by zero<br>
            > <br>
            > I assume this means I have to set nspin /= 0 somewhere,
            but I cannot<br>
            > figure out how to do so.<br>
            <br>
            This is a bug, I am sorry for it. The dp_tools package is
            apparently not<br>
            aware about the non-colinear output format. While we are
            fixing it, you<br>
            can convert the band.out file using the following
            shell-command:<br>
            <br>
            sed 's/KPT[ ]*\([^ ]*\)/KPT \1 SPIN 1/g' band.out >
            band_fixed.out<br>
            <br>
            If you run dp_tools on band_fixed.out, it should work.<br>
            <br>
              Best regards,<br>
            <br>
              Bálint<br>
            <br>
            -- <br>
            Dr. Bálint Aradi<br>
            Bremen Center for Computational Materials Science,
            University of Bremen<br>
            <a moz-do-not-send="true"
              href="http://www.bccms.uni-bremen.de/cms/people/b-aradi/">http://www.bccms.uni-bremen.de/cms/people/b-aradi/</a><br>
            <br>
            <br>
          </div>
        </span></font>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
DFTB-Plus-User mailing list
<a class="moz-txt-link-abbreviated" href="mailto:DFTB-Plus-User@mailman.zfn.uni-bremen.de">DFTB-Plus-User@mailman.zfn.uni-bremen.de</a>
<a class="moz-txt-link-freetext" href="https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user">https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user</a></pre>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
      Dr. B. Hourahine, SUPA, Department of Physics,
    University of Strathclyde, John Anderson Building,
            107 Rottenrow, Glasgow G4 0NG, UK.
    +44 141 548 2325, <a class="moz-txt-link-abbreviated" href="mailto:benjamin.hourahine@strath.ac.uk">benjamin.hourahine@strath.ac.uk</a>

2013/4 THE Awards Entrepreneurial University of the Year
      2012/13 THE Awards UK University of the Year

   The University of Strathclyde is a charitable body,
        registered in Scotland, number SC015263</pre>
  </body>
</html>