Dear Dr. Cococcioni,<div><br></div><div>Thanks a lot for the reply. It is very clear. I appreciate it. <br><br><div class="gmail_quote">On Sun, Dec 2, 2012 at 12:25 PM, Matteo Cococcioni <span dir="ltr"><<a href="mailto:matteo@umn.edu" target="_blank">matteo@umn.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear Peng,<br>
<br>
I'm not entirely sure about what you mean by "site symmetry".<br>
Anyway, if two atomic positions are not equivalent by (point group)<br>
symmetry, then the U will likely be different and should be recomputed<br>
for each of them. This is the case of Fayalite.<br>
Some times you can avoid recomputing the U if the only difference<br>
between the atomic sites is the sign of magnetization. This is the<br>
case of FeO or other transition-metal oxides with the same<br>
stoiciometry. In these systems you have two TM atoms in the unit cell<br>
one with spin up the other with spin down (but the same absolute<br>
magnetization: in fact the material is AF).<br>
If you are not sure about what to do, a good way to decide whether you<br>
need to recompute U or not is to look at the atomic occupations<br>
printed by the code (which are, in fact symmetrized). If two atoms<br>
have different occupations (the difference has to be larger than<br>
possible numerical noise, usually affecting occupations at the 4th<br>
decimal point) and their occupations cannot be obtained one from the<br>
other just interchanging majority with minority spins, then you need<br>
to recompute U.<br>
<br>
Regards,<br>
<br>
Matteo<br>
<div><div class="h5"><br>
On Sun, Dec 2, 2012 at 11:02 AM, Peng Chen <<a href="mailto:pchen@ion.chem.utk.edu">pchen@ion.chem.utk.edu</a>> wrote:<br>
> Dear All,<br>
><br>
> In the calculations of U, I remember that I got some advice from the forum<br>
> that the equivalence of the ions is determined by space group symmetry not<br>
> site symmetry. E.g AB2O3, B ions has 2 different site symmetries (B(1)<br>
> B(2)). We should only run the calculation on perturbing B ions. But from<br>
> Dr. Matteo Cococcioni's paper (PRB 71, 035105 (2005)), both Fe1 and Fe2 of<br>
> fayalite are perturbed and have different U values. I think Fe1 and Fe2 are<br>
> different in site symmetry. So I am just curious which is the best way to<br>
> do the same kind of calculations.<br>
><br>
> --<br>
> Best Regards.<br>
> Peng<br>
><br>
</div></div>> _______________________________________________<br>
> Pw_forum mailing list<br>
> <a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
> <a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
<span class="HOEnZb"><font color="#888888"><br>
<br>
<br>
--<br>
Matteo Cococcioni<br>
Department of Chemical Engineering and Materials Science,<br>
University of Minnesota<br>
421 Washington Av. SE<br>
Minneapolis, MN 55455<br>
Tel. <a href="tel:%2B1%20612%20624%209056" value="+16126249056">+1 612 624 9056</a> Fax <a href="tel:%2B1%20612%20626%207246" value="+16126267246">+1 612 626 7246</a><br>
_______________________________________________<br>
Pw_forum mailing list<br>
<a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
<a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br> Best Regards.<br> Peng <br>
</div>