<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p><font face="Times New Roman">Not an expert on the subject, but below
are my comments.<br>
</font></p>
<p><font face="Times New Roman">For </font><font face="Times New
Roman"><i>I</i>*DOS(<i>E</i>_F), whether you use</font></p>
<p><font face="Times New Roman"><font face="Times New Roman"><i>I</i></font>
in eV and </font><font face="Times New Roman"><font face="Times
New Roman">DOS(<i>E</i>_F)</font> in 1/eV [1,2]</font><br>
</p>
or<br>
<br>
<font face="Times New Roman"><font face="Times New Roman"><font
face="Times New Roman"><i>I</i></font></font> in </font>(eV*spin*f.u.)/states
and <font face="Times New Roman">DOS(<i>E</i>_F)</font><font
face="Times New Roman"> in </font>states/(eV*spin*f.u.) [<span
class="doi-field"></span>3]<br>
<br>
or<br>
<br>
another unit variant such as<font face="Times New Roman"> </font>with
energy in Ry [<font face="Times New Roman">4</font>]. It does not
look like it matters as long as <font face="Times New Roman"><i>I</i>*DOS(<i>E</i>_F)
</font>comes out unitless for the dimensional analysis and has
appropriate values/constants.<br>
<br>
Note: For natural units (atomic units of WIEN2k [5]), while they
have the advantage of simplicity, their can sometimes be a loss of
information for dimensional analysis [6]. Thus, you may want to use
SI units [7].<br>
<br>
For <i><i>U</i>*</i>D<i>(E_</i>F<i>) > </i>(1/v) with D<i>(E_</i>F<i>)</i>
in atom/volume/energy and perhaps U in 1/(atom/energy), if you
multiple both sides of the equation by v:<br>
<br>
<i><i>U</i>*</i>D<i>(E_</i>F<i>)*v > </i>(1/v)*v<br>
<br>
<i><i>U</i>*</i>D<i>(E_</i>F<i>)*v > </i>1<br>
<br>
then<br>
<br>
<i><i>U</i>*</i>D<i>(E_</i>F<i>) > </i>1<br>
<br>
with D<i>(E_</i>F<i>)</i> now in atom/energy. <br>
<br>
[1] <font face="Times New Roman"><a class="moz-txt-link-freetext" href="https://hal.archives-ouvertes.fr/hal-00785221">https://hal.archives-ouvertes.fr/hal-00785221</a>
, <a class="moz-txt-link-freetext" href="https://doi.org/10.1016/j.solidstatesciences.2012.12.003">https://doi.org/10.1016/j.solidstatesciences.2012.12.003</a><br>
[2] </font><font face="Times New Roman"><font face="Times New
Roman"><span class="doi-field"><font face="Times New Roman"><font
face="Times New Roman"><a class="moz-txt-link-freetext" href="http://susi.theochem.tuwien.ac.at/reg_user/textbooks/WIEN2k_lecture-notes_2011/Schwarz_magnetism.pdf">http://susi.theochem.tuwien.ac.at/reg_user/textbooks/WIEN2k_lecture-notes_2011/Schwarz_magnetism.pdf</a>
(slide 7) ,
<a class="moz-txt-link-freetext" href="https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05254.html">https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05254.html</a><br>
</font></font></span></font>[3] </font><font face="Times
New Roman"><a class="moz-txt-link-freetext" href="http://yclept.ucdavis.edu/Publ/tiau-prb-20april.pdf">http://yclept.ucdavis.edu/Publ/tiau-prb-20april.pdf</a> , <span
class="doi-field"><a class="moz-txt-link-freetext" href="https://doi.org/10.1103/PhysRevB.95.205124">https://doi.org/10.1103/PhysRevB.95.205124</a></span></font><font
face="Times New Roman"><span class="doi-field"><font face="Times
New Roman"><font face="Times New Roman"><br>
[4] </font></font></span></font><font face="Times New
Roman"><span class="doi-field"><font face="Times New Roman"><font
face="Times New Roman"><font face="Times New Roman"><a class="moz-txt-link-freetext" href="https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05127.html">https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05127.html</a><br>
[5]
<a class="moz-txt-link-freetext" href="https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15660.html">https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15660.html</a><br>
[6] <a class="moz-txt-link-freetext" href="https://en.wikipedia.org/wiki/Natural_units">https://en.wikipedia.org/wiki/Natural_units</a><br>
[7] <a class="moz-txt-link-freetext" href="https://en.wikipedia.org/wiki/SI_base_unit">https://en.wikipedia.org/wiki/SI_base_unit</a> <br>
[8] No Stoner parameter in WIEN2k output file:
<a class="moz-txt-link-freetext" href="http://zeus.theochem.tuwien.ac.at/pipermail/wien/2014-April/020942.html">http://zeus.theochem.tuwien.ac.at/pipermail/wien/2014-April/020942.html</a><br>
<br>
</font></font></font></span></font>
<div class="moz-cite-prefix">On 12/7/2017 7:51 PM, Alex wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAEDv6u5vAk9VSUiFRRtAsa7-ZOB-onO0ZW2zy+U4hwT3qxudZw@mail.gmail.com">
<div dir="ltr">
<div>
<div>Dear all,</div>
<div><br>
</div>
<div>I was wondering that what the correct units of Stoner
parameter (<i>I</i>) and DOS(<i>E</i>_F) (density of state
at Fermi level) used in the Stoner criteria of magnetism (<i>I</i>*DOS(<i>E</i>_F)
> 1) is?</div>
<div>Does DOS(<i>E</i>_F) should have a unit of (atom/energy)
or (atom/volume/energy) ... and then similar change for the
unit of <i>I</i>? <br>
</div>
<div>Different defined units for them in different literatures
cause confusion (for me), or for example when Coulomb
repulsion (<i>U</i>) is used instead of <i>I,</i> the
Stoner criteria change to <i><i>U</i>*</i>D<i>(E_</i>F<i>)
> </i>(1/v) where v is the unit cell volume<i>.</i></div>
<div>Or for instance inspection of the formulas of 14--27 of
below reference [1] convinces me that the unit of volume is
missing somewhere.</div>
<div><i><br>
</i></div>
<div>[1]. <a
href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.92.9357&rep=rep1&type=pdf"
moz-do-not-send="true">http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.92.9357&rep=rep1&type=pdf</a><i><br>
</i></div>
<div><i><br>
</i></div>
<div>In my specific case I consider an simple FCC Ni, and then
in non-magnetic phase I calculated the DOS(<i>E</i>_F) for
one of the Ni atoms such that the integration of the
corresponding DOS curve until the Fermi level gives around
10 electron.<i><br>
</i></div>
<div><i><br>
</i></div>
<div>Thanks for you help.</div>
<div><br>
</div>
<div>Regards,</div>
<div>Alex<br>
</div>
</div>
</div>
</blockquote>
</body>
</html>