Page 1 of 1

Non-collinear DOS

Posted: Mon Dec 08, 2008 7:06 am
by duncan2
I conducted a noncollinear calculation to see some magnetic properties depending on the direction of the magnetic moments. Conditions are as follows:

LNONCOLLINEAR = T
ISPIN = 2
MAGMOM properly set.
LORBIT = default (not set in INCAR)

And I got corresponding DOSCAR file, where three strange point found I.

1. DOSCAR contains only one DOS and IDOS in the first paragraph. It seems not like ISPIN=2 case, which shows two DOS and two IDOS lines.

2. 12 colums in atomic DOS (you know, the next paragrah followed by the first paragrah in DOSCAR. I hope you know what I'm saying...) I don't know what these columns stand for.
(eg. sx, sy, sz, px, py, pz, dx, dy, dz, tx, ty, tz)

3. I checked some columns has both positive and negative DOS value. I want to know what it means.

Thanks~
<span class='smallblacktext'>[ Edited ]</span>

Non-collinear DOS

Posted: Wed Jan 21, 2009 3:25 pm
by admin
for non-collinear systems, ISPIN=2 is not appropriate because there is not just spin up and down.
Thererfore the first part of DOSCAR only contains the total and the integrated DOS.
in the second part, you will find 4 components for each s, py,.... : these are the (s-like,...) total dos and mx, my, mz (m...magnetization)

Non-collinear DOS

Posted: Wed Jan 21, 2009 3:48 pm
by admin
for non-collinear systems, ISPIN=2 is not appropriate because there is not just spin up and down.
Thererfore the first part of DOSCAR only contains the total and the integrated DOS.
in the second part, you will find 4 components for each s, py,.... : these are the (s-like,...) total dos and mx, my, mz (m...magnetization)