forum@abinit.org
Subject: The ABINIT Users Mailing List ( CLOSED )
List archive
- From: Gerd Steinle-Neumann <g.steinle-neumann@uni-bayreuth.de>
- To: forum@abinit.org
- Cc: vojtech.vlcek@uni-bayreuth.de
- Subject: [abinit-forum] linear response for conventional/super cell
- Date: Wed, 03 Mar 2010 20:34:42 +0100
Dear all --
We are trying to do linear response calculations for the conventional unit cell of MgO, and encounter a sunstantial problem on the way (probably more to come).
The goal is to move to a supercell built on the conventional cell and model the influence of impurities on the phonon spectrum. Obviously we need the pure MgO end-member to have a comparison to (a) the phonon spectrum for MgO from the primitive cell and (b) a baseline for the impurity study.
The in-file is given below.
The error we get is for the ddk computation with the following problem:
occeig : ERROR -Initially we had a similar problem with the GS calculations, and were able to trick the code to do what we wanted by using the brvlatt 3 option (see below). It appears that the brvlatt does not apply to the ddk calculation. Any idea on either (a) how we can get the code to do we want to do here, or (b) a completely different approach on doing this for the conventional cell.
In a non-metallic case (occopt<3), for a RF calculation,
if the eigenvalues are degenerate, the occupation numbers must also be degener
ate.
However, the following pair of states gave :
k -state, band number 16, occ= 2.000000E+00, eigenvalue= 3.432344E-01,
kq-state, band number 17, occ= 0.000000E+00, eigenvalue= 3.432344E-01.
Action : change occopt, consistently, in GS and RF calculations.
leave_new : decision taken to exit ...
Thanks, Vojtech + Gerd.
*.in-file
ndtset 2
#GS-SC
getwfk1 0
kptopt1 2
tolvrs1 1.0d-18 # from tutorial
rfphon1 0
nqpt1 0
# 2 MgO calculation - d/dk calculation
getwfk2 1
iscf2 -3
kptopt2 2
tolwfr2 1.0d-22
tolvrs2 0
rfelfd2 2
rfphon2 0
rfatpol2 1 8
rfdir2 1 1 1
#qpts
nqpt 1
qpt2 0.00000000E+00 0.00000000E+00 0.00000000E+00
acell 3*7.89862
brvltt 3
chkprim 0
natom 8
ntypat 2
znucl 8 12
typat 1 1 1 1 2 2 2 2
nband 20 # oxygen = 6; magnesium = 2
rprim 1.0 0.0 0.0 # primitive
0.0 1.0 0.0
0.0 0.0 1.0
xred 0.5 0.5 0.5 0.5 0.5 0.0
0.5 0.0 0.5
0.0 0.5 0.5
0.0 0.0 0.0
0.5 0.0 0.0
0.0 0.5 0.0
0.0 0.0 0.5
ngkpt 4 4 4
nshiftk 1 # recommended in abinit tutorial
shiftk 0.5 0.5 0.5 # recommended in abinit tutorial
ecut 40
#scf
nstep 25
- [abinit-forum] linear response for conventional/super cell, Gerd Steinle-Neumann, 03/03/2010
- Re: [abinit-forum] linear response for conventional/super cell, Josef Zwanziger, 03/03/2010
- Re: [abinit-forum] linear response for conventional/super cell, Xavier Gonze, 03/03/2010
Archive powered by MHonArc 2.6.16.