[Pw_forum] question about bands calculation
lukethulin at netscape.net
lukethulin at netscape.net
Mon Oct 23 17:12:20 CEST 2006
hai-ping,
I'm not sure, your input file looks pretty good to me. You can try this path that I've used in past.
K_POINTS tpiba
23
0.000000000 0.000000000 0.400000000 1.000000000
0.000000000 0.000000000 0.350000000 1.000000000
0.000000000 0.000000000 0.300000000 1.000000000
0.000000000 0.000000000 0.250000000 1.000000000
0.000000000 0.000000000 0.200000000 1.000000000
0.000000000 0.000000000 0.150000000 1.000000000
0.000000000 0.000000000 0.100000000 1.000000000
0.000000000 0.000000000 0.050000000 1.000000000
0.000000000 0.000000000 0.000000000 1.000000000
0.050000000 0.050000000 0.000000000 1.000000000
0.100000000 0.100000000 0.000000000 1.000000000
0.150000000 0.150000000 0.000000000 1.000000000
0.200000000 0.200000000 0.000000000 1.000000000
0.250000000 0.250000000 0.000000000 1.000000000
0.300000000 0.300000000 0.000000000 1.000000000
0.350000000 0.350000000 0.000000000 1.000000000
0.400000000 0.400000000 0.000000000 1.000000000
0.450000000 0.450000000 0.000000000 1.000000000
0.500000000 0.500000000 0.000000000 1.000000000
0.500000000 0.500000000 0.050000000 1.000000000
0.500000000 0.500000000 0.100000000 1.000000000
0.500000000 0.500000000 0.150000000 1.000000000
0.500000000 0.500000000 0.200000000 1.000000000
My bands output file gave:
Highest occupied, lowest unoccupied level (ev): 7.2352 9.3408
Luke
----- Original Message -----
From: lan haiping
To: pw_forum at pwscf.org
Sent: Monday, October 23, 2006 10:14 AM
Subject: Re: [Pw_forum] question about bands calculation
Dear Luke,
thanks for you pacient explaination and suggest.
i recalculated this anatase tio2 in succession again with input below, and found again there were no output about highest occupied and lowest unoccupied level in the 'bands' output. i just think it is so strange to understand, though i could calculate correctly-like bands structure.
kpoints for 'scf' is using mp grid 8x8x8 without offsets.and kpoint path for 'bands' calculations is \Gamma[0,0,0] to X [0,0,0.5], from x[0,0,0.5] to [0.0.5,0.5], from [0.0, 0.5,0.5] to [0.5,0.5,0.5] ,from[0.5,0.5 ,0.5] to[0.0,0.0 ,0.5]
Do you think any fault related to my kpoint path consideration,or any other parameters in
the input file?
&CONTROL
title = 'TiO2' ,
calculation = 'bands',
verbosity = 'default',
tprnfor = .true.,
outdir = '/home/haiping/tmp/',
prefix = 'tio2',
disk_io = 'default',
nstep = 200,
pseudo_dir = '/home/haiping/espresso-3.0/pseudo/',
etot_conv_thr= 1.0D-4
forc_conv_thr= 1.0D-4
/
&SYSTEM
ibrav= 7,
celldm(1)=7.1526162323759346,celldm(3)=2.5136063408190221,
nat = 6 ,
ntyp = 2 ,
nbnd = 30 ,
ecutwfc = 30.0000000000,
ecutrho = 240.00
occupations = 'fixed' ,
/
&ELECTRONS
electron_maxstep = 100,
conv_thr = 1.0e-12,
mixing_mode = 'plain' ,
mixing_beta = 0.300000000,
mixing_ndim = 8,
diagonalization = 'cg' ,
/
ATOMIC_SPECIES
O 15.999 O.pbe-rrkjus.UPF
Ti 47.867 Ti.pbe-sp-van_ak.UPF
ATOMIC_POSITIONS angstrom
Ti 0.000000000 -0.946250000 -1.189250000
Ti 0.000000000 0.946250000 1.189250000
O 0.000000000 0.946250000 -0.796426921
O 0.000000000 -0.946250000 0.796426921
O 1.892500000 0.946250000 1.582073079
O -1.892500000 -0.946250000 -1.582073079
K_POINTS crystal
41
0.0 0.0 0.0 1
0.0 0.0 0.1 1
0.0 0.0 0.2 1
0.0 0.0 0.3 1
0.0 0.0 0.4 1
0.0 0.0 0.5 1
0.0 0.05 0.5 1
0.0 0.10 0.5 1
0.0 0.15 0.5 1
0.0 0.20 0.5 1
0.0 0.25 0.5 1
0.0 0.30 0.5 1
0.0 0.35 0.5 1
0.0 0.40 0.5 1
0.0 0.45 0.5 1
0.0 0.50 0.5 1
0.05 0.50 0.5 1
0.10 0.50 0.5 1
0.15 0.50 0.5 1
0.20 0.50 0.50 1
0.25 0.50 0.50 1
0.30 0.50 0.50 1
0.35 0.50 0.50 1
0.40 0.50 0.50 1
0.45 0.50 0.50 1
0.50 0.50 0.50 1
0.467 0.467 0.50 1
0.433 0.433 0.50 1
0.400 0.400 0.50 1
0.367 0.367 0.50 1
0.333 0.333 0.50 1
0.30 0.30 0.50 1
0.267 0.267 0.50 1
0.233 0.233 0.50 1
0.20 0.20 0.50 1
0.167 0.167 0.50 1
0.133 0.133 0.50 1
0.100 0.100 0.50 1
0.067 0.067 0.50 1
0.033 0.033 0.50 1
0.00 0.00 0.50 1
Yours,
hai-ping
On 10/23/06, lukethulin at netscape.net <lukethulin at netscape.net> wrote:
hai-ping,
I've successfully done band structure modeling for anatase, I'll go back and find what I chose for high symmetry lines, or find the reference I used. I guess I just thought it was strange that the nscf calculations were being done separately, when you could just pick several paths and do them all in one calculation.
Luke
----- Original Message -----
From: lan haiping
To: pw_forum at pwscf.org
Sent: Monday, October 23, 2006 12:38 AM
Subject: Re: [Pw_forum] question about bands calculation
Dear Luke,
thanks for your reply. my structure is anatase titania.
Actually, i set the k-point paths according to high symmetry lines such as \Gamma[0,0,0] to Z [0,0,0.5] , Z to Q[0,0.5,0.5] etc.
Is it right ?
Regards,
hai-ping
On 10/22/06, lukethulin at netscape.net < lukethulin at netscape.net> wrote:
Hai-ping,
I don't think you're going about a band structure calculation correctly. The scf calculation should only use automatic k-point selection of (4,4,4) or something like that. The following nscf calculation needs you to input k-paths along high symmetry lines that vary by crystal structure. What form of titania are you modeling, anatase, rutile?
Luke
----- Original Message -----
From: lan haiping
To: pw_forum
Sent: Saturday, October 21, 2006 2:31 PM
Subject: [Pw_forum] question about bands calculation
Hi!
i have a question about bands calculation .
Just after finished a 'scf' calculation for tio2, i performed a series (about 5) 'nscf' calculations with different k-grid paths. Except k-grid path, all settings are the same for these series ' nscf ' calculation.
In some nscf-output, the values of "Highest occupied level and lowest unoccuped level" are found. But these values are not found in the other nscf-output . I donot know what the reasons are related to . Morever, i also came to a calculation failed with complains "Too many eigenvalues are not converged".
Regards,
hai-ping
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20061023/698ccfa0/attachment.html>
More information about the users
mailing list