# [Pw_forum] absorption spectrum &dielectric constant of turboTDDFT

Timrov Iurii iurii.timrov at epfl.ch
Fri Jul 22 12:45:41 CEST 2016

Dear Sunghwan Choi,

http://www.quantum-espresso.org/forum/)<http://www.quantum-espresso.org/forum/>

1. In order to solve your problem, do not specify the parameter "nbnd" in the input file for PWscf calculation. In this case, the code will compute only occupied states. The turboTDDFPT code (turbo_lanczos.x) needs only occupied states. But if one still includes some empty states in the calculations, the turbo_lanczos.x code still must work, but it does not in 5.4.0 version (and probably in some older versions), because there is a bug. This bug will be fixed soon and it will be committed to the SVN trunk on qe-forge.org. In your case, just do not specify "nbnd" and everything should be fine.

2. Your input for the PWscf code requires careful inspection, because there are several weird things. I comment only about one of them. Namely, you specified just "smearing='fermi-dirac'" probably thinking that your system is metallic. But your system is not metallic (and you can realize this by inspecting the output of the PWscf calculation), because in the PWscf you must also specify occupations = 'smearing', and of course the value of the broadening parameter "degauss = XXX" for the Fermi-Dirac smearing (because in your case you have smearing='fermi-dirac').

3. Even when you solve the problem in step 2, the turboTDDFT code does not currently support metallic systems. If you try to run it you will get an error message "turboTDDFT is not extended to metals".

Thank you for reporting about your problem, because it helped to find a bug.

HTH

Best regards,

Iurii

--
Dr. Iurii Timrov
Postdoctoral Researcher
Swiss Federal Institute of Technology Lausanne (EPFL)
Laboratory of Theory and Simulation of Materials (THEOS)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334

________________________________
From: pw_forum-bounces at pwscf.org <pw_forum-bounces at pwscf.org> on behalf of SungHwan Choi <sunghwanchoi91 at gmail.com>
Sent: Friday, July 22, 2016 7:00 AM
To: PWSCF Forum
Subject: Re: [Pw_forum] absorption spectrum &dielectric constant of turboTDDFT

Dear all,
As a one of newbies, I am happy to see active discussion.
I am now trying to obtain some spectrum of metallic quantum dot.
For a test, I put two gold atoms but I failed to run it

pwscf/Au2> cat input.pw<http://input.pw>
&control
calculation='scf'
pseudo_dir ='./',
outdir='./'
title='scf Au'
prefix='Au'
tprnfor=.true.
disk_io='low', restart_mode='from_scratch'
/
&system
ibrav=1, celldm(1) =30.0, nat=2, ntyp= 1, nbnd=50, smearing='fermi-dirac'
ecutwfc =24.0, nosym=.true.
/
&electrons
mixing_mode='plain'
mixing_beta = 0.5,
electron_maxstep=500,
diago_david_ndim=30,
conv_thr = 1.0d-6
/
ATOMIC_SPECIES
Au 196.97 Au.pbe-theos.UPF
ATOMIC_POSITIONS {bohr}
Au   13.0            15.0            15.0
Au   17.0            15.0            15.0
K_POINTS {gamma}

pwscf/Au2> cat input.lr<http://input.lr>
&lr_input
restart=.false.
restart_step=100
prefix='Au'
outdir='./'
lr_verbosity=1
/
&lr_control
itermax=1000
ipol=4
/

input.pw<http://input.pw> work fine but input.lr<http://input.lr> failed to run. output firstly work fine but at some point it just ended.

Program turboTDDFT v.5.4.0 starts on 22Jul2016 at 13:49:57

This program is part of the open-source Quantum ESPRESSO suite
for quantum simulation of materials; please cite
"P. Giannozzi et al., J. Phys.:Condens. Matter 21 395502 (2009);
URL http://www.quantum-espresso.org",
in publications or presentations arising from this work. More details at
http://www.quantum-espresso.org/quote

Parallel version (MPI), running on    48 processors
R & G space division:  proc/nbgrp/npool/nimage =      48

./Au.save

Info: using nr1, nr2, nr3 values from input

Info: using nr1, nr2, nr3 values from input

IMPORTANT: XC functional enforced from input :
Exchange-correlation      =  SLA  PW   PBX  PBC ( 1  4  3  4 0 0)
Any further DFT definition will be discarded
Please, verify this is what you really want

Parallelization info
--------------------
sticks:   dense  smooth     PW     G-vecs:    dense   smooth      PW
Min         142     142     34                 8926     8926    1110
Max         144     144     36                 8936     8936    1124
Sum        6869    6869   1725               428689   428689   53715
Tot        3435    3435    863

Subspace diagonalization in iterative solution of the eigenvalue problem:
one sub-group per k-point group (pool) will be used
scalapack distributed-memory algorithm (size of sub-group:  4*  4 procs)

Warning: There are virtual states in the input file, trying to disregard in response calculation

Gamma point algorithm

with such a error message
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source

Stack trace terminated abnormally.
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source
turbo_lanczos.x    0000000000AC5C31  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000AC4387  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A6A554  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A6A366  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A00E8F  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A081BD  Unknown               Unknown  Unknown
libc.so.6          00002AAAAF88F0B8  Unknown               Unknown  Unknown
libc.so.6          00002AAAAF8911F7  Unknown               Unknown  Unknown
libmkl_core.so     00002AAAACAC3460  Unknown               Unknown  Unknown

Stack trace terminated abnormally.
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source
turbo_lanczos.x    0000000000AC5C31  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000AC4387  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A6A554  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A6A366  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A00E8F  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A081BD  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A1FC11  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A1D6A1  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000775A72  davcio_                   382  io_files.f90
turbo_lanczos.x    0000000000406BD9  MAIN__                    118  lr_main.f90
turbo_lanczos.x    000000000040698E  Unknown               Unknown  Unknown
libc.so.6          00002AAAAF831C36  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000406899  Unknown               Unknown  Unknown
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source
turbo_lanczos.x    0000000000AC5C31  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000AC4387  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A6A554  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A6A366  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A00E8F  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A081BD  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A1FC11  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000A1D6A1  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000775A72  davcio_                   382  io_files.f90
turbo_lanczos.x    0000000000406BD9  MAIN__                    118  lr_main.f90
turbo_lanczos.x    000000000040698E  Unknown               Unknown  Unknown
libc.so.6          00007FFFCA207C36  Unknown               Unknown  Unknown
turbo_lanczos.x    0000000000406899  Unknown               Unknown  Unknown

Could you help me? please do not hesitate to tell me if you have any clues
best,
Sunghwan Choi

2016-07-22 6:54 GMT+09:00 Manu Hegde <mhegde at uwaterloo.ca<mailto:mhegde at uwaterloo.ca>>:
Hi Dario,
Thanks. My unit cell already big contains 80 atoms, I will try to calculate using supercell.

Regards,
Manu

On Thu, Jul 21, 2016 at 4:02 PM, dario rocca <roccad at gmail.com<mailto:roccad at gmail.com>> wrote:
Hi Manu,
The last release of the code does not allow for k-point sampling, that's why you get an error message.
You would need to use a supercell. If you have done a gamma only calculation for your system that's probably why you get a spectrum
that does not match at all experiments.
Additionally, the Bethe-Salpeter equation is considered the most accurate method for bulk solids.
Best,
Dario

On Thu, Jul 21, 2016 at 9:42 AM, Manu Hegde <mhegde at uwaterloo.ca<mailto:mhegde at uwaterloo.ca>> wrote:
Hi Dario,
Thank you very much. I have done some calculations for UV-Vis (see attached) and it is no where matching my experimental results in case of bcc-In2O3. Is it due to pseudopotential (I have used PBE USPP)?. Also I am unable to BZ sampling on k-mesh and TDDFT shown me an error.

Regards,
Manu
(University of Waterloo)

On Wed, Jul 20, 2016 at 8:09 PM, dario rocca <roccad at gmail.com<mailto:roccad at gmail.com>> wrote:
Dear Manu,
You can simply extract it with the command
\$ grep "S(E)" prefix.plot_chi.dat  > file_for_plot
and use your favorite program to plot it.
Best,
Dario

On Wed, Jul 20, 2016 at 10:49 AM, Manu Hegde <mhegde at uwaterloo.ca<mailto:mhegde at uwaterloo.ca>> wrote:
Hello Dario,

I have done the similar calculations, it went well, after running turbo_spectrum.x  Do you have any code to extract S(E) or I have to do it manually?

Manu
(University of Waterloo)

On Tue, Jul 19, 2016 at 8:24 PM, dario rocca <roccad at gmail.com<mailto:roccad at gmail.com>> wrote:
Dear SungHwan,
http://arxiv.org/pdf/1402.0486.pdf
and
http://urania.sissa.it/xmlui/bitstream/handle/1963/5797/turboTDDFT-CPC.pdf?sequence=1
for more details.

In the output of the postprocessing chi_1_1 is the xx component of the dynamical polarizability, chi_1_2 the xy component, etc.
Often chi is indicated as \alpha in the papers.
Specifically you would have

Component      energy of the perturbation      real part of chi                          imaginary
(indicated either as E or \omega)
chi_1_1=          0.000000000000000E+00     0.955769346791472E+01  0.000000000000000E+00
.........

As you correctly say the absorption spectrum is related to the imaginary part of the average of chi multiplied by E (the energy E is often
indicated as \omega).
This averaged quantity is contained in S(E). Specifically S(E) is the strength function defined as
S(E)=2m/(\pi*e^2*\hbar) * E*Im[chi(E)], where chi has been averaged over the 3 spatial directions.
This definition is convenient because S(E) satisfies the f-sum rule.
So if you extract S(E) and plot it you will have the absorption spectrum.
Let me know if it's still not clear
Best,
Dario

On Tue, Jul 19, 2016 at 4:01 PM, SungHwan Choi <sunghwanchoi91 at gmail.com<mailto:sunghwanchoi91 at gmail.com>> wrote:
hi, I am a newbie for turboTDDFT. Now, I ran a simple example. It was assume. Now I have prefix.plot_chi.dat  file. then I don't know how to generate absorption spectrum from that file and dielectric constant values. As far as I understood, the absorption spectrum is related to imaginary part of average \chi value. but in the file there are 9 values
chi_1_1
chi_2_1
chi_3_1
chi_1_2
chi_2_2
chi_3_2
chi_1_3
chi_2_3
chi_3_3
This is 3*3*3 tensor. how do I get an absorption spectrum and dieletric constant values? by the way S(E) means what?

Sincerely
Sunghwan Choi

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20160722/1a3d700f/attachment.html>