[QE-users] [SPAM] Re: Occupation calculation - epsilon.x code
Dear Dr. Andrea, Thanks a lot for your help, it was very useful! I've paid attention and ran the pw.x calculation imposing non-symmetries. It is good to hear that the Epsilon.x code was updated. Is there a possibility of getting the newer version of it, before the inclusion in the official distribution? "epsilon.x should be fast enough, though" It really is. My problem is with the pw.x. I had to reduce the energy convergency threshold considerably, being imposed on increase the wave function energy cutoff to preserve S-Matrix positivity. Therefore, even for a small system, the computation time grew. My last attempt was conv_thres=1e-14, the next step will be 1e-15. Any suggestions, shortcuts? Cheers. Anibal ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Occupation calculation - epsilon.x code
Dear QE users and developers, The epsilon.x manual (in the PP/DOC folder) shows the possibility of calculating occupations using the key "occ" within the epsilon.x. It is emphasized to be a good tool to analyze convergency against the broadening parameter and the k points sampling. However, the "occ" calculation is not implemented (at least in the version I'm using - 6.4). Such a calculation was implemented with other packages? If not, is there a way of verifying the convergence other than explicitly changing the broadening and k points sampling? I'm working with an AuAl alloy, trying to evaluate the dielectric function. Using epsilon.x I've got anisotropic behavior that I was not expecting for. Working with pure systems (Au and Al) I concluded that reducing conv_thr increases the epsilon.x output precision, returning the isotropic behavior of the dielectric function. Therefore, to the alloy (with 12 atoms in the cell), I increased both the conv_thr (1e-13) and k points (14 14 14). I still got the anisotropy. Should I go further (calculations with my actual computing power are becoming very time and memory consuming)? Thanks in advance!! Anibal Bezerra The Federal University of Alfenas ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Sum rule violation - turbo_eels
Dear QE users and developers, I'm using turbo_eels + turbo_spectrum to get the dielectric function of gold and aluminum alloys. The calculations finish with no major problems, however, the output file showed a violation of around 17% of the sum rule. I've increased the k points sampling with no appreciable changes to the sum rule. I've changed the method between TDDFT and IPA, with no significant changes to the rule. Is it a problem with the pseudopotential (I've tried NC and USPP from SSSP Material Cloud)? I'm using a small q vector (~0,001) to get the minimum perturbation to the system, could it be the problem? Thanks in advance Anibal Bezerra The Federal University of Alfenas Alfenas - MG Brazil ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Determining anisotropic dielectric function with turbo_eels.x and turbo_espectrum.x
Dear Quantum Espresso users and developers, I've followed the discussion in a previous set of messages sent to this forum about using turbo_eels.x and turbo_espctrum.x to determine the dielectric function. Taking a look at the documentation, I've seen that the programs return a "single" component of the *chi* and *eps* functions, depending on the transferred momentum vector. Is it possible to check for dielectric function anisotropic response by using the different transferred momentum vectors? I mean: -> x response Q=(qx,0,0) -> y response Q=(0,qy,0) -> z response Q=(0,0,qz) Taking a look at the hands-on presentation " Time-Dependent Density Functional Perturbation Theory:" by Iurii Timrov, he compared turboDavidson and turboLanczos. In particular, he showed that at turboLanczos we can choose polarization by the variable ipol = 1,2,3, and 4. Example 01 of TDDFPT packages does Lanczos calculation using ipol=4, but the results (eps and chi) have only imaginary and real parts of a single component. I would expect different responses for each direction (x,y,z, since ipol=4), is it right? Tanks in advance! Anibal Bezerra Federal University of Alfenas Alfenas-MG Brazil ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Epsilon.x code returns anisotropic dieletric function for FCC Aluminium
Dear Quantum Espresso users and developers,
I am trying to use epsilon.x package to get dielectric function for
metallic alloys. I had no problems with golde-silver. However the same was
not true to aluminum.
Having Al an FCC lattice, I would expect its optical response to be
isotropic. I ran VC-RELAX, SCF, NSCF, and Bands calculations, with no major
difficulties in the pure Al system (ibrav =2, 1 atom at unit cell). The
results were in agreement with literature and the examples found at QE
examples folders. Then I moved to epsilon.x and had no success.
I've tried to increase convergence parameters (conv_thr = 1d-09). As we
have to use NC pseudopotentials with epsilon.x, I've increased the wave
function cutoff (Ecutwfc > 200 Ry). Being a metallic system I've played
around with smearing (tested both gaussian and mv), and degauss. Dielectric
function showed a minor change with changing degauss.
I don't know what else to try! Any suggestions?
Follows the input files for the vc-relaxed structure:
title = 'Au66Al33'
prefix = 'Au66Al33' ,
calculation = 'nscf',
disk_io = 'low',
outdir = './tmp/' ,
pseudo_dir = './pseudo/'
verbosity = 'high' ,
etot_conv_thr = 1.0d-8 ,
forc_conv_thr = 1.0d-7 ,
tstress = .true. ,
tprnfor = .true. ,
/
ibrav=0,
celldm(1)=12.93669207d0,
nat=12,
ntyp=2,
ecutwfc=200,
ecutrho=600,
input_dft='pbe',
occupations='smearing',
smearing='mv',
degauss=0.02d0,
noinv = .true.,
nosym = .true.,
/
electron_maxstep = 1000,
conv_thr = 1d-09 ,
mixing_mode = 'plain' ,
mixing_beta = 0.5d0 ,
/
/
/
ATOMIC_SPECIES
Al 26.981500d0 Al.UPF
Au 196.967000d0 Au.UPF
K_POINTS {automatic}
9 9 9 1 1 1
CELL_PARAMETERS {alat}
0.988224309 0.0 0.23347
0.0 0.477626328 0.0
0.22090 0.0 1.334026170
ATOMIC_POSITIONS {crystal}
Al 0.628198968 0.25000 0.277143747
Al 0.128140017 0.25000 0.222904681
Al 0.371801032 0.75000 0.722856253
Al 0.871859983 0.75000 0.777095319
Au 0.147502860 0.25000 0.920025162
Au 0.647627196 0.25000 0.579877017
Au 0.852497140 0.75000 0.079974838
Au 0.352372804 0.75000 0.420122983
Au 0.592675840 0.25000 0.886918451
Au 0.092652971 0.25000 0.613011168
Au 0.407324160 0.75000 0.113081549
Au 0.907347029 0.75000 0.386988832
***
outdir="./tmp",
prefix="Au66Al33",
calculation="eps",
/
_GRID
smeartype="gauss",
intersmear=0.136d0,
intrasmear=0.0d0,
wmax=10.0d0,
wmin=0.0d0,
nw=500,
shift=0.0d0,
/
Thanks in advance
Anibal Bezerra
Federal University of Alfenas
Alfenas-MG
Brazil
Em sex., 8 de mai. de 2020 às 11:27, Anibal Bezerra <
anibal.beze...@unifal-mg.edu.br> escreveu:
> Dear Quantum Espresso users and developers,
>
> I am trying to use epsilon.x package to get dielectric function for
> metallic alloys. I had no problems with golde-silver. However the same was
> not true to aluminum.
>
> Having Al an FCC lattice, I would expect its optical response to be
> isotropic. I ran VC-RELAX, SCF, NSCF, and Bands calculations, with no major
> difficulties in the pure Al system (ibrav =2, 1 atom at unit cell). The
> results were in agreement with literature and the examples found at QE
> examples folders. Then I moved to epsilon.x and had no success.
>
> I've tried to increase convergence parameters (conv_thr = 1d-09). As we
> have to use NC pseudopotentials with epsilon.x, I've increased the wave
> function cutoff (Ecutwfc > 200 Ry). Being a metallic system I've played
> around with smearing (tested both gaussian and mv), and degauss. Dielectric
> function showed a minor change with changing degauss.
>
> I don't know what else to try! Any suggestions?
>
> Follows the input files for the vc-relaxed structure:
>
>
>title = 'Au66Al33'
> prefix = 'Au66Al33' ,
> calculation = 'nscf',
> disk_io = 'low',
>outdir = './tmp/' ,
> pseudo_dir = './pseudo/'
>verbosity = 'high' ,
>etot_conv_thr = 1.0d-8 ,
>forc_conv_thr = 1.0d-7 ,
> tstress = .true. ,
> tprnfor = .true. ,
> /
>
> ibrav=0,
> celldm(1)=12.93669207d0,
> nat=12,
> ntyp=2,
> ecutwfc=200,
> ecutrho=600,
> input_dft='pbe',
> occupations='smearing',
> smearing='mv',
Re: [QE-users] non unifrom kpt grid at epsilon.x
Dear Fariba I've been using Norm-Conserving Pseudo-Potentials from SG15 ONCV Potentials ( http://www.quantum-simulation.org/potentials/sg15_oncv/) They work fairly good! Best Wishes! Anibal ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Projwfc issue
Dear QE experts, We are trying to retrieve the PDoS from an Au/AG alloy with no success. The projwfc.x routine returns: Error in routine do_projwfc (1): Cannot project on zero atomic wavefunctions Searching on QE forum we found the following message: The projection of the calculated Kohn-Sham orbitals can only be done on the wave functions read from the pseudo potential file(s). The message tells you that, probably, the UPF files you’re using do not have a PP_PSWFC section, so the variable natomwfc is set to 0 by the pw.x initialisation. The solution is either tell the code you used to generate the pseudopotential files to write the atomic wave functions or to download the pseudopotential files from http://www.quantum-espresso.org/pseudopotentials/ or to use pslibrary. Giovanni We believe that could be the problem, however changing the pseudopotential is not a possibility. We are also working with the dielectric tensor via eps.x routine. Such a program requires norm-conserving pseudopotentials that we were no able to find ar http://www.quantum-espresso.org/pseudopotentials/. Therefore we used Optimized Norm Conserving Vanderbilt PP (ONCVPSP) from http://www.quantum-simulation.org/potentials/sg15_oncv/. Is there another way in order to retrieve the PDOS information still using such norm-conserving PPs? Thanks in advance! Dr. Anibal Thiago Bezerra ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Epsilon.x - metalcalc variable.
Dear Quantum espresso developers, We are facing some trouble with epsilon.x routine of quantum espresso suite when dealing with metallic alloys. Taking a close look to epsilon.f90 routine, we observed the 'metalcalc' variable set as .FALSE. As the alloy we are working with is metallic (verified by the band structure), in order to retrieve the dielectric tensor of such an alloy, is it necessary to set 'metalcalc' variable as .TRUE.? Best regards! ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] epsilon.x returns infinity and NAN in epsr and epsi
Dear quantum espresso experts,
We are trying to evaluate the dielectric tensor of a 50% Ag 50% Ag alloy
using epsilon.x (eps calculation). We had already relaxed the structure,
calculated scf, nscf, and bands without errors. However, after successfully
ran epsilon.x it keeps returning epsr.dat, epsri.dat files fulled with NaN
and infinity.
We had used options in the nscf calculation nosym = .true.,
noinv = .true. , as well as norm-conserving pseudopotentials for Ag and Au.
Thanks in advance, and looking forward to your answers!!
Anibal Bezerra
Federal University of Alfenas
MInas Gerais
Brasil
Em qua, 2 de out de 2019 às 07:01,
escreveu:
> Send users mailing list submissions to
> users@lists.quantum-espresso.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://lists.quantum-espresso.org/mailman/listinfo/users
> or, via email, send a message with subject or body 'help' to
> users-requ...@lists.quantum-espresso.org
>
> You can reach the person managing the list at
> users-ow...@lists.quantum-espresso.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of users digest..."
>
>
> Today's Topics:
>
>1. Re: visualising vibrations from the results of phonon
> calculations (Nicola Marzari)
>2. ??? unit of dynamical matrix (=?gb18030?B?yebQ8Q==?=)
>3. Degauss vlaue (Pooja Vyas)
>4. ??: Degauss vlaue (Zhou Jianqiang)
>5. degauss and smearing (Pooja Vyas)
>6. ??: degauss and smearing (Zhou Jianqiang)
>7. Re: degauss and smearing (Nicola Marzari)
>8. ??? unit of dynamical matrix (=?gb18030?B?yebQ8Q==?=)
>9. Question about Quantum Espresso (Pooja Vyas)
> 10. Re: Question about Quantum Espresso (Giuseppe Mattioli)
>
>
> --
>
> Message: 1
> Date: Tue, 1 Oct 2019 19:39:33 +0200
> From: Nicola Marzari
> To: y.uem...@uu.nl, Quantum ESPRESSO users Forum
> , Yohei Uemura
> , "nicola.colo...@psi.ch"
>
> Subject: Re: [QE-users] visualising vibrations from the results of
> phonon calculations
> Message-ID:
> Content-Type: text/plain; charset=utf-8; format=flowed
>
>
>
> Dear Yohei,
>
> Nicola Colonna, here in cc, is a MARVEL researcher at PSI; one of the
> goals is also to help experimental colleagues there - since he is a
> Quantum ESPRESSO expert, you'd be wlelcome to get in touch with him.
>
> Re the visualization of phonons, indeed the materials cloud work tool,
> as suggested, would be the one to use.
>
> nicola
>
>
>
> On 30/09/2019 15:59, Yohei Uemura wrote:
> > Dear users,
> >
> > I'm new for quantum espresso (phonon.x).
> > I calculated the vibrations of CO2 or CH4.
> > I would like to visualise the motion of atoms in CO2 or CH4.
> >
> > As someone discussed before, XCrysDen did not work to show the motions
> > of atoms.("[Pw_forum] Phonon axsf does not move in XCrysden")
> >
> > Does anyone know how to visualise the motions of atoms?
> > I have the output file from dynmat.x.
> >
> > Best regards,
> > Yohei
> > --
> >
> > Yohei Uemura (Ph.D)
> > Affiliation : Paul Scherrer Institut (PSI),
> > ? ? ? ? ? ? ? ? ? Laboratory for Environmental Chemistry (LUC)
> > Address : OFLB 106 (PSI Ost), PSI, Forschungsstrasse 111,
> > ? ? ? ? ? ? ? ? ?5232 Villigen, Switzerland
> > Phone: +41 (0)5-6310-4571
> >
> >
> > ?? : Paul Scherrer Institut (PSI),
> > ? ? ? ? ? Laboratory for Environmental Chemistry (LUC)
> > ?? : OFLB 106 (PSI Ost), PSI, Forschungsstrasse 111,
> > ? ? ? ? ? 5232 Villigen, Switzerland
> > ?? : +41 (0)5-6310-4571
> >
> >
> > ___
> > Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso
> )
> > users mailing list users@lists.quantum-espresso.org
> > https://lists.quantum-espresso.org/mailman/listinfo/users
> >
>
>
> --
> --
> Prof Nicola Marzari, Chair of Theory and Simulation of Materials, EPFL
> Director, National Centre for Competence in Research NCCR MARVEL, EPFL
> http://theossrv1.epfl.ch/Main/Contact http://nccr-marvel.ch/en/project
>
>
> --
>
> Message: 2
> Date: Wed, 2 Oct 2019 09:26:00 +0800
> From: "=?gb18030?B?yebQ8Q==?=" <2268083...@qq.com>
> To: "=?gb18030?B?UXVhbnR1bSBFU1BSR
[QE-users] Tran-Blaha modified Becke-Johnson potential (TB-mBJ)
Dear Quantum Espresso experts Is there any implementation of Tran-Blaha modified Becke-Johnson potential (TB-mBJ) on Quantum Espresso? Thanks a Lot!! Em sex, 25 de jan de 2019 às 09:01, < users-requ...@lists.quantum-espresso.org> escreveu: > Send users mailing list submissions to > users@lists.quantum-espresso.org > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.quantum-espresso.org/mailman/listinfo/users > or, via email, send a message with subject or body 'help' to > users-requ...@lists.quantum-espresso.org > > You can reach the person managing the list at > users-ow...@lists.quantum-espresso.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of users digest..." > > > Today's Topics: > >1. Tutorial on writing reproducible workflows for computational > materials science using AiiDA (Leopold Talirz) > > > -- > > Message: 1 > Date: Thu, 24 Jan 2019 19:15:46 +0100 > From: Leopold Talirz > To: users@lists.quantum-espresso.org > Subject: [QE-users] Tutorial on writing reproducible workflows for > computational materials science using AiiDA > Message-ID: > < > cajnv91hmcxb5g1kapb8abwxbt2k56vqmshwjfvqjzmdq5ob...@mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Dear mailing list, > > we'll be hosting a tutorial on writing reproducible workflows for > computational materials science from May 21st, 2019 (9:00) until May 24th, > 2019 (13:00) at EPFL, Lausanne, Switzerland. > > This 3.5-day tutorial is designed to get Master students, PhD students and > Postdocs from the field of computational materials science started with > writing reproducible workflows. Participants will be introduced to the > state of the art in workflow management and high-throughput computations by > experts in the field, and gain in-depth hands-on experience using a tool > that they can directly apply to their own research. > > Our tool of choice is the AiiDA framework (see aiida.net) for workflow > management and provenance tracking, which is backed by a significant > community of users and developers, and has interfaces to more than 20 > materials science codes (see plugin registry), including to the ab initio > codes Quantum ESPRESSO, VASP, cp2k, Castep, Siesta, Fleur, Crystal, NWChem, > Wannier90, and Yambo. AiiDA?s permissive open source license (MIT) enables > participants to use it both in academic and commercial settings. By virtue > of its general design and flexible plugin system, AiiDA is easily extended > to new codes and new use cases. > > For more information, please head over to the tutorial web site: > http://www.aiida.net/news/tutorial-writing-reproducible-workflows/ > > Looking forward to meeting you in Lausanne! > Leopold > -- next part -- > An HTML attachment was scrubbed... > URL: < > http://lists.quantum-espresso.org/pipermail/users/attachments/20190124/e1a8e778/attachment-0001.html > > > > -- > > Subject: Digest Footer > > ___ > users mailing list > users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users > > -- > > End of users Digest, Vol 138, Issue 24 > ** > ___ users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Problems to run QE in parallel
Dear Paulo, Thanks a lot for the quick response!! Even using the command "mpirun" when I complied QE using Intel's MPI, I got the same output. I've started the compiling and making process from the beginning, after uninstalled Intel's MPI, Mpich, installed Mpich's MPI again re-configured QE. When I tried to run the silicon test, using both "mpiexec" and "mpirun" I got replicated Parallel version(MPI), running on1 processors MPI processes distributed on 1 nodes. Is there anything else I could check for? Thanks again!! Anibal Message: 5 Date: Tue, 10 Jul 2018 22:54:57 +0200 From: Paolo Giannozzi To: Quantum Espresso users Forum Subject: Re: [QE-users] Problems to run QE in parallel Message-ID: Content-Type: text/plain; charset="utf-8" If you link intel mpi, you should use the version of "mpirun" provided by intel to run the parallel code Paolo On Tue, Jul 10, 2018 at 8:02 PM, Anibal Bezerra < anibal.beze...@unifal-mg.edu.br> wrote: > Dear all, > > I'm new on Quantum Espresso and I'm in trouble with configuring it to run > in parallel. I've installed the mpich MPI and added the bin path to bashrc > file. I've tested the MPI and it's working fine. However, after > successfully configuring QE, detecting parallel environment, I've tried to > run an example using mpiexec command (mpiexec -n 10...), and I got > > Parallel version(MPI), running on1 processors > MPI processes distributed on 1 nodes. > > Replicated 10 times on the out file. > > I've tried Intel's MPI, the same occurred. I've tried explicitly give the > MPI's path to the configure without success. > > I really don't know how to proceed. > > Thanks in advance! ___ users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] Problems to run QE in parallel
Dear all, I'm new on Quantum Espresso and I'm in trouble with configuring it to run in parallel. I've installed the mpich MPI and added the bin path to bashrc file. I've tested the MPI and it's working fine. However, after successfully configuring QE, detecting parallel environment, I've tried to run an example using mpiexec command (mpiexec -n 10...), and I got Parallel version(MPI), running on1 processors MPI processes distributed on 1 nodes. Replicated 10 times on the out file. I've tried Intel's MPI, the same occurred. I've tried explicitly give the MPI's path to the configure without success. I really don't know how to proceed. Thanks in advance! ___ users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
