[Pw_forum] compilation error
On Dec 9, 2009, at 20:58 , Shane Taylor wrote: > > In file included from /usr/include/stdio.h:910, > from /usr/include/malloc.h:27, > from memstat.c:25: > /usr/include/bits/stdio.h: In function ?memstat_?: > /usr/include/bits/stdio.h:38: error: nested function ?vprintf? > declared ?extern? [...] this should be the same problem as here: http://www.democritos.it/pipermail/pw_forum/2009-October/014531.html Modified clib/memstat.c file attached. Will be fixed in all subsequent releases. P. -- next part -- A non-text attachment was scrubbed... Name: memstat.c Type: application/octet-stream Size: 848 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20091209/c89504c7/attachment.obj -- next part -- --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
[Pw_forum] nondegenerate electrons in DFT.
> > Correct - you would use Fermi-Dirac for the electronic occupations, and > include > enough bands to make sure the the highest bands are almost empty. > Thanks a lot. Therefore, we can deal with the nondegenerate electrons using this way. However, i think the convergence and the computational speed should be a problem in this case. > This is the canonical extension to DFT according to Mermin (look at his > Phys Rev papers > from the late 60s) - note that the true xc functional would be > temperature-dependent, but > you would use a 0 temperature one. There might be early work (Singwi ?) > on the matter. > About the effect of xc functional in MD, there is a work recently talking about it(High Energy Density Physics,5,2009,74). And the found small influence. Therefore, i think we can use a zero temperature xc functional here. Thanks. Jiayu -- --- Jiayu Dai Department of Physics National University of Defense Technology, Changsha, 410073, P R China -
[Pw_forum] hyperfine calculation
Dear Gregor, I'm just calculating HF tensor of an isolated Si atom, so I don't think magnetic response matters. Below is summary of calculation. QE version 4.1.1 Spin-polarization: 3 up and 1 down Simulation cell: 10-angstrom cubic simulation cell Ewfccut: 30 Ry PP of Si:obtained at http://www-int.impmc.upmc.fr/~software/gipaw/pseudopotentials.html Besides Si atom, I also got zero dipole term for C, N, and O. Regards, Jones On Wed, Dec 9, 2009 at 3:02 PM, Gregor Mali wrote: > Dear Jones Wan > >> >> HF of Si: >> >> ?Si ? 1: g_n = -1.110574 ? ? ? ? ? ? ? ? ?MHz >> ?Si ? 1 ?hfi_dipole ? ? ? 0.00 ? ? ? 0.00 ? ? ? 0.00 >> ?Si ? 1 ?hfi_dipole ? ? ? 0.00 ? ? ? 0.00 ? ? ? 0.00 >> ?Si ? 1 ?hfi_dipole ? ? ? 0.00 ? ? ? 0.00 ? ? ? 0.00 >> >> ?Si ? 1 ?hfi_dipole ? ? ? 0.00 ? ? ? 0.00 ? ? ? 0.00 >> >> Fermi contact term: ? ? ? ?bare ? ? reconstruction ? ? ?total >> ? ? ? ? ? ? ? ? ? ? ? ? ?0.00 ? ? ?45.266354 ? ? ?45.266354 >> ? ZORA >> ? ? ? ? ? ? ? ? ? ? ? ? ?0.00 ? ? ?42.470825 ? ? ?42.470825 > > So far I've never obtained dipole coupling equal to zero! Are you sure > that there are no mistakes in your spin-polarized scf calculation? > > (For example, the materials that I study are paramagnetic at room > temperature, and antiferromagnetic at low temperature. However, if I > want to obtain information on hyperfine coupling related to the > paramagnetic state, I start with ferromagnetic arrangement of spins. > This arrangement more accurately resembles the state of my material at > room temperature and strong external magnetic field. > I mention this because one reason for zero dipole term might be > antiferromagnetic arrangement of spins. In such an arrangement > contributions of neighbouring spins to dipole tensor could cancel one > another.) > > Best regards. > > Gregor > > Gregor Mali > National Institute of Chemistry > Ljubljana, Slovenia > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum
[Pw_forum] compilation error
esn't work and how I might go about fixing it? Shane Taylor -- next part -- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091209/d3b67bfb/attachment.htm
[Pw_forum] sorting phonon bands based on symmetry/eigenvectors
Dear Stefano, Thank you very much for the suggestion. I'll see what I can find regarding the vibrational Wannier functions and I will be glad to report back if I come across some interesting options. Regards, Derek Derek Stewart, Ph. D. Scientific Computation Associate ** New Webpage ** http://sites.google.com/site/dft4nano/ 250 Duffield Hall Cornell Nanoscale Facility (CNF) Ithaca, NY 14853 stewart (at) cnf.cornell.edu (607) 255-2856 Stefano Baroni wrote: > Derek: I do not think/know that/if anything is implemented already, > but you problem was elegantly addressed some time ago by Karin Rabe > who introduced the concept of "vibrational Wannier function" (or > something like that). You may want to browse the literature and report > to us if you find anything meaningful/interesting. I maintain, though, > that the process you want to automate can be automated using IFC as > well. Cheers - Stefano > > On Dec 8, 2009, at 4:07 PM, Derek Stewart wrote: > >> Hi Miguel, >> >> Thank you for your quick response and advice. I agree that with the >> cheap cost of calculating phonon points once you have the IFC, a visual >> inspection of the plot is usually enough to order the phonon dispersion >> along a given symmetry line in the Brillouin Zone. >> >> However, I would like to be able to sort the phonon frequencies for a >> large number of q-vectors in the Brillouin zone (not just restricted to >> a single symmetry line) and for this it would be very helpful if I can >> automate the process. >> >> I know that for electronic bands, it is possible to determine the >> symmetry character of a given electronic band using sym_band.f90. Would >> it be possible to use the same subroutines to determine the symmetry >> character of the phonon bands? I could then sort based on this output. >> If anyone has some thoughts on this, I would be very interested. >> >> Thanks, >> >> Derek >> >> >> Derek Stewart, Ph. D. >> Scientific Computation Associate >> ** New Webpage ** >> http://sites.google.com/site/dft4nano/ >> 250 Duffield Hall >> Cornell Nanoscale Facility (CNF) >> Ithaca, NY 14853 >> stewart (at) cnf.cornell.edu >> (607) 255-2856 >> >> >> >> >> Miguel Mart?nez wrote: >>> Dear Derek, >>> >>> I recently discovered that plotband.x is also able to plot phonon bands >>> as well as band structures. Just set all the references to 0. In any >>> case, if you have the IFC you shouldn't have issues distinguishing >>> crossing from anticrossing. Adding more points to the dispersion bands >>> is almost free, after all. At least it is in the smallish systems I've >>> done phonon calculations for. >>> >>> Regards, >>> >>> Miguel >>> >>> On Mon, 07 Dec 2009 13:42:50 -0500 Derek Stewart >>> wrote: >>> Hi everyone, I was wondering if anyone has written a script or program to sort the phonon frequencies based on their band index or branch. This would be very helpful for both plotting and post-analysis. Currently when you use matdyn for a given set of q-vectors, it prints out the phonon frequencies in order from lowest to highest. However, this does not take into account the possibility of band crossings which can lead to changes in the order of the bands. Matdyn also prints out the eigenvectors for the phonons so it should be possible to analyze this and then sort them. If someone has already addressed this issue, I would be very interested in any tools that can handle this. Thanks, Derek Derek Stewart, Ph. D. Senior Research Associate ** New Webpage ** http://sites.google.com/site/dft4nano/ 250 Duffield Hall Cornell Nanoscale Facility (CNF) Ithaca, NY 14853 stewart (at) cnf.cornell.edu (607) 255-2856 >> >> >> ___ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / > stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une morale > de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum >
[Pw_forum] hyperfine calculation
Dear Davide, I read the post before. I guess we actually have two problems: the Fermi-contact term and dipole-dipole term. Regarding the Fermi contact term, can you send me the code you mentioned so that I give it a try? Also, I don't think that the core-relaxation would completely cancel the dipole-dipole term. Regards, Jones On Wed, Dec 9, 2009 at 3:01 AM, Davide Ceresoli wrote: > Jones Tsz-Kai Wan wrote: >> Dear pwscf users >> >> Does anyone here have experience on HF calculation using gipaw? >> >> I made extensive test on hyperfine tensor calculation using the gipaw >> code, and I couldn't get reasonable result except free hydrogen atom. >> For H atom, the Fermi contact term (A_s) obtained (~1480) is within >> 10% of published work [e.g. Van de Walle and Blochl, PRB 47, 4244 >> (1993)]. However, the results for Si is unreasonable. The Fermi >> contact is only ~ 1% of the published data, and the dipole term (A_p) >> is zero, which is impossible. >> >> Jones Wan >> >> Published result: >> A_s ~ 1400 for H, A_s ~ -4200 and A_p ~ -100 [Van de Walle, PRB 47, >> 4244 (1993)]. > Dear Jones Wan, > ? ? because in Si the spin-density comes from p orbitals. The exchange > potential generated by the p orbitals, polarizes slightly the 3s orbital, > strongly the 2s electrons, which are not included in the pseudopotential. > One has to perform the core-relaxation. For further details, see the > following post: > http://www.democritos.it/pipermail/pw_forum/2009-September/014475.html > > Davide > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum
[Pw_forum] sorting phonon bands based on symmetry/eigenvectors
Derek: I do not think/know that/if anything is implemented already, but you problem was elegantly addressed some time ago by Karin Rabe who introduced the concept of "vibrational Wannier function" (or something like that). You may want to browse the literature and report to us if you find anything meaningful/interesting. I maintain, though, that the process you want to automate can be automated using IFC as well. Cheers - Stefano On Dec 8, 2009, at 4:07 PM, Derek Stewart wrote: > Hi Miguel, > > Thank you for your quick response and advice. I agree that with the > cheap cost of calculating phonon points once you have the IFC, a visual > inspection of the plot is usually enough to order the phonon dispersion > along a given symmetry line in the Brillouin Zone. > > However, I would like to be able to sort the phonon frequencies for a > large number of q-vectors in the Brillouin zone (not just restricted to > a single symmetry line) and for this it would be very helpful if I can > automate the process. > > I know that for electronic bands, it is possible to determine the > symmetry character of a given electronic band using sym_band.f90. Would > it be possible to use the same subroutines to determine the symmetry > character of the phonon bands? I could then sort based on this output. > If anyone has some thoughts on this, I would be very interested. > > Thanks, > > Derek > > > Derek Stewart, Ph. D. > Scientific Computation Associate > ** New Webpage ** > http://sites.google.com/site/dft4nano/ > 250 Duffield Hall > Cornell Nanoscale Facility (CNF) > Ithaca, NY 14853 > stewart (at) cnf.cornell.edu > (607) 255-2856 > > > > > Miguel Mart?nez wrote: >> Dear Derek, >> >> I recently discovered that plotband.x is also able to plot phonon bands >> as well as band structures. Just set all the references to 0. In any >> case, if you have the IFC you shouldn't have issues distinguishing >> crossing from anticrossing. Adding more points to the dispersion bands >> is almost free, after all. At least it is in the smallish systems I've >> done phonon calculations for. >> >> Regards, >> >> Miguel >> >> On Mon, 07 Dec 2009 13:42:50 -0500 Derek Stewart >> wrote: >> >>> Hi everyone, >>> >>> I was wondering if anyone has written a script or program to sort the >>> phonon frequencies based on their band index or branch. This would >>> be very helpful for both plotting and post-analysis. Currently when >>> you use matdyn for a given set of q-vectors, it prints out the phonon >>> frequencies in order from lowest to highest. However, this does not >>> take into account the possibility of band crossings which can lead to >>> changes in the order of the bands. Matdyn also prints out the >>> eigenvectors for the phonons so it should be possible to analyze this >>> and then sort them. If someone has already addressed this issue, I >>> would be very interested in any tools that can handle this. >>> >>> Thanks, >>> >>> Derek >>> >>> >>> Derek Stewart, Ph. D. >>> Senior Research Associate >>> ** New Webpage ** >>> http://sites.google.com/site/dft4nano/ >>> 250 Duffield Hall >>> Cornell Nanoscale Facility (CNF) >>> Ithaca, NY 14853 >>> stewart (at) cnf.cornell.edu >>> (607) 255-2856 >>> >>> >>> >>> >>> >>> > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -- next part -- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091209/54a4ba03/attachment.htm
[Pw_forum] GIPAW and spin-density maps
Dear PWscf/GIPAW users, GIPAW module of QE employs (GI)PAW approach to reconstruct wavefunctuions in the core region and then calculates different physical observables that depend a lot on the electronic wavefuction at or close to the nucleus (e.g. chemical shift, hyperfine coupling, ...). Are these reconstructed wavefunctions saved and can they be used by postprocessing routines to generate spin-density maps? If yes, is the procedure then simply: 1. scf calculation, 2. gipaw calculation, 3. postprocessing, where 'prefix' is kept the same in all input files? Thank you. Gregor Mali Gregor Mali National Institute of Chemistry Ljubljana
[Pw_forum] hyperfine calculation
Dear Jones Wan > > HF of Si: > > Si 1: g_n = -1.110574 MHz > Si 1 hfi_dipole 0.00 0.00 0.00 > Si 1 hfi_dipole 0.00 0.00 0.00 > Si 1 hfi_dipole 0.00 0.00 0.00 > > Si 1 hfi_dipole 0.00 0.00 0.00 > > Fermi contact term:bare reconstruction total > 0.00 45.266354 45.266354 > ZORA > 0.00 42.470825 42.470825 So far I've never obtained dipole coupling equal to zero! Are you sure that there are no mistakes in your spin-polarized scf calculation? (For example, the materials that I study are paramagnetic at room temperature, and antiferromagnetic at low temperature. However, if I want to obtain information on hyperfine coupling related to the paramagnetic state, I start with ferromagnetic arrangement of spins. This arrangement more accurately resembles the state of my material at room temperature and strong external magnetic field. I mention this because one reason for zero dipole term might be antiferromagnetic arrangement of spins. In such an arrangement contributions of neighbouring spins to dipole tensor could cancel one another.) Best regards. Gregor Gregor Mali National Institute of Chemistry Ljubljana, Slovenia
[Pw_forum] hyperfine calculation
Dear pwscf users Does anyone here have experience on HF calculation using gipaw? I made extensive test on hyperfine tensor calculation using the gipaw code, and I couldn't get reasonable result except free hydrogen atom. For H atom, the Fermi contact term (A_s) obtained (~1480) is within 10% of published work [e.g. Van de Walle and Blochl, PRB 47, 4244 (1993)]. However, the results for Si is unreasonable. The Fermi contact is only ~ 1% of the published data, and the dipole term (A_p) is zero, which is impossible. Jones Wan Published result: A_s ~ 1400 for H, A_s ~ -4200 and A_p ~ -100 [Van de Walle, PRB 47, 4244 (1993)]. My calculation results: HF of H: H1: g_n = 5.585693 MHz H1 hfi_dipole 0.00 0.00 0.00 H1 hfi_dipole 0.00 0.00 0.00 H1 hfi_dipole 0.00 0.00 0.00 H1 hfi_dipole 0.00 0.00 0.00 Fermi contact term:bare reconstruction total 0.001484.3280211484.328021 ZORA 0.001485.0260081485.026008 HF of Si: Si 1: g_n = -1.110574 MHz Si 1 hfi_dipole 0.00 0.00 0.00 Si 1 hfi_dipole 0.00 0.00 0.00 Si 1 hfi_dipole 0.00 0.00 0.00 Si 1 hfi_dipole 0.00 0.00 0.00 Fermi contact term:bare reconstruction total 0.00 45.266354 45.266354 ZORA 0.00 42.470825 42.470825
