From sclauzer at sissa.it Thu Oct 1 08:52:20 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 01 Oct 2009 08:52:20 +0200 Subject: [Pw_forum] Projected DOS In-Reply-To: References: Message-ID: <4AC451A4.1080401@sissa.it> Dear Yongduo Liu, ??? wrote: > Dear all > > > Does any body know what is the setting of Wigner Seitz radii in > the DOS projection/? I don't think there is such parameter in the program which computes Projected DOS (projwfc.x), since it uses atomic wavefunctions to build the projections. These wavefunctions extend over the whole space (although, for practical purposes, they may be considered to be vanishing far enough from the atom - and its periodic replicas - on which they are centered), hence the overlap between a KS eigenstate and an atomic wavefunction is computed by integrating their product over the whole unit-cell. Regards, GS > What is the default value and how to reset it? > > > Thanks very much! > > > Yongduo Liu > > Materials Science & Engineering > University of California, Los Angeles > / > ------------------------------------------------------------------------ > ??????????MSN??? ????? > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Thu Oct 1 09:01:52 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 01 Oct 2009 09:01:52 +0200 Subject: [Pw_forum] symmetry of defect state In-Reply-To: <290030.71709.qm@web112507.mail.gq1.yahoo.com> References: <290030.71709.qm@web112507.mail.gq1.yahoo.com> Message-ID: <4AC453E0.8000203@sissa.it> Dear Ali, ali kazempour wrote: > Hi > Does anyone knows how to determine the symmetry of the defect state > introduced in gap or how to join the the relaxation of atoms around the > defect by the changing point group? Honestly, your question is not much clear to me (and maybe also to the others, since you received no reply yet), but I would like to help you to make it more specific. What symmetry do you want to determine? The symmetry of a specific KS eigenstate, or the point group of a system after removing an atom to create a defect? In the first case you can use the program bands.x with lsym=.TRUE., in the second it is done automatically by pw.x at the beginning of the run (if you specify verbosity='high' it will also print all symmetry operations belonging to the point group). I think that if you introduce a defect and you want to relax your structure, you should slightly, randomly displace some atoms near the defect in order to further reduce the symmetry and allow atomic coordinates to span a larger phase space to reach a better minimum. Remember that very large supercells may be needed in these kind of studies. HTH, GS > thanks a lot > > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From daniel.forrer at unipd.it Thu Oct 1 10:00:01 2009 From: daniel.forrer at unipd.it (Daniel Forrer) Date: Thu, 1 Oct 2009 10:00:01 +0200 Subject: [Pw_forum] bug in mm_dispersion.f90 In-Reply-To: <9A4683A5-EEAB-4680-BE1B-D1443C8195C9@democritos.it> References: <4AC218C7.9080805@sissa.it> <9A4683A5-EEAB-4680-BE1B-D1443C8195C9@democritos.it> Message-ID: Dear Paolo, 2009/9/29 Paolo Giannozzi > > On Sep 29, 2009, at 17:12 , Javier Fuhr wrote: > > The parallelization is done to all processes (nproc), and >> if I understand correctly the sum should be performed >> over all processes. Looking better at the different >> parallelizations, I think the correct way is to replace, >> in mm_dispersion.f90, intra_pool_comm by world_comm >> > > I think you are (almost) right. "image" parallelization, useful > for NEB calculations, implies different atomic positions on > different groups of processors. So the correct communicator > is "intra_image_comm", containing nproc_image processors. > Attached the (hopefully) correct Modules/mm_dispersion.f90. > Thank you for reporting this bug and its (quasi-)fix > i'm not expert in parallelization, but it seems to me that the processor index "mpime" should be replaced by "me_image" if we use the "intra_image_comm" communicator. Is it correct? daniel -- ---------------------------------- Daniel Forrer Dipartimento di Scienze Chimiche Universit? degli Studi di Padova V. Marzolo 1, Padova ---------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091001/09f1f0c8/attachment.htm From giannozz at democritos.it Thu Oct 1 10:51:20 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 01 Oct 2009 10:51:20 +0200 Subject: [Pw_forum] bug in mm_dispersion.f90 In-Reply-To: References: <4AC218C7.9080805@sissa.it> <9A4683A5-EEAB-4680-BE1B-D1443C8195C9@democritos.it> Message-ID: <4AC46D88.2050200@democritos.it> Daniel Forrer wrote: > i'm not expert in parallelization nor am I! > but it seems to me that the processor index "mpime" > should be replaced by "me_image" if we use the > "intra_image_comm" communicator. Is it correct? it is (I think). For a single image, it makes no difference because the two are equal (I think). Thank you, it will be fixed P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From silveto at gmail.com Thu Oct 1 12:02:53 2009 From: silveto at gmail.com (Silvia Bakalova) Date: Thu, 1 Oct 2009 11:02:53 +0100 Subject: [Pw_forum] epsilon.x uniaxial material Message-ID: Hi, I am using epsilon.x to obtain the dielectric function of uniaxial material (ibrav= 5 with trigonal symmetry). However, e.g. the file epsi.dat contains three different values epsi_x epsi_y epsi_z, while there should be two independent components (for E perpendicular and parallel to the optical axis, e.g. epsi_x=epsi_y). Could you, please, think of what the reason could be? Many thanks, Silvia ---------------------- Dr. Silvia Bakalova, Post Doctoral Researcher, HH Wills Physics Laboratory, Bristol, BS8 1TL, UK http://spectra.phy.bris.ac.uk/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091001/7090e48d/attachment.htm From ferretti at MIT.EDU Thu Oct 1 17:14:34 2009 From: ferretti at MIT.EDU (Andrea Ferretti) Date: Thu, 1 Oct 2009 11:14:34 -0400 (EDT) Subject: [Pw_forum] epsilon.x uniaxial material In-Reply-To: Message-ID: Hi Silvia, the dielectric tensors computed by epsilon are defined on cartesian coordinates, and symmetry is not used to reduce the number of components to be computed. Of course, this does not mean that the proper symmetry of the dielectric tensor is missing (and this is typically a good check to be done a posteriori) Andrea > > I am using epsilon.x to obtain the dielectric function of uniaxial material > (ibrav= 5 with trigonal symmetry). > > However, e.g. the file epsi.dat contains three different values epsi_x > epsi_y epsi_z, while there should be two independent components (for E > perpendicular and parallel to the optical axis, e.g. epsi_x=epsi_y). Could > you, please, think of what the reason could be? > > Many thanks, > > Silvia > > > ---------------------- > Dr. Silvia Bakalova, > Post Doctoral Researcher, > HH Wills Physics Laboratory, > Bristol, BS8 1TL, UK > http://spectra.phy.bris.ac.uk/ > -- Andrea Ferretti MIT, Dept Material Science & Engineering bldg 13-4078, 77, Massachusetts Ave, Cambridge, MA Tel: +1 617-452-2455; Skype: andrea_ferretti URL: http://quasiamore.mit.edu Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From duchl06 at 163.com Thu Oct 1 20:03:53 2009 From: duchl06 at 163.com (duchl06) Date: Fri, 2 Oct 2009 02:03:53 +0800 (CST) Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: References: Message-ID: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> dear pwscf users, i found in some papers, they always say "To examine the stability of the local magnetic state, we calculate the energy difference between the diamagnetic(DM) antiferromagnetic (AFM) and ferromagnetic(FM) states." i know that in the case of FM, I need to do only spin-polarized calculation with start magntic moment, while for AF, I should define 2 different atoms having equal but opposite sign magnetic moments. is this right? but i really donot know how to set up a cell for the diamagnetic material, and how to set the start magnetic moment? thank you! best regards! graduate student: chen xiaguanying university -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/090c566d/attachment.htm From ttduyle at gmail.com Thu Oct 1 22:11:43 2009 From: ttduyle at gmail.com (Duy Le) Date: Thu, 1 Oct 2009 16:11:43 -0400 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> References: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> Message-ID: <8974d3b20910011311s598f7805wb0729ff2f4173563@mail.gmail.com> You just treat them as 2 different types of atoms, say Co1 Co2 then use starting_magnetization(1)=3 starting_magnetization(2)=-3 2009/10/1 duchl06 > > > dear pwscf users, > > i found in some papers, they always say "To examine the stability of the > local magnetic state, we calculate the energy difference between the > diamagnetic(DM) antiferromagnetic (AFM) and ferromagnetic(FM) states." > i know that in the case of FM, I need to do only spin-polarized > calculation with start magntic moment, while for AF, I should define 2 > different atoms having equal but opposite sign magnetic moments. is this > right? > but i really donot know how to set up a cell for the diamagnetic > material, and how to set the start magnetic moment? > thank you! > > best regards! > > > graduate student: chen > xiaguanying university > > > ------------------------------ > ?????????????? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091001/b7923560/attachment-0001.htm From giannozz at democritos.it Thu Oct 1 22:27:07 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 1 Oct 2009 22:27:07 +0200 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <8974d3b20910011311s598f7805wb0729ff2f4173563@mail.gmail.com> References: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <8974d3b20910011311s598f7805wb0729ff2f4173563@mail.gmail.com> Message-ID: <0269C775-AE9E-442A-BA4E-C67A463EFACA@democritos.it> On Oct 1, 2009, at 22:11 , Duy Le wrote: > starting_magnetization(1)=3 > starting_magnetization(2)=-3 actually starting_magnetization should be between -1 and 1. The code will "renormalize" them to 1 or -1 anyway P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From nkxirainbow at gmail.com Fri Oct 2 04:41:52 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Fri, 2 Oct 2009 02:41:52 +0000 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> References: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> Message-ID: <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> Dear Chen? > but i really donot know how to set up a cell for the diamagnetic material, and how to set >the start magnetic moment? Hear is an example about graphene ribbon antiferromagnetic calculation. Remember: different "starting_magnetization" atoms are treated as totally different atoms. Input file:graphene.rx.in &CONTROL calculation = 'relax' , .................. / &SYSTEM ibrav = 8, celldm(1) = 4.67689278, celldm(2) = 8.623989813, celldm(3) = 4.048794087, .................... nat = 12, ntyp = 2, nspin = 2 , starting_magnetization(1) = 1.0, starting_magnetization(2) = -1.0, / &ELECTRONS ................... / &IONS .................... / ATOMIC_SPECIES C1 12.00000 C.pbe-rrkjus.UPF C2 12.00000 C.pbe-rrkjus.UPF ATOMIC_POSITIONS crystal C1 0.000000000 0.236465542 0.000000000 C2 0.500000000 0.265456486 0.000000000 C1 0.500000000 0.334092417 0.000000000 C2 0.000000000 0.366772814 0.000000000 C1 0.000000000 0.433673757 0.000000000 C2 0.500000000 0.466640035 0.000000000 C1 0.500000000 0.533360168 0.000000000 C2 0.000000000 0.566326232 0.000000000 C1 0.000000000 0.633227130 0.000000000 C2 0.500000000 0.665907475 0.000000000 C1 0.500000000 0.734543349 0.000000000 C2 0.000000000 0.763534596 0.000000000 K_POINTS automatic 13 1 1 0 0 0 -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/0dfd8692/attachment.htm From duchl06 at 163.com Fri Oct 2 04:57:29 2009 From: duchl06 at 163.com (duchl06) Date: Fri, 2 Oct 2009 10:57:29 +0800 (CST) Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> References: <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> Message-ID: <27999742.165361254452249345.JavaMail.coremail@app165.163.com> thanks every one's reply! hi, Hui Wang the example you supplied is a calculation for a antiferromagnetic material, i mean how to set the start magnetic moment for the diamagnetic material, best regards! graduate student: chen xiaguanying university ?2009-10-02?xirainbow ??? Dear Chen? > but i really donot know how to set up a cell for the diamagnetic material, and how to set >the start magnetic moment? Hear is an example about graphene ribbon antiferromagnetic calculation. Remember: different "starting_magnetization" atoms are treated as totally different atoms. Input file:graphene.rx.in &CONTROL calculation = 'relax' , .................. / &SYSTEM ibrav = 8, celldm(1) = 4.67689278, celldm(2) = 8.623989813, celldm(3) = 4.048794087, .................... nat = 12, ntyp = 2, nspin = 2 , starting_magnetization(1) = 1.0, starting_magnetization(2) = -1.0, / &ELECTRONS ................... / &IONS .................... / ATOMIC_SPECIES C1 12.00000 C.pbe-rrkjus.UPF C2 12.00000 C.pbe-rrkjus.UPF ATOMIC_POSITIONS crystal C1 0.000000000 0.236465542 0.000000000 C2 0.500000000 0.265456486 0.000000000 C1 0.500000000 0.334092417 0.000000000 C2 0.000000000 0.366772814 0.000000000 C1 0.000000000 0.433673757 0.000000000 C2 0.500000000 0.466640035 0.000000000 C1 0.500000000 0.533360168 0.000000000 C2 0.000000000 0.566326232 0.000000000 C1 0.000000000 0.633227130 0.000000000 C2 0.500000000 0.665907475 0.000000000 C1 0.500000000 0.734543349 0.000000000 C2 0.000000000 0.763534596 0.000000000 K_POINTS automatic 13 1 1 0 0 0 -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/a6755c04/attachment.htm From nkxirainbow at gmail.com Fri Oct 2 05:16:36 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Fri, 2 Oct 2009 03:16:36 +0000 Subject: [Pw_forum] pwcond with spin-orbital effect Message-ID: <21fbc4790910012016s44ce0ddet6a722188204e5e21@mail.gmail.com> Dear all? I know the example 22 give a transmission with spin-orbital effect. Is the spin-orbital effect considered as perturbation? Does the pwcond.x use "Green function" method? Can the pwcond.x deal with "voltage bias" condition? Thanks in advance?? -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/69067e62/attachment-0001.htm From nkxirainbow at gmail.com Fri Oct 2 05:26:55 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Fri, 2 Oct 2009 03:26:55 +0000 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <27999742.165361254452249345.JavaMail.coremail@app165.163.com> References: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> Message-ID: <21fbc4790910012026x6f7e7edctea80dc6193a8554b@mail.gmail.com> Dear Chen:You can use " nspin = 2 , starting_magnetization(1) = 1.0,starting_magnetization(2) = -1.0,". Then PWscf will change magnetization during scf or relaxation calculation. If the start state is antiferromagnetic state(zero initial magnetization), the scf result would be zero magnetization(antiferromagnetic or paramagnetic). 2009/10/2 duchl06 > thanks every one's reply! > hi, Hui Wang > the example you supplied is a calculation for a antiferromagnetic > material, i mean how to set the start magnetic moment for the diamagnetic > material, > > > > best regards! > > graduate student: chen > xiaguanying university > > ?2009-10-02?xirainbow ??? > > Dear Chen? > > > but i really donot know how to set up a cell for the diamagnetic > material, and how to set >the start magnetic moment? > > > Hear is an example about graphene ribbon antiferromagnetic calculation. > Remember: different "starting_magnetization" atoms are treated as totally > different atoms. > > Input file:graphene.rx.in > &CONTROL > calculation = 'relax' , > .................. > / > &SYSTEM > ibrav = 8, > celldm(1) = 4.67689278, > celldm(2) = 8.623989813, > celldm(3) = 4.048794087, > .................... > nat = 12, > ntyp = 2, > nspin = 2 , > starting_magnetization(1) = 1.0, > starting_magnetization(2) = -1.0, > / > &ELECTRONS > ................... > / > &IONS > .................... > / > ATOMIC_SPECIES > C1 12.00000 C.pbe-rrkjus.UPF > C2 12.00000 C.pbe-rrkjus.UPF > ATOMIC_POSITIONS crystal > C1 0.000000000 0.236465542 0.000000000 > C2 0.500000000 0.265456486 0.000000000 > C1 0.500000000 0.334092417 0.000000000 > C2 0.000000000 0.366772814 0.000000000 > C1 0.000000000 0.433673757 0.000000000 > C2 0.500000000 0.466640035 0.000000000 > C1 0.500000000 0.533360168 0.000000000 > C2 0.000000000 0.566326232 0.000000000 > C1 0.000000000 0.633227130 0.000000000 > C2 0.500000000 0.665907475 0.000000000 > C1 0.500000000 0.734543349 0.000000000 > C2 0.000000000 0.763534596 0.000000000 > K_POINTS automatic > 13 1 1 0 0 0 > > > -- > ____________________________________ > Hui Wang > School of physics, Nankai University, Tianjin, China > > > > ------------------------------ > "????",????60??? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/c2aff623/attachment.htm From ttduyle at gmail.com Fri Oct 2 07:02:05 2009 From: ttduyle at gmail.com (Duy Le) Date: Fri, 2 Oct 2009 01:02:05 -0400 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <21fbc4790910012026x6f7e7edctea80dc6193a8554b@mail.gmail.com> References: <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> <21fbc4790910012026x6f7e7edctea80dc6193a8554b@mail.gmail.com> Message-ID: <8974d3b20910012202p5eee1e1y9bf122f831a69a10@mail.gmail.com> Oh, I see what you meant now. Diamagnetic materials are either AF or FM. Just do spin polarized calculation with starting_magnetization. I don't see any different between diamagnetic VS. AF/FM unless you have them in external magnetic file. 2009/10/1 xirainbow > Dear Chen:You can use " nspin = 2 , starting_magnetization(1) = 1.0,starting_magnetization(2) > = -1.0,". > Then PWscf will change magnetization during scf or relaxation calculation. > If the start state is antiferromagnetic state(zero initial > magnetization), the scf result would be zero > magnetization(antiferromagnetic or paramagnetic). > > 2009/10/2 duchl06 > >> thanks every one's reply! >> hi, Hui Wang >> the example you supplied is a calculation for a antiferromagnetic >> material, i mean how to set the start magnetic moment for the diamagnetic >> material, >> >> >> >> best regards! >> >> graduate student: chen >> xiaguanying university >> >> ?2009-10-02?xirainbow ??? >> >> Dear Chen? >> >> > but i really donot know how to set up a cell for the diamagnetic >> material, and how to set >the start magnetic moment? >> >> >> Hear is an example about graphene ribbon antiferromagnetic calculation. >> Remember: different "starting_magnetization" atoms are treated as totally >> different atoms. >> >> Input file:graphene.rx.in >> &CONTROL >> calculation = 'relax' , >> .................. >> / >> &SYSTEM >> ibrav = 8, >> celldm(1) = 4.67689278, >> celldm(2) = 8.623989813, >> celldm(3) = 4.048794087, >> .................... >> nat = 12, >> ntyp = 2, >> nspin = 2 , >> starting_magnetization(1) = 1.0, >> starting_magnetization(2) = -1.0, >> / >> &ELECTRONS >> ................... >> / >> &IONS >> .................... >> / >> ATOMIC_SPECIES >> C1 12.00000 C.pbe-rrkjus.UPF >> C2 12.00000 C.pbe-rrkjus.UPF >> ATOMIC_POSITIONS crystal >> C1 0.000000000 0.236465542 0.000000000 >> C2 0.500000000 0.265456486 0.000000000 >> C1 0.500000000 0.334092417 0.000000000 >> C2 0.000000000 0.366772814 0.000000000 >> C1 0.000000000 0.433673757 0.000000000 >> C2 0.500000000 0.466640035 0.000000000 >> C1 0.500000000 0.533360168 0.000000000 >> C2 0.000000000 0.566326232 0.000000000 >> C1 0.000000000 0.633227130 0.000000000 >> C2 0.500000000 0.665907475 0.000000000 >> C1 0.500000000 0.734543349 0.000000000 >> C2 0.000000000 0.763534596 0.000000000 >> K_POINTS automatic >> 13 1 1 0 0 0 >> >> >> -- >> ____________________________________ >> Hui Wang >> School of physics, Nankai University, Tianjin, China >> >> >> >> ------------------------------ >> "????",????60??? >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > > -- > ____________________________________ > Hui Wang > School of physics, Nankai University, Tianjin, China > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/7c4507b5/attachment-0001.htm From sclauzer at sissa.it Fri Oct 2 08:51:32 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 02 Oct 2009 08:51:32 +0200 Subject: [Pw_forum] pwcond with spin-orbital effect In-Reply-To: <21fbc4790910012016s44ce0ddet6a722188204e5e21@mail.gmail.com> References: <21fbc4790910012016s44ce0ddet6a722188204e5e21@mail.gmail.com> Message-ID: <4AC5A2F4.20106@sissa.it> xirainbow wrote: > Dear all? > > I know the example 22 give a transmission with spin-orbital effect. > > Is the spin-orbital effect considered as perturbation? No, it is not correct to say this in my point of view. The SO effect is included through the use of so-called fully-relativistic pseudo potentials, as it is done in PWscf (see Andrea Dal Corso and Adriano Mosca Conte, Phys. Rev. B 71 115106 (2005) for details). > Does the pwcond.x use "Green function" method? No, it uses a scattering state approach, which integrates numerically a scattering equation in real space along the direction of transport (while in the perpendicular directions the PW basis of the PWscf calculation is retained). See: Hyoung Joon Choi and Jisoon Ihm, Phys. Rev. B 59, 2267 (1999) for the method originally developed for norm-conserving PPs and: Andrea Dal Corso, Alexander Smogunov, and Erio Tosatti, Phys. Rev. B 74, 045429 (2006) for the extension to US-PPs with spin-orbit coupling. > Can the pwcond.x deal with "voltage bias" condition? No, it is built within the Landauer-Buttiker linear response theory, which gives the low bias limit of the conductance. Regards, GS > > Thanks in advance?? > > -- > ____________________________________ > Hui Wang > School of physics, Nankai University, Tianjin, China > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Fri Oct 2 09:02:47 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 02 Oct 2009 09:02:47 +0200 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <27999742.165361254452249345.JavaMail.coremail@app165.163.com> References: <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> Message-ID: <4AC5A597.5090804@sissa.it> duchl06 wrote: > thanks every one's reply! > hi, Hui Wang > the example you supplied is a calculation for a antiferromagnetic > material, i mean how to set the start magnetic moment for the > diamagnetic material, Diamagnetic material in absence of an external magnetic field should have no net magnetic moment per atom, if I'm not wrong. So it is not clear to me what do you want to study and if all these suggestions are pointing you to the correct solution. If you want to study a diamagnetic material in absence of magnetic field there should be no need for a spin-polarized calculation. So, just to agree on the starting point of your question: what you mean with diamagnetic material is really what people usually refer to? Can you give us more details on the system you want to address? What properties do you want to study? ... Regards, GS > > > > best regards! > > > graduate student: chen > xiaguanying university > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From duchl06 at 163.com Fri Oct 2 08:52:38 2009 From: duchl06 at 163.com (duchl06) Date: Fri, 2 Oct 2009 14:52:38 +0800 (CST) Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <8974d3b20910012202p5eee1e1y9bf122f831a69a10@mail.gmail.com> References: <8974d3b20910012202p5eee1e1y9bf122f831a69a10@mail.gmail.com> <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> <21fbc4790910012026x6f7e7edctea80dc6193a8554b@mail.gmail.com> Message-ID: <22803458.189711254466358853.JavaMail.coremail@bj163app33.163.com> hi, thank you for the explanations. i read some book and found that the diamagnetism is a normal, universal characteristic of the material, it is covered by other magnetism in non-diamagnetic materials. and some other told me that: paramagnetism is random orientation of spins throughout the material, Diamagnetism is local magnetic moment (no spins).So non spin-magnetisation calculation is indeed the closest thing to diamagnetism simulation. Duy Le said Diamagnetic materials are either AFM or FM, but there are diamagnetic materials, like Cu, Ag. so, could I comprehend like this: when do a calculation for AFM and FM materials, just do as Duy Le, Hui Wang and Paolo Giannozzi said. when do a calculation for DM materials or (paramagnetism)PM materials, just turn the spin polarized off? graduate student: chen xiaguanying university ?2009-10-02?"Duy Le" ??? Oh, I see what you meant now. Diamagnetic materials are either AF or FM. Just do spin polarized calculation with starting_magnetization. I don't see any different between diamagnetic VS. AF/FM unless you have them in external magnetic file. 2009/10/1 xirainbow Dear Chen: You can use " nspin = 2 , starting_magnetization(1) = 1.0,starting_magnetization(2) = -1.0,". Then PWscf will change magnetization during scf or relaxation calculation. If the start state is antiferromagnetic state(zero initial magnetization), the scf result would be zero magnetization(antiferromagnetic or paramagnetic). 2009/10/2 duchl06 thanks every one's reply! hi, Hui Wang the example you supplied is a calculation for a antiferromagnetic material, i mean how to set the start magnetic moment for the diamagnetic material, best regards! graduate student: chen xiaguanying university ?2009-10-02?xirainbow ??? Dear Chen? > but i really donot know how to set up a cell for the diamagnetic material, and how to set >the start magnetic moment? Hear is an example about graphene ribbon antiferromagnetic calculation. Remember: different "starting_magnetization" atoms are treated as totally different atoms. Input file:graphene.rx.in &CONTROL calculation = 'relax' , .................. / &SYSTEM ibrav = 8, celldm(1) = 4.67689278, celldm(2) = 8.623989813, celldm(3) = 4.048794087, .................... nat = 12, ntyp = 2, nspin = 2 , starting_magnetization(1) = 1.0, starting_magnetization(2) = -1.0, / &ELECTRONS ................... / &IONS .................... / ATOMIC_SPECIES C1 12.00000 C.pbe-rrkjus.UPF C2 12.00000 C.pbe-rrkjus.UPF ATOMIC_POSITIONS crystal C1 0.000000000 0.236465542 0.000000000 C2 0.500000000 0.265456486 0.000000000 C1 0.500000000 0.334092417 0.000000000 C2 0.000000000 0.366772814 0.000000000 C1 0.000000000 0.433673757 0.000000000 C2 0.500000000 0.466640035 0.000000000 C1 0.500000000 0.533360168 0.000000000 C2 0.000000000 0.566326232 0.000000000 C1 0.000000000 0.633227130 0.000000000 C2 0.500000000 0.665907475 0.000000000 C1 0.500000000 0.734543349 0.000000000 C2 0.000000000 0.763534596 0.000000000 K_POINTS automatic 13 1 1 0 0 0 -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China "????",????60??? _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -- -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/699b8ffc/attachment-0001.htm From kazempoor2000 at yahoo.com Fri Oct 2 13:21:07 2009 From: kazempoor2000 at yahoo.com (ali kazempour) Date: Fri, 2 Oct 2009 04:21:07 -0700 (PDT) Subject: [Pw_forum] defect Message-ID: <906593.14449.qm@web112505.mail.gq1.yahoo.com> Hi Does anyone knows how to determine the symmetry of the defect state introduced in gap or how to join the the relaxation of atoms around the defect by the changing point group? thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/f55b470a/attachment.htm From sclauzer at sissa.it Fri Oct 2 13:30:33 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 02 Oct 2009 13:30:33 +0200 Subject: [Pw_forum] defect In-Reply-To: <906593.14449.qm@web112505.mail.gq1.yahoo.com> References: <906593.14449.qm@web112505.mail.gq1.yahoo.com> Message-ID: <4AC5E459.2020400@sissa.it> Dear Ali, I guess that you have some problems with your mailing program/server/provider, since this is the third time you posted the same question and ignored my reply (although it was somehow vague, as I felt that your question is not much clear... to me at least) http://www.democritos.it/pipermail/pw_forum/2009-October/014495.html GS ali kazempour wrote: > > Hi > Does anyone knows how to determine the symmetry of the defect state > introduced in gap or how to join the the relaxation of atoms around the > defect by the changing point group? > thanks a lot > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From kazempoor2000 at yahoo.com Fri Oct 2 13:32:46 2009 From: kazempoor2000 at yahoo.com (ali kazempour) Date: Fri, 2 Oct 2009 04:32:46 -0700 (PDT) Subject: [Pw_forum] Fw: defect Message-ID: <472899.45790.qm@web112502.mail.gq1.yahoo.com> Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 ----- Forwarded Message ---- From: ali kazempour To: pw Sent: Friday, October 2, 2009 1:21:07 PM Subject: defect Hi Does anyone knows how to determine the symmetry of the defect state introduced in gap or how to join the the relaxation of atoms around the defect by the changing point group? thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/08deff1c/attachment.htm From sclauzer at sissa.it Fri Oct 2 13:43:47 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 02 Oct 2009 13:43:47 +0200 Subject: [Pw_forum] Fw: defect In-Reply-To: <472899.45790.qm@web112502.mail.gq1.yahoo.com> References: <472899.45790.qm@web112502.mail.gq1.yahoo.com> Message-ID: <4AC5E773.9080407@sissa.it> Please stop! (see mail below...) Maybe you are not receiving mails from the forum since you turned on digest mode. Please check your subscription at http://www.democritos.it/mailman/options/pw_forum GS Dear Ali, I guess that you have some problems with your mailing program/server/provider, since this is the third time you posted the same question and ignored my reply (although it was somehow vague, as I felt that your question is not much clear... to me at least) http://www.democritos.it/pipermail/pw_forum/2009-October/014495.html GS ali kazempour wrote: > > > > Hi > > Does anyone knows how to determine the symmetry of the defect state > > introduced in gap or how to join the the relaxation of atoms around the > > defect by the changing point group? > > thanks a lot > > Ali Kazempour > > Physics department, Isfahan University of Technology > > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 > > > > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum ali kazempour wrote: > > > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 > > > ----- Forwarded Message ---- > *From:* ali kazempour > *To:* pw > *Sent:* Friday, October 2, 2009 1:21:07 PM > *Subject:* defect > > > Hi > Does anyone knows how to determine the symmetry of the defect state > introduced in gap or how to join the the relaxation of atoms around the > defect by the changing point group? > thanks a lot > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From duchl06 at 163.com Fri Oct 2 14:10:12 2009 From: duchl06 at 163.com (duchl06) Date: Fri, 2 Oct 2009 20:10:12 +0800 (CST) Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <4AC5A597.5090804@sissa.it> References: <4AC5A597.5090804@sissa.it> <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> Message-ID: <18823439.239381254485412186.JavaMail.coremail@bj163app79.163.com> hi,Gabriele Sclauzero thanks for your advice and hoping your further guide. i am reading a paper "Charge-Transfer-Based Mechanism for Half-Metallicity and Ferromagnetism in one-Dimensional OrganometallicSandwich Molecular Wires" http://pubs.acs.org/doi/abs/10.1021/ja804053a in this paper they consider a sandwich molecular wires (such as 1D (VBz)? wire, (MnCp)? wire, Bz is benzene, Cp is cyclopentadienyl), and they said "Furthermore, we investigate the effect of doubling the unit cell size along the SMW direction to evaluate the ground state of ferromagnetic (FM), antiferromagnetic (AFM), and diamagnetic (DM) con?gurations in our spin-polarized calculations with structural optimization, and assess possibility of the Peierls transition." (and i also found in some other papers when they do not know the groud state is FM or DM, they would always say "To examine the stability of the local magnetic state, we calculate the energy difference between the diamagnetic and ferromagnetic states". ?2009-10-02?"Gabriele Sclauzero" ??? > > >duchl06 wrote: >> thanks every one's reply! >> hi, Hui Wang >> the example you supplied is a calculation for a antiferromagnetic >> material, i mean how to set the start magnetic moment for the >> diamagnetic material, > >Diamagnetic material in absence of an external magnetic field should have no net magnetic >moment per atom, if I'm not wrong. So it is not clear to me what do you want to study and >if all these suggestions are pointing you to the correct solution. If you want to study a >diamagnetic material in absence of magnetic field there should be no need for a >spin-polarized calculation. >So, just to agree on the starting point of your question: what you mean with diamagnetic >material is really what people usually refer to? Can you give us more details on the >system you want to address? What properties do you want to study? ... > >Regards, > > >GS > > >> >> >> >> best regards! >> >> >> graduate student: chen >> xiaguanying university >> > >-- > > >o ------------------------------------------------ o >| Gabriele Sclauzero, PhD Student | >| c/o: SISSA & CNR-INFM Democritos, | >| via Beirut 2-4, 34014 Trieste (Italy) | >| email: sclauzer at sissa.it | >| phone: +39 040 3787 511 | >| skype: gurlonotturno | >o ------------------------------------------------ o >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/b631d3cf/attachment.htm From sclauzer at sissa.it Fri Oct 2 14:59:57 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 02 Oct 2009 14:59:57 +0200 Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <18823439.239381254485412186.JavaMail.coremail@bj163app79.163.com> References: <4AC5A597.5090804@sissa.it> <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> <18823439.239381254485412186.JavaMail.coremail@bj163app79.163.com> Message-ID: <4AC5F94D.10307@sissa.it> duchl06 wrote: > hi,Gabriele Sclauzero > thanks for your advice and hoping your further guide. > i am reading a paper "Charge-Transfer-Based Mechanism for > Half-Metallicity and Ferromagnetism in one-Dimensional > OrganometallicSandwich Molecular Wires" > http://pubs.acs.org/doi/abs/10.1021/ja804053a > in this paper they consider a sandwich molecular wires (such as 1D > (VBz)? wire, (MnCp)? wire, Bz is benzene, Cp** is cyclopentadienyl), and > they said "Furthermore, we investigate the effect of doubling the unit > cell size along the SMW direction to evaluate the ground state of > ferromagnetic (FM), antiferromagnetic (AFM), and diamagnetic (DM) > con?gurations in our spin-polarized calculations with structural > optimization, In this paper, I think they refer to a non-magnetic ground-state as "DM configuration". A priori you do not know if the DFT ground state (which does not necessarily always correspond to the true physical GS) will be FM, AFM or non-magnetic. You have to allow the system to "explore" an enlarged phase space with non-zero magnetization and see if the energy of a magnetic configuration (either FM or AFM) is lower than the non-magnetic solution. Actually, if the GS is non-magnetic, you should end to the non-magnetic GS even if you start from a magnetic configuration (unless you get trapped in some local minimum). I think that's why \Delta E in table 1 of the paper you pointed out is always negative (or not given): if the GS is magnetic, it will have a lower energy than the lowest energy non-magnetic solution, otherwise the GS is not magnetic and a spin-polarized calculation will give the same GS of a non-spin polarized one. HTH GS > and assess possibility of the Peierls transition." > > > (and i also found in some other papers when they do not know the groud > state is FM or DM, they would always say "To examine the stability of > the local magnetic state, we calculate the energy difference between the > diamagnetic and ferromagnetic states". > > ?2009-10-02?"Gabriele Sclauzero" ??? >> >> >>duchl06 wrote: >>> thanks every one's reply! >>> hi, Hui Wang >>> the example you supplied is a calculation for a antiferromagnetic >>> material, i mean how to set the start magnetic moment for the >>> diamagnetic material, >> >>Diamagnetic material in absence of an external magnetic field should have no net magnetic >>moment per atom, if I'm not wrong. So it is not clear to me what do you want to study and >>if all these suggestions are pointing you to the correct solution. If you want to study a >>diamagnetic material in absence of magnetic field there should be no need for a >>spin-polarized calculation. >>So, just to agree on the starting point of your question: what you mean with diamagnetic >>material is really what people usually refer to? Can you give us more details on the >>system you want to address? What properties do you want to study? ... >> >>Regards, >> >> >>GS >> -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From nkxirainbow at gmail.com Fri Oct 2 15:04:53 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Fri, 2 Oct 2009 21:04:53 +0800 Subject: [Pw_forum] pwcond with spin-orbital effect In-Reply-To: <4AC5A2F4.20106@sissa.it> References: <21fbc4790910012016s44ce0ddet6a722188204e5e21@mail.gmail.com> <4AC5A2F4.20106@sissa.it> Message-ID: <21fbc4790910020604s7e9d1252me4639ce3b78e23d2@mail.gmail.com> Dear Gabriele Sclauzero: It is very kind of you to give me so detailed explanation ?? Thanks again ?? On Fri, Oct 2, 2009 at 2:51 PM, Gabriele Sclauzero wrote: > > > xirainbow wrote: > > Dear all? > > > > I know the example 22 give a transmission with spin-orbital effect. > > > > Is the spin-orbital effect considered as perturbation? > > No, it is not correct to say this in my point of view. The SO effect is > included through > the use of so-called fully-relativistic pseudo potentials, as it is done in > PWscf (see > Andrea Dal Corso and Adriano Mosca Conte, Phys. Rev. B 71 115106 (2005) for > details). > > > Does the pwcond.x use "Green function" method? > > No, it uses a scattering state approach, which integrates numerically a > scattering > equation in real space along the direction of transport (while in the > perpendicular > directions the PW basis of the PWscf calculation is retained). > See: > Hyoung Joon Choi and Jisoon Ihm, Phys. Rev. B 59, 2267 (1999) > > for the method originally developed for norm-conserving PPs and: > Andrea Dal Corso, Alexander Smogunov, and Erio Tosatti, Phys. Rev. B 74, > 045429 (2006) > > for the extension to US-PPs with spin-orbit coupling. > > > Can the pwcond.x deal with "voltage bias" condition? > > No, it is built within the Landauer-Buttiker linear response theory, which > gives the low > bias limit of the conductance. > > Regards, > > GS > > > > > ____________________________________ > Hui Wang > School of physics, Nankai University, Tianjin, China > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/3a09bb61/attachment.htm From duchl06 at 163.com Fri Oct 2 16:10:39 2009 From: duchl06 at 163.com (duchl06) Date: Fri, 2 Oct 2009 22:10:39 +0800 (CST) Subject: [Pw_forum] how to build cell for diamagnetic material? In-Reply-To: <4AC5F94D.10307@sissa.it> References: <4AC5F94D.10307@sissa.it> <4AC5A597.5090804@sissa.it> <21fbc4790910011941t4d2cd9dlf1819309fe4106eb@mail.gmail.com> <1175614.75251254420233978.JavaMail.coremail@bj163app36.163.com> <27999742.165361254452249345.JavaMail.coremail@app165.163.com> <18823439.239381254485412186.JavaMail.coremail@bj163app79.163.com> Message-ID: <786061.380231254492639434.JavaMail.coremail@app188.163.com> thanks for everyone's reply, especially Gabriele Sclauzero's detailed explanation. best regards! graduate student: chen xiaguanying university ?2009-10-02?"Gabriele Sclauzero" ??? > > >duchl06 wrote: >> hi,Gabriele Sclauzero >> thanks for your advice and hoping your further guide. >> i am reading a paper "Charge-Transfer-Based Mechanism for >> Half-Metallicity and Ferromagnetism in one-Dimensional >> OrganometallicSandwich Molecular Wires" >> http://pubs.acs.org/doi/abs/10.1021/ja804053a >> in this paper they consider a sandwich molecular wires (such as 1D >> (VBz)? wire, (MnCp)? wire, Bz is benzene, Cp** is cyclopentadienyl), and >> they said "Furthermore, we investigate the effect of doubling the unit >> cell size along the SMW direction to evaluate the ground state of >> ferromagnetic (FM), antiferromagnetic (AFM), and diamagnetic (DM) >> con?gurations in our spin-polarized calculations with structural >> optimization, > >In this paper, I think they refer to a non-magnetic ground-state as "DM configuration". >A priori you do not know if the DFT ground state (which does not necessarily always >correspond to the true physical GS) will be FM, AFM or non-magnetic. You have to allow the >system to "explore" an enlarged phase space with non-zero magnetization and see if the >energy of a magnetic configuration (either FM or AFM) is lower than the non-magnetic >solution. >Actually, if the GS is non-magnetic, you should end to the non-magnetic GS even if you >start from a magnetic configuration (unless you get trapped in some local minimum). >I think that's why \Delta E in table 1 of the paper you pointed out is always negative (or >not given): if the GS is magnetic, it will have a lower energy than the lowest energy >non-magnetic solution, otherwise the GS is not magnetic and a spin-polarized calculation >will give the same GS of a non-spin polarized one. > >HTH > >GS > > > >> and assess possibility of the Peierls transition." >> >> >> (and i also found in some other papers when they do not know the groud >> state is FM or DM, they would always say "To examine the stability of >> the local magnetic state, we calculate the energy difference between the >> diamagnetic and ferromagnetic states". >> >> ?2009-10-02?"Gabriele Sclauzero" ??? >>> >>> >>>duchl06 wrote: >>>> thanks every one's reply! >>>> hi, Hui Wang >>>> the example you supplied is a calculation for a antiferromagnetic >>>> material, i mean how to set the start magnetic moment for the >>>> diamagnetic material, >>> >>>Diamagnetic material in absence of an external magnetic field should have no net magnetic >>>moment per atom, if I'm not wrong. So it is not clear to me what do you want to study and >>>if all these suggestions are pointing you to the correct solution. If you want to study a >>>diamagnetic material in absence of magnetic field there should be no need for a >>>spin-polarized calculation. >>>So, just to agree on the starting point of your question: what you mean with diamagnetic >>>material is really what people usually refer to? Can you give us more details on the >>>system you want to address? What properties do you want to study? ... >>> >>>Regards, >>> >>> >>>GS >>> > >-- > > >o ------------------------------------------------ o >| Gabriele Sclauzero, PhD Student | >| c/o: SISSA & CNR-INFM Democritos, | >| via Beirut 2-4, 34014 Trieste (Italy) | >| email: sclauzer at sissa.it | >| phone: +39 040 3787 511 | >| skype: gurlonotturno | >o ------------------------------------------------ o >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091002/e8795759/attachment.htm From miguel.martinez at ehu.es Sat Oct 3 17:18:55 2009 From: miguel.martinez at ehu.es (Miguel =?UTF-8?B?TWFydMOtbmV6?=) Date: Sat, 3 Oct 2009 17:18:55 +0200 Subject: [Pw_forum] Error building q-e 4.1 with Ubuntu 9.10 beta Message-ID: <20091003171855.0a7b3549@lcpybm> Fellow q-e users, Given the increased strictness seen by recent gcc versions, I tried to compile quantum-espresso 4.1 in ubuntu 9.10 beta. Yes, I know it's a development version and that nothing is guaranteed, but it's also a good way to see improvements in the 4.4 branch of gcc. However, when using gfortran and gcc 4.4, with or without atlas and lapack, with or without openmpi, the build fails. That is after a successful configure. I'll put the build error later. First I thought that gcc 4.4 was to blame, but the 4.4 branch compiles q-e successfully in a Debian system. The last lines of compilation are: http://pastebin.org/33895 It seems that if I comment out the test regarding HAVE_MALLINFO in configure (lines 4085 to 4090) and reconfigure, the compilation will produce useful binaries. Does anybody know what's happening here? Is this (another) ubuntu bug? I'll also put the configuration log for anyone interested (without the fix): http://pastebin.org/33900 Thanks for your comments, Miguel PS: I hope nobody's bothered by the pastebin links (they expire in one month). It seemed somewhat unpolite to send 40k of plain text most people won't be interested in. -- ---------------------------------------- Miguel Mart?nez Canales Condensed Matter Physics Dpt. UPV/EHU Faculty of Science and Technology Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5326 ---------------------------------------- "If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas." George Bernard Shaw From giannozz at democritos.it Sat Oct 3 17:42:40 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 3 Oct 2009 17:42:40 +0200 Subject: [Pw_forum] Error building q-e 4.1 with Ubuntu 9.10 beta In-Reply-To: <20091003171855.0a7b3549@lcpybm> References: <20091003171855.0a7b3549@lcpybm> Message-ID: <6CA082B5-781A-43DE-B3EE-BAE19D3A1796@democritos.it> On Oct 3, 2009, at 17:18 , Miguel Mart?nez wrote: > It seems that if I comment out the test regarding HAVE_MALLINFO in > configure (lines 4085 to 4090) and reconfigure, the compilation will > produce useful binaries. Does anybody know what's happening here? I don't. Since configure finds the struct mallinfo.arena (whatever it means): configure:3982: checking for struct mallinfo.arena ... configure:4083: result: yes it is correct to define HAVE_MALLINFO. If you #undef-ine it, however, nothing serious happens: you do not have the info on memory usage. Just remove it from include/c_defs.h (produced by configure) P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From sunshoutian168 at yahoo.com.cn Mon Oct 5 03:44:39 2009 From: sunshoutian168 at yahoo.com.cn (shoutian sun) Date: Mon, 5 Oct 2009 09:44:39 +0800 (CST) Subject: [Pw_forum] (no subject) Message-ID: <961088.44126.qm@web15302.mail.cnb.yahoo.com> Dear all, I want to calculate IR spectra(without external electronic field), should I have to set: tefield = .TRUE. ? Or could anyone tell me how to use cp.x to calculate IR? And which paprameters should I have to care about? Thank you in advance, SUN ST ___________________________________________________________ ????????????????? http://card.mail.cn.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091005/f987eabb/attachment.htm From sunshoutian168 at yahoo.com.cn Mon Oct 5 03:52:57 2009 From: sunshoutian168 at yahoo.com.cn (shoutian sun) Date: Mon, 5 Oct 2009 09:52:57 +0800 (CST) Subject: [Pw_forum] IR calculation Message-ID: <795320.93369.qm@web15307.mail.cnb.yahoo.com> >Dear all, > >I want to calculate IR spectra(without external electronic field), should I have to set: tefield = .TRUE. ? >Or could anyone tell me how to use cp.x to calculate IR? And which paprameters should I have to >care about? > >Thank you in advance, >SUN ST > ___________________________________________________________ ????????????????? http://card.mail.cn.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091005/67fd6dd4/attachment.htm From baroni at sissa.it Mon Oct 5 09:06:54 2009 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 5 Oct 2009 09:06:54 +0200 Subject: [Pw_forum] (no subject) In-Reply-To: <961088.44126.qm@web15302.mail.cnb.yahoo.com> References: <961088.44126.qm@web15302.mail.cnb.yahoo.com> Message-ID: <86619BC0-1F05-4628-98A7-49EA34BDEB51@sissa.it> Dear Sun (please provide your affiliation). IR spectra can be obtained from the Born effective charges which are calculated by the phonon code. SB On Oct 5, 2009, at 3:44 AM, shoutian sun wrote: > Dear all, > > I want to calculate IR spectra(without external electronic field), > should I have to set: tefield = .TRUE. ? > Or could anyone tell me how to use cp.x to calculate IR? And which > paprameters should I have to > care about? > > Thank you in advance, > SUN ST > > > > ????????????????? > _______________________________________________ > 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/20091005/3f40c059/attachment.htm From carlo.nervi at unito.it Mon Oct 5 11:20:27 2009 From: carlo.nervi at unito.it (Carlo Nervi) Date: Mon, 05 Oct 2009 11:20:27 +0200 Subject: [Pw_forum] openmpi 1.3.3 Message-ID: <4AC9BA5B.2000601@unito.it> Dear forum users, I tried to compile QE (snapshot 25-09-09 version) under linux with ifort 11.0 and MKL libraries 10.1.1.019 using OpenMPI 1.3.3 As far as I read OpenMPI 1.3 has improved performances over the previous versions... I (apparently) successfully compiled OpenMPI with Intel ifort (the hello.World program run on all 8 core of my linux pc). If I run pw.x in serial mode (OMP_NUM_THREADS=1) I got the correct results, but if I try to run pw.x using OMP_NUM_THREADS=8 either it runs forever (no convergence) reporting erroneous results, or it crash. If I run it by "mpirun -np 8 pw.x -in scf.it > scf.out &" (using OMP_NUM_THREAD=1), i get always the following message: "MPI_ABORT was invoked on rank 5 in communicator MPI_COMM_WORLD" I found the following url: http://software.intel.com/en-us/forums/intel-math-kernel-library/topic/68207/ where they explain that "...Since Open MPI considers MPI_COMM_WORLD to be a pointer it turns out to be 64-bit long. Whereas Cluster FFT was designed in times where sizeof(MPI_Comm) used to be 32-bit. In order to work correctly with Open MPI you just need to wrap the communicator as follows: DftiCreateDescriptorDM(MPI_Comm_c2f(MPI_COMM_WORLD),&desc,DFTI_DOUBLE,DFTI_COMPLEX,1,len);" I don't have enough experience to understand if this could be really relevant to the present topic, could be useful to the whole QE community that would like to compile QE using ifort and MKL, or it is totally irrelevant and unuseful... Somebody have tried to successfully use OpenMPI v 1.3.3? Maybe with ifort and MKL? I would greatly appreciate any comments. Thanks! Carlo -- ------------------------------------------------------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ From Giovanni.Cantele at na.infn.it Mon Oct 5 11:50:25 2009 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Mon, 05 Oct 2009 11:50:25 +0200 Subject: [Pw_forum] openmpi 1.3.3 In-Reply-To: <4AC9BA5B.2000601@unito.it> References: <4AC9BA5B.2000601@unito.it> Message-ID: <4AC9C161.9030802@na.infn.it> Carlo Nervi wrote: > Dear forum users, > I tried to compile QE (snapshot 25-09-09 version) under linux with ifort > 11.0 and MKL libraries 10.1.1.019 using OpenMPI 1.3.3 > As far as I read OpenMPI 1.3 has improved performances over the previous > versions... > I (apparently) successfully compiled OpenMPI with Intel ifort (the > hello.World program run on all 8 core of my linux pc). > > If I run pw.x in serial mode (OMP_NUM_THREADS=1) I got the correct > results, but if I try to run pw.x using OMP_NUM_THREADS=8 either it runs > forever (no convergence) reporting erroneous results, or it crash. > If I run it by "mpirun -np 8 pw.x -in scf.it > scf.out &" (using > OMP_NUM_THREAD=1), i get always the following message: > "MPI_ABORT was invoked on rank 5 in communicator MPI_COMM_WORLD" > > I found the following url: > http://software.intel.com/en-us/forums/intel-math-kernel-library/topic/68207/ > where they explain that > "...Since Open MPI considers MPI_COMM_WORLD to be a pointer it turns out > to be 64-bit long. Whereas Cluster FFT was designed in times where > sizeof(MPI_Comm) used to be 32-bit. In order to work correctly with Open > MPI you just need to wrap the communicator as follows: > DftiCreateDescriptorDM(MPI_Comm_c2f(MPI_COMM_WORLD),&desc,DFTI_DOUBLE,DFTI_COMPLEX,1,len);" > > I don't have enough experience to understand if this could be really > relevant to the present topic, could be useful to the whole QE community > that would like to compile QE using ifort and MKL, or it is totally > irrelevant and unuseful... > > Somebody have tried to successfully use OpenMPI v 1.3.3? > Maybe with ifort and MKL? > I would greatly appreciate any comments. Thanks! > Carlo > Hi, I succesfully run (but always with OMP_NUM_THREADS=1) QE from 3.2.3 to 4.1 under: ifort (IFORT) 10.1 20080112 MKL 10.0.1.014 mpirun (Open MPI) 1.3.2 OpenMPI (not sure whether this is relevant!) was configured using: ./configure CC=icc F77=ifort F90=ifort FC=ifort CXX=icpc OBJC=icc FCFLAGS=-i-dynamic CFLAGS=-i-dynamic CXXFLAGS=-i-dynamic --with-tm=/usr/local --prefix=/opt/openmpi/1.3.2/ifort (--with-tm is irrelevant, used only if you want support for PBS/Torque) I don't know if switching from 1.3.2 to 1.3.3 might give problems. However, I previously installed also openmpi-1.2.5, and even in that case it worked. Regarding the errors under OMP_NUM_THREADS=8 : did you follow these suggestions (found in a previous thread on this forum)? DFLAGS=...-D__OPENMP _D__FFTW + ifort -openmp + mkl is tested and safe; other combinations may still run into trouble due to conflicts between "internal" OpenMP and autothreading libraries. I've not experience with using OMP_NUM_THREADS>1, though. Hope this helps, Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it Skype contact: giocan74 From giannozz at democritos.it Mon Oct 5 11:59:38 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 05 Oct 2009 11:59:38 +0200 Subject: [Pw_forum] openmpi 1.3.3 In-Reply-To: <4AC9BA5B.2000601@unito.it> References: <4AC9BA5B.2000601@unito.it> Message-ID: <4AC9C38A.9030305@democritos.it> Carlo, OpenMPI /= OpenMP (or, if you prefer C syntax: OpenMPI != OpenMP). OMP_NUM_THREADs is for the latter; mpirun -np N for the former. When you link MKL, by default you link the multithreaded (OpenMP) version. In order to avoid conflicts with MPI, you should set OMP_NUM_THREADS to 1. You can use both parallelization paradigms, but only in some special cases (large multicore parallel machines) and if you have control of where the MPI processes run and where OpenMP threads are spawned. In all other cases, stay clear of mixed MPI-OpenMP parallelization. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From carlo.nervi at unito.it Mon Oct 5 16:04:52 2009 From: carlo.nervi at unito.it (Carlo Nervi) Date: Mon, 05 Oct 2009 16:04:52 +0200 Subject: [Pw_forum] openmpi 1.3.3 In-Reply-To: <4AC9C161.9030802@na.infn.it> References: <4AC9BA5B.2000601@unito.it> <4AC9C161.9030802@na.infn.it> Message-ID: <4AC9FD04.9010405@unito.it> Hi, sorry to bother again. I recompiled OpenMPI with the suggested options, but I still have problems. I think I have something wrong in my make.sys file. First I was trying to use fftw3xf_intel libs, but ... are they incompatible with MPI? In a previous forum post I read that user were unable to use Intel FFTW3 + OpenMPI, so i switched to the internal FFTW. Paolo, thank you for the message. I know that OpenMPI and OpenMP are different, and after your suggestion I tried to recompile all (after a make clean) using only OpenMPI and do not mix them: ./configure CC=icc F77=ifort F90=ifort FC=ifort CPP=icpc FFLAGS=-xHost FCFLAGS=-xHost CFLAGS=-xHost CPPFLAGS=-xHost (the -xHost tell to Intel compiler to optimize the code for the current CPU), and I got the following part of make.sys: DFLAGS = -D__INTEL -D__FFTW -D__MPI -D__PARA IFLAGS = -I../include MOD_FLAG = -I MPIF90 = mpif90 #F90 = ifort CC = icc F77 = ifort CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) CFLAGS = -O3 -xHost $(DFLAGS) $(IFLAGS) F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) $(IFLAGS) $(MODFLAGS) FFLAGS = -O2 -xHost -assume byterecl FFLAGS_NOOPT = -O0 -assume byterecl -xHost LD = mpif90 LDFLAGS = -static-intel -openmp LD_LIBS = BLAS_LIBS = -lmkl_em64t FLIB_TARGETS = all LAPACK_LIBS = -lmkl_em64t SCALAPACK_LIBS = The -openmp (and -parallel ?) direction to ifort should be used or not? As far as i understood only the -openmp should be passed to the linker. And only in the case I'd like to use OpenMP I should use the -openmp directive for ifort compiler. Is it? I tryed also to use "-lmkl_intel_lp64 -lmkl_sequential -lmkl_core" instead of "-lmkl_em64t": no success... Thank you, Carlo -- ------------------------------------------------------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ From giannozz at democritos.it Mon Oct 5 16:40:30 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 05 Oct 2009 16:40:30 +0200 Subject: [Pw_forum] openmpi 1.3.3 In-Reply-To: <4AC9FD04.9010405@unito.it> References: <4AC9BA5B.2000601@unito.it> <4AC9C161.9030802@na.infn.it> <4AC9FD04.9010405@unito.it> Message-ID: <4ACA055E.3020708@democritos.it> Carlo Nervi wrote: > I think I have something wrong in my make.sys file. I think you have something wrong in your MPI libraries. Just in case, try the stable version of QE as well (for the very unlikely event that some recent change in the cvs version broke something in the parallelization) > First I was trying to use fftw3xf_intel libs, but ... are they > incompatible with MPI? no, they aren't > ./configure CC=icc F77=ifort F90=ifort FC=ifort CPP=icpc FFLAGS=-xHost > FCFLAGS=-xHost CFLAGS=-xHost CPPFLAGS=-xHost if you have a properly installed system, you should use "./configure" and nothing else > As far as i understood only the -openmp should be passed to the linker. > And only in the case I'd like to use OpenMP I should use the -openmp > directive for ifort compiler. Is it? it is. > I tryed also to use "-lmkl_intel_lp64 -lmkl_sequential -lmkl_core" > instead of "-lmkl_em64t": no success... how to link the non-openMP version of MKL should be explained in the documentation of MKL. In any case, linking "plain vanilla" MKL and using it together with MPI is perfectly fine, as long as you do not try to use multithreading. Anyway: QE compiles without any external library apart MPI. Compile it without anything apart MPI and see what happens Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From siyouber at yahoo.fr Mon Oct 5 17:50:36 2009 From: siyouber at yahoo.fr (Bertrand SITAMTZE) Date: Mon, 5 Oct 2009 15:50:36 +0000 (GMT) Subject: [Pw_forum] Band structure calculation Message-ID: <676336.7383.qm@web26505.mail.ukl.yahoo.com> Hi all, I which to calculate the Band structure for an ionic conductor. But it seems as there is something wrong.The structure has 222 electrons and 133 kohn-sham states. The problem is that I dont know if I have to indicate the number of bounds (nbnd) during the band calculation. According to me, I should allow it and the code will consider it as the half of electrons(Which could be very high in this case). But when I do? that, the code runs until the line: "Band structure calculation" and continue running without writing anything Please can somebody give me some hint about this? *********************************** Bertrand SITAMTZE YOUMBI Laboratory of material Sciences Department of Physics University of Yaound? I-Cameroon PO Box 812 Yaound?-Cameroon ***************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091005/e6021f9d/attachment.htm From paulatto at sissa.it Mon Oct 5 18:04:16 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 05 Oct 2009 18:04:16 +0200 Subject: [Pw_forum] Band structure calculation In-Reply-To: <676336.7383.qm@web26505.mail.ukl.yahoo.com> References: <676336.7383.qm@web26505.mail.ukl.yahoo.com> Message-ID: In data 05 ottobre 2009 alle ore 17:50:36, Bertrand SITAMTZE ha scritto: > I which to calculate the Band structure for an ionic conductor. But it > seems as there is something wrong.The structure has 222 electrons and > 133 kohn-sham states. The problem is that I dont know if I have to > indicate the number of bounds (nbnd) during the band calculation. Dear Bertrand, the band structure calculation is a non-self-consistent calculation, where the ground-state charge density is used to compute the Hamiltonian eigenvalues (i.e. the bands) at the k-points you request. If you don't specify the number of *bands* using the nbnd keyword the default value will be used, its exact value depend on a few factors which you can find in the documentation. > According to me, I should allow it and the code will consider it as the > half of electrons(Which could be very high in this case). But when I do > that, the code runs until the line: > > "Band structure calculation" > > and continue running without writing anything > > Please can somebody give me some hint about this? Maybe the calculation is just slower than you expected! The current implementation of non-self-consistent calculation is not extremely efficient, although there are plans to improve it. Normally you can expect it to take as long as the first self-consistent iteration (not a full cycle, just the initial iteration). If you increase the number of bands the required time will increase with its third power. Furthermore it will scale linearly with the number of k-points that describe the k-space path on which you want the bands to be computed. Finally, the empty bands are diagonalized to a higher precision in the band calculation than in the scf calculation, if you have many empty bands it will take even longer. Maybe a small increasing of these factor is enough to fill your system RAM memory and force the operating system to swap to disk, this would decrease performance dramatically! Please have a look at your setup: how many bands and k-points have you requested in the band calculation with respect to the scf calculation? best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From jjr19 at zips.uakron.edu Mon Oct 5 18:54:25 2009 From: jjr19 at zips.uakron.edu (James J Ramsey) Date: Mon, 5 Oct 2009 12:54:25 -0400 Subject: [Pw_forum] Error building q-e 4.1 with Ubuntu 9.10 beta In-Reply-To: <20091003171855.0a7b3549@lcpybm> References: <20091003171855.0a7b3549@lcpybm> Message-ID: <46e014700910050954w70b4e82ax3fd36cc380bdb511@mail.gmail.com> 2009/10/3 Miguel Mart?nez : > It seems that if I comment out the test regarding HAVE_MALLINFO in > configure (lines 4085 to 4090) and reconfigure, the compilation will > produce useful binaries. Does anybody know what's happening here? Is > this (another) ubuntu bug? The problem seems to happen in memstat.c, and there's one part of it that looks fishy to me: void F77_FUNC(memstat,MEMSTAT)(int *kilobytes) { #if defined (__SVR4) && defined (__sun) #define SUN_MALLINFO #endif #if defined(HAVE_MALLINFO) && !defined(__QK_USER__) && !defined(SUN__MALLINFO) #include struct mallinfo info; info = mallinfo(); #if defined(__AIX) *kilobytes = (info.arena) / 1024 ; #else *kilobytes = (info.arena + info.hblkhd) / 1024 ; #endif #else *kilobytes = -1; #endif } The #include statement is *inside* a function, and while I'm no expert in C, I think that explains the error messages about nested functions declared "extern". I noticed, for example, that in my copy of /usr/include/bits/stdio.h on Ubuntu 8.04, the getchar definition looks like this: /* Read a character from stdin. */ __STDIO_INLINE int getchar (void) { return _IO_getc (stdin); } and the macro __STDIO_INLINE is defined as follows: #ifndef __extern_inline # define __STDIO_INLINE inline #else # define __STDIO_INLINE __extern_inline #endif My guess, but one that could easily be wrong, is that the #include statement inside the function was tolerated before because the macro __STDIO_INLINE was expanded to "inline," but now the function ends up declared "extern". In any case, judging from a post on the gcl-devel list (http://www.mail-archive.com/gcl-devel at gnu.org/msg02038.html), that #include statement inside the function was a time bomb. From the post: Then GCL is buggy. Standard headers are never meant to be included inside of a function. E.g. ISO C99 says in 7.1.2/4: ... "If used, a header shall be included outside of any external declaration or definition, and it shall first be included before the first reference to any of the functions or objects it declares, or to any of the types or macros it defines."... I'd say that memstat.c needs to be fixed. University of Akron Computer Modeling & Simulation Civil Engineering Dept. From paulatto at sissa.it Mon Oct 5 19:16:27 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 05 Oct 2009 19:16:27 +0200 Subject: [Pw_forum] Error building q-e 4.1 with Ubuntu 9.10 beta In-Reply-To: <46e014700910050954w70b4e82ax3fd36cc380bdb511@mail.gmail.com> References: <20091003171855.0a7b3549@lcpybm> <46e014700910050954w70b4e82ax3fd36cc380bdb511@mail.gmail.com> Message-ID: In data 05 ottobre 2009 alle ore 18:54:25, James J Ramsey ha scritto: > The #include statement is *inside* a function, and while I'm no expert > in C, I think that explains the error messages about nested functions > declared "extern". I think you are right, I've moved the #include outside the function (i.e. just under the other include) on my copy QE and it *still* works, could you please try to see if it solves the problem in ubuntu 9.04? best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From jjr19 at zips.uakron.edu Mon Oct 5 21:17:20 2009 From: jjr19 at zips.uakron.edu (James J Ramsey) Date: Mon, 5 Oct 2009 15:17:20 -0400 Subject: [Pw_forum] Error building q-e 4.1 with Ubuntu 9.10 beta In-Reply-To: References: <20091003171855.0a7b3549@lcpybm> <46e014700910050954w70b4e82ax3fd36cc380bdb511@mail.gmail.com> Message-ID: <46e014700910051217g1ca2a841u6ec0c0bd3d036bcd@mail.gmail.com> On Mon, Oct 5, 2009 at 1:16 PM, Lorenzo Paulatto wrote: > In data 05 ottobre 2009 alle ore 18:54:25, James J Ramsey > ha scritto: >> The #include statement is *inside* a function, and while I'm no expert >> in C, I think that explains the error messages about nested functions >> declared "extern". > > I think you are right, I've moved the #include outside the function (i.e. > just under the other include) on my copy QE and it *still* works, could > you please try to see if it solves the problem in ubuntu 9.04? I have Ubuntu 8.04, and my compiler appears to be more forgiving that Mr. Mart?nez, and QE compiles fine for me. I don't think I can help in the testing. From baroni at sissa.it Mon Oct 5 21:41:05 2009 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 5 Oct 2009 21:41:05 +0200 Subject: [Pw_forum] Band structure calculation In-Reply-To: <676336.7383.qm@web26505.mail.ukl.yahoo.com> References: <676336.7383.qm@web26505.mail.ukl.yahoo.com> Message-ID: <62BFC5BB-6FA5-4688-A1CA-FFA7136D75DD@sissa.it> why don't you try to practice a little bit with a simple system, and learn therein the tricks of the trade? SB On Oct 5, 2009, at 5:50 PM, Bertrand SITAMTZE wrote: > Hi all, > > I which to calculate the Band structure for an ionic conductor. But > it seems as there is something wrong.The structure has 222 electrons > and 133 kohn-sham states. The problem is that I dont know if I have > to indicate the number of bounds (nbnd) during the band calculation. > According to me, I should allow it and the code will consider it as > the half of electrons(Which could be very high in this case). But > when I do that, the code runs until the line: > > "Band structure calculation" > > and continue running without writing anything > > Please can somebody give me some hint about this? > > *********************************** > Bertrand SITAMTZE YOUMBI > Laboratory of material Sciences > Department of Physics > University of Yaound? I-Cameroon > PO Box 812 Yaound?-Cameroon > ***************************** > > > > > > _______________________________________________ > 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/20091005/baa6bff7/attachment.htm From sunshoutian168 at yahoo.com.cn Tue Oct 6 02:44:02 2009 From: sunshoutian168 at yahoo.com.cn (shoutian sun) Date: Tue, 6 Oct 2009 08:44:02 +0800 (CST) Subject: [Pw_forum] IR Message-ID: <361096.56850.qm@web15305.mail.cnb.yahoo.com> Dear all, I want to calculate IR spectrum with cp MD (cp.x) at the temperature of 100K(without elec-field). And I just make a change the input file of example30 as follows. But in the ourput file the temperature value: tempp changed every step. Can anyone tell me how can I calculate IR with cp MD. Thanks in advance. >&CONTROL > calculation = 'cp', > restart_mode = 'from_scratch', >nstep = 10000, > iprint = 100, > isave = 100, > tstress = .FALSE., >tprnfor = .TRUE., > tefield = .TRUE., > prefix = 'mgo', > dt=5., > pseudo_dir = '/ds13/zfliu/stsun/espresso-4.1/pseudo/', > outdir='/ds13/zfliu/stsun/espresso-4.1/examples/example30/results/tmp' >/ > >&SYSTEM > ibrav = 1, > celldm(1) = 15.912, > celldm(2) = 0.0, > celldm(3) = 0.0, > celldm(4) = 0.0, > celldm(5) = 0.0, > celldm(6) = 0.0, > nbnd=128, > nelec=256, > nat =64, > ntyp =2, > ecutwfc = 40, > ecutrho = 250, > nr1b= 30, nr2b = 30, nr3b = 30, > nspin=1, >/ > >&ELECTRONS > emass = 250.d0, > emass_cutoff = 3.d0, > electron_dynamics='damp', ortho_max=30 > electron_damping=0.05 > startingwfc = 'random', > ampre = 0.02, > efield = 0.0 > epol = 3 >/ > >&IONS > ion_dynamics = 'verlet', > ion_temperature = 'rescaling', > tempw = 100, >/ > >&CELL > cell_dynamics = 'none', >/ > >ATOMIC_SPECIES >O 1.0 O.pz-rrkjus.UPF >Mg 1.0 Mg.pz-n-vbc.UPF >ATOMIC_POSITIONS (bohr) >Mg 0.066300000 0.066300000 0.066300000 >........... ___________________________________________________________ ????????????????? http://card.mail.cn.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091006/10fc98c1/attachment.htm From vtmtrinh at caltech.edu Tue Oct 6 03:53:54 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Mon, 5 Oct 2009 18:53:54 -0700 (PDT) Subject: [Pw_forum] Information on new release Message-ID: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> Dear PWSCF Users, If I would like to know the information on what has been changed from the old versions to the new version of PWSCF, where should I look at. I searched around, but I still could not find it. Thank you very much, MyTrinh Vo JPL/CalTech From nkxirainbow at gmail.com Tue Oct 6 04:04:59 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Tue, 6 Oct 2009 02:04:59 +0000 Subject: [Pw_forum] Information on new release In-Reply-To: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> References: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> Message-ID: <21fbc4790910051904o3c14f183p259887bff6fd8bba@mail.gmail.com> Dear: I use 4.0.4 version. I can find the change iformation at ?\espresso-4.0.5\Doc\ChangeLog-4.0? file. I think you can find it in other version ?? On Tue, Oct 6, 2009 at 1:53 AM, wrote: > Dear PWSCF Users, > > If I would like to know the information on what has been changed from the > old versions to the new version of PWSCF, where should I look at. I > searched around, but I still could not find it. > > Thank you very much, > > MyTrinh Vo > JPL/CalTech > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091006/f8cc6475/attachment.htm From ttduyle at gmail.com Tue Oct 6 04:46:19 2009 From: ttduyle at gmail.com (Duy Le) Date: Mon, 5 Oct 2009 22:46:19 -0400 Subject: [Pw_forum] Information on new release In-Reply-To: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> References: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> Message-ID: <8974d3b20910051946o1c2a7677r9926a55df0e68f6e@mail.gmail.com> It should be in $ESPRESSO_DIR/Doc/Changlog.somthing.. -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. On Mon, Oct 5, 2009 at 9:53 PM, wrote: > Dear PWSCF Users, > > If I would like to know the information on what has been changed from the > old versions to the new version of PWSCF, where should I look at. I > searched around, but I still could not find it. > > Thank you very much, > > MyTrinh Vo > JPL/CalTech > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091005/602ba0ea/attachment-0001.htm From vtmtrinh at caltech.edu Tue Oct 6 06:25:14 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Mon, 5 Oct 2009 21:25:14 -0700 (PDT) Subject: [Pw_forum] Information on new release In-Reply-To: <8974d3b20910051946o1c2a7677r9926a55df0e68f6e@mail.gmail.com> References: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> <8974d3b20910051946o1c2a7677r9926a55df0e68f6e@mail.gmail.com> Message-ID: <4873.137.78.73.28.1254803114.squirrel@webmail.caltech.edu> Thanks, MyTrinh > It should be in $ESPRESSO_DIR/Doc/Changlog.somthing.. > -------------------------------------------------- > Duy Le > PhD Student > Department of Physics > University of Central Florida. > > > On Mon, Oct 5, 2009 at 9:53 PM, wrote: > >> Dear PWSCF Users, >> >> If I would like to know the information on what has been changed from >> the >> old versions to the new version of PWSCF, where should I look at. I >> searched around, but I still could not find it. >> >> Thank you very much, >> >> MyTrinh Vo >> JPL/CalTech >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From vtmtrinh at caltech.edu Tue Oct 6 06:33:09 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Mon, 5 Oct 2009 21:33:09 -0700 (PDT) Subject: [Pw_forum] Information on new release In-Reply-To: <21fbc4790910051904o3c14f183p259887bff6fd8bba@mail.gmail.com> References: <4686.137.78.73.28.1254794034.squirrel@webmail.caltech.edu> <21fbc4790910051904o3c14f183p259887bff6fd8bba@mail.gmail.com> Message-ID: <4904.137.78.73.28.1254803589.squirrel@webmail.caltech.edu> Thanks! (I looked around on pwscf website, and that was why could not find it :-) ). MyTrinh > Dear: > I use 4.0.4 version. > I can find the change iformation at > ???\espresso-4.0.5\Doc\ChangeLog-4.0??? > file. > I think you can find it in other version ?????? > > On Tue, Oct 6, 2009 at 1:53 AM, wrote: > >> Dear PWSCF Users, >> >> If I would like to know the information on what has been changed from >> the >> old versions to the new version of PWSCF, where should I look at. I >> searched around, but I still could not find it. >> >> Thank you very much, >> >> MyTrinh Vo >> JPL/CalTech >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > > -- > ____________________________________ > Hui Wang > School of physics, Nankai University, Tianjin, China > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From baroni at sissa.it Tue Oct 6 08:53:30 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 6 Oct 2009 08:53:30 +0200 Subject: [Pw_forum] IR In-Reply-To: <361096.56850.qm@web15305.mail.cnb.yahoo.com> References: <361096.56850.qm@web15305.mail.cnb.yahoo.com> Message-ID: Hi. Would you be so kind as to sign your posts, with your name and academic affiliation? MD conserves energy. It is natural that the temperature fluctuates. I am not an expert in MD, but I would also say that in order to calculate dynamical properties (such as the time correlations needed to get the IR spectrum) it is not a good idea to rescale the temperature, as "ion_temperature = 'rescaling' " seems to imply. Finally, it does not seem to me a great idea either to use MD to calculate the IR spectrum of MgO at such a low temperature. Even when when you'll have learnt all the tricks of MD, I would expect the dynamics to be very harmonic, and hence not ergodic, thus requiring long (and difficult to estimate) equilibration times. Why not using standard lattice dynamics? (vibrational frequencies + effective charges, both calculated bu "phonon"). Stefano Baroni On Oct 6, 2009, at 2:44 AM, shoutian sun wrote: > Dear all, > I want to calculate IR spectrum with cp MD (cp.x) at the temperature > of 100K(without elec-field). > And I just make a change the input file of example30 as follows. But > in the ourput file > the temperature value: tempp changed every step. Can anyone tell me > how can I > calculate IR with cp MD. Thanks in advance. > > >&CONTROL > > calculation = 'cp', > > restart_mode = 'from_scratch', > > nstep = 10000, > > iprint = 100, > > isave = 100, > > tstress = .FALSE., > >tprnfor = .TRUE., > > tefield = .TRUE., > > prefix = 'mgo', > > dt=5., > > pseudo_dir = '/ds13/zfliu/stsun/espresso-4.1/pseudo/', > > outdir='/ds13/zfliu/stsun/espresso-4.1/examples/example30/results/ > tmp' > >/ > > > >&SYSTEM > > ibrav = 1, > > celldm(1) = 15.912, > > celldm(2) = 0.0, > > celldm(3) = 0.0, > > celldm(4) = 0.0, > > celldm(5) = 0.0, > > celldm(6) = 0.0, > > nbnd=128, > > nelec=256, > > nat =64, > > ntyp =2, > > ecutwfc = 40, > > ecutrho = 250, > > nr1b= 30, nr2b = 30, nr3b = 30, > > nspin=1, > >/ > > > >&ELECTRONS > > emass = 250.d0, > > emass_cutoff = 3.d0, > > electron_dynamics='damp', ortho_max=30 > > electron_damping=0.05 > > startingwfc = 'random', > > ampre = 0.02, > > efield = 0.0 > > epol = 3 > >/ > > > >&IONS > > ion_dynamics = 'verlet', > > ion_temperature = 'rescaling', > > tempw = 100, > >/ > > > >&CELL > > cell_dynamics = 'none', > >/ > > > >ATOMIC_SPECIES > >O 1.0 O.pz-rrkjus.UPF > >Mg 1.0 Mg.pz-n-vbc.UPF > >ATOMIC_POSITIONS (bohr) > >Mg 0.066300000 0.066300000 0.066300000 > >........... > > ????????????????? > _______________________________________________ > 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 From somesh.kb at gmail.com Tue Oct 6 09:39:20 2009 From: somesh.kb at gmail.com (Somesh Kumar Bhattacharya) Date: Tue, 6 Oct 2009 09:39:20 +0200 Subject: [Pw_forum] IR In-Reply-To: <361096.56850.qm@web15305.mail.cnb.yahoo.com> References: <361096.56850.qm@web15305.mail.cnb.yahoo.com> Message-ID: Hi Shoutian Sun, With CPMD one can calculate the IR spectra by taking the fourier transform of the velocity autocorrelation function (VAF). For this, one has to run a very long simulation ( time scale few picoseconds) and then calculate the VAF. You can read a bit about VAF from any standard statistical mechanics book ( plenty of materials are available online). Also refer to the paper : J. Phys. Chem. A * 111*, 12810 (2007) regarding the expression for the fourier transform. Regarding your problem of tempp changing at every step, try using the nose thermostat. Regards, Somesh 2009/10/6 shoutian sun > Dear all, > I want to calculate IR spectrum with cp MD (cp.x) at the temperature of > 100K(without elec-field). > And I just make a change the input file of example30 as follows. But in the > ourput file > the temperature value: tempp changed every step. Can anyone tell me how can > I > calculate IR with cp MD. Thanks in advance. > > >&CONTROL > > calculation = 'cp', > > restart_mode = 'from_scratch', > > nstep = 10000, > > iprint = 100, > > isave = 100, > > tstress = .FALSE., > >tprnfor = .TRUE., > > tefield = .TRUE., > > prefix = 'mgo', > > dt=5., > > pseudo_dir = '/ds13/zfliu/stsun/espresso-4.1/pseudo/', > > outdir='/ds13/zfliu/stsun/espresso-4.1/examples/example30/results/tmp' > >/ > > > >&SYSTEM > > ibrav = 1, > > celldm(1) = 15.912, > > celldm(2) = 0.0, > > celldm(3) = 0.0, > > celldm(4) = 0.0, > > celldm(5) = 0.0, > > celldm(6) = 0.0, > > nbnd=128, > > nelec=256, > > nat =64, > > ntyp =2, > > ecutwfc = 40, > > ecutrho = 250, > > nr1b= 30, nr2b = 30, nr3b = 30, > > nspin=1, > >/ > > > >&ELECTRONS > > emass = 250.d0, > > emass_cutoff = 3.d0, > > electron_dynamics='damp', ortho_max=30 > > electron_damping=0.05 > > startingwfc = 'random', > > ampre = 0.02, > > efield = 0.0 > > epol = 3 > >/ > > > >&IONS > > ion_dynamics = 'verlet', > > ion_temperature = 'rescaling', > > tempw = 100, > >/ > > > >&CELL > > cell_dynamics = 'none', > >/ > > > >ATOMIC_SPECIES > >O 1.0 O.pz-rrkjus.UPF > >Mg 1.0 Mg.pz-n-vbc.UPF > >ATOMIC_POSITIONS (bohr) > >Mg 0.066300000 0.066300000 0.066300000 > >........... > > ------------------------------ > ????????????????? > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Somesh Kr. Bhattacharya Post Doctoral Fellow Room No. 263, Leonardo Building, The Abdus Salam International Centre for Theoretical Physics Strada Costiera, 11 I-34014 Trieste Italy Phone: +39-040-2240399 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091006/8a85d31c/attachment.htm From miguel.martinez at ehu.es Tue Oct 6 10:14:38 2009 From: miguel.martinez at ehu.es (Miguel =?UTF-8?B?TWFydMOtbmV6?=) Date: Tue, 6 Oct 2009 10:14:38 +0200 Subject: [Pw_forum] Error building q-e 4.1 with Ubuntu 9.10 beta In-Reply-To: References: <20091003171855.0a7b3549@lcpybm> <46e014700910050954w70b4e82ax3fd36cc380bdb511@mail.gmail.com> Message-ID: <20091006101438.74446311@lcpybm> Hello again, First, thanks for all the comments. And following the advice to move the include statement together with the other include allows q-e to build successfully on ubuntu 9.10. I've run a few examples and all seems fine. Please note that compiling under ubuntu 9.04 was fine. Back on topic, the code bits mentioned by James inside /usr/include/bits/stdio.h are exactly the same in my copy of that file under Ubuntu 9.10 and in a Debian Testing pc. I fear it's just a change in the strictness of gcc. Regards, Miguel On Mon, 05 Oct 2009 19:16:27 +0200 "Lorenzo Paulatto" wrote: > In data 05 ottobre 2009 alle ore 18:54:25, James J Ramsey > ha scritto: > > The #include statement is *inside* a function, and while I'm no > > expert in C, I think that explains the error messages about nested > > functions declared "extern". > > I think you are right, I've moved the #include outside the function > (i.e. just under the other include) on my copy QE and it *still* > works, could you please try to see if it solves the problem in ubuntu > 9.04? > > best regards > -- ---------------------------------------- Miguel Mart?nez Canales Condensed Matter Physics Dpt. UPV/EHU Faculty of Science and Technology Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5326 ---------------------------------------- "If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas." George Bernard Shaw From udayagiri3 at gmail.com Tue Oct 6 11:34:33 2009 From: udayagiri3 at gmail.com (udayagiri sai babu) Date: Tue, 6 Oct 2009 15:04:33 +0530 Subject: [Pw_forum] scf calculation of isolated atom Message-ID: Dear all I am performing a scf cacultion of Cr isolated atom. Below is my input file. I got an error saying *"strange occupations:number of electrons from occupations is wrong"*. I undestood like my occupation card is not matching with my "nbnd" value but i could not correct it can somebody give me some suggestion. 2) Can i write occupations card for chromium as 1 1 1 1 1 1 1 0 0 0 0 0 4s1 3d5. 3) usually what should be the nbnd value? -- &control calculation = 'scf', restart_mode='from_scratch', outdir='/home/sai/softwares/espresso-4.0.1/workspace/temp', prefix='chromium' pseudo_dir='/home/sai/softwares/espresso-4.0.1/pseudo/', / &system ibrav=1 celldm(1)=15, nat=1, ntyp=1 ecutwfc=50, ecutrho=500, occupations='from_input', nspin=2 starting_magnetization(1)=0.2, nbnd=13 nosym=.true., / &electrons conv_thr=1.D-8, diagonalization='david', mixing_mode='plain', mixing_beta= 0.3 electron_maxstep=10000 / ATOMIC_SPECIES Cr 51.9961 Cr.pw91-sp-van.UPF ATOMIC_POSITIONS (crystal) Cr 0.00000000 0.00000000 0.00000000 K_POINTS {automatic} 1 1 1 0 0 0 OCCUPATIONS 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 U.Saibabu PhD student, Deformation mechanisms modeling group, Materials engineering department, IISc Bangalore, India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091006/faa09bf9/attachment.htm From sclauzer at sissa.it Tue Oct 6 14:21:23 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 06 Oct 2009 14:21:23 +0200 Subject: [Pw_forum] scf calculation of isolated atom In-Reply-To: References: Message-ID: <4ACB3643.8070608@sissa.it> Dear Saibabu, udayagiri sai babu wrote: > Dear all > I am performing a scf cacultion of Cr isolated atom. Below is my input > file. I got an error saying *"strange occupations:number of electrons > from occupations is wrong"*. I undestood like my occupation card is not > matching with my "nbnd" value but i could not correct it can somebody > give me some suggestion. I tried your input file with the CVS version of QE and it works fine. It seems that the piece of code parsing the OCCUPATIONS input card has been rewritten in version 4.1. Please upgrade your code to the newest version and report to the forum if you find any problem. Thanks. > > 2) Can i write occupations card for chromium as > 1 1 1 1 1 1 > 1 0 0 0 0 0 > 4s1 3d5. Not like this, but the way you wrote them in the input file below is OK. You must specify as many rows as nspin and, for each row, as many occupation factors as nbnd. > > 3) usually what should be the nbnd value? For non-magnetic insulators it can be as low as nelec/2, while in this case you need to specify a higher value since your spin population is unbalanced (the code does not allow a different number of spin-up bands and spin-down bands). Cheers GS > > -- &control > calculation = 'scf', > restart_mode='from_scratch', > outdir='/home/sai/softwares/espresso-4.0.1/workspace/temp', > prefix='chromium' > pseudo_dir='/home/sai/softwares/espresso-4.0.1/pseudo/', > / > &system > ibrav=1 > celldm(1)=15, > nat=1, > ntyp=1 > ecutwfc=50, > ecutrho=500, > occupations='from_input', > nspin=2 > starting_magnetization(1)=0.2, > nbnd=13 > nosym=.true., > / > &electrons > conv_thr=1.D-8, > diagonalization='david', > mixing_mode='plain', > mixing_beta= 0.3 > electron_maxstep=10000 > / > ATOMIC_SPECIES > Cr 51.9961 Cr.pw91-sp-van.UPF > ATOMIC_POSITIONS (crystal) > Cr 0.00000000 0.00000000 0.00000000 > K_POINTS {automatic} > 1 1 1 0 0 0 > OCCUPATIONS > 1 1 1 1 1 1 1 1 1 1 0 0 0 > 1 1 1 1 0 0 0 0 0 0 0 0 0 > > U.Saibabu > PhD student, > Deformation mechanisms modeling group, > Materials engineering department, > IISc Bangalore, > India. > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From paulatto at sissa.it Tue Oct 6 14:48:58 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 06 Oct 2009 14:48:58 +0200 Subject: [Pw_forum] scf calculation of isolated atom In-Reply-To: References: Message-ID: In data 06 ottobre 2009 alle ore 11:34:33, udayagiri sai babu ha scritto: > I got an error saying *"strange occupations:number of electrons from > occupations is wrong"*. I undestood like my occupation card is not > matching with my "nbnd" value but i could not correct it can somebody > give me some > suggestion. In QE versions 4.0.x and before the occupations you had to specify at most ten occupations per line, you input would work this way: OCCUPATIONS 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 QE 4.1 and newer do not have this limitation. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From marcello.rosini at unimore.it Tue Oct 6 17:29:07 2009 From: marcello.rosini at unimore.it (Marcello Rosini) Date: Tue, 06 Oct 2009 17:29:07 +0200 Subject: [Pw_forum] error in atomic coordinates Message-ID: <4ACB6243.3040608@unimore.it> hallo everybody I launched a calculation wit pw.x 4.1 and I got the following error %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 0 from card_atomic_positions : error # 1 Error while parsing atomic position card. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% while launching with version 4.0.5 is not present. I see that there are some modifications in file modules/read_cards.f90, but I didn't manage to find out where the problem comes from.... anyway.... I launched with 4.0.5, copy the initial coordinate of the output file into the input, and now the 4.1 runs fine. Any suggestion? please find here the original and the modified input files http://cdm.unimo.it/home/fisica/rosini.marcello/inputs.txt many thanks, Marcello -- ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Dipartimento di Fisica, Universit? di Modena e Reggio Emilia and S3 National Research Centre of CNR-INFM via campi 213/a - 41125 Modena - Italy tel: +39 059.205.5067 email: marcello.rosini at unimore.it www.nanomodelling.unimore.it ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ From Giovanni.Cantele at na.infn.it Tue Oct 6 17:42:58 2009 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Tue, 06 Oct 2009 17:42:58 +0200 Subject: [Pw_forum] error in atomic coordinates In-Reply-To: <4ACB6243.3040608@unimore.it> References: <4ACB6243.3040608@unimore.it> Message-ID: <4ACB6582.70706@na.infn.it> Marcello Rosini wrote: > hallo everybody > I launched a calculation wit pw.x 4.1 and I got the following error > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > task # 0 > from card_atomic_positions : error # 1 > Error while parsing atomic position card. > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > while launching with version 4.0.5 is not present. > I see that there are some modifications in file modules/read_cards.f90, > but I didn't manage to find out where the problem comes from.... > anyway.... I launched with 4.0.5, copy the initial coordinate of the > output file into the input, and now the 4.1 runs fine. > Any suggestion? > > please find here the original and the modified input files > http://cdm.unimo.it/home/fisica/rosini.marcello/inputs.txt > > many thanks, > Marcello > > I was able to bot reproduce your error and get rid of it. It was due to the fact that, provided the copy and paste from the above link correctly reproduced all characters, there were some "tabs" in your file. I just replaced them with spaces and everything worked fine. Probably, tabs are "unwanted" characters in the new version. Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it Skype contact: giocan74 From paulatto at sissa.it Tue Oct 6 17:50:43 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 06 Oct 2009 17:50:43 +0200 Subject: [Pw_forum] error in atomic coordinates In-Reply-To: <4ACB6582.70706@na.infn.it> References: <4ACB6243.3040608@unimore.it> <4ACB6582.70706@na.infn.it> Message-ID: In data 06 ottobre 2009 alle ore 17:42:58, Giovanni Cantele ha scritto: > I was able to bot reproduce your error and get rid of it. It was due to > the fact that, provided the copy and paste from the above link correctly > reproduced all characters, there were some "tabs" in your file. I just > replaced them with spaces and everything worked fine. > > Probably, tabs are "unwanted" characters in the new version. Tabs are usually ok, what the code does not like are a "+" at the beginning of a coordinate or dos-style line terminations, tabs are bad only at the beginning of the line... There are still a few quirks with the coordinates parser, I am planning to fix them soon, but have no time at the moment. Sorry for the inconvenience. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From vtmtrinh at caltech.edu Tue Oct 6 23:00:08 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Tue, 6 Oct 2009 14:00:08 -0700 (PDT) Subject: [Pw_forum] write on local disk Message-ID: <1843.137.78.73.28.1254862808.squirrel@webmail.caltech.edu> Dear PWSCF Users, If I would like to write wavefunction files on local disk (to avoid network file system daemon), besides creating scratch directory in each node, do I need to do additional configuration. Thank you very much, MyTrinh JPL/CalTech From paulatto at sissa.it Tue Oct 6 23:51:14 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 6 Oct 2009 23:51:14 +0200 (CEST) Subject: [Pw_forum] write on local disk In-Reply-To: <1843.137.78.73.28.1254862808.squirrel@webmail.caltech.edu> References: <1843.137.78.73.28.1254862808.squirrel@webmail.caltech.edu> Message-ID: <34068.78.12.158.197.1254865874.squirrel@webmail.sissa.it> On Tue, October 6, 2009 23:00, vtmtrinh at caltech.edu wrote: > If I would like to write wavefunction files on local disk (to avoid > network file system daemon), besides creating scratch directory in each > node, do I need to do additional configuration. Dear MyTrinh, the outdir variable, or at least the wfcdir variable, must point to the local scratch directory: their default value is $HOME/tmp. You can also use disk_io='none' to completely prevent the output of wavefunctions to disk. It may see a drastic solution, but if you don't have the possibility to restart an interrupted job on the same nodes it does not really make much difference. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From vtmtrinh at caltech.edu Wed Oct 7 02:22:22 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Tue, 6 Oct 2009 17:22:22 -0700 (PDT) Subject: [Pw_forum] write on local disk In-Reply-To: <34068.78.12.158.197.1254865874.squirrel@webmail.sissa.it> References: <1843.137.78.73.28.1254862808.squirrel@webmail.caltech.edu> <34068.78.12.158.197.1254865874.squirrel@webmail.sissa.it> Message-ID: <2279.137.78.73.28.1254874942.squirrel@webmail.caltech.edu> Dear Lorenzo, Thank you. I'll try it and see how it goes. Best, MyTrinh > > On Tue, October 6, 2009 23:00, vtmtrinh at caltech.edu wrote: >> If I would like to write wavefunction files on local disk (to avoid >> network file system daemon), besides creating scratch directory in each >> node, do I need to do additional configuration. > > Dear MyTrinh, > the outdir variable, or at least the wfcdir variable, must point to the > local scratch directory: their default value is $HOME/tmp. > > You can also use disk_io='none' to completely prevent the output of > wavefunctions to disk. It may see a drastic solution, but if you don't > have the possibility to restart an interrupted job on the same nodes it > does not really make much difference. > > best regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From bmalone at berkeley.edu Wed Oct 7 03:08:33 2009 From: bmalone at berkeley.edu (Brad Malone) Date: Tue, 6 Oct 2009 18:08:33 -0700 Subject: [Pw_forum] Tips for vc-relax when far from equilibrium Message-ID: Hi, I am trying to relax (via vc-relax) structures that are, at least most likely, very far from equilibrium. I was wondering if there were any tips on how to make this more robust, as I'm currently experiencing frequent crashes for these systems. The main problem I'm having is shown below: After a *couple successful relaxation steps*, I'll find the following extrapolated charge 6.83616, renormalised to 8.00000 > > total cpu time spent up to now is 614.83 secs > > per-process dynamical memory: 28.6 Mb > > Self-consistent Calculation > > iteration # 1 ecut= 20.00 Ry beta=0.10 > CG style diagonalization > WARNING pzsteqr, convergence not achieved INFO = 3 > c_bands: 4 eigenvalues not converged > WARNING pzsteqr, convergence not achieved INFO = 3 > c_bands: 4 eigenvalues not converged > WARNING pzsteqr, convergence not achieved INFO = 3 > c_bands: 4 eigenvalues not converged > c_bands: 1 eigenvalues not converged > 4 processes killed (possibly by Open MPI) > So the first thing is that my extrapolated charge is very far off from the value of 8 expected (I have two Si atoms in the unit cell). Secondly, the self-consistent calculation fails and kills the job. I've tried using davidson diagonalization as well and it also crashes. Any suggestions on how I might avoid this? Thanks for your time. Best, Brad UC Berkeley -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091006/064b83d4/attachment.htm From nkxirainbow at gmail.com Wed Oct 7 04:11:59 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Wed, 7 Oct 2009 02:11:59 +0000 Subject: [Pw_forum] Tips for vc-relax when far from equilibrium In-Reply-To: References: Message-ID: <21fbc4790910061911m66851fbbw79a92041d86dde48@mail.gmail.com> Dear Brad Malone: You can opitmize structure using empirical or semi-empirical method(for example: tight-binding method) with other software?? After that, the pre-optimized structure will be more suitable for DFT calculation?? -- ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091007/f1ecf0d5/attachment.htm From vtmtrinh at caltech.edu Wed Oct 7 09:07:18 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Wed, 7 Oct 2009 00:07:18 -0700 (PDT) Subject: [Pw_forum] phonon errors Message-ID: <2940.137.78.73.28.1254899238.squirrel@webmail.caltech.edu> Dear PWSCF Users, I have 2 questions: 1. I did a test phonon calculation for Si bulk, 2atom-cell, following example_06. I first did an scf calculation with k-point mesh 10x10x10 Si. Next, I wanted to do a phonon calculation along [001] direction (i.e q=0, q=0.1, 0.2,...). If I specified 'ldisp=.true.', it was supposed to do nscf calculation and then phonon calculation in a series. However, it stopped with errors. When I checked the output file, after it had finished calculation at Gamma point and nscf calculations, when it entered the phonon calculations for q !=0, it stopped with the error: task # 8 from openfilq : error # 1 file si.wfc not found The run is supposed to run continously as I specified ldisp=.true. Now suddenly, it stopped with that error, so I could not understand why. When I tried with different q-point mesh, e.g. 4x4x4. I ahd the same problem. Here is my input for the phonon calculation: phonons of Si bulk &inputph tr2_ph=1.0d-14, prefix='si', ldisp=.true., nq1=1, nq2=1, nq3=10 !(or nq1=4, nq2=4, nq3=4 for second test) amass(1)=28.086, outdir='./', fildyn='si.dyn', / 2. If I run each single q-point,i.e. did scf, then Gamma calculation, then nscf and phonon calculation for each single q-point, it was fine. I obtained 6 dyn files for 6 q-points from 0 to 0.5. However, I had problem when I used q2r.x. In the input file, q2r.in, if I specified the q-point mesh: 1 1 10, and below it I provided a list of 6 dyn files that I calculated above, I got an error: task # 0 from init : error # 1 q not allowed I am not sure what is the correct input for the q-point mesh in q2r.in that I should use in this case. Below is my q2r.in: &input fildyn='si.dyn', zasr='crystal', flfrc='si.fc' / 1 1 10 6 si.dyn1 si.dyn2 si.dyn3 si.dyn4 si.dyn5 si.dyn5 I am using QE version 4.0.3. Thank you very much, MyTrinh JPL/CakTech From xinhuolin at gmail.com Wed Oct 7 09:09:50 2009 From: xinhuolin at gmail.com (Huolin Xin) Date: Wed, 7 Oct 2009 03:09:50 -0400 Subject: [Pw_forum] PDOS on different Oxygen atoms Message-ID: Hi all, I have a layered structure ...,AO,AO,BO,BO,.... Does projwfc.x gives seperate PDOS for the Oxygen in AO layers and BO layers automatically? H. Xin Cornell University From udayagiri3 at gmail.com Wed Oct 7 09:19:17 2009 From: udayagiri3 at gmail.com (udayagiri sai babu) Date: Wed, 7 Oct 2009 12:49:17 +0530 Subject: [Pw_forum] scf calculation of isolated atom In-Reply-To: References: Message-ID: Thank you for the answer GS and Lorenzo. I am using 10 occupations per line in the occupations card as you suggested and It is not giving any error now. On Tue, Oct 6, 2009 at 6:18 PM, Lorenzo Paulatto wrote: > In data 06 ottobre 2009 alle ore 11:34:33, udayagiri sai babu > ha scritto: > > I got an error saying *"strange occupations:number of electrons from > > occupations is wrong"*. I undestood like my occupation card is not > > matching with my "nbnd" value but i could not correct it can somebody > > give me some > > suggestion. > > In QE versions 4.0.x and before the occupations you had to specify at most > ten occupations per line, you input would work this way: > > OCCUPATIONS > 1 1 1 1 1 1 1 1 1 1 > 0 0 0 > 1 1 1 1 0 0 0 0 0 0 > 0 0 0 > > QE 4.1 and newer do not have this limitation. > > best regards > > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- U.Saibabu PhD student, Deformation mechanisms modeling group, Materials engineering department, IISc Bangalore, India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091007/8de40c7c/attachment.htm From miguel.martinez at ehu.es Wed Oct 7 09:24:16 2009 From: miguel.martinez at ehu.es (Miguel =?UTF-8?B?TWFydMOtbmV6?=) Date: Wed, 7 Oct 2009 09:24:16 +0200 Subject: [Pw_forum] phonon errors In-Reply-To: <2940.137.78.73.28.1254899238.squirrel@webmail.caltech.edu> References: <2940.137.78.73.28.1254899238.squirrel@webmail.caltech.edu> Message-ID: <20091007092416.2c4c123a@lcpybm> Dear Mytrinh, As far as I know, q2r.x requires a regular mesh (i.e. 4 4 4), and that would explain why it's crashing in your second question. Regarding your first crash, I'd make sure outdirs are identical in both the pw.x and ph.x inputs and then check why ph.x can't read the files in the outdir specified to pw.x. On Wed, 7 Oct 2009 00:07:18 -0700 (PDT) vtmtrinh at caltech.edu wrote: > Dear PWSCF Users, > > I have 2 questions: > > 1. I did a test phonon calculation for Si bulk, 2atom-cell, following > example_06. I first did an scf calculation with k-point mesh > 10x10x10 Si. Next, I wanted to do a phonon calculation along [001] > direction (i.e q=0, q=0.1, 0.2,...). If I specified 'ldisp=.true.', > it was supposed to do nscf calculation and then phonon calculation in > a series. However, it stopped with errors. When I checked the > output file, after it had finished calculation at Gamma point and > nscf calculations, when it entered the phonon calculations for q !=0, > it stopped with the error: task # 8 > from openfilq : error # 1 > file si.wfc not found > The run is supposed to run continously as I specified ldisp=.true. > Now suddenly, it stopped with that error, so I could not understand > why. When I tried with different q-point mesh, e.g. 4x4x4. I ahd > the same problem. > > Here is my input for the phonon calculation: > > phonons of Si bulk > &inputph > tr2_ph=1.0d-14, > prefix='si', > ldisp=.true., > nq1=1, nq2=1, nq3=10 !(or nq1=4, nq2=4, nq3=4 for second test) > amass(1)=28.086, > outdir='./', > fildyn='si.dyn', > / > > 2. If I run each single q-point,i.e. did scf, then Gamma calculation, > then nscf and phonon calculation for each single q-point, it was > fine. I obtained 6 dyn files for 6 q-points from 0 to 0.5. However, I > had problem when I used q2r.x. In the input file, q2r.in, if I > specified the q-point mesh: 1 1 10, and below it I provided a list of > 6 dyn files that I calculated above, I got an error: > > task # 0 > from init : error # 1 > q not allowed > > I am not sure what is the correct input for the q-point mesh in q2r.in > that I should use in this case. > > Below is my q2r.in: > > &input > fildyn='si.dyn', zasr='crystal', flfrc='si.fc' > / > 1 1 10 > 6 > si.dyn1 > si.dyn2 > si.dyn3 > si.dyn4 > si.dyn5 > si.dyn5 > > I am using QE version 4.0.3. > > Thank you very much, > > MyTrinh > JPL/CakTech > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ---------------------------------------- Miguel Mart?nez Canales Condensed Matter Physics Dpt. UPV/EHU Faculty of Science and Technology Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5326 ---------------------------------------- "If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas." George Bernard Shaw From udayagiri3 at gmail.com Wed Oct 7 09:24:28 2009 From: udayagiri3 at gmail.com (udayagiri sai babu) Date: Wed, 7 Oct 2009 12:54:28 +0530 Subject: [Pw_forum] occupations fixed Message-ID: In the archive I have seen a discussion suggesting use of occupations=fixed to find out the ground state energy of isolated oxygen atom accurately. Can I use the same (occupations=fixed) for metal also. Because in the pw document it is told to use only for insulators. -- U.Saibabu PhD student, Deformation mechanisms modeling group, Materials engineering department, IISc Bangalore, India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091007/02a58abd/attachment.htm From mastermik at gmail.com Wed Oct 7 09:39:03 2009 From: mastermik at gmail.com (Mikiyas Tsegaye) Date: Wed, 7 Oct 2009 03:39:03 -0400 Subject: [Pw_forum] phonon errors (vtmtrinh@caltech.edu) In-Reply-To: References: Message-ID: <3B28E8A6-BCDB-4223-82A4-805689ED2418@gmail.com> On Oct 7, 2009, at 3:24 AM, pw_forum-request at pwscf.org wrote: > > Message: 7 > Date: Wed, 7 Oct 2009 00:07:18 -0700 (PDT) > From: vtmtrinh at caltech.edu > Subject: [Pw_forum] phonon errors > To: pw_forum at pwscf.org > Message-ID: > <2940.137.78.73.28.1254899238.squirrel at webmail.caltech.edu> > Content-Type: text/plain;charset=iso-8859-1 > > Dear PWSCF Users, > > I have 2 questions: > > 1. I did a test phonon calculation for Si bulk, 2atom-cell, following > example_06. I first did an scf calculation with k-point mesh > 10x10x10 Si. > Next, I wanted to do a phonon calculation along [001] direction (i.e > q=0, > q=0.1, 0.2,...). If I specified 'ldisp=.true.', it was supposed to do > nscf calculation and then phonon calculation in a series. However, it > stopped with errors. When I checked the output file, after it had > finished calculation at Gamma point and nscf calculations, when it > entered > the phonon calculations for q !=0, it stopped with the error: > task # 8 > from openfilq : error # 1 > file si.wfc not found > The run is supposed to run continously as I specified ldisp=.true. > Now > suddenly, it stopped with that error, so I could not understand > why. When > I tried with different q-point mesh, e.g. 4x4x4. I ahd the same > problem. > > Here is my input for the phonon calculation: > > phonons of Si bulk > &inputph > tr2_ph=1.0d-14, > prefix='si', > ldisp=.true., > nq1=1, nq2=1, nq3=10 !(or nq1=4, nq2=4, nq3=4 for second test) > amass(1)=28.086, > outdir='./', > fildyn='si.dyn', > / > > 2. If I run each single q-point,i.e. did scf, then Gamma > calculation, then > nscf and phonon calculation for each single q-point, it was fine. I > obtained 6 dyn files for 6 q-points from 0 to 0.5. However, I had > problem > when I used q2r.x. In the input file, q2r.in, if I specified the q- > point > mesh: 1 1 10, and below it I provided a list of 6 dyn files that I > calculated above, I got an error: > > task # 0 > from init : error # 1 > q not allowed > > I am not sure what is the correct input for the q-point mesh in q2r.in > that I should use in this case. > > Below is my q2r.in: > > &input > fildyn='si.dyn', zasr='crystal', flfrc='si.fc' > / > 1 1 10 > 6 > si.dyn1 > si.dyn2 > si.dyn3 > si.dyn4 > si.dyn5 > si.dyn5 > > I am using QE version 4.0.3. > > Thank you very much, > > MyTrinh > JPL/CakTech > > > Dear, It is essentially complaining that the data from your first pw.x run is not found in the outdir under that prefix ("si"). Have you checked that the outdir matches in both input files? Also, you can manually make sure that Si.wfc does exist in the outdir. Another helpful thing to do is to clear the outdir before you do these runs. - Mikiyas Tsegaye University of Virginia Charlottesville, VA, USA mastermik at gmail.com From paulatto at sissa.it Wed Oct 7 09:39:56 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 07 Oct 2009 09:39:56 +0200 Subject: [Pw_forum] occupations fixed In-Reply-To: References: Message-ID: In data 07 ottobre 2009 alle ore 09:24:28, udayagiri sai babu ha scritto: > In the archive I have seen a discussion suggesting use of > occupations=fixed to find out the ground state energy of isolated oxygen > atom accurately. Can I use the same (occupations=fixed) for metal also. > Because in the pw > document it is told to use only for insulators. Dear U, for isolated atoms you can always use fixed occupations: as you use only use the Gamma point, bands are actually flat. Nevertheless it may be difficult to reach the ground state in some cases, I suggest using a small smearing which should give exactly the same result but makes convergence faster. I see you have already asked how to compute the ground state for a Ti atom (e.g. ), which actually is a metal... why don't you tell us your problem, instead of your guess for a solution? best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From sclauzer at sissa.it Wed Oct 7 10:01:20 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Wed, 07 Oct 2009 10:01:20 +0200 Subject: [Pw_forum] PDOS on different Oxygen atoms In-Reply-To: References: Message-ID: <4ACC4AD0.5030505@sissa.it> Huolin Xin wrote: > Hi all, > > I have a layered structure ...,AO,AO,BO,BO,.... Does projwfc.x gives > seperate PDOS for the Oxygen in AO layers and BO layers automatically? projwfc.x gives you separate PDOS for each atom in your system, numbering the corresponding files according to the order which is shown in the list of atoms at the beginning of the output of the preceding pw.x calculation. After identifying the number of the atom(s) of interest (e.g. by looking at the coordinates reported in the list), you can easily match the file containing the PDOS of that atom. I think it is more difficult to explain than to try it directly. Simply run projwfc.x and I think you will see clearly (also give a look at Doc/INPUT_PROJWFC.txt, if you haven't already). HTH GS > > H. Xin > Cornell University > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From udayagiri3 at gmail.com Wed Oct 7 10:32:28 2009 From: udayagiri3 at gmail.com (udayagiri sai babu) Date: Wed, 7 Oct 2009 14:02:28 +0530 Subject: [Pw_forum] occupations fixed In-Reply-To: References: Message-ID: Dear Lorenzo thanks for the answer. My problem is simple it is to find out the cohesive energy of Cr,V,Ti. I posted this question earlier also. I was told to try out with occupations=from_input instead of degauss value to get the accurate value of the ground state energy because i am getting an error of around 10% in Ec (i was wondering if that can be acceptable) meanwhile i came across the post which says to use ocuupations=fixed. On Wed, Oct 7, 2009 at 1:09 PM, Lorenzo Paulatto wrote: > In data 07 ottobre 2009 alle ore 09:24:28, udayagiri sai babu > ha scritto: > > > In the archive I have seen a discussion suggesting use of > > occupations=fixed to find out the ground state energy of isolated oxygen > > atom accurately. Can I use the same (occupations=fixed) for metal also. > > Because in the pw > > document it is told to use only for insulators. > > > Dear U, > for isolated atoms you can always use fixed occupations: as you use only > use the Gamma point, bands are actually flat. Nevertheless it may be > difficult to reach the ground state in some cases, I suggest using a small > smearing which should give exactly the same result but makes convergence > faster. > > I see you have already asked how to compute the ground state for a Ti atom > (e.g. > ), > which actually is a metal... why don't you tell us your problem, instead > of your guess for a solution? > > best regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- U.Saibabu PhD student, Deformation mechanisms modeling group, Materials engineering department, IISc Bangalore, India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091007/df679726/attachment.htm From paulatto at sissa.it Wed Oct 7 10:39:46 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 07 Oct 2009 10:39:46 +0200 Subject: [Pw_forum] occupations fixed In-Reply-To: References: Message-ID: In data 07 ottobre 2009 alle ore 10:32:28, udayagiri sai babu ha scritto: > Dear Lorenzo thanks for the answer. My problem is simple it is to find > out the cohesive energy of Cr,V,Ti. I posted this question earlier also. > I was told to try out with occupations=from_input instead of degauss > value to get the accurate value of the ground state energy because i am > getting an error of around 10% in Ec (i was wondering if that can be > acceptable) meanwhile i came across the post which says to use > ocuupations=fixed. Isolated Cr, V and Ti all have magnetic ground state, you have to take care in making the calculation for the isolated atom with spin and with some initial magnetization. It would be even better to compute it with and without spin, so you can see the difference. V and Ti have a ground state which is NOT spherical, you absolutely must use *nosym=.true.*, and break the symmetry somehow, e.g. by specifying occupations from input, or tot_magnetization or multiplicity; or taking a slightly non-cubic unit cell. Best regards. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From daijiayu at nudt.edu.cn Wed Oct 7 13:47:35 2009 From: daijiayu at nudt.edu.cn (Jiayu Dai) Date: Wed, 07 Oct 2009 19:47:35 +0800 Subject: [Pw_forum] How to consider more valence electrons in Pseudopotentials? Message-ID: <454916055.22499@nudt.edu.cn> Dear Users, I have a question about the generation of pseudopotential(PP): how to consider more electrons as valence electrons? For example, the configurations of Al is 1s2 2s2 2p6 3s2 3p1, and we usually consider 3s2 3p1 as valence electrons. Now, i want consider 2s2 and 2p6 as valence electrons also. Then, i should generate new PP. I used the configuration in inputfile of ld1.x [He] 2s2 sp6 3s2 3p1 3d-2 5 2S 1 0 2.00 0.00 1.60 1.60 2P 2 1 6.00 0.00 1.60 1.60 3S 3 0 2.00 0.00 2.40 2.40 3P 4 1 1.00 0.05 2.60 2.60 3D 5 2 -2.00 0.25 2.30 2.30 And the errors tell me ps-label wrong. I knew if i treat the 2s and 2p electrons as semicore states, this propblem can be considered partly. But the question is how to consider them as semicore states? Can they be like this? [He] 2s2 2p6 3s0 3p0 3d0 3 2S 1 0 2.00 0.00 1.60 1.60 2P 2 1 6.00 0.00 1.60 1.60 3D 3 2 0.00 0.05 2.30 2.30 Here, the cutoff energy will be tested after the generation of PPs. Thanks in advance. Jiayu ------------------------------ ------------------------------------------- Jiayu Dai Department of Physics National University of Defense Technology, Changsha, 410073, P R China ----------------------------------------- From paulatto at sissa.it Wed Oct 7 14:19:21 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 07 Oct 2009 14:19:21 +0200 Subject: [Pw_forum] How to consider more valence electrons in Pseudopotentials? In-Reply-To: <454916055.22499@nudt.edu.cn> References: <454916055.22499@nudt.edu.cn> Message-ID: In data 07 ottobre 2009 alle ore 13:47:35, Jiayu Dai ha scritto: > [He] 2s2 sp6 3s2 3p1 3d-2 > 5 > 2S 1 0 2.00 0.00 1.60 1.60 > 2P 2 1 6.00 0.00 1.60 1.60 > 3S 3 0 2.00 0.00 2.40 2.40 > 3P 4 1 1.00 0.05 2.60 2.60 > 3D 5 2 -2.00 0.25 2.30 2.30 > > And the errors tell me ps-label wrong. It is not very clear to me what's your problem, please remember to ALWAYS attach the full output of the calculation. Anyway, you are requesting the 2S pseudo-wavefunction to have zero node and the 3S one to have 2 nodes, which is at least awkward. The situation is even worst for the 3P, as for no apparent reason you have set their principal quantum number to 4. Finally, you have put electrons in the 3P norm-conserving orbital and also specified a positive energy for it, that will make it unbound. It is not necessarily wrong, but I would not do it before having explored any other possibility. Change the input as following and it may work: 5 2S 1 0 2.00 0.00 1.60 1.60 2P 2 1 6.00 0.00 1.60 1.60 3S 2 0 2.00 0.00 2.40 2.40 3P 3 1 1.00 0.00 2.60 2.60 3D 5 2 -2.00 0.25 2.30 2.30 -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From carlo.nervi at unito.it Wed Oct 7 14:30:11 2009 From: carlo.nervi at unito.it (Carlo Nervi) Date: Wed, 07 Oct 2009 14:30:11 +0200 Subject: [Pw_forum] again on OpenMPI 1.3.3 Message-ID: <4ACC89D3.702@unito.it> Hi again, I am sorry to bother all the community with my compiling problems, that a little OT, but they are quite unusual. Certainly something is wrong in my machine (Linux Gentoo on dual Xeon 5345), but I cannot guess what. After many tests and compiling I found that pw.x run perfectly using "mpirun -np 2", but fail with "mpirun -np 8". The error is "MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD with errorcode 0". This really I cannot understand. Is there anyone that could give me a hint? I succesfully compiled the "pingpong" code below with mpif90 (wrapper to ifort). Also in this case mpirun -np 2 works, but if I put -np 4, 6 or 8 the program crash with the following message: *** Process received signal *** Signal: Segmentation fault (11) Signal code: Address not mapped (1) Failing at address: (nil) [ 0] /lib/libpthread.so.0 [0x2b48da46b400] [ 1] /lib/libc.so.6(fputs+0x1e) [0x2b48da6d9d0e] [ 2] ./pingpong(main+0x206) [0x402536] [ 3] /lib/libc.so.6(__libc_start_main+0xf4) [0x2b48da6965e4] [ 4] ./pingpong [0x4022b9] *** End of error message *** I was thinking that ssh does not propagate the environment variables (so the libraries cannot be found), but it runs on 2 cpus! Any helps would be greatly appreciated. Carlo ---------------- /* pingpong - measure effective bandwidth and latency */ #include "mpi.h" #include #include #include #include #include #define MAXSIZE (1024*1024) #define MINSIZE (0) #define REPEAT 50 #define INCSIZE (2) #define INCOP *= #define CALIBRATION_LOOPS 100 #define TAG_PING 1 #define TAG_PONG 2 /* define DETAIL if you want to create histogramms by measuring the latency of each single ping-pong transfer */ #if 0 #define DETAIL #include "getus.h" #endif #ifdef linux #define longlong_t long long #endif char *buffer; char *exename; int min_size, max_size, inc_size, repeats; int myrank, mysize; static FILE* fpGlobal = NULL; static FILE* fpDetail = NULL; void ping (int to, int from); void pong (int to); int main(int argc, char **argv) { MPI_Status status; int first; char fname[128]; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &myrank); MPI_Comm_size(MPI_COMM_WORLD, &mysize); if (myrank == 0) { strcpy(fname, argv[0]); strcat(fname,".dat"); fpGlobal = fopen(fname,"w"); } exename = argv[0]; if(mysize % 2 != 0) { printf ("pingpong must be used with an even number of processes.\n"); MPI_Finalize(); exit (1); } /* set run parameters */ if (argc != 4) { printf ("usage: pingpong min_size max_size repeats\n"); printf ("using default values for this run\n"); min_size = MINSIZE; max_size = MAXSIZE; inc_size = INCSIZE; repeats = REPEAT; } else { min_size = atoi( argv[1] ); max_size = atoi( argv[2] ); inc_size = INCSIZE; repeats = atoi( argv[3] ); } buffer = (char *)malloc (max_size); /* find ping and pong processes */ if ( myrank < mysize/2 ) { if (myrank % 2 == 0) ping( myrank + mysize/2, myrank); else pong( myrank + mysize/2 ); } else { first = (mysize/2) % 2; if (myrank % 2 == first) pong( myrank - mysize/2 ); else ping( myrank - mysize/2, myrank ); } if (myrank == 0) { fclose(fpGlobal); } free (buffer); MPI_Finalize(); } void ping( int to, int from ) { MPI_Status status; double starttime, totaltime; double getticks_overhead; #ifdef DETAIL longlong_t hr_start, hr_end; longlong_t calibration = 0; longlong_t *timings; #endif char fname[128]; char bytes[128]; int i, j; int firstrun = 1; #ifdef DETAIL fprintf (stderr, "Calibrating..."); for (i = 0; i < CALIBRATION_LOOPS; i++) { GETTICKS(&hr_start); GETTICKS(&hr_end); calibration += hr_end - hr_start; } getticks_overhead = ((double)calibration)/(CALIBRATION_LOOPS); fprintf (stderr, "gethrtime() overhead is %6.3f\n", getticks_overhead); timings = (longlong_t *)malloc (repeats*sizeof(longlong_t)); #endif printf("pingpong from %d to %d\n\n", from, to); fprintf(fpGlobal, "# msgsize[byte] repeats bandwidth[MB/s] latency[us]\n"); fflush(stdout); for( i = min_size; i <= max_size; i INCOP inc_size) { if ((!firstrun) && (i == 0)) { i++; if (i > max_size) break; } -- ------------------------------------------------------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ From skegin at ph2.uni-koeln.de Wed Oct 7 14:31:22 2009 From: skegin at ph2.uni-koeln.de (Boris Skegin) Date: Wed, 07 Oct 2009 14:31:22 +0200 Subject: [Pw_forum] worth recompiling ? Message-ID: <1254918682.6632.20.camel@studi-acer.lfap.ph2.uni-koeln.de> Hallo. I managed to compile espresso-4.1 on our Linux PC cluster using intel/9.1 mkl/9.1 vltmpi/mvapich_0.9.4 as it is said in http://www.pwscf.org/user_guide/node13.html#SECTION00036600000000000000 : "The latest (July 2006) 32-bit version of ifort 9.1 works flawlessy." And espresso-4.1 does work in the parallel version on our cluster, at least, if checked with simple examples. Besides intel/9.1 and mkl/9.1 we have also intel/11.0 intel/10.0 intel/10.1 mkl/10.0 mkl/10.1 mkl/9.1 As I have some free time now, I'd like to ask you if it is worth to recompile, say, with intel/11.0 and mkl/10.0 in order to achieve even better performance for complicated jobs? Regards, Boris Skegin -- Telefon: +49-(0)221-470-6392 Universit?t zu K?ln II. Physikalisches Institut Z?lpicher Str. 77 50937 K?ln Deutschland From carlo.nervi at unito.it Wed Oct 7 15:02:30 2009 From: carlo.nervi at unito.it (Carlo Nervi) Date: Wed, 07 Oct 2009 15:02:30 +0200 Subject: [Pw_forum] worth recompiling ? In-Reply-To: <1254918682.6632.20.camel@studi-acer.lfap.ph2.uni-koeln.de> References: <1254918682.6632.20.camel@studi-acer.lfap.ph2.uni-koeln.de> Message-ID: <4ACC9166.7090402@unito.it> Hello, when i was able to compile everything and all was working, I recompiled with intel 11.0 and especially with MKL 10 libraries. There is a switch in intel 11 compiler (-xHost) that choose the best code for the CPU you have. This switch improved (at least for me) the speed. From my point of view I would give to it a try. However, there others in this forum that could properly answer (as you have probably read, now I'm in still trouble and cannot compile anymore QE...) Carlo Boris Skegin ha scritto: > Hallo. > > I managed to compile espresso-4.1 on our Linux PC cluster using > intel/9.1 > mkl/9.1 > vltmpi/mvapich_0.9.4 > > as it is said in > http://www.pwscf.org/user_guide/node13.html#SECTION00036600000000000000 : > > "The latest (July 2006) 32-bit version of ifort 9.1 works flawlessy." > > And espresso-4.1 does work in the parallel version on our cluster, at > least, if checked with simple examples. > > > Besides intel/9.1 and mkl/9.1 we have also > > intel/11.0 > intel/10.0 > intel/10.1 > > mkl/10.0 > mkl/10.1 > mkl/9.1 > > As I have some free time now, I'd like to ask you if it is worth to > recompile, say, with intel/11.0 and mkl/10.0 in order to achieve even > better performance for complicated jobs? > > Regards, > Boris Skegin > > > > -- ------------------------------------------------------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ From Peter.Pallister at nrc-cnrc.gc.ca Wed Oct 7 17:10:11 2009 From: Peter.Pallister at nrc-cnrc.gc.ca (Pallister, Peter) Date: Wed, 7 Oct 2009 11:10:11 -0400 Subject: [Pw_forum] GIPAW pseudopotentials Message-ID: Hi all, I am attempting to generate a gipaw pseudopotentials, in this case for Mg, but I am having tremendous difficulty. The wavefunctions produced with the input below (or minute variations thereof) continuously show discontinuities, specifically at a radius of 15bohrs. I have looked at the previous entries in the forum on this topic as well as looked at the tutorials available on Dr. Seitsonen's webpage and the QE summer school media but I still cannot successfully generate one of these pseudopotentials. Any help is greatly appreciated. Below is a typical input I use for the ld1.x module: &input title = 'Mg-oct09' prefix = 'Mg-oct09' zed = 12.0 rel = 0 nld = 3 rlderiv = 5.0 eminld = -6.0 emaxld = -1.0 deld = 0.01 config = '1s2 2s2 2p6 3s2 3p-1 3d-1' iswitch = 3 dft = 'PBE' / &inputp pseudotype = 1 lloc = 2 file_pseudopw = 'Mg.pbe-gipaw-oct09.UPF' file_recon = 'Mg.pbe-gipaw-oct09.recon' lgipaw_reconstruction = .true. tm = .true. / 3 2S 1 0 2.00 0.00 1.600 1.600 2P 2 1 6.00 0.00 1.600 1.600 3D 3 2 -1.00 -0.30 1.600 1.600 &test / 4 2S 1 0 2.00 0.00 1.600 1.600 3S 2 0 2.00 0.00 1.600 1.600 2P 2 1 6.00 0.00 1.600 1.600 3P 3 1 -1.00 -0.10 1.600 1.600 Thanks, Peter Pallister Steacie Institute for Molecular Sciences National Research Council Ottawa, Ontario Canada -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091007/c2c54ec2/attachment.htm From daijiayu at nudt.edu.cn Wed Oct 7 17:14:40 2009 From: daijiayu at nudt.edu.cn (Jiayu Dai) Date: Wed, 07 Oct 2009 23:14:40 +0800 Subject: [Pw_forum] How to consider more valence electrons inPseudopotentials? Message-ID: <454928480.24310@nudt.edu.cn> Thank you, Lorenzo, now it can work as your suggestion. But there is another question: I tested the PP considering the lattice constant of Al crystal. The total energy became lower when the lattice constant in input file is larger. We know the energy shuold increase when the lattice constant is greater than some value, but it did not happen in my case. I tried to increase the rcut to 20.0, but it seems to be the same. So, what's the proplem here? The input of PP is following: &input title='Al', zed=13.0, rel=1, config='[He] 2s2 2p6 3s2 3p1 3d-2' iswitch=3, dft='PBE', / &inputp lloc=2, pseudotype=2, nlcc = .true. tm = .true. , file_pseudopw='Al.pbe-tm-nc.UPF', author='DJY', / 5 2S 1 0 2.00 0.00 2.00 2.00 2P 2 1 6.00 0.00 2.00 2.00 3S 2 0 2.00 0.00 2.40 2.40 3P 3 1 1.00 0.00 2.70 2.70 3D 5 2 -2.00 0.25 2.30 2.30 In your mail: >From: "Lorenzo Paulatto" >Reply-To: >To: "Jiayu Dai" , "PWSCF Forum" >Subject: Re: [Pw_forum] How to consider more valence electrons in Pseudopotentials? >Date:Wed, 07 Oct 2009 14:19:21 +0200 > >In data 07 ottobre 2009 alle ore 13:47:35, Jiayu Dai ha scritt> [He] 2s2 sp6 3s2 3p1 3d-2 >> 5 >> 2S 1 0 2.00 0.00 1.60 1.6> 2P 2 1 6.00 0.00 1.60 1.6> 3S 3 0 2.00 0.00 2.40 2.4> 3P 4 1 1.00 0.05 2.60 2.6> 3D 5 2 -2.00 0.25 2.30 2.3> > > And the errors tell me ps-label wrong. > It is not very clear to me what's your problem, please remember to ALWAYS >attach the full output of the calculation. Anyway, you are requesting the >2S pseudo-wavefunction to have zero node and the 3S one to have 2 nodes,which is at least awkward. The situation is even worst for the 3P, as for >no apparent reason you have set their principal quantum number to 4. > Finally, you have put electrons in the 3P norm-conserving orbital and also specified a positive energy for it, that will make it unbound. It is notnecessarily wrong, but I would not do it before having explored any other >possibility. > Change the input as following and it may wor5 > 2S 1 0 2.00 0.00 1.60 1.60 >2P 2 1 6.00 0.00 1.60 1.60 >3S 2 0 2.00 0.00 2.40 2.40 >3P 3 1 1.00 0.00 2.60 2.60 >3D 5 2 -2.00 0.25 2.30 2.30 > > > -- >Lorenzo PaulattoSISSA & DEMOCRITOS (Triestphone: +39 040 3787 511 >skype: paulawww: http://people.sissa.it/~paulatto/ > *** save italian brains *** http://saveitalianbrains.wordpress.com > From giannozz at democritos.it Wed Oct 7 22:11:11 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 7 Oct 2009 22:11:11 +0200 Subject: [Pw_forum] Tips for vc-relax when far from equilibrium In-Reply-To: References: Message-ID: <3A6C77D0-7840-4100-B7AC-0F8AC8A37027@democritos.it> On Oct 7, 2009, at 3:08 , Brad Malone wrote: > extrapolated charge 6.83616, renormalised to 8.00000 do not use either wavefunction or potential extrapolation other than the default together with bfgs. You might also try damped dynamics with a relatively small time step in extreme cases Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Wed Oct 7 22:16:15 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 7 Oct 2009 22:16:15 +0200 Subject: [Pw_forum] phonon errors In-Reply-To: <2940.137.78.73.28.1254899238.squirrel@webmail.caltech.edu> References: <2940.137.78.73.28.1254899238.squirrel@webmail.caltech.edu> Message-ID: <77AFEA50-D814-405F-9362-E6CF392455C3@democritos.it> Try to find out if the problem you have is related to the usage of local disks as scratch directories. It is quite tricky, because you have no control on which filesystem the various pieces are written to. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From manoj at phys.ufl.edu Wed Oct 7 23:23:25 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Wed, 7 Oct 2009 17:23:25 -0400 (EDT) Subject: [Pw_forum] Relaxation of low symmetry lattices Message-ID: Dear PWSCF users, I have a question regarding relaxation of low symmetry lattices. I did relaxation of 5 atoms FCC unit cell of copper. This is pure copper, no interface, no defect or anything. The unit cell is just not conventional, but a different way of choosing bulk FCC. For the default value of convergence the calculation finishes at the end of 1st scf cycle, but I see different forces when I describe nosym=.TRUE. or cell_parameters {cubic}. Following is the input file. &control calculation='relax' pseudo_dir = '/home/manoj/espresso-4.0.4/pseudo', outdir='./', prefix='lcu', wf_collect=.TRUE. / &system ibrav =0, celldm(1)=6.81650937063832 nat= 5, ntyp= 1, ecutwfc =50.0, occupations='smearing', smearing='gaussian', degauss=0.02, ecutrho=500 / &electrons conv_thr = 1.0e-8 mixing_beta=0.7 / &ions ion_dynamics='bfgs' pot_extrapolation='second_order' wfc_extrapolation='second_order' / ATOMIC_SPECIES Cu 63.55 Cu.pz-d-rrkjus.UPF ATOMIC_POSITIONS Cu 0.0 0.0 0.0 Cu 0.5 0.5 0.0 Cu 0.0 1.0 0.0 Cu 1.0 1.0 0.0 Cu 0.5 1.5 0.0 K_POINTS (automatic) 8 8 8 0 0 0 CELL_PARAMETERS {cubic} 1.5 0.5 0.0 -0.5 1.5 0.0 0.0 0.5 0.5 Result of relaxation : The covergence is achieved with follwing forces on the atoms. Forces acting on atoms (Ry/au): Forces acting on atoms (Ry/au): atom 1 type 1 force = 0.00000000 0.00000000 0.00000000 atom 2 type 1 force = 0.00000567 0.00000185 -0.00000172 atom 3 type 1 force = 0.00001155 0.00000377 -0.00000372 atom 4 type 1 force = -0.00001155 -0.00000377 0.00000372 atom 5 type 1 force = -0.00000567 -0.00000185 0.00000172 I realized that I should not have chosen cubic symmetry group, as my unit cell is not cubic. So, as described in the input file parameters, for the low symmetry lattices, I specifed nosym=.TRUE., and again the relaxation stopped after 1 scf cycle, acheiving convergence. But the forces on the atoms are - Forces acting on atoms (Ry/au): atom 1 type 1 force = 0.00001355 0.00000524 -0.00000516 atom 2 type 1 force = -0.00001493 -0.00000395 0.00000395 atom 3 type 1 force = 0.00000253 -0.00000043 0.00000045 atom 4 type 1 force = 0.00000820 0.00000230 -0.00000239 atom 5 type 1 force = -0.00000935 -0.00000316 0.00000315 So, still forces are .001 ev/a.u, which i think is higher for a perfect FCC lattice. I can specify stricter convergence of force, but that would move atoms around and break symmetry. So, as long as the lattice parameter is correct, I think for perfect FCC lattice, forces on atoms should be smaller than what they are presently. I have checked the lattice constant for bulk FCC with a different calculation, so lattice constant is not an issue. Another possibility is symmetry group, as you can see with the same input variables, changing symmetry group changes results.I was wondering if there is a way i can choose correct symmetry group for this low symmetry lattice which might give me smaller forces? Are there any other parameters I need to specify? On a broad view I am curious to know how to do relaxation of low symmetry lattices. Regards, Manoj Srivastava University of Florida Gainesville, USA From vtmtrinh at caltech.edu Wed Oct 7 23:53:18 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Wed, 7 Oct 2009 14:53:18 -0700 (PDT) Subject: [Pw_forum] phonon errors (vtmtrinh@caltech.edu) In-Reply-To: <3B28E8A6-BCDB-4223-82A4-805689ED2418@gmail.com> References: <3B28E8A6-BCDB-4223-82A4-805689ED2418@gmail.com> Message-ID: <3849.137.78.73.28.1254952398.squirrel@webmail.caltech.edu> Dear Miguel and Mikiyas, Thanks for your help. Yes, I am sure that the outdirs are in the same directory. My main purpose is to calculate the phonon dispersion curve from Gamma to X for Si. I'd like to reproduce the Si phonon dispersion curve from Gamma to X. I thought that after the scf calculation for 10x10x10, I can run phonon calculation directly for a series of q points from Gamma to X (let say 5 points). Then from there, I use q2r.x and then dynmat.x to calculate intermediate points in order to fix the non-zero frequency at Gamma point (applying ASR) and to get a smooth dispersion curve curve. However, the phonon calculation (the first crash) crashed all the times. Now I am testing by running phonon calculation for a mesh of 4x4x4 to see if it still crashes. If it works, this means that direct calculation of phonon by specifying directly q points along G-X does not work. However, this will be expensive calculation if to get phonon dispersion a long one direction, I have to do the 3D-messh phonon calculation first. If I did only a single q-point calculation one at a time, which would be much faster than does the approach above, how could I specify q-grid in q2r.x? Also, I have another question, if I would like to apply ASR to fix the phonon freq. at Gamma point, is it correct to just specify the q-grid in the input files for q2r.x and matdyn.x as follows: &input fildyn='si.dyn', zasr='crystal', flfrc='si.fc' / 1 1 1 1 si.dyn1 !this the *.dyn file at gamma Best, MyTrinh > > On Oct 7, 2009, at 3:24 AM, pw_forum-request at pwscf.org wrote: > >> >> Message: 7 >> Date: Wed, 7 Oct 2009 00:07:18 -0700 (PDT) >> From: vtmtrinh at caltech.edu >> Subject: [Pw_forum] phonon errors >> To: pw_forum at pwscf.org >> Message-ID: >> <2940.137.78.73.28.1254899238.squirrel at webmail.caltech.edu> >> Content-Type: text/plain;charset=iso-8859-1 >> >> Dear PWSCF Users, >> >> I have 2 questions: >> >> 1. I did a test phonon calculation for Si bulk, 2atom-cell, following >> example_06. I first did an scf calculation with k-point mesh >> 10x10x10 Si. >> Next, I wanted to do a phonon calculation along [001] direction (i.e >> q=0, >> q=0.1, 0.2,...). If I specified 'ldisp=.true.', it was supposed to do >> nscf calculation and then phonon calculation in a series. However, it >> stopped with errors. When I checked the output file, after it had >> finished calculation at Gamma point and nscf calculations, when it >> entered >> the phonon calculations for q !=0, it stopped with the error: >> task # 8 >> from openfilq : error # 1 >> file si.wfc not found >> The run is supposed to run continously as I specified ldisp=.true. >> Now >> suddenly, it stopped with that error, so I could not understand >> why. When >> I tried with different q-point mesh, e.g. 4x4x4. I ahd the same >> problem. >> >> Here is my input for the phonon calculation: >> >> phonons of Si bulk >> &inputph >> tr2_ph=1.0d-14, >> prefix='si', >> ldisp=.true., >> nq1=1, nq2=1, nq3=10 !(or nq1=4, nq2=4, nq3=4 for second test) >> amass(1)=28.086, >> outdir='./', >> fildyn='si.dyn', >> / >> >> 2. If I run each single q-point,i.e. did scf, then Gamma >> calculation, then >> nscf and phonon calculation for each single q-point, it was fine. I >> obtained 6 dyn files for 6 q-points from 0 to 0.5. However, I had >> problem >> when I used q2r.x. In the input file, q2r.in, if I specified the q- >> point >> mesh: 1 1 10, and below it I provided a list of 6 dyn files that I >> calculated above, I got an error: >> >> task # 0 >> from init : error # 1 >> q not allowed >> >> I am not sure what is the correct input for the q-point mesh in q2r.in >> that I should use in this case. >> >> Below is my q2r.in: >> >> &input >> fildyn='si.dyn', zasr='crystal', flfrc='si.fc' >> / >> 1 1 10 >> 6 >> si.dyn1 >> si.dyn2 >> si.dyn3 >> si.dyn4 >> si.dyn5 >> si.dyn5 >> >> I am using QE version 4.0.3. >> >> Thank you very much, >> >> MyTrinh >> JPL/CakTech >> >> >> > > Dear, > > It is essentially complaining that the data from your first pw.x run > is not found in the outdir under that prefix ("si"). Have you checked > that the outdir matches in both input files? Also, you can manually > make sure that Si.wfc does exist in the outdir. Another helpful thing > to do is to clear the outdir before you do these runs. > > - Mikiyas Tsegaye > University of Virginia > Charlottesville, VA, USA > mastermik at gmail.com > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From eyvaz_isaev at yahoo.com Thu Oct 8 01:01:28 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 7 Oct 2009 16:01:28 -0700 (PDT) Subject: [Pw_forum] phonon errors (vtmtrinh@caltech.edu) In-Reply-To: <3849.137.78.73.28.1254952398.squirrel@webmail.caltech.edu> Message-ID: <577828.32791.qm@web65702.mail.ac4.yahoo.com> Hi, --- On Thu, 10/8/09, vtmtrinh at caltech.edu wrote: > My main purpose is to calculate the phonon dispersion curve from Gamma > to X for Si. I'd like to reproduce the Si phonon dispersion curve from > Gamma to X.? In order to do this one correctly first you should calculate dynamical matrices for a given q-mesh, then perform FFT by q2r.x, then apply matdyn.x to calculate phonons along given directions (specifying a list of q-points). In this way you apply ASR for both q2r.x and matdyn.x calculations. If you have only \Gamma-point phonons you can use dynmat.x to apply ASR. > I thought that after the scf calculation for 10x10x10, I can > run phonon calculation directly for a series of q points from Gamma to > X (let say 5 points).? Then from there, I use q2r.x and then > dynmat.x to calculate intermediate points in order to fix the non-zero > frequency at Gamma point (applying ASR) and to get a smooth dispersion > curve curve.? This is a wrong way to calculate phonons. Of course, you can calculate phonons for any q-point you like, but they are not oblidged to form a proper mesh for Fast-Fourier-Transformation. In this way you always will meet troubles with q2r.x (FFT). > However, the phonon calculation (the first crash) crashed all the >times.? Now I am testing by running phonon calculation for a mesh of >4x4x4 to see if it still crashes. If it works, this means that direct > calculation of phonon by specifying directly q points along G-X does not > work.? However, this will be expensive calculation if to get phonon > dispersion a long one direction, I have to do the 3D-messh phonon > calculation first. > > If I did only a single q-point calculation one at a time, > which would be much faster than does the approach above, Sorry, I am not sure I exactly understand what you mean, but calculations performed in the right way (see above) give you Interatomic Force Constants (IFC) matrix which allows you to evaluate many thermodynamical properties, besides phonons along high symmetry directions. >how could I specify q-grid in q2r.x? You should not specify q-grid in q2r.x's input file. It (q2r.x) reads Sysname.dyn0 file generated by ph.x which contains info about the q-grid (divisions along each basis vectors, total number of q-points and q-points itself. Of course, you could calculate dynamical matrices in the old manner, specifying a list of q-points (but they have to be a result of KxLxM mesh, and calculated before you start phonon calculations). In this case you can follow your example given below. > > Also, I have another question, if I would like to apply ASR > to fix the phonon freq. at Gamma point, is it correct to just specify > the q-grid in the input files for q2r.x and matdyn.x as follows: > > &input > ? ? fildyn='si.dyn', zasr='crystal', > flfrc='si.fc' > ? / > 1 1 1 > 1 > si.dyn1? ? !this the *.dyn file at gamma I do not think so, as q2r.x calculates IFC Matrix, not phonon modes as matdyn.x does this job. How to apply ASR for Gamma-point phonons see above. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From vtmtrinh at caltech.edu Thu Oct 8 01:14:34 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Wed, 7 Oct 2009 16:14:34 -0700 (PDT) Subject: [Pw_forum] phonon errors (vtmtrinh@caltech.edu) In-Reply-To: <577828.32791.qm@web65702.mail.ac4.yahoo.com> References: <577828.32791.qm@web65702.mail.ac4.yahoo.com> Message-ID: <3973.137.78.73.28.1254957274.squirrel@webmail.caltech.edu> Dear Prof. Isaev, Thank you very much for your detailed answer. It is clear for me now. MyTrinh > > Hi, > > --- On Thu, 10/8/09, vtmtrinh at caltech.edu wrote: > >> My main purpose is to calculate the phonon dispersion curve from Gamma >> to X for Si. I'd like to reproduce the Si phonon dispersion curve from >> > Gamma to X.? > > In order to do this one correctly first you should calculate dynamical > matrices for a given q-mesh, then perform FFT by q2r.x, then apply > matdyn.x to calculate phonons along given directions (specifying a list > of q-points). In this way you apply ASR for both q2r.x and matdyn.x > calculations. > > If you have only \Gamma-point phonons you can use dynmat.x to apply ASR. > >> I thought that after the scf calculation for 10x10x10, I can >> run phonon calculation directly for a series of q points from Gamma to >> X (let say 5 points).? Then from there, I use q2r.x and then >> dynmat.x to calculate intermediate points in order to fix the non-zero >> > frequency at Gamma point (applying ASR) and to get a smooth >> dispersion > curve curve.? > > This is a wrong way to calculate phonons. Of course, you can calculate > phonons for any q-point you like, but they are not oblidged to form a > proper mesh for Fast-Fourier-Transformation. In this way you always will > meet troubles with q2r.x (FFT). > >> However, the phonon calculation (the first crash) crashed all the >> >times.? Now I am testing by running phonon calculation for a mesh of >> >4x4x4 to see if it still crashes. If it works, this means that direct >> calculation of phonon by specifying directly q points along G-X does >> not >> work.? However, this will be expensive calculation if to get phonon >> dispersion a long one direction, I have to do the 3D-messh phonon >> calculation first. >> >> If I did only a single q-point calculation one at a time, >> which would be much faster than does the approach above, > > Sorry, I am not sure I exactly understand what you mean, but calculations > performed in the right way (see above) give you Interatomic Force > Constants (IFC) matrix which allows you to evaluate many thermodynamical > properties, besides phonons along high symmetry directions. > >>how could I specify q-grid in q2r.x? > > You should not specify q-grid in q2r.x's input file. It (q2r.x) reads > Sysname.dyn0 file generated by ph.x which contains info about the q-grid > (divisions along each basis vectors, total number of q-points and q-points > itself. > > Of course, you could calculate dynamical matrices in the old manner, > specifying a list of q-points (but they have to be a result of KxLxM mesh, > and calculated before you start phonon calculations). In this case you can > follow your example given below. > >> >> Also, I have another question, if I would like to apply ASR >> to fix the phonon freq. at Gamma point, is it correct to just specify >> the q-grid in the input files for q2r.x and matdyn.x as follows: >> >> &input >> ? ? fildyn='si.dyn', zasr='crystal', >> flfrc='si.fc' >> ? / >> 1 1 1 >> 1 >> si.dyn1? ? !this the *.dyn file at gamma > > I do not think so, as q2r.x calculates IFC Matrix, not phonon modes as > matdyn.x does this job. > How to apply ASR for Gamma-point phonons see above. > > Bests, > Eyvaz. > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, > Russia, > Department of Physics, Chemistry, and Biology (IFM), Linkoping University, > Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From vtmtrinh at caltech.edu Thu Oct 8 07:53:05 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Wed, 7 Oct 2009 22:53:05 -0700 (PDT) Subject: [Pw_forum] phonon errors In-Reply-To: <77AFEA50-D814-405F-9362-E6CF392455C3@democritos.it> References: <2940.137.78.73.28.1254899238.squirrel@webmail.caltech.edu> <77AFEA50-D814-405F-9362-E6CF392455C3@democritos.it> Message-ID: <4512.137.78.73.28.1254981185.squirrel@webmail.caltech.edu> Thanks, Paolo. I'll chack on this too. MyTrinh > Try to find out if the problem you have is related to the > usage of local disks as scratch directories. It is quite > tricky, because you have no control on which filesystem > the various pieces are written to. > > Paolo > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From ttduyle at gmail.com Thu Oct 8 08:55:29 2009 From: ttduyle at gmail.com (Duy Le) Date: Thu, 8 Oct 2009 02:55:29 -0400 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: References: Message-ID: <8974d3b20910072355h2f2fda94y662a5a71209d78ef@mail.gmail.com> Is your question "Why do you get non-zero force for this fcc system with your given Cell_parameters and Atomic_positions?" If you have a perfect fcc Cu structure, you should get 0 force. I think there are some numerical errors, it likes <<<2*0.5*6.81650937063832 .NE.6.81650937063832>>>. This kind of error "moves" the atoms off their ideal positions. I guess if you use crystal coordinate instead of Cartesian coordinate, you will get 0 force. It's just my humble thought, I don't know for sure how pwscf read and process your coordinates. -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. On Wed, Oct 7, 2009 at 5:23 PM, Manoj Srivastava wrote: > ATOMIC_POSITIONS > Cu 0.0 0.0 0.0 > Cu 0.5 0.5 0.0 > Cu 0.0 1.0 0.0 > Cu 1.0 1.0 0.0 > Cu 0.5 1.5 0.0 > > K_POINTS (automatic) > 8 8 8 0 0 0 > CELL_PARAMETERS {cubic} > 1.5 0.5 0.0 > -0.5 1.5 0.0 > 0.0 0.5 0.5 > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091008/b0d36044/attachment.htm From baroni at sissa.it Thu Oct 8 10:37:39 2009 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 8 Oct 2009 10:37:39 +0200 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: References: Message-ID: <126E8F47-A02C-4056-AA89-9729CDE91B93@sissa.it> Hi Manoj. the forces you are getting seem really small to me. If you enforce cubic symmetry, forces vanish exactly. If you do not, the code itself has to "figure out" that the atoms are indeed at their proper equilibrium positions, and this can only be achieved within the given overall numerical accuracy. One thing that may explicitly (although very slightly) break cubic symmetry in your supercell is tha FFT grid. Not that it makes any physical sense, but if you really want to overkill zero fores, you may want to play a little bit with the kinetic-energy cutoff and/or FFT grid. SB On Oct 7, 2009, at 11:23 PM, Manoj Srivastava wrote: > Dear PWSCF users, > I have a question regarding relaxation of low symmetry lattices. I did > relaxation of 5 atoms FCC unit cell of copper. This is pure copper, no > interface, no defect or anything. The unit cell is just not > conventional, > but a different way of choosing bulk FCC. For the default value of > convergence the calculation finishes at the end of 1st scf cycle, > but I > see different forces when I describe nosym=.TRUE. or cell_parameters > {cubic}. Following is the input file. > > &control > calculation='relax' > pseudo_dir = '/home/manoj/espresso-4.0.4/pseudo', > outdir='./', > prefix='lcu', > wf_collect=.TRUE. > / > &system > ibrav =0, > celldm(1)=6.81650937063832 > nat= 5, > ntyp= 1, > ecutwfc =50.0, > occupations='smearing', > smearing='gaussian', > degauss=0.02, > ecutrho=500 > / > &electrons > conv_thr = 1.0e-8 > mixing_beta=0.7 > / > &ions > ion_dynamics='bfgs' > pot_extrapolation='second_order' > wfc_extrapolation='second_order' > / > ATOMIC_SPECIES > Cu 63.55 Cu.pz-d-rrkjus.UPF > ATOMIC_POSITIONS > Cu 0.0 0.0 0.0 > Cu 0.5 0.5 0.0 > Cu 0.0 1.0 0.0 > Cu 1.0 1.0 0.0 > Cu 0.5 1.5 0.0 > > K_POINTS (automatic) > 8 8 8 0 0 0 > CELL_PARAMETERS {cubic} > 1.5 0.5 0.0 > -0.5 1.5 0.0 > 0.0 0.5 0.5 > > Result of relaxation : The covergence is achieved with follwing > forces on > the atoms. > Forces acting on atoms (Ry/au): > > Forces acting on atoms (Ry/au): > > atom 1 type 1 force = 0.00000000 0.00000000 > 0.00000000 > atom 2 type 1 force = 0.00000567 0.00000185 > -0.00000172 > atom 3 type 1 force = 0.00001155 0.00000377 > -0.00000372 > atom 4 type 1 force = -0.00001155 -0.00000377 > 0.00000372 > atom 5 type 1 force = -0.00000567 -0.00000185 > 0.00000172 > > > I realized that I should not have chosen cubic symmetry group, as my > unit > cell is not cubic. So, as described in the input file parameters, > for the > low symmetry lattices, I specifed nosym=.TRUE., and again the > relaxation > stopped after 1 scf cycle, acheiving convergence. But the forces on > the > atoms are - > > Forces acting on atoms (Ry/au): > > atom 1 type 1 force = 0.00001355 0.00000524 > -0.00000516 > atom 2 type 1 force = -0.00001493 -0.00000395 > 0.00000395 > atom 3 type 1 force = 0.00000253 -0.00000043 > 0.00000045 > atom 4 type 1 force = 0.00000820 0.00000230 > -0.00000239 > atom 5 type 1 force = -0.00000935 -0.00000316 > 0.00000315 > > So, still forces are .001 ev/a.u, which i think is higher for a > perfect > FCC lattice. I can specify stricter convergence of force, but that > would > move atoms around and break symmetry. So, as long as the lattice > parameter > is correct, I think for perfect FCC lattice, forces on atoms should be > smaller than what they are presently. I have checked the lattice > constant > for bulk FCC with a different calculation, so lattice constant is > not an > issue. Another possibility is symmetry group, as you can see with > the same > input variables, changing symmetry group changes results.I was > wondering > if there is a way i can choose correct symmetry group for this low > symmetry lattice which might give me smaller forces? Are there any > other > parameters I need to specify? On a broad view I am curious to know > how to > do relaxation of low symmetry lattices. > > Regards, > Manoj Srivastava > University of Florida > Gainesville, USA > > _______________________________________________ > 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/20091008/895d2070/attachment.htm From sclauzer at sissa.it Thu Oct 8 10:46:13 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 08 Oct 2009 10:46:13 +0200 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: References: Message-ID: <4ACDA6D5.5070406@sissa.it> Dear Manoj, Manoj Srivastava wrote: > Result of relaxation : The covergence is achieved with follwing forces on > the atoms. > Forces acting on atoms (Ry/au): > > Forces acting on atoms (Ry/au): > > atom 1 type 1 force = 0.00000000 0.00000000 0.00000000 > atom 2 type 1 force = 0.00000567 0.00000185 -0.00000172 > atom 3 type 1 force = 0.00001155 0.00000377 -0.00000372 > atom 4 type 1 force = -0.00001155 -0.00000377 0.00000372 > atom 5 type 1 force = -0.00000567 -0.00000185 0.00000172 > > > I realized that I should not have chosen cubic symmetry group, as my unit > cell is not cubic. So, as described in the input file parameters, for the in my understanding, the option cubic does not mean that you use a cubic symmetry group, but that you search for symmetry operations starting from the cubic ones (and removing those which do not hold in your system). If you try with hexagonal it will find only 2 symmetries (identity and inversion), if you use cubic it should find 2 more, but they are not compatible with your FFT grid (see at the beginning of the output). > low symmetry lattices, I specifed nosym=.TRUE., and again the relaxation > stopped after 1 scf cycle, acheiving convergence. But the forces on the > atoms are - > > Forces acting on atoms (Ry/au): > > atom 1 type 1 force = 0.00001355 0.00000524 -0.00000516 > atom 2 type 1 force = -0.00001493 -0.00000395 0.00000395 > atom 3 type 1 force = 0.00000253 -0.00000043 0.00000045 > atom 4 type 1 force = 0.00000820 0.00000230 -0.00000239 > atom 5 type 1 force = -0.00000935 -0.00000316 0.00000315 > > So, still forces are .001 ev/a.u, which i think is higher for a perfect From your data I see that all components of the forces are lower than 0.0002 ev/au, which is pretty small for practical purposes. > FCC lattice. I can specify stricter convergence of force, but that would > move atoms around and break symmetry. So, as long as the lattice parameter Typically, a good convergence threshold for forces in relaxations is around 0.01 ev/au, which is 2 orders of magnitude higher than your residual forces. > is correct, I think for perfect FCC lattice, forces on atoms should be > smaller than what they are presently. I have checked the lattice constant You have to take into account that you are using several approximations in the code (as in any other numeric solver), and you are using a finite basis set to represent wavefunctions and charge density, a finite number of k-point for sampling the BZ,... > for bulk FCC with a different calculation, so lattice constant is not an > issue. Another possibility is symmetry group, as you can see with the same sure, lattice constant cannot be an issue in this case, since it simply rescales proportionally the lattice vectors lengths > input variables, changing symmetry group changes results.I was wondering > if there is a way i can choose correct symmetry group for this low > symmetry lattice which might give me smaller forces? Are there any other The code gives you the correct symmetry group of your system, which depends on the symmetry of the Bravais lattice and can be lowered if you have more than one atom per cell (sometimes it gives you a subgroup for the FFT incompatibility issues I mentioned). I don't see any other group that you could use other than the correct one... for sure not one of higher order: that makes no sense to me. > parameters I need to specify? On a broad view I am curious to know how to > do relaxation of low symmetry lattices. Usually you lower the symmetry by purpose in order to let atoms explore non-symmetric configurations (surfaces, defects, nanocontacts, ...), so I don't understand you aim of chosing a low symmetry lattice to describe a very high symmetry system. > > Regards, > Manoj Srivastava > University of Florida > Gainesville, USA > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From dalcorso at sissa.it Thu Oct 8 10:52:52 2009 From: dalcorso at sissa.it (Dal Corso Andrea) Date: Thu, 08 Oct 2009 10:52:52 +0200 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: <126E8F47-A02C-4056-AA89-9729CDE91B93@sissa.it> References: <126E8F47-A02C-4056-AA89-9729CDE91B93@sissa.it> Message-ID: <1254991972.3263.4.camel@dhpc-5-36.sissa.it> On Thu, 2009-10-08 at 10:37 +0200, Stefano Baroni wrote: > Hi Manoj. the forces you are getting seem really small to me. > If you enforce cubic symmetry, forces vanish exactly. If you do not, > the code itself has to "figure out" that the atoms are indeed at their > proper equilibrium positions, and this can only be achieved within the > given overall numerical accuracy. One thing that may explicitly > (although very slightly) break cubic symmetry in your supercell is tha > FFT grid. Not that it makes any physical sense, but if you really want > to overkill zero fores, you may want to play a little bit with the > kinetic-energy cutoff and/or FFT grid. SB I would also decrease the threshold of the self-consistency conv_thr. HTH Andrea > > On Oct 7, 2009, at 11:23 PM, Manoj Srivastava wrote: > > > Dear PWSCF users, > > I have a question regarding relaxation of low symmetry lattices. I > > did > > relaxation of 5 atoms FCC unit cell of copper. This is pure copper, > > no > > interface, no defect or anything. The unit cell is just not > > conventional, > > but a different way of choosing bulk FCC. For the default value of > > convergence the calculation finishes at the end of 1st scf cycle, > > but I > > see different forces when I describe nosym=.TRUE. or cell_parameters > > {cubic}. Following is the input file. > > > > &control > > calculation='relax' > > pseudo_dir = '/home/manoj/espresso-4.0.4/pseudo', > > outdir='./', > > prefix='lcu', > > wf_collect=.TRUE. > > / > > &system > > ibrav =0, > > celldm(1)=6.81650937063832 > > nat= 5, > > ntyp= 1, > > ecutwfc =50.0, > > occupations='smearing', > > smearing='gaussian', > > degauss=0.02, > > ecutrho=500 > > / > > &electrons > > conv_thr = 1.0e-8 > > mixing_beta=0.7 > > / > > &ions > > ion_dynamics='bfgs' > > pot_extrapolation='second_order' > > wfc_extrapolation='second_order' > > / > > ATOMIC_SPECIES > > Cu 63.55 Cu.pz-d-rrkjus.UPF > > ATOMIC_POSITIONS > > Cu 0.0 0.0 0.0 > > Cu 0.5 0.5 0.0 > > Cu 0.0 1.0 0.0 > > Cu 1.0 1.0 0.0 > > Cu 0.5 1.5 0.0 > > > > K_POINTS (automatic) > > 8 8 8 0 0 0 > > CELL_PARAMETERS {cubic} > > 1.5 0.5 0.0 > > -0.5 1.5 0.0 > > 0.0 0.5 0.5 > > > > Result of relaxation : The covergence is achieved with follwing > > forces on > > the atoms. > > Forces acting on atoms (Ry/au): > > > > Forces acting on atoms (Ry/au): > > > > atom 1 type 1 force = 0.00000000 0.00000000 > > 0.00000000 > > atom 2 type 1 force = 0.00000567 0.00000185 > > -0.00000172 > > atom 3 type 1 force = 0.00001155 0.00000377 > > -0.00000372 > > atom 4 type 1 force = -0.00001155 -0.00000377 > > 0.00000372 > > atom 5 type 1 force = -0.00000567 -0.00000185 > > 0.00000172 > > > > > > I realized that I should not have chosen cubic symmetry group, as my > > unit > > cell is not cubic. So, as described in the input file parameters, > > for the > > low symmetry lattices, I specifed nosym=.TRUE., and again the > > relaxation > > stopped after 1 scf cycle, acheiving convergence. But the forces on > > the > > atoms are - > > > > Forces acting on atoms (Ry/au): > > > > atom 1 type 1 force = 0.00001355 0.00000524 > > -0.00000516 > > atom 2 type 1 force = -0.00001493 -0.00000395 > > 0.00000395 > > atom 3 type 1 force = 0.00000253 -0.00000043 > > 0.00000045 > > atom 4 type 1 force = 0.00000820 0.00000230 > > -0.00000239 > > atom 5 type 1 force = -0.00000935 -0.00000316 > > 0.00000315 > > > > So, still forces are .001 ev/a.u, which i think is higher for a > > perfect > > FCC lattice. I can specify stricter convergence of force, but that > > would > > move atoms around and break symmetry. So, as long as the lattice > > parameter > > is correct, I think for perfect FCC lattice, forces on atoms should > > be > > smaller than what they are presently. I have checked the lattice > > constant > > for bulk FCC with a different calculation, so lattice constant is > > not an > > issue. Another possibility is symmetry group, as you can see with > > the same > > input variables, changing symmetry group changes results.I was > > wondering > > if there is a way i can choose correct symmetry group for this low > > symmetry lattice which might give me smaller forces? Are there any > > other > > parameters I need to specify? On a broad view I am curious to know > > how to > > do relaxation of low symmetry lattices. > > > > Regards, > > Manoj Srivastava > > University of Florida > > Gainesville, USA > > > > _______________________________________________ > > 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 -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34151 Trieste (Italy) e-mail: dalcorso at sissa.it From paulatto at sissa.it Thu Oct 8 10:59:25 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 08 Oct 2009 10:59:25 +0200 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: <4ACDA6D5.5070406@sissa.it> References: <4ACDA6D5.5070406@sissa.it> Message-ID: In data 08 ottobre 2009 alle ore 10:46:13, Gabriele Sclauzero ha scritto: > if you use cubic it should find 2 more, but they are not compatible with > your FFT grid (see at the beginning of the output). The code will only use the 2 additional operations if nr3 is a multiple of nr1 and nr2, which is weird as it is, by default, much smaller in your case. Anyway, if you set nr3 equal to nr1 it will use all the symmetry operations. For the symmetry experts: could you please check if the code behaviour is correct? cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From aqswde30 at gmail.com Thu Oct 8 16:38:03 2009 From: aqswde30 at gmail.com (Luciano C.) Date: Thu, 8 Oct 2009 11:38:03 -0300 Subject: [Pw_forum] Use of Pseudo potential for functional blyp in the calculation with B3LYP functional Message-ID: Hello, I have a question about the use of pseudopotenciais. I could use a pseudo potential generated for the functional blyp in the calculation with the B3LYP functional ??? Grateful Luciano -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091008/6aa29e03/attachment.htm From ttduyle at gmail.com Thu Oct 8 17:17:45 2009 From: ttduyle at gmail.com (Duy Le) Date: Thu, 8 Oct 2009 11:17:45 -0400 Subject: [Pw_forum] Use of Pseudo potential for functional blyp in the calculation with B3LYP functional In-Reply-To: References: Message-ID: <8974d3b20910080817o13c56eeblbda3cbed1c76997b@mail.gmail.com> BLYP was not mentioned in what can qe do (QE website) -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. On Thu, Oct 8, 2009 at 10:38 AM, Luciano C. wrote: > Hello, > > I have a question about the use of pseudopotenciais. I could use a pseudo > potential generated for the functional blyp in the calculation with the > B3LYP functional ??? > > Grateful > > Luciano > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091008/ceb8e567/attachment.htm From wparker at mps.ohio-state.edu Thu Oct 8 17:01:43 2009 From: wparker at mps.ohio-state.edu (William Parker) Date: Thu, 8 Oct 2009 11:01:43 -0400 (EDT) Subject: [Pw_forum] Use of Pseudo potential for functional blyp in the calculation with B3LYP functional In-Reply-To: References: Message-ID: Dear Luciano, Where are you from, and what are you trying to do? You can use a pseudopotential generated by any means with the B3LYP functional (use the input_dft='b3lyp' in your input file). Mismatching with the pseudopotential exchange-correlation functional is inconsistent, but you can control for this by trying pseudopotentials from a few different exchange-correlation functionals to estimate how large of an error the mismatch creates for your calculation. I know of no pseudopotential generator that has B3LYP in it, but I would be happy to hear otherwise from someone else. --William ====================================================================== William Parker Tel : +1 (614) 292-2887 Graduate Research Associate Fax : +1 (614) 292-7557 Ohio State University Department of Physics 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker U.S.A. Office: 2025 Physics Research Building ====================================================================== On Thu, 8 Oct 2009, Luciano C. wrote: > Hello, > > I have a question about the use of pseudopotenciais. I could use a pseudo > potential generated for the functional blyp in the calculation with the > B3LYP functional ??? > > Grateful > > Luciano > From giannozz at democritos.it Thu Oct 8 18:27:39 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 08 Oct 2009 18:27:39 +0200 Subject: [Pw_forum] Use of Pseudo potential for functional blyp in the calculation with B3LYP functional In-Reply-To: <8974d3b20910080817o13c56eeblbda3cbed1c76997b@mail.gmail.com> References: <8974d3b20910080817o13c56eeblbda3cbed1c76997b@mail.gmail.com> Message-ID: <4ACE12FB.10209@democritos.it> Duy Le wrote: > BLYP was not mentioned in what can qe do (QE website) it wasn't, but it is now P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From manoj at phys.ufl.edu Fri Oct 9 09:26:43 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Fri, 9 Oct 2009 03:26:43 -0400 (EDT) Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: Message-ID: Dear users, Thank you all for quick reply. Through all the replies I am under the impression that the forces can never be actually zero, even for perfect FCC lattice. I dont know the exact reason yet, whether it is just numerical or something bad on my side. So, I would like to give it one more shot, and i have some questions on that. Duy Le: I will try using crystal and report to the forum. Although relaxation stops after completing one scf cycle, so atoms actually dont move, but still nonzero forces. S. Baroni: When you use cubic symmetry, do you get forces on all atoms zero. Did you do the calculation or is this your intution? also, can one use cubic symmetry for low symmetry unit cell as in my case? I increased the kinetic energy cutoff to a really high value 150Ry !! forces are - atom 1 type 1 force = -0.00000074 -0.00000315 0.00000310 atom 2 type 1 force = 0.00000084 -0.00000391 0.00000354 atom 3 type 1 force = -0.00000128 0.00000470 -0.00000401 atom 4 type 1 force = 0.00000111 0.00000044 -0.00000083 atom 5 type 1 force = 0.00000008 0.00000193 -0.00000181 which is one order of magnitude smaller than before :) but the KE cutoff is ridiculously high! I also increase k points with keeping KE cutoff 50 Ry, and that gave me - atom 1 type 1 force = -0.00000103 0.00000086 -0.00000038 atom 2 type 1 force = 0.00000535 -0.00000155 -0.00000025 atom 3 type 1 force = 0.00000643 0.00000039 -0.00000244 atom 4 type 1 force = -0.00000603 -0.00000141 0.00000310 atom 5 type 1 force = -0.00000471 0.00000171 -0.00000003 which is also one order of magnitude smaller than before, so i am guessing there might me some other parameters that are not set up correctly and causing symmetry breaking? Would someone mind to tell me? I dont know much about FFT grid, Isn't FFT dimension determined by the KE cutoff? Is there anyother way of changing it? Gabriele: Would you mind to tell me how can i get rid of FFT incompatibility issue? Andrea: I changed it from e(-8) to e(-10), still no improvement. Once again, thank you all for your help. Regards, Manoj Srivastava University of Florida Gainesville, FL On Thu, 8 Oct 2009, Lorenzo Paulatto wrote: > In data 08 ottobre 2009 alle ore 10:46:13, Gabriele Sclauzero > ha scritto: > > if you use cubic it should find 2 more, but they are not compatible with > > your FFT grid (see at the beginning of the output). > > The code will only use the 2 additional operations if nr3 is a multiple of > nr1 and nr2, which is weird as it is, by default, much smaller in your > case. Anyway, if you set nr3 equal to nr1 it will use all the symmetry > operations. > > For the symmetry experts: could you please check if the code behaviour is > correct? > > cheers > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From paulatto at sissa.it Fri Oct 9 09:46:52 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 9 Oct 2009 09:46:52 +0200 (CEST) Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: References: Message-ID: <60090.78.12.158.197.1255074412.squirrel@webmail.sissa.it> On Fri, October 9, 2009 09:26, Manoj Srivastava wrote: > which is one order of magnitude smaller than before :) but the KE cutoff > is ridiculously high! I also increase k points with keeping KE cutoff 50 > Ry, and that gave me - Your initial choice of k-point grid is not very wise: as you cell is much longer in the x and y direction than in the z direction you will need more points in z than in x and y. I would say about 2 - 3 times as many. e.g. you could choose: K_POINTS automatic 6 6 14 0 0 0 > Gabriele: Would you mind to tell me how can i get rid of FFT > incompatibility issue? I just told you, in my previous email. Yet I don't think it's a good idea, and the forces won't be zero anyway. Your lattice have lower symmetry than the FCC lattice (it is actually a supercell), hence you can only have zero forces up to the numerical accuracy. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From paulatto at sissa.it Fri Oct 9 10:20:03 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 9 Oct 2009 10:20:03 +0200 (CEST) Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: References: Message-ID: <48251.78.12.158.197.1255076403.squirrel@webmail.sissa.it> On Wed, October 7, 2009 17:10, Pallister, Peter wrote: > Hi all, > > I am attempting to generate a gipaw pseudopotentials, in this case for > Mg, but I am having tremendous difficulty. The wavefunctions produced > with the input below (or minute variations thereof) continuously show > discontinuities, specifically at a radius of 15bohrs. Dear Peter, the wavefunctions are generated on a logarithmic radial grid, around 15 bohr the points are very coarse, btw how are you examining the wavfunctions? (there is not file_wfcncgen='..' in your input). best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From dimpy.sharma at tyndall.ie Fri Oct 9 11:56:24 2009 From: dimpy.sharma at tyndall.ie (Dimpy Sharma) Date: Fri, 9 Oct 2009 10:56:24 +0100 Subject: [Pw_forum] Relax calculation not converging Message-ID: Hi Quantum espresso users, I have been trying to perform a relax calculation in my system with 99 atoms (semiconductor system) and I am running my calculation in 48 processors. The calculation has finished 3 scf steps and finished 100 iterations, but after this it stops and it is showing me the following message . ' convergence NOT achieved after 100 iterations: stopping' I have restart my calculation, still I got the same error message after 100 iteration steps. Can any body give me any suggestion? Thanks Dimpy -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/1191e362/attachment.htm From mighfar at jncasr.ac.in Fri Oct 9 12:38:39 2009 From: mighfar at jncasr.ac.in (Mighfar Imam) Date: Fri, 9 Oct 2009 16:08:39 +0530 (IST) Subject: [Pw_forum] Relax calculation not converging In-Reply-To: References: Message-ID: <49076.172.16.1.1.1255084719.squirrel@172.16.1.1> Convergence might be slow or oscillatory, try changing mixing beta. mighfar imam JNCASR, B'lore. India. Dimpy Sharma > > Hi Quantum espresso users, > > I have been trying to perform a relax calculation in my system with 99 > atoms (semiconductor system) and I am running my calculation in 48 > processors. The calculation has finished 3 scf steps and finished 100 > iterations, but after this it stops and it is showing me the following > message . > ' convergence NOT achieved after 100 iterations: stopping' > > I have restart my calculation, still I got the same error message after > 100 iteration steps. > > Can any body give me any suggestion? > > Thanks > > Dimpy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From aryjunior at gmail.com Fri Oct 9 13:02:17 2009 From: aryjunior at gmail.com (Ary Junior) Date: Fri, 9 Oct 2009 08:02:17 -0300 Subject: [Pw_forum] Relax calculation not converging In-Reply-To: <49076.172.16.1.1.1255084719.squirrel@172.16.1.1> References: <49076.172.16.1.1.1255084719.squirrel@172.16.1.1> Message-ID: <4c9c61c20910090402h37425735rbb10f857d2f0ce55@mail.gmail.com> Did you make any convergence tests for ecut? The set of PW that you are using is enough? And about your k-points sample? [] Ary Junior On Fri, Oct 9, 2009 at 7:38 AM, Mighfar Imam wrote: > Convergence might be slow or oscillatory, try changing mixing beta. > > mighfar imam > JNCASR, B'lore. > India. > > > Dimpy Sharma > > > > Hi Quantum espresso users, > > > > I have been trying to perform a relax calculation in my system with 99 > > atoms (semiconductor system) and I am running my calculation in 48 > > processors. The calculation has finished 3 scf steps and finished 100 > > iterations, but after this it stops and it is showing me the following > > message . > > ' convergence NOT achieved after 100 iterations: stopping' > > > > I have restart my calculation, still I got the same error message after > > 100 iteration steps. > > > > Can any body give me any suggestion? > > > > Thanks > > > > Dimpy > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- http://lattes.cnpq.br/8221674673413336 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/f543d972/attachment.htm From giannozz at democritos.it Fri Oct 9 13:54:32 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 9 Oct 2009 13:54:32 +0200 Subject: [Pw_forum] Relax calculation not converging In-Reply-To: References: Message-ID: <4BEA21B6-6BEA-4473-8328-8039DDB3AF53@democritos.it> On Oct 9, 2009, at 11:56 , Dimpy Sharma wrote: > ' convergence NOT achieved after 100 iterations: stopping' > please see here: http://www.quantum-espresso.org/user_guide/ node76.html#SECTION00096020000000000000 P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From sclauzer at sissa.it Fri Oct 9 14:15:16 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 09 Oct 2009 14:15:16 +0200 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: References: Message-ID: <4ACF2954.5030401@sissa.it> Manoj Srivastava wrote: > Dear users, > Thank you all for quick reply. Through all the replies I am under the > impression that the forces can never be actually zero, even for perfect > FCC lattice. I dont know the exact reason yet, whether it is just > numerical or something bad on my side. So, I would like to give it one > more shot, and i have some questions on that. > > Duy Le: I will try using crystal and report to the forum. Although > relaxation stops after completing one scf cycle, so atoms actually dont > move, but still nonzero forces. > > S. Baroni: When you use cubic symmetry, do you get forces on all atoms > zero. Did you do the calculation or is this your intution? also, can one > use cubic symmetry for low symmetry unit cell as in my case? As I wrote you in my previous mail, it is not that you use cubic symmetry, but you start from cubic symmetry operations to find the actual symmetries (by removing those which do not hold anymore depending on the lattice and atomic positions). You could do the exercise to find out by hand what the symmetry should be in your case and check that the symmetry operations correspond to those found by pw.x > > which is also one order of magnitude smaller than before, so i am > guessing there might me some other parameters that are not set up > correctly and causing symmetry breaking? Would someone mind to tell me? I > dont know much about FFT grid, Isn't FFT dimension determined by the KE > cutoff? Is there anyother way of changing it? I don't know what the cause could be, anyway, since you seem to be very curious and to have some spare time for testing, I can suggest the following: in forces.f90 uncomment the section between #if defined (DEBUG) and #endif and look at the various contributions to the forces (this is present in the CVS version, I don't know if also in the released one). By looking at the single terms you might find why they do not cancel exactly. Cheers GS > > Regards, > Manoj Srivastava > University of Florida > Gainesville, FL > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From xinhuolin at gmail.com Fri Oct 9 15:59:17 2009 From: xinhuolin at gmail.com (Huolin Xin) Date: Fri, 9 Oct 2009 09:59:17 -0400 Subject: [Pw_forum] Relax calculation not converging In-Reply-To: <4c9c61c20910090402h37425735rbb10f857d2f0ce55@mail.gmail.com> References: <49076.172.16.1.1.1255084719.squirrel@172.16.1.1> <4c9c61c20910090402h37425735rbb10f857d2f0ce55@mail.gmail.com> Message-ID: For such big unit cells, even Gamma point should work, right? On Fri, Oct 9, 2009 at 7:02 AM, Ary Junior wrote: > Did you make any convergence tests for ecut? The set of PW that you are > using is enough? And about your k-points sample? > > [] > > Ary Junior > > > On Fri, Oct 9, 2009 at 7:38 AM, Mighfar Imam wrote: > >> Convergence might be slow or oscillatory, try changing mixing beta. >> >> mighfar imam >> JNCASR, B'lore. >> India. >> >> >> Dimpy Sharma >> > >> > Hi Quantum espresso users, >> > >> > I have been trying to perform a relax calculation in my system with 99 >> > atoms (semiconductor system) and I am running my calculation in 48 >> > processors. The calculation has finished 3 scf steps and finished 100 >> > iterations, but after this it stops and it is showing me the following >> > message . >> > ' convergence NOT achieved after 100 iterations: stopping' >> > >> > I have restart my calculation, still I got the same error message after >> > 100 iteration steps. >> > >> > Can any body give me any suggestion? >> > >> > Thanks >> > >> > Dimpy >> > _______________________________________________ >> > Pw_forum mailing list >> > Pw_forum at pwscf.org >> > http://www.democritos.it/mailman/listinfo/pw_forum >> > >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > > -- > http://lattes.cnpq.br/8221674673413336 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/d1e55813/attachment-0001.htm From ttduyle at gmail.com Fri Oct 9 16:13:31 2009 From: ttduyle at gmail.com (Duy Le) Date: Fri, 9 Oct 2009 10:13:31 -0400 Subject: [Pw_forum] Relax calculation not converging In-Reply-To: References: <49076.172.16.1.1.1255084719.squirrel@172.16.1.1> <4c9c61c20910090402h37425735rbb10f857d2f0ce55@mail.gmail.com> Message-ID: <8974d3b20910090713s47e8fcbk82c942ee40673b87@mail.gmail.com> If it is not converged, and you restart it again without any modification (or dirty trick), it should not be converged. 1. Grep total energy of last few iteration, if they are converging, you may change the electron_maxstep to 200, 300, 5000 :) or you may increase the conv_th (after getting the relaxed structure, restart the calculation with new structure and read the potential, wavefunction, lower the conv_th...) 2. Play with Mixing_mode and mixing_beta 3. Change diagonalization -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. On Fri, Oct 9, 2009 at 9:59 AM, Huolin Xin wrote: > For such big unit cells, even Gamma point should work, right? > > > On Fri, Oct 9, 2009 at 7:02 AM, Ary Junior wrote: > >> Did you make any convergence tests for ecut? The set of PW that you are >> using is enough? And about your k-points sample? >> >> [] >> >> Ary Junior >> >> >> On Fri, Oct 9, 2009 at 7:38 AM, Mighfar Imam wrote: >> >>> Convergence might be slow or oscillatory, try changing mixing beta. >>> >>> mighfar imam >>> JNCASR, B'lore. >>> India. >>> >>> >>> Dimpy Sharma >>> > >>> > Hi Quantum espresso users, >>> > >>> > I have been trying to perform a relax calculation in my system with 99 >>> > atoms (semiconductor system) and I am running my calculation in 48 >>> > processors. The calculation has finished 3 scf steps and finished 100 >>> > iterations, but after this it stops and it is showing me the following >>> > message . >>> > ' convergence NOT achieved after 100 iterations: stopping' >>> > >>> > I have restart my calculation, still I got the same error message after >>> > 100 iteration steps. >>> > >>> > Can any body give me any suggestion? >>> > >>> > Thanks >>> > >>> > Dimpy >>> > _______________________________________________ >>> > Pw_forum mailing list >>> > Pw_forum at pwscf.org >>> > http://www.democritos.it/mailman/listinfo/pw_forum >>> > >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> >> >> >> -- >> http://lattes.cnpq.br/8221674673413336 >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/bd5a92a3/attachment.htm From giannozz at democritos.it Fri Oct 9 16:20:21 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 9 Oct 2009 16:20:21 +0200 Subject: [Pw_forum] Relax calculation not converging In-Reply-To: <8974d3b20910090713s47e8fcbk82c942ee40673b87@mail.gmail.com> References: <49076.172.16.1.1.1255084719.squirrel@172.16.1.1> <4c9c61c20910090402h37425735rbb10f857d2f0ce55@mail.gmail.com> <8974d3b20910090713s47e8fcbk82c942ee40673b87@mail.gmail.com> Message-ID: On Oct 9, 2009, at 16:13 , Duy Le wrote: > 1. Grep total energy of last few iteration, if they are converging, > you may change the electron_maxstep to 200, 300, 5000 :) not a good advice. Self-consistency should converge in less than 100 iterations (and it does, in non-pathological cases) P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From silveto at gmail.com Fri Oct 9 16:53:18 2009 From: silveto at gmail.com (Silvia Bakalova) Date: Fri, 9 Oct 2009 15:53:18 +0100 Subject: [Pw_forum] energy-degenerate states with different irrep labels Message-ID: Hi, I have one question: for spin-orbit calculations, the energy bands are labelled with double point group notation (D3d? in my case). Some of the energy-degenerate states have different irrep labels and I wonder why? e.g. the valence band top: e( 45 - 46) = 7.09176 eV 2 --> G_5+ L_4+ e( 45 - 46) = 7.09176 eV 2 --> G_6+ L_5+ I would be grateful for your reply or some literature reference, as I am not familiar with the group theory. Many thanks, Silvia ---------------------- Dr. Silvia Bakalova, Post Doctoral Researcher, HH Wills Physics Laboratory, Bristol, BS8 1TL, UK http://spectra.phy.bris.ac.uk/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/f4dcc61b/attachment.htm From sclauzer at sissa.it Fri Oct 9 16:59:49 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 09 Oct 2009 16:59:49 +0200 Subject: [Pw_forum] energy-degenerate states with different irrep labels In-Reply-To: References: Message-ID: <4ACF4FE5.1090307@sissa.it> Silvia Bakalova wrote: > Hi, > > I have one question: for spin-orbit calculations, the energy bands are > labelled with double point group notation (D3d? in my case). > > Some of the energy-degenerate states have different irrep labels and I > wonder why? > At which k-point? I think that at Gamma they have to be degenerate if time reversal symmetry holds (i.e. no magnetic field). GS > e.g. the valence band top: > > e( 45 - 46) = 7.09176 eV 2 --> G_5+ L_4+ > > e( 45 - 46) = 7.09176 eV 2 --> G_6+ L_5+ > > I would be grateful for your reply or some literature reference, as I am > not familiar with the group theory. > > Many thanks, > > Silvia > > > ---------------------- > Dr. Silvia Bakalova, > Post Doctoral Researcher, > HH Wills Physics Laboratory, > Bristol, BS8 1TL, UK > http://spectra.phy.bris.ac.uk/ > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From silveto at gmail.com Fri Oct 9 17:56:02 2009 From: silveto at gmail.com (Silvia Bakalova) Date: Fri, 9 Oct 2009 16:56:02 +0100 Subject: [Pw_forum] energy-degenerate states with different irrep labels In-Reply-To: References: Message-ID: Thank you for the reply, Gabriele. Yes, this is @Gamma, the states are degenerate, but I wonder why they have different irreducible representations (G_5+ and G_6+)? Silvia Bakalova wrote: >* Hi, *>* *>* I have one question: for spin-orbit calculations, the energy bands are *>* labelled with double point group notation (D3d? in my case). *>* *>* Some of the energy-degenerate states have different irrep labels and I *>* wonder why? *>* * At which k-point? I think that at Gamma they have to be degenerate if time reversal symmetry holds (i.e. no magnetic field). GS >* e.g. the valence band top: *>* *>* e( 45 - 46) = 7.09176 eV 2 --> G_5+ L_4+ *>* *>* e( 45 - 46) = 7.09176 eV 2 --> G_6+ L_5+ *>* *>* I would be grateful for your reply or some literature reference, as I am *>* not familiar with the group theory. *>* *>* Many thanks, *>* *>* Silvia *>* *>* *>* ---------------------- *>* Dr. Silvia Bakalova, *>* Post Doctoral Researcher, *>* HH Wills Physics Laboratory, *>* Bristol, BS8 1TL, UK *>* http://spectra.phy.bris.ac.uk/ *>* *>* *>* *>* ------------------------------------------------------------------------ *>* *>* _______________________________________________ *>* Pw_forum mailing list *>* Pw_forum at pwscf.org *>* http://www.democritos.it/mailman/listinfo/pw_forum * -- o ------------------------------------------------ o |* Gabriele Sclauzero, PhD Student | *|* c/o: SISSA & CNR-INFM Democritos, | *|* via Beirut 2-4, 34014 Trieste (Italy) | *|* email: sclauzer at sissa.it | *|* phone: +39 040 3787 511 | *|* skype: gurlonotturno | *o ------------------------------------------------ o -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/d63ab4be/attachment.htm From manoj at phys.ufl.edu Fri Oct 9 19:35:16 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Fri, 9 Oct 2009 13:35:16 -0400 (EDT) Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: <4ACF2954.5030401@sissa.it> Message-ID: Dear Lorenzo and Gabriele, Thank you very much for your suggestions. I will try them out. Regards, Manoj On Fri, 9 Oct 2009, Gabriele Sclauzero wrote: > > > Manoj Srivastava wrote: > > Dear users, > > Thank you all for quick reply. Through all the replies I am under the > > impression that the forces can never be actually zero, even for perfect > > FCC lattice. I dont know the exact reason yet, whether it is just > > numerical or something bad on my side. So, I would like to give it one > > more shot, and i have some questions on that. > > > > Duy Le: I will try using crystal and report to the forum. Although > > relaxation stops after completing one scf cycle, so atoms actually dont > > move, but still nonzero forces. > > > > S. Baroni: When you use cubic symmetry, do you get forces on all atoms > > zero. Did you do the calculation or is this your intution? also, can one > > use cubic symmetry for low symmetry unit cell as in my case? > > As I wrote you in my previous mail, it is not that you use cubic symmetry, but you start > from cubic symmetry operations to find the actual symmetries (by removing those which do > not hold anymore depending on the lattice and atomic positions). > You could do the exercise to find out by hand what the symmetry should be in your case and > check that the symmetry operations correspond to those found by pw.x > > > > > > which is also one order of magnitude smaller than before, so i am > > guessing there might me some other parameters that are not set up > > correctly and causing symmetry breaking? Would someone mind to tell me? I > > dont know much about FFT grid, Isn't FFT dimension determined by the KE > > cutoff? Is there anyother way of changing it? > > I don't know what the cause could be, anyway, since you seem to be very curious and to > have some spare time for testing, I can suggest the following: in forces.f90 uncomment the > section between #if defined (DEBUG) and #endif and look at the various contributions to > the forces (this is present in the CVS version, I don't know if also in the released one). > By looking at the single terms you might find why they do not cancel exactly. > > Cheers > > GS > > > > > Regards, > > Manoj Srivastava > > University of Florida > > Gainesville, FL > > > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From ZChen at cau.edu Fri Oct 9 21:19:38 2009 From: ZChen at cau.edu (Chen, Zhifan) Date: Fri, 9 Oct 2009 15:19:38 -0400 Subject: [Pw_forum] use cvs Message-ID: Hi, everyone I installed RE 4.1 in a cluster with 40 processes using compiler mpif90 and pgf90. Machine type is 86_64. OS is Linux. There is no problem during the compile. However, I received the error message when I run first example01 #Error in iotk_scan (iotk_scan.spp:717) # CVS Revision:1.7 # direction Control=2 # Error in iotk_close_read (iotk_files.spp:589) #CVS Revision:1.7 # From IOTK Library, version 1.1.0 development # Unrecoverable error (ierr=2) And more .... May be I can solve the problem by login CVS, I set the environment parameter: Export CVS_RSH=ssh Export CVSROOT=:pserver:anonymous at scm.qe-forge.org:/cvsroot/q-e Then cvs login I press Enter. I did not login but back to my own director. Zhifan Chen Dept of Physics Clark Atlanta University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091009/40407729/attachment.htm From giannozz at democritos.it Sat Oct 10 09:30:08 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 10 Oct 2009 09:30:08 +0200 Subject: [Pw_forum] use cvs In-Reply-To: References: Message-ID: On Oct 9, 2009, at 21:19 , Chen, Zhifan wrote: > #Error in iotk_scan (iotk_scan.spp:717) > you have a buggy compiler > May be I can solve the problem by login CVS > unlikely > cvs login > > I press Enter. I did not login but back to my own director. > that's the correct behavior. Please read the developers guide: there is a section on how to use cvs P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From shypirate at gmail.com Sat Oct 10 14:07:35 2009 From: shypirate at gmail.com (shypirate) Date: Sat, 10 Oct 2009 20:07:35 +0800 Subject: [Pw_forum] What is the meaning of dipfield=.true. ? Message-ID: <200910102007318751954@gmail.com> Dear All, Sorry to send this e-mail to everyone. I want to use saw-tooth electric potential to investigate behaviour of thin film slab under electric field. My system indeed behaves differently under different electric field amplitude. But my question is, what is the meaning if I set dipfield=.true. ? If I use dipfield=.false., all results are wrong, is not it? Any advice from you is helpful to me. Best Wishes, Y.P.Wang 2009-10-10 shypirate -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091010/32fc4ec3/attachment.htm From giannozz at democritos.it Sat Oct 10 15:21:31 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 10 Oct 2009 15:21:31 +0200 Subject: [Pw_forum] What is the meaning of dipfield=.true. ? In-Reply-To: <200910102007318751954@gmail.com> References: <200910102007318751954@gmail.com> Message-ID: <910CA46D-9C22-4621-9771-A5BF5D9FFF41@democritos.it> On Oct 10, 2009, at 14:07 , shypirate wrote: > But my question is, what is the meaning if I set dipfield=.true. ? did you read the documentation for variable dipfield? If .TRUE. and tefield=.TRUE. a dipole correction is also added to the bare ionic potential - implements the recipe of L. Bengtsson, PRB 59, 12301 (1999). See variables edir, emaxpos, eopreg for the form of the correction, that must be used only in a slab geometry, for surface calculations, with the discontinuity in the empty space. > If I use dipfield=.false., all results are wrong, is not it? if you use options without knowing what they do, there are good changes that all your results are wrong P. -- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From shypirate at gmail.com Sat Oct 10 15:59:11 2009 From: shypirate at gmail.com (shypirate) Date: Sat, 10 Oct 2009 21:59:11 +0800 Subject: [Pw_forum] What is the meaning of dipfield=.true. ? References: <200910102007318751954@gmail.com> Message-ID: <200910102159084218647@gmail.com> Thanks fro your quick reply! In my understanding, dipole correction is electrostatic interaction between slabs. After set dipfield=.true. and other settings remain unchanged, self-consistent calculation has trouble to converge. In each self-consistent calculation step, there are several "c_bands: * eigenvalues not converged" warning messages. And, after several self-consistent steps, usually 3 steps, program stops with error message: "too many eigenvalues not converged". Could you give me some advice about this error? Where may be the error in my input file? number of k-points ? My input file for pw.x is like this: &control calculation = 'relax', nstep=200, tprnfor = .TRUE., tefield = .TRUE., dipfield = .TRUE., prefix = 'relax', pseudo_dir = '/home/ypwang/espresso/pseudo/', outdir = './out', / &system ibrav = 6, celldm(1) = 7.35, celldm(3) = 8, nat = 20, ntyp = 5, ecutwfc = 40.0, ecutrho = 480, occupations = 'smearing', degauss = 0.005, smearing = 'mv', edir =3, emaxpos = 0.9, eopreg = 0.1, eamp = 0.00 / &electrons diagonalization = 'cg' mixing_mode = 'local-TF' mixing_beta = 0.3 conv_thr = 1.0d-8 / &IONS bfgs_ndim = 7 / ATOMIC_SPECIES ... K_POINTS automatic 4 4 2 1 1 1 Thanks in advance! 2009-10-10 shypirate ???? Paolo Giannozzi ????? 2009-10-10 21:21:47 ???? PWSCF Forum ??? ??? Re: [Pw_forum] What is the meaning of dipfield=.true. ? On Oct 10, 2009, at 14:07 , shypirate wrote: > But my question is, what is the meaning if I set dipfield=.true. ? did you read the documentation for variable dipfield? If .TRUE. and tefield=.TRUE. a dipole correction is also added to the bare ionic potential - implements the recipe of L. Bengtsson, PRB 59, 12301 (1999). See variables edir, emaxpos, eopreg for the form of the correction, that must be used only in a slab geometry, for surface calculations, with the discontinuity in the empty space. > If I use dipfield=.false., all results are wrong, is not it? if you use options without knowing what they do, there are good changes that all your results are wrong P. -- 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 mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091010/301e2eb4/attachment.htm From decboy9 at gmail.com Sat Oct 10 20:58:06 2009 From: decboy9 at gmail.com (dev sharma) Date: Sun, 11 Oct 2009 00:28:06 +0530 Subject: [Pw_forum] fhi2upf.x Message-ID: hi 2 all, I have read the forum and found that Abinit Norm conserving pseudo potential (fhi ) can be converted with the upftool fhi2upf.x I tried it with the command /home/dev/espresso-4.0.2/upftools/fhi2upf.x <39-Y.LDA.fhi> 39-Y.LDA.UPF but it is giving an error in terminal At line 26 of file fhi2upf.f90 (unit = 1, file = '') Fortran runtime error: File ' 537 0.76677159662378E+02 0.00000000000000E+00 -.39125079922229E-01' does not exist. Sir, How to solve this problem ???? Please Help. Thanks Dev Sharma, University of Delhi -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091011/2906ccbe/attachment.htm From paulatto at sissa.it Sat Oct 10 21:04:04 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 10 Oct 2009 21:04:04 +0200 (CEST) Subject: [Pw_forum] fhi2upf.x In-Reply-To: References: Message-ID: <40672.78.12.170.175.1255201444.squirrel@webmail.sissa.it> On Sat, October 10, 2009 20:58, dev sharma wrote: > /home/dev/espresso-4.0.2/upftools/fhi2upf.x <39-Y.LDA.fhi> 39-Y.LDA.UPF Dear Dev, fhi2upf does not work like that. The correct syntax is: /whatever/fhi2upf.x 39-Y.LDA.fhi which would create 39-Y.LDA.fhi.UPF regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From baroni at sissa.it Sat Oct 10 22:07:01 2009 From: baroni at sissa.it (Stefano Baroni) Date: Sat, 10 Oct 2009 22:07:01 +0200 Subject: [Pw_forum] Relaxation of low symmetry lattices In-Reply-To: <60090.78.12.158.197.1255074412.squirrel@webmail.sissa.it> References: <60090.78.12.158.197.1255074412.squirrel@webmail.sissa.it> Message-ID: <49C89098-36EC-490E-83A8-C2CA45D5A5A8@sissa.it> On Oct 9, 2009, at 9:46 AM, Lorenzo Paulatto wrote: > > On Fri, October 9, 2009 09:26, Manoj Srivastava wrote: >> which is one order of magnitude smaller than before :) but the KE >> cutoff >> is ridiculously high! I also increase k points with keeping KE >> cutoff 50 >> Ry, and that gave me - > > Your initial choice of k-point grid is not very wise: as you cell is > much > longer in the x and y direction than in the z direction you will > need more > points in z than in x and y. I would say about 2 - 3 times as many. > e.g. > you could choose: > K_POINTS automatic > 6 6 14 0 0 0 > >> Gabriele: Would you mind to tell me how can i get rid of FFT >> incompatibility issue? > > I just told you, in my previous email. Yet I don't think it's a good > idea, > and the forces won't be zero anyway. Your lattice have lower > symmetry than > the FCC lattice (it is actually a supercell), hence you can only > have zero > forces up to the numerical accuracy. all in all, "zero" in the physical sciences means essentially two conceptually different things 1) a quantity is zero because a physical law forbids it to be different from zero. typically, "physical" laws of this kind are not quite "physical", but rather "geometrical" (i.e. dictated by symmetry, selection rules, and the like). 2) a quantity is _much smaller_ (to within numerical/experimental) accuracy than detectable by any physical means. Forces in an FCC lattice where cubic symmetry is explicitly imposed are zero because of 1. If cubic symmetry is not explicitly imposed, they may "vanish" (if ever) because of 2. Would you now please tell us for which purpose the forces that you calculate (and that do not vanish for reasons that you understand now) are too large? SB --- 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/20091010/2668f9e6/attachment-0001.htm From xinhuolin at gmail.com Sun Oct 11 01:41:10 2009 From: xinhuolin at gmail.com (Huolin Xin) Date: Sat, 10 Oct 2009 19:41:10 -0400 Subject: [Pw_forum] Electron Energy-Loss Near-Edge Fine Structure and Near Edge X-ray Absorption Fine Structure Message-ID: Hi all: Is anyone considering to write a module to calculate Electron Energy-Loss Near-Edge Fine Structure and NEXAFS (Near Edge X-ray Absorption Fine Structure) for Quantum Espresso? H Xin Cornell University From decboy9 at gmail.com Sun Oct 11 08:54:44 2009 From: decboy9 at gmail.com (dev sharma) Date: Sun, 11 Oct 2009 12:24:44 +0530 Subject: [Pw_forum] fhi2upf.x In-Reply-To: <40672.78.12.170.175.1255201444.squirrel@webmail.sissa.it> References: <40672.78.12.170.175.1255201444.squirrel@webmail.sissa.it> Message-ID: Dear sir , Thanks , I successfully made a UPF format, Please have a look on the wave functions i include. Is these correct ???? [dev at dev try1]$ /home/dev/espresso-4.0.2/upftools/fhi2upf.x 39-Y.LDA.fhi read_fhi: assuming abinit format Pseudopotential without NLCC successfully read Wavefunction # 1: label, occupancy > 5S. 2 Wavefunction # 2: label, occupancy > 5p, 0 Wavefunction # 3: label, occupancy > 4d, 1 Wavefunction # 4: label, occupancy > 4f, 0 Pseudopotential successfully converted Output PP file in UPF format : 39-Y.LDA.fhi.UPF Please have a look. Thanks Yours sincerely, Dev , University of Delhi, India On Sun, Oct 11, 2009 at 12:34 AM, Lorenzo Paulatto wrote: > > On Sat, October 10, 2009 20:58, dev sharma wrote: > > /home/dev/espresso-4.0.2/upftools/fhi2upf.x <39-Y.LDA.fhi> 39-Y.LDA.UPF > > Dear Dev, > fhi2upf does not work like that. The correct syntax is: > /whatever/fhi2upf.x 39-Y.LDA.fhi > which would create 39-Y.LDA.fhi.UPF > > regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091011/4f252367/attachment.htm From siyouber at yahoo.fr Sun Oct 11 14:09:14 2009 From: siyouber at yahoo.fr (Bertrand SITAMTZE) Date: Sun, 11 Oct 2009 12:09:14 +0000 (GMT) Subject: [Pw_forum] Band structure calculation Message-ID: <667610.66999.qm@web26501.mail.ukl.yahoo.com> Dear all, I calculated the bands structure of an ionic conductor with bands.x. But when I plot it with plotband.x, the Fermi level is exactly at the middle of the forbidden gap. I think as a conductor, the Fermi level should be in the conduction band. Has something gone wrong? Thanks for your help ****************************************** Bertrand SITAMTZE YOUMBI Laboratory of Material sciences Department of physics University of Yaound? I-Cameroon PO Box 812 Yaound?-Cameroon ******************************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091011/10622185/attachment.htm From shypirate at gmail.com Sun Oct 11 14:17:11 2009 From: shypirate at gmail.com (=?utf-8?B?c2h5cGlyYXRl?=) Date: Sun, 11 Oct 2009 20:17:11 +0800 Subject: [Pw_forum] =?utf-8?q?Band_structure_calculation?= References: <667610.66999.qm@web26501.mail.ukl.yahoo.com> Message-ID: <200910112017085934186@gmail.com> I think you should read the Fermi energy from output of pw.x 2009-10-11 shypirate ???? Bertrand SITAMTZE ????? 2009-10-11 20:09:32 ???? pw code ??? ??? [Pw_forum] Band structure calculation Dear all, I calculated the bands structure of an ionic conductor with bands.x. But when I plot it with plotband.x, the Fermi level is exactly at the middle of the forbidden gap. I think as a conductor, the Fermi level should be in the conduction band. Has something gone wrong? Thanks for your help ****************************************** Bertrand SITAMTZE YOUMBI Laboratory of Material sciences Department of physics University of Yaound? I-Cameroon PO Box 812 Yaound?-Cameroon ******************************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091011/e6b839b5/attachment.htm From degironc at sissa.it Sun Oct 11 14:31:02 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sun, 11 Oct 2009 14:31:02 +0200 Subject: [Pw_forum] Band structure calculation In-Reply-To: <667610.66999.qm@web26501.mail.ukl.yahoo.com> References: <667610.66999.qm@web26501.mail.ukl.yahoo.com> Message-ID: <4AD1D006.2090709@sissa.it> An ionic conductor is a material where the charge carriers are ions NOT electrons... Molten NaCl will be able to sustain an electrical current but it will still be an insulator from the electronic point of view.. stefano Bertrand SITAMTZE wrote: > Dear all, > I calculated the bands structure of an ionic conductor with bands.x. > But when I plot it with plotband.x, the Fermi level is exactly at the > middle of the forbidden gap. > I think as a conductor, the Fermi level should be in the conduction band. > Has something gone wrong? > > Thanks for your help > > > > ****************************************** > Bertrand SITAMTZE YOUMBI > Laboratory of Material sciences > Department of physics > University of Yaound? I-Cameroon > PO Box 812 Yaound?-Cameroon > ******************************************** > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From rezaebraahimi at yahoo.com Sun Oct 11 19:50:30 2009 From: rezaebraahimi at yahoo.com (Reza Ebraahimi) Date: Sun, 11 Oct 2009 10:50:30 -0700 (PDT) Subject: [Pw_forum] About the input file for matdyn.x Message-ID: <314781.21235.qm@web59401.mail.ac4.yahoo.com> Hi i have a question. imagine we have defined?a structure?by using?celldm(1)="a_0" and cell_parameters(1,1)="b",?so?the first bravais lattice vector's value?becomes "a_0*b". In the file matdyn.f90 is written "nq: q-points in 2pi/a units", is "a" in the mentioned description, the "a_0" or "a_0*b" ? Thank You Reza Ebraahimi Graduate Student- Tehran University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091011/4ec92fe0/attachment.htm From paulatto at sissa.it Sun Oct 11 19:57:21 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sun, 11 Oct 2009 19:57:21 +0200 (CEST) Subject: [Pw_forum] About the input file for matdyn.x In-Reply-To: <314781.21235.qm@web59401.mail.ac4.yahoo.com> References: <314781.21235.qm@web59401.mail.ac4.yahoo.com> Message-ID: <59744.78.12.170.175.1255283841.squirrel@webmail.sissa.it> On Sun, October 11, 2009 19:50, Reza Ebraahimi wrote: > a_0" or "a_0*b" ? Dear Reza, it is a_0, celldm(1) that is always used as internal unit of measure regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From shypirate at gmail.com Mon Oct 12 02:32:17 2009 From: shypirate at gmail.com (shypirate) Date: Mon, 12 Oct 2009 08:32:17 +0800 Subject: [Pw_forum] problem of scf calculation under electric field Message-ID: <200910120832144217120@gmail.com> Dear all, Sorry to bother you again. I want to calculate band structure under electric field. After set dipfield=.true. and other settings remain unchanged, self-consistent calculation has trouble to converge. In each self-consistent calculation step, there are several "c_bands: * eigenvalues not converged" warning messages. And, after several self-consistent steps, usually 3 steps, program stops with error message: "too many eigenvalues not converged". Could you give me some advice about this error? Where may be the error in my input file? number of k-points ? My input file for pw.x is like this: &control calculation = 'relax', nstep=200, tprnfor = .TRUE., tefield = .TRUE., dipfield = .TRUE., prefix = 'relax', pseudo_dir = '/home/ypwang/espresso/pseudo/', outdir = './out', / &system ibrav = 6, celldm(1) = 7.35, celldm(3) = 8, nat = 20, ntyp = 5, ecutwfc = 40.0, ecutrho = 480, occupations = 'smearing', degauss = 0.005, smearing = 'mv', edir =3, emaxpos = 0.9, eopreg = 0.1, eamp = 0.00 / &electrons diagonalization = 'cg' mixing_mode = 'local-TF' mixing_beta = 0.3 conv_thr = 1.0d-8 / &IONS bfgs_ndim = 7 / ATOMIC_SPECIES ... K_POINTS automatic 4 4 2 1 1 1 Thanks in advance! 2009-10-12 shypirate -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091012/262a83aa/attachment-0001.htm From decboy9 at gmail.com Mon Oct 12 08:04:11 2009 From: decboy9 at gmail.com (dev sharma) Date: Mon, 12 Oct 2009 11:34:11 +0530 Subject: [Pw_forum] k points in epsilon.x Message-ID: hi 2 all, I have one question. In epsilon.x, the k points is not automatic option. Sir, means how to give the K points option i.e. we have to give only irreducible k points or like for the calculation ='bands' , we have give the path (like G{0,0,0}-->X(0.5,0,0)). If we have to give the irreducible points , how to calculate those points ???? can i do this , that run the pw.x with a mesh like 4*4*4 grid and the number of k points in the scf output are the required k points like in a scf out number of k points= 8 cart. coord. in units 2pi/a_0 k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = 0.0312500 k( 2) = ( 0.0000000 0.0000000 0.2829528), wk = 0.0625000 k( 3) = ( 0.0000000 0.0000000 -0.5659056), wk = 0.0312500 k( 4) = ( 0.0000000 0.2500000 0.0000000), wk = 0.0625000 k( 5) = ( 0.0000000 0.2500000 0.2829528), wk = 0.0625000 k( 6) = ( 0.0000000 0.2500000 -0.5659056), wk = 0.0625000 k( 7) = ( 0.0000000 -0.5000000 0.0000000), wk = 0.0312500 k( 8) = ( 0.0000000 -0.5000000 0.2829528), wk = 0.0625000 , means i have 2 give these k points with their weight set to equal to each other. And in out of these i have to choose also because like 0 0 0.282 and 0 0 -0.282 will yield the same results ?? M i right ??? Please advice ???? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091012/249510e4/attachment.htm From sclauzer at sissa.it Mon Oct 12 09:10:27 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 12 Oct 2009 09:10:27 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: References: Message-ID: <4AD2D663.9000402@sissa.it> dev sharma wrote: > hi 2 all, > > I have one question. In epsilon.x, the k points is not automatic > option. Sir, means how to give the K points option i.e. we have to give > only irreducible k points or like for the calculation ='bands' , we have > give the path (like G{0,0,0}-->X(0.5,0,0)). If we have to give the > irreducible points , how to calculate those points ???? can i do this , > that run the pw.x with a mesh like 4*4*4 grid and the number of k > points in the scf output are the required k points Never used this utility, so maybe my advice is useless: are you sure that epsilon.x does need a k-point grid on input? For what purpose are you using the program? Do you know what it computes? If not, please look at the documentation in Doc/eps_max.tex. GS PS: Please always sign your mails with your name and affiliation. Thanks -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Mon Oct 12 09:20:18 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 12 Oct 2009 09:20:18 +0200 Subject: [Pw_forum] problem of scf calculation under electric field In-Reply-To: <200910120832144217120@gmail.com> References: <200910120832144217120@gmail.com> Message-ID: <4AD2D8B2.8060402@sissa.it> shypirate wrote: > Dear all, > Sorry to bother you again. > I want to calculate band structure under electric field. After set > dipfield=.true. and other settings remain unchanged, self-consistent > calculation has trouble to converge. In each self-consistent calculation > step, there are several "c_bands: * eigenvalues not converged" warning > messages. And, after several self-consistent steps, usually 3 steps, > program stops with error message: "too many eigenvalues not converged". > Could you give me some advice about this error? Where may be the error > in my input file? number of k-points ? My input file for pw.x is like this: Are you sure you set up correctly the sawtooth electric field potential? It must have the inversion region in the vacuum region. Since you don't report the atomic positions in your message, this makes it impossible to be checked by other, more experienced users (who's not me on this topic, sorry). Please always sign you mails with your full name and affiliation! Thanks! GS > > &control > calculation = 'relax', > nstep=200, > tprnfor = .TRUE., > tefield = .TRUE., > dipfield = .TRUE., > prefix = 'relax', > pseudo_dir = '/home/ypwang/espresso/pseudo/', > outdir = './out', > / > &system > ibrav = 6, celldm(1) = 7.35, celldm(3) = 8, nat = 20, ntyp = 5, > ecutwfc = 40.0, ecutrho = 480, > occupations = 'smearing', degauss = 0.005, smearing = 'mv', > edir =3, > emaxpos = 0.9, > eopreg = 0.1, > eamp = 0.00 > / > &electrons > diagonalization = 'cg' > mixing_mode = 'local-TF' > mixing_beta = 0.3 > conv_thr = 1.0d-8 > / > &IONS > bfgs_ndim = 7 > / > ATOMIC_SPECIES > ... > > K_POINTS automatic > 4 4 2 1 1 1 > > Thanks in advance! > > 2009-10-12 > ------------------------------------------------------------------------ > shypirate > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Mon Oct 12 09:37:30 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 12 Oct 2009 09:37:30 +0200 Subject: [Pw_forum] energy-degenerate states with different irrep labels In-Reply-To: References: Message-ID: <4AD2DCBA.6060004@sissa.it> Dear Silvia, Silvia Bakalova wrote: > Thank you for the reply, Gabriele. > Yes, this is @Gamma, the states are degenerate, but I wonder why they > have different irreducible representations (G_5+ and G_6+)? Because they actually belong to different representations of the double group... Andrea and me some time ago kind of understood why in presence of time-reversal there are couples of bands which can be matched, in the sense that they are degenerate not in the usual sense, but in the following. For each k, if there is an eigenvalue e_{k,v} belonging to a band v of, say, G_5 symmetry, there must be at -k an eigenvalue e_{-k,v'} with the same magnitude belonging to the matching band v' of, say, G_6 symmetry (and viceversa). We checked this for the double group C_{2v}, which was the one of our case study. Please have a look at the band structure here http://people.sissa.it/sclauzer/Data/COsu7PtFR_G3G4.pdf You probably will understand better what I was saying above. Since at Gamma k=-k, the bands must be degenerate there (in the usual sense). I never checked what happens with other double groups, but you may confirm (or not) that the situation is the same. The demonstration of such property of the bands structure in presence of time reversal can be found in this book (I don't remember the page, I don't have it at hand now): Bassani F.; Pastori Parravicini G. (1975). Electronic states and optical transitions in solids. Pergamon Press, Oxford. HTH GS > > Silvia Bakalova wrote: >>/ Hi, > > />/ > />/ I have one question: for spin-orbit calculations, the energy bands are > />/ labelled with double point group notation (D3d? in my case). > />/ > />/ Some of the energy-degenerate states have different irrep labels and I > > />/ wonder why? > />/ > / > At which k-point? I think that at Gamma they have to be degenerate if time reversal > symmetry holds (i.e. no magnetic field). > > > GS > >>/ e.g. the valence band top: > > />/ > />/ e( 45 - 46) = 7.09176 eV 2 --> G_5+ L_4+ > />/ > />/ e( 45 - 46) = 7.09176 eV 2 --> G_6+ L_5+ > />/ > />/ I would be grateful for your reply or some literature reference, as I am > > />/ not familiar with the group theory. > />/ > />/ Many thanks, > />/ > />/ Silvia -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From decboy9 at gmail.com Mon Oct 12 11:11:44 2009 From: decboy9 at gmail.com (dev sharma) Date: Mon, 12 Oct 2009 14:41:44 +0530 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: <4AD2D663.9000402@sissa.it> References: <4AD2D663.9000402@sissa.it> Message-ID: Sir, i want to calculate dielectric constant as a function of frequency. I have read the manual And it is written there that Epsilon.x doesn't support the reduction of the k-points grid into the unreducible Brillouin zone, so the previous PW runs must be performed with a uniform k-points grid and all k-points weights must be equal to each other, i.e. in the k-points card the k-points coordinates must be given manually in \emph{crystal} or \emph{alat} or \emph{bohr}, but not with the \emph{automatic} option. Thanks On Mon, Oct 12, 2009 at 12:40 PM, Gabriele Sclauzero wrote: > > dev sharma wrote: > > hi 2 all, > > > > I have one question. In epsilon.x, the k points is not automatic > > option. Sir, means how to give the K points option i.e. we have to give > > only irreducible k points or like for the calculation ='bands' , we have > > give the path (like G{0,0,0}-->X(0.5,0,0)). If we have to give the > > irreducible points , how to calculate those points ???? can i do this , > > that run the pw.x with a mesh like 4*4*4 grid and the number of k > > points in the scf output are the required k points > > Never used this utility, so maybe my advice is useless: are you sure that > epsilon.x does > need a k-point grid on input? For what purpose are you using the program? > Do you know what > it computes? If not, please look at the documentation in Doc/eps_max.tex. > > > GS > > > PS: Please always sign your mails with your name and affiliation. Thanks > > > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091012/6d8bd0d0/attachment.htm From marsamos at democritos.it Mon Oct 12 12:04:44 2009 From: marsamos at democritos.it (marsamos at democritos.it) Date: Mon, 12 Oct 2009 12:04:44 +0200 Subject: [Pw_forum] Relax calculation not converging In-Reply-To: References: Message-ID: <20091012120444.3da8i6kwtz40kkc0@mail.democritos.it> Dear Dimpy, the default value for scf steps is 100. The message means that the code does not achieve the electronic convergence after 100 scf steps. This usually means that your initial electronic configuration is too far from the scf one and the algorithm does not manage to converge in 100 step. Or, that there exist "two" solutions and the algorithm is oscillating. If it is just a question of been to far there are two possibilities 1) the algorithm need just more scf steps, so you just need to increase electron_maxstep (default=100) or to decrease mixing_beta (or both) (attention the decrease of mixing beta might produce that the algorithm need also more electron step). 2) you are too far and "two" solutions are accessible. This is for instance what happens in SiO2 for the charged -1 and -2 oxygen interstitial, or the Si vacancy. For treating such a problem you need to first do a pre-relaxation with some trick. For the insterstitial it is better to start from a neutral relaxation, for the Si vacancy one need to relax first with a substituted Al. Sometimes it can also bee usefull to rpe-relax with electronic temperature to facilitate convergence (ad a broadening). hope it can help you, bests Layla Quoting Dimpy Sharma : > > Hi Quantum espresso users, > > I have been trying to perform a relax calculation in my system with > 99 atoms (semiconductor system) and I am running my calculation in 48 > processors. The calculation has finished 3 scf steps and finished 100 > iterations, but after this it stops and it is showing me the > following message . > ' convergence NOT achieved after 100 iterations: stopping' > > I have restart my calculation, still I got the same error message > after 100 iteration steps. > > Can any body give me any suggestion? > > Thanks > > Dimpy > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From sclauzer at sissa.it Mon Oct 12 12:18:45 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 12 Oct 2009 12:18:45 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: References: <4AD2D663.9000402@sissa.it> Message-ID: <4AD30285.4080004@sissa.it> dev sharma wrote: > Sir, i want to calculate dielectric constant as a function of > frequency. I have read the manual And it is written there that > Epsilon.x doesn't support the reduction of the k-points grid into the > unreducible Brillouin zone, so the previous PW runs must be performed > with a uniform k-points grid and all k-points weights must be equal to > each other, i.e. in the k-points card the k-points coordinates must be > given manually in \emph{crystal} or \emph{alat} or \emph{bohr}, but not > with the \emph{automatic} option. In your first mail you were speaking about the automatic keyword as if it was an option for a k-point list to be supplied to epsilon.x. As you have now reported from the manual, epsilon.x does not need a k-point grid, since it uses the k-points from the underlying data produced by the previous pw.x run. For a cubic symmetry system, for instance, you can supply manually a N x N x N mesh of k-points (using crystal option might be the easiest solution, in my opinion), which must provide a good sampling the full BZ (it is not difficult to produce one, see the source code of pwtools/kpoints.f or PW/kpoint_grid.f90). Be careful and check the code does not reduce the mesh by symmetry (which is the default, good behaviour of the code, that in this specific case you want to avoid). Use the nosym option if needed (as written in the manual). You can also try to generate the grid with the automatic option of K_POINTS after setting to true nosym and noinv (see Doc/INPUT_PW.txt), but I'm not sure if this always works. Regards, GS > Thanks > > On Mon, Oct 12, 2009 at 12:40 PM, Gabriele Sclauzero > wrote: > > > dev sharma wrote: > > hi 2 all, > > > > I have one question. In epsilon.x, the k points is not automatic > > option. Sir, means how to give the K points option i.e. we have > to give > > only irreducible k points or like for the calculation ='bands' , > we have > > give the path (like G{0,0,0}-->X(0.5,0,0)). If we have to give the > > irreducible points , how to calculate those points ???? can i do > this , > > that run the pw.x with a mesh like 4*4*4 grid and the number of k > > points in the scf output are the required k points > > Never used this utility, so maybe my advice is useless: are you sure > that epsilon.x does > need a k-point grid on input? For what purpose are you using the > program? Do you know what > it computes? If not, please look at the documentation in > Doc/eps_max.tex. > > > GS > > > PS: Please always sign your mails with your name and affiliation. Thanks > > > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it > | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From marsamos at democritos.it Mon Oct 12 12:29:44 2009 From: marsamos at democritos.it (marsamos at democritos.it) Date: Mon, 12 Oct 2009 12:29:44 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: <4AD30285.4080004@sissa.it> References: <4AD2D663.9000402@sissa.it> <4AD30285.4080004@sissa.it> Message-ID: <20091012122944.xd46mkwpw04w0kks@mail.democritos.it> Dear User, for generating a kpoint grid supported by epsilon you can use the following small program (for pwscf input you have to specify nosym=.true., and KPOINTS crystal) program kpoints_gen integer :: ikz, iky, ikx, nk real :: weight real :: shiftx, shifty, shiftz read(*,*) nkx, nky, nkz, shiftx, shifty, shiftz nk=nkx*nky*nkz weight=1.0/real(nk) write(*,*) nk do ikz=1,nkz do iky=1,nky do ikx=1,nkx write(*,"(4f15.9)") (real(ikx)-real(1.0)+shiftx)/real(nkx), (real(iky)-real(1.0)+shifty)/real(nky), & (real(ikz)-real(1.0)+shiftz)/real(nkz), weight enddo enddo enddo end program kpoints_gen BUT I strongly suggest to you to use or the utilities inside SaX (see www.sax-project.org) or yambo (www.yambo-code.org) for calculating the dielectric constant. epsilon.x is not complete (it does not include the nonlocal part of the pseudo potential and is not able to include in the calculation the non-local field effects and excitonic effects). If YOU WORK WITH SEMICONDUCTORS OR INSULATORS YOU WILL NOT BE ABLE TO COMPARE WITH EXPERIMENTS. You should now which kind of RESPONSE you want to reproduce. bests layla Quoting Gabriele Sclauzero : > > > dev sharma wrote: >> Sir, i want to calculate dielectric constant as a function of >> frequency. I have read the manual And it is written there that >> Epsilon.x doesn't support the reduction of the k-points grid into the >> unreducible Brillouin zone, so the previous PW runs must be performed >> with a uniform k-points grid and all k-points weights must be equal to >> each other, i.e. in the k-points card the k-points coordinates must be >> given manually in \emph{crystal} or \emph{alat} or \emph{bohr}, but not >> with the \emph{automatic} option. > > In your first mail you were speaking about the automatic keyword as > if it was an option > for a k-point list to be supplied to epsilon.x. As you have now > reported from the manual, > epsilon.x does not need a k-point grid, since it uses the k-points > from the underlying > data produced by the previous pw.x run. > For a cubic symmetry system, for instance, you can supply manually a > N x N x N mesh of > k-points (using crystal option might be the easiest solution, in my > opinion), which must > provide a good sampling the full BZ (it is not difficult to produce > one, see the source > code of pwtools/kpoints.f or PW/kpoint_grid.f90). > Be careful and check the code does not reduce the mesh by symmetry > (which is the default, > good behaviour of the code, that in this specific case you want to > avoid). Use the nosym > option if needed (as written in the manual). > You can also try to generate the grid with the automatic option of > K_POINTS after setting > to true nosym and noinv (see Doc/INPUT_PW.txt), but I'm not sure if > this always works. > > > Regards, > > GS > > > >> Thanks >> >> On Mon, Oct 12, 2009 at 12:40 PM, Gabriele Sclauzero > > wrote: >> >> >> dev sharma wrote: >> > hi 2 all, >> > >> > I have one question. In epsilon.x, the k points is not automatic >> > option. Sir, means how to give the K points option i.e. we have >> to give >> > only irreducible k points or like for the calculation ='bands' , >> we have >> > give the path (like G{0,0,0}-->X(0.5,0,0)). If we have to give the >> > irreducible points , how to calculate those points ???? can i do >> this , >> > that run the pw.x with a mesh like 4*4*4 grid and the number of k >> > points in the scf output are the required k points >> >> Never used this utility, so maybe my advice is useless: are you sure >> that epsilon.x does >> need a k-point grid on input? For what purpose are you using the >> program? Do you know what >> it computes? If not, please look at the documentation in >> Doc/eps_max.tex. >> >> >> GS >> >> >> PS: Please always sign your mails with your name and affiliation. Thanks >> >> >> >> -- >> >> >> o ------------------------------------------------ o >> | Gabriele Sclauzero, PhD Student | >> | c/o: SISSA & CNR-INFM Democritos, | >> | via Beirut 2-4, 34014 Trieste (Italy) | >> | email: sclauzer at sissa.it >> | >> | phone: +39 040 3787 511 | >> | skype: gurlonotturno | >> o ------------------------------------------------ o >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From decboy9 at gmail.com Mon Oct 12 13:59:50 2009 From: decboy9 at gmail.com (dev sharma) Date: Mon, 12 Oct 2009 17:29:50 +0530 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: <20091012122944.xd46mkwpw04w0kks@mail.democritos.it> References: <4AD2D663.9000402@sissa.it> <4AD30285.4080004@sissa.it> <20091012122944.xd46mkwpw04w0kks@mail.democritos.it> Message-ID: Thanks SIr, i will study Yambo or Sax. Thanks, Yours sincerely, Dev , University of Delhi, On Mon, Oct 12, 2009 at 3:59 PM, wrote: > Dear User, for generating a kpoint grid supported by epsilon you can > use the following small program (for pwscf input you have to specify > nosym=.true., and KPOINTS crystal) > > program kpoints_gen > > integer :: ikz, iky, ikx, nk > real :: weight > real :: shiftx, shifty, shiftz > > read(*,*) nkx, nky, nkz, shiftx, shifty, shiftz > > nk=nkx*nky*nkz > weight=1.0/real(nk) > write(*,*) nk > do ikz=1,nkz > do iky=1,nky > do ikx=1,nkx > > write(*,"(4f15.9)") (real(ikx)-real(1.0)+shiftx)/real(nkx), > (real(iky)-real(1.0)+shifty)/real(nky), & > (real(ikz)-real(1.0)+shiftz)/real(nkz), weight > > enddo > enddo > enddo > > > end program kpoints_gen > > > BUT I strongly suggest to you to use or the utilities inside SaX (see > www.sax-project.org) or yambo (www.yambo-code.org) for calculating the > dielectric constant. epsilon.x is not complete (it does not include the > nonlocal part of the pseudo potential and is not able to include in the > calculation the non-local field effects and excitonic effects). If YOU > WORK WITH SEMICONDUCTORS OR INSULATORS YOU WILL NOT BE ABLE TO COMPARE > WITH EXPERIMENTS. You should now which kind of RESPONSE you want to > reproduce. > > bests > > layla > > > > Quoting Gabriele Sclauzero : > > > > > > > dev sharma wrote: > >> Sir, i want to calculate dielectric constant as a function of > >> frequency. I have read the manual And it is written there that > >> Epsilon.x doesn't support the reduction of the k-points grid into the > >> unreducible Brillouin zone, so the previous PW runs must be performed > >> with a uniform k-points grid and all k-points weights must be equal to > >> each other, i.e. in the k-points card the k-points coordinates must be > >> given manually in \emph{crystal} or \emph{alat} or \emph{bohr}, but not > >> with the \emph{automatic} option. > > > > In your first mail you were speaking about the automatic keyword as > > if it was an option > > for a k-point list to be supplied to epsilon.x. As you have now > > reported from the manual, > > epsilon.x does not need a k-point grid, since it uses the k-points > > from the underlying > > data produced by the previous pw.x run. > > For a cubic symmetry system, for instance, you can supply manually a > > N x N x N mesh of > > k-points (using crystal option might be the easiest solution, in my > > opinion), which must > > provide a good sampling the full BZ (it is not difficult to produce > > one, see the source > > code of pwtools/kpoints.f or PW/kpoint_grid.f90). > > Be careful and check the code does not reduce the mesh by symmetry > > (which is the default, > > good behaviour of the code, that in this specific case you want to > > avoid). Use the nosym > > option if needed (as written in the manual). > > You can also try to generate the grid with the automatic option of > > K_POINTS after setting > > to true nosym and noinv (see Doc/INPUT_PW.txt), but I'm not sure if > > this always works. > > > > > > Regards, > > > > GS > > > > > > > >> Thanks > >> > >> On Mon, Oct 12, 2009 at 12:40 PM, Gabriele Sclauzero >> > wrote: > >> > >> > >> dev sharma wrote: > >> > hi 2 all, > >> > > >> > I have one question. In epsilon.x, the k points is not automatic > >> > option. Sir, means how to give the K points option i.e. we have > >> to give > >> > only irreducible k points or like for the calculation ='bands' , > >> we have > >> > give the path (like G{0,0,0}-->X(0.5,0,0)). If we have to give > the > >> > irreducible points , how to calculate those points ???? can i do > >> this , > >> > that run the pw.x with a mesh like 4*4*4 grid and the number of > k > >> > points in the scf output are the required k points > >> > >> Never used this utility, so maybe my advice is useless: are you sure > >> that epsilon.x does > >> need a k-point grid on input? For what purpose are you using the > >> program? Do you know what > >> it computes? If not, please look at the documentation in > >> Doc/eps_max.tex. > >> > >> > >> GS > >> > >> > >> PS: Please always sign your mails with your name and affiliation. > Thanks > >> > >> > >> > >> -- > >> > >> > >> o ------------------------------------------------ o > >> | Gabriele Sclauzero, PhD Student | > >> | c/o: SISSA & CNR-INFM Democritos, | > >> | via Beirut 2-4, 34014 Trieste (Italy) | > >> | email: sclauzer at sissa.it > >> | > >> | phone: +39 040 3787 511 | > >> | skype: gurlonotturno | > >> o ------------------------------------------------ o > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> http://www.democritos.it/mailman/listinfo/pw_forum > >> > >> > >> > >> ------------------------------------------------------------------------ > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > > > > o ------------------------------------------------ o > > | Gabriele Sclauzero, PhD Student | > > | c/o: SISSA & CNR-INFM Democritos, | > > | via Beirut 2-4, 34014 Trieste (Italy) | > > | email: sclauzer at sissa.it | > > | phone: +39 040 3787 511 | > > | skype: gurlonotturno | > > o ------------------------------------------------ o > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > ---------------------------------------------------------------- > This message was sent using IMP, the Internet Messaging Program. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091012/0f824c6e/attachment.htm From silveto at gmail.com Mon Oct 12 16:05:29 2009 From: silveto at gmail.com (Silvia Bakalova) Date: Mon, 12 Oct 2009 15:05:29 +0100 Subject: [Pw_forum] energy-degenerate states with different irrep labels In-Reply-To: References: Message-ID: Dear Gabriele, Thank you for your help and really useful reply. I think that you are right and this is the case also for D_3d (-3m) double group in my QE calculations, as most of the states remain doubly degenerate, but with the same energy. What I found is that this type of degeneracy is Kramers degeneracy and there is a proof of the at least double degeneracy of the ground state when the spin is included, based on the Kramers degeneracy theorem: http://arxiv.org/abs/0809.4471v1 Thanks again. I just wanted to ask, because everywhere they prove this degeneracy for odd number of electrons, and I have even number of electrons in my case, is there any difference and could really double point group be used to describe the electronic states? Cheers, Silvia Silvia Bakalova wrote: >* Thank you for the reply, Gabriele. *>* Yes, this is @Gamma, the states are degenerate, but I wonder why they *>* have different irreducible representations (G_5+ and G_6+)? * Because they actually belong to different representations of the double group... Andrea and me some time ago kind of understood why in presence of time-reversal there are couples of bands which can be matched, in the sense that they are degenerate not in the usual sense, but in the following. For each k, if there is an eigenvalue e_{k,v} belonging to a band v of, say, G_5 symmetry, there must be at -k an eigenvalue e_{-k,v'} with the same magnitude belonging to the matching band v' of, say, G_6 symmetry (and viceversa). We checked this for the double group C_{2v}, which was the one of our case study. Please have a look at the band structure here http://people.sissa.it/sclauzer/Data/COsu7PtFR_G3G4.pdf You probably will understand better what I was saying above. Since at Gamma k=-k, the bands must be degenerate there (in the usual sense). I never checked what happens with other double groups, but you may confirm (or not) that the situation is the same. The demonstration of such property of the bands structure in presence of time reversal can be found in this book (I don't remember the page, I don't have it at hand now): Bassani F.; Pastori Parravicini G. (1975). Electronic states and optical transitions in solids. Pergamon Press, Oxford. HTH GS On Fri, Oct 9, 2009 at 3:53 PM, Silvia Bakalova wrote: > Hi, > > I have one question: for spin-orbit calculations, the energy bands are > labelled with double point group notation (D3d? in my case). > > Some of the energy-degenerate states have different irrep labels and I > wonder why? > > e.g. the valence band top: > > e( 45 - 46) = 7.09176 eV 2 --> G_5+ L_4+ > > e( 45 - 46) = 7.09176 eV 2 --> G_6+ L_5+ > > I would be grateful for your reply or some literature reference, as I am > not familiar with the group theory. > > Many thanks, > > Silvia > > > ---------------------- > Dr. Silvia Bakalova, > Post Doctoral Researcher, > HH Wills Physics Laboratory, > Bristol, BS8 1TL, UK > http://spectra.phy.bris.ac.uk/ > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091012/5766310d/attachment.htm From sclauzer at sissa.it Mon Oct 12 16:28:49 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 12 Oct 2009 16:28:49 +0200 Subject: [Pw_forum] energy-degenerate states with different irrep labels In-Reply-To: References: Message-ID: <4AD33D21.6030607@sissa.it> Dear Silvia, Silvia Bakalova wrote: > Dear Gabriele, > > Thank you for your help and really useful reply. I think that you are > right and this is the case also for D_3d (-3m) double group in my QE > calculations, as most of the states remain doubly degenerate, but with > the same energy. What I found is that this type of degeneracy is Kramers > degeneracy and there is a proof of the at least double degeneracy of the > ground state when the spin is included, based on the Kramers degeneracy > theorem: > > http://arxiv.org/abs/0809.4471v1 I think you are right on this point > > Thanks again. I just wanted to ask, because everywhere they prove this > degeneracy for odd number of electrons, and I have even number of > electrons in my case, is there any difference and could really double > point group be used to describe the electronic states? I'm not sure of this, but since KS equations are single particle equations, then you should treat this case as a single electron one (your KS eigenstates are one electron wavefunctions, not many-body wavefunctions). Cheers Gabriele > > Cheers, > > Silvia > > > Silvia Bakalova wrote: >>/ Thank you for the reply, Gabriele. > />/ Yes, this is @Gamma, the states are degenerate, but I wonder why they > />/ have different irreducible representations (G_5+ and G_6+)? > > / > Because they actually belong to different representations of the double group... > > Andrea and me some time ago kind of understood why in presence of time-reversal there are > couples of bands which can be matched, in the sense that they are degenerate not in the > > usual sense, but in the following. > For each k, if there is an eigenvalue e_{k,v} belonging to a band v of, say, G_5 symmetry, > there must be at -k an eigenvalue e_{-k,v'} with the same magnitude belonging to the > > matching band v' of, say, G_6 symmetry (and viceversa). > > We checked this for the double group C_{2v}, which was the one of our case study. Please > have a look at the band structure here > http://people.sissa.it/sclauzer/Data/COsu7PtFR_G3G4.pdf > > > You probably will understand better what I was saying above. > Since at Gamma k=-k, the bands must be degenerate there (in the usual sense). I never > checked what happens with other double groups, but you may confirm (or not) that the > > situation is the same. The demonstration of such property of the bands structure in > presence of time reversal can be found in this book (I don't remember the page, I don't > have it at hand now): > > Bassani F.; Pastori Parravicini G. (1975). Electronic states and optical transitions > > in solids. Pergamon Press, Oxford. > > HTH > > GS > > > On Fri, Oct 9, 2009 at 3:53 PM, Silvia Bakalova > wrote: > > Hi, > > I have one question: for spin-orbit calculations, the energy bands > are labelled with double point group notation (D3d? in my case). > > Some of the energy-degenerate states have different irrep labels and > I wonder why? > > e.g. the valence band top: > > e( 45 - 46) = 7.09176 eV 2 --> G_5+ L_4+ > > e( 45 - 46) = 7.09176 eV 2 --> G_6+ L_5+ > > I would be grateful for your reply or some literature reference, as > I am not familiar with the group theory. > > Many thanks, > > Silvia > > > ---------------------- > Dr. Silvia Bakalova, > Post Doctoral Researcher, > HH Wills Physics Laboratory, > Bristol, BS8 1TL, UK > http://spectra.phy.bris.ac.uk/ > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From ferretti at MIT.EDU Mon Oct 12 16:51:29 2009 From: ferretti at MIT.EDU (Andrea Ferretti) Date: Mon, 12 Oct 2009 10:51:29 -0400 (EDT) Subject: [Pw_forum] k points in epsilon.x In-Reply-To: Message-ID: Hi, > Sir, i want to calculate dielectric constant as a function of frequency. I > have read the manual And it is written there that > Epsilon.x doesn't support the reduction of the k-points grid into the > unreducible Brillouin zone, so the previous PW runs must be performed with a > uniform k-points grid and all k-points weights must be equal to each other, > i.e. in the k-points card the k-points coordinates must be given manually in > \emph{crystal} or \emph{alat} or \emph{bohr}, but not with the > \emph{automatic} option. > Thanks > I think this is correct.. a good procedure could be first to run a scf calculation using whatever symmetrized kpt-mesh. Then you can run a nscf calculation with a suitable number of empty bands (according to your needs) and a regular mesh of kpts over the whole BZ. andrea DMSE, MIT Massachusetts From ZChen at cau.edu Mon Oct 12 17:32:48 2009 From: ZChen at cau.edu (Chen, Zhifan) Date: Mon, 12 Oct 2009 11:32:48 -0400 Subject: [Pw_forum] use cvs In-Reply-To: References: Message-ID: Thanks Paora. I also compiled espresso-4.0 in parallel using mpif90. I can calculate do the scf and band structure calculation in example01. However, in analyze symmetry it stops with error message: R & G space division Proc/pool=4 P0_17713: p4 error: interrupt SIGSEGV:11 By the way is there any place I can find the meaning of the error message such as SIGSEGV:11. Zhifan Chen Dept. of Physics Clark Atlanta Univ. Atlanta GA 30314 USA -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Paolo Giannozzi Sent: Saturday, October 10, 2009 3:30 AM To: PWSCF Forum Subject: Re: [Pw_forum] use cvs On Oct 9, 2009, at 21:19 , Chen, Zhifan wrote: > #Error in iotk_scan (iotk_scan.spp:717) > you have a buggy compiler > May be I can solve the problem by login CVS > unlikely > cvs login > > I press Enter. I did not login but back to my own director. > that's the correct behavior. Please read the developers guide: there is a section on how to use cvs P. --- 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 mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From wparker at mps.ohio-state.edu Mon Oct 12 17:34:11 2009 From: wparker at mps.ohio-state.edu (William Parker) Date: Mon, 12 Oct 2009 11:34:11 -0400 (EDT) Subject: [Pw_forum] Reference for Fock Exchange Minimization Scheme? Message-ID: Hello everyone, Is there a publication that contains a more detailed description of the Fock exchange minimization scheme in QE than the reference paper appendix or the README in examples/EXX_example? --William ====================================================================== William Parker Tel : +1 (614) 292-2887 Graduate Research Associate Fax : +1 (614) 292-7557 Ohio State University Department of Physics 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker U.S.A. Office: 2025 Physics Research Building ====================================================================== From giannozz at democritos.it Mon Oct 12 17:50:48 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 12 Oct 2009 17:50:48 +0200 Subject: [Pw_forum] use cvs In-Reply-To: References: Message-ID: <4AD35058.9000505@democritos.it> Chen, Zhifan wrote: > By the way is there any place I can find the meaning of the error > message such as SIGSEGV:11. SIGSEGV 11 is a (in)famous error message meaning no more than "something very wrong happened". Again: with very high probability, this is a compiler or mpi library problem. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From baroni at sissa.it Tue Oct 13 08:08:30 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 13 Oct 2009 08:08:30 +0200 Subject: [Pw_forum] Reference for Fock Exchange Minimization Scheme? In-Reply-To: References: Message-ID: <053EE3E5-F11C-4787-A540-10C260077FCE@sissa.it> On Oct 12, 2009, at 5:34 PM, William Parker wrote: > Hello everyone, > > Is there a publication that contains a more detailed description of > the > Fock exchange minimization scheme in QE than the reference paper > appendix > or the README in examples/EXX_example? something (not much) is here: http://stacks.iop.org/0953-8984/21/395502 have you tried it out? SB --- 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/20091013/0733b9e3/attachment.htm From decboy9 at gmail.com Tue Oct 13 08:57:27 2009 From: decboy9 at gmail.com (dev sharma) Date: Tue, 13 Oct 2009 12:27:27 +0530 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: References: Message-ID: Dear Madam, May be my question was not clear to all. originally my question was the same . That as u suggested do a scf and then a nscf with the nbnd and regular mesh of k points. What i want to know , that in nscf calculation i have to give kpoints {automatic} 4 4 4 or like Kpoints 72 0 0 0 0 0.1 0 / / 0.5 0 0 and i asked that if i have to give k points in second option, How to give that ???? can i give these 72 k points generated by the scf output ???? Thnaks Dev Sharma, Unoveristy of Delhi On Mon, Oct 12, 2009 at 8:21 PM, Andrea Ferretti wrote: > > > Hi, > > > Sir, i want to calculate dielectric constant as a function of frequency. > I > > have read the manual And it is written there that > > Epsilon.x doesn't support the reduction of the k-points grid into the > > unreducible Brillouin zone, so the previous PW runs must be performed > with a > > uniform k-points grid and all k-points weights must be equal to each > other, > > i.e. in the k-points card the k-points coordinates must be given manually > in > > \emph{crystal} or \emph{alat} or \emph{bohr}, but not with the > > \emph{automatic} option. > > Thanks > > > > I think this is correct.. > a good procedure could be first to run a scf calculation using whatever > symmetrized kpt-mesh. Then you can run a nscf calculation with a suitable > number of empty bands (according to your needs) and a regular mesh of kpts > over the whole BZ. > > andrea > > DMSE, MIT > Massachusetts > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091013/453bb0e6/attachment.htm From paulatto at sissa.it Tue Oct 13 09:30:07 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 13 Oct 2009 09:30:07 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: References: Message-ID: In data 13 ottobre 2009 alle ore 08:57:27, dev sharma ha scritto: > and i asked that if i have to give k points in second option, How to give > that ???? can i give these 72 k points generated by the scf output ???? Dear Dev, It is just the same! You can use automatic, but you *must* specify nosym=.true. noinv=.true. in the system namelist for the nscf calculation. If you *really* want to specify the point by hand you can generate a list with program kpoints.x that you can find in pwtools/ directory. best regards P.S. > Dear Madam, "Andrea" is a masculine name in Italian (it is feminine in English, German, Spanish and others), on the other hand "Layla" is feminine in every language I'm aware of. You can avoid this kind of mistakes by calling people directly with their names, instead of using Sir/Madam. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From sclauzer at sissa.it Tue Oct 13 09:30:39 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 13 Oct 2009 09:30:39 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: References: Message-ID: <4AD42C9F.2010705@sissa.it> First: Andrea Ferretti is Sir, Layla Martin Samos is Madam, not the opposite... dev sharma wrote: > Dear Madam, > > May be my question was not clear to all. originally my question was the > same . That as u suggested do a scf and then a nscf with the nbnd and > regular mesh of k points. What i want to know , that in nscf calculation > i have to give > kpoints {automatic} > 4 4 4 > or like > Kpoints > 72 > 0 0 0 > 0 0.1 0 > / > / > 0.5 0 0 > > and i asked that if i have to give k points in second option, How to > give that ???? can i give these 72 k points generated by the scf > output ???? Second: Many people has already replied you, giving different possible solutions to your problem. I'm quite astonished that after a dozen of mails we have not been able to close this thread. Maybe there is some problem in communication/understanding which I cannot capture. I summarize what are the possible solutions if you still want to use epsilon.x and don't want to try Yambo or SaX and co. In the nscf run, either use nosym=.TRUE. noinv=.TRUE. K_POINTS AUTOMATIC 4 4 4 0 0 0 Such that the code will NOT use symmetry to reduce the 4x4x4 k-point mesh. Or use: nosym=.TRUE. K_POINTS CRYSTAL and then append a list of k-points corresponding to a 4x4x4 regular grid of key points that you can easily generate either using the piece of code given by Layla, or adapting the code in pwtools/kpoints.f or PW/kpoint_grid.f90. Obviously you must test for convergence with respect to the BZ sampling (i.e. number of k-points) HTH GS > Thnaks > Dev Sharma, > Unoveristy of Delhi > > On Mon, Oct 12, 2009 at 8:21 PM, Andrea Ferretti > wrote: > > > > Hi, > > > Sir, i want to calculate dielectric constant as a function of > frequency. I > > have read the manual And it is written there that > > Epsilon.x doesn't support the reduction of the k-points grid into the > > unreducible Brillouin zone, so the previous PW runs must be > performed with a > > uniform k-points grid and all k-points weights must be equal to > each other, > > i.e. in the k-points card the k-points coordinates must be given > manually in > > \emph{crystal} or \emph{alat} or \emph{bohr}, but not with the > > \emph{automatic} option. > > Thanks > > > > I think this is correct.. > a good procedure could be first to run a scf calculation using whatever > symmetrized kpt-mesh. Then you can run a nscf calculation with a > suitable > number of empty bands (according to your needs) and a regular mesh > of kpts > over the whole BZ. > > andrea > > DMSE, MIT > Massachusetts > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From Peter.Pallister at nrc-cnrc.gc.ca Tue Oct 13 15:04:00 2009 From: Peter.Pallister at nrc-cnrc.gc.ca (Pallister, Peter) Date: Tue, 13 Oct 2009 09:04:00 -0400 Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: <48251.78.12.158.197.1255076403.squirrel@webmail.sissa.it> Message-ID: Hi, Actually, one of the tutorials on the QE site has a step-by-step pseudopotential generation tutorial showing how to generate the AE calculation, examine the logarithmic derivatives, etc. This tutorial suggests looking at the file (well, in my case) 'Mg-oct09ps.wfc' to look at the pseudo-generated orbitals. Is this incorrect? What do the individual file/s mean in the "file_wfcncgen=" input? Thanks! Peter Pallister -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Lorenzo Paulatto Sent: October 9, 2009 4:20 AM To: PWSCF Forum Subject: Re: [Pw_forum] GIPAW pseudopotentials On Wed, October 7, 2009 17:10, Pallister, Peter wrote: > Hi all, > > I am attempting to generate a gipaw pseudopotentials, in this case for > Mg, but I am having tremendous difficulty. The wavefunctions produced > with the input below (or minute variations thereof) continuously show > discontinuities, specifically at a radius of 15bohrs. Dear Peter, the wavefunctions are generated on a logarithmic radial grid, around 15 bohr the points are very coarse, btw how are you examining the wavfunctions? (there is not file_wfcncgen='..' in your input). best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From wparker at mps.ohio-state.edu Tue Oct 13 16:03:43 2009 From: wparker at mps.ohio-state.edu (William Parker) Date: Tue, 13 Oct 2009 10:03:43 -0400 (EDT) Subject: [Pw_forum] Reference for Fock Exchange Minimization Scheme? In-Reply-To: <053EE3E5-F11C-4787-A540-10C260077FCE@sissa.it> References: <053EE3E5-F11C-4787-A540-10C260077FCE@sissa.it> Message-ID: That is the "reference paper appendix" I was referring to, but it's worth pointing out that a little bit is there. Anywhere else? --William ====================================================================== William Parker Tel : +1 (614) 292-2887 Graduate Research Associate Fax : +1 (614) 292-7557 Ohio State University Department of Physics 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker U.S.A. Office: 2025 Physics Research Building ====================================================================== On Tue, 13 Oct 2009, Stefano Baroni wrote: > > On Oct 12, 2009, at 5:34 PM, William Parker wrote: > > > Hello everyone, > > > > Is there a publication that contains a more detailed description of > > the > > Fock exchange minimization scheme in QE than the reference paper > > appendix > > or the README in examples/EXX_example? > > something (not much) is here: http://stacks.iop.org/0953-8984/21/395502 > > have you tried it out? > > SB > > --- > 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 > > > > > > > > From Ari.P.Seitsonen at iki.fi Tue Oct 13 17:30:37 2009 From: Ari.P.Seitsonen at iki.fi (Ari P Seitsonen) Date: Tue, 13 Oct 2009 17:30:37 +0200 (CEST) Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: References: Message-ID: Dear Peter et al, Sorry for the long time it took me to answer!! :( The *.wfc files are generated by default, as far as I know. There is a special issue concerning the wave functions for the GIPAW reconstruction: They are _not_ generated in the normal procedure how the ld1.x/atomic code works, but for GIPAW they are calculated _after_ the *.wfc files have been written: the standard part of the code calculates them once, with some value (possibly zero) as the reference energy; for GIPAW one then recalculates them, but with the reference energies taken from the '&test' section. This is why the wave functions for the GIPAW reconstruction can be different from the ones in the *.wfc files. And this is also why one _has to_ define a reference energy different from zero, at least for unbound states (ie. whose eigenvalue would be positive) in the section '&test'. So, in practise it doesn't matter if the wave functions in the *.wfc corresponding to empty orbitals do not make sense, but you can check that they look ok in the GIPAW section at the end of the UPF file. I attach an unofficial script for writing out the GIPAW data in the UPF file (it is a very small script file). You can probably modify the output to show the properties you want to have. If you still have problems with the wave functions coming out from this script please check that you use a non-zero reference energy for unbound states, otherwise please contact me again. Greetings from Zurich, apsi -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Physikalisch-Chemisches Institut der Universit?t Z?rich Tel: +41 44 63 55 44 97 / Mobile: +41 79 71 90 935 On Tue, 13 Oct 2009, Pallister, Peter wrote: > Hi, > Actually, one of the tutorials on the QE site has a step-by-step > pseudopotential generation tutorial showing how to generate the AE > calculation, examine the logarithmic derivatives, etc. This tutorial > suggests looking at the file (well, in my case) 'Mg-oct09ps.wfc' to look > at the pseudo-generated orbitals. Is this incorrect? What do the > individual file/s mean in the "file_wfcncgen=" input? Thanks! > > Peter Pallister > > -----Original Message----- > From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On > Behalf Of Lorenzo Paulatto > Sent: October 9, 2009 4:20 AM > To: PWSCF Forum > Subject: Re: [Pw_forum] GIPAW pseudopotentials > > > On Wed, October 7, 2009 17:10, Pallister, Peter wrote: >> Hi all, >> >> I am attempting to generate a gipaw pseudopotentials, in this case > for >> Mg, but I am having tremendous difficulty. The wavefunctions produced >> with the input below (or minute variations thereof) continuously show >> discontinuities, specifically at a radius of 15bohrs. > > Dear Peter, > the wavefunctions are generated on a logarithmic radial grid, around 15 > bohr the points are very coarse, btw how are you examining the > wavfunctions? (there is not file_wfcncgen='..' in your input). > > best regards > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- A non-text attachment was scrubbed... Name: upf2plot.sh Type: application/x-shellscript Size: 3500 bytes Desc: Url : http://www.democritos.it/pipermail/pw_forum/attachments/20091013/0db4f5b5/attachment-0001.bin From lex at phys.ufl.edu Tue Oct 13 18:52:00 2009 From: lex at phys.ufl.edu (Lex Kemper) Date: Tue, 13 Oct 2009 12:52:00 -0400 Subject: [Pw_forum] Wavefunction projections Message-ID: <4AD4B030.9070206@phys.ufl.edu> Hello PWSCF community, I have a brief question. projwfc.x projects the charge density onto atomic orbitals, which are Ylms time a radial coordinate. The brief question is what direction the Ylms are oriented with respect to. Suppose I have a crystal structure described by: CELL_PARAMETERS (alat) 5.614610000 -5.57421000 0.000000000 5.614610000 5.574210000 0.000000000 0.000000000 0.000000000 12.94533 then if I look at the d orbital, and the projection onto d_{xz}, is that along a1 (5.61,-5.57,0) or some other direction? Thanks, Lex Kemper Department of Physics University of Florida From degironc at sissa.it Tue Oct 13 20:16:48 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 13 Oct 2009 20:16:48 +0200 Subject: [Pw_forum] Wavefunction projections In-Reply-To: <4AD4B030.9070206@phys.ufl.edu> References: <4AD4B030.9070206@phys.ufl.edu> Message-ID: <4AD4C410.1010402@sissa.it> In your example a1 = 5.614610000 x -5.57421000 y + 0.000000000 z a2 = 5.614610000 x +5.574210000 y + 0.000000000 z a3 = 0.000000000 x + 0.000000000 y + 12.94533 z ans d_xy is x*y ... that is it has large positive lobes for x=y (ROUGHLY in the a2 direction) and large negative lobes for x=-y (ROUGHLY in the a1 direction). stefano Lex Kemper wrote: > Hello PWSCF community, > > I have a brief question. projwfc.x projects the charge density onto > atomic orbitals, which are Ylms time a radial coordinate. The brief > question is what direction the Ylms are oriented with respect to. > Suppose I have a crystal structure described by: > > CELL_PARAMETERS (alat) > 5.614610000 -5.57421000 0.000000000 > 5.614610000 5.574210000 0.000000000 > 0.000000000 0.000000000 12.94533 > > > then if I look at the d orbital, and the projection onto d_{xz}, is that > along a1 (5.61,-5.57,0) or some other direction? > > Thanks, > > Lex Kemper > Department of Physics > University of Florida > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From baroni at sissa.it Tue Oct 13 21:12:34 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 13 Oct 2009 21:12:34 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: References: Message-ID: <2D5BCDEF-D387-4A23-A31E-D0E9661985E5@sissa.it> On Oct 13, 2009, at 8:57 AM, dev sharma wrote: > Dear Madam, Just as a note of lore, Andrea Ferretti (as well as Nicola Marzari, for that matter) is Italian. In Italian "Andrea" (as well as "Nicola") are masculine names (the feminine counterparts being "Andreina" and "Nicoletta"). This is being said, I myselfe I am totally unable to guess the gender of many (virtually, all, but a few) foreign languages. Stefano B ... > > On Mon, Oct 12, 2009 at 8:21 PM, Andrea Ferretti > wrote: ... > > > andrea > > DMSE, MIT > Massachusetts > > _______________________________________________ > Pw_forum mailing l --- 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/20091013/7347e94d/attachment.htm From baroni at sissa.it Tue Oct 13 21:14:07 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 13 Oct 2009 21:14:07 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: <4AD42C9F.2010705@sissa.it> References: <4AD42C9F.2010705@sissa.it> Message-ID: <4664FA4B-1F30-4955-BD0C-C8E792A023CD@sissa.it> oops ... many other picky Italian speaking members of this list ... S. On Oct 13, 2009, at 9:30 AM, Gabriele Sclauzero wrote: > First: > > Andrea Ferretti is Sir, Layla Martin Samos is Madam, not the > opposite... > > dev sharma wrote: >> Dear Madam, >> >> May be my question was not clear to all. originally my question was >> the >> same . That as u suggested do a scf and then a nscf with the nbnd >> and >> regular mesh of k points. What i want to know , that in nscf >> calculation >> i have to give >> kpoints {automatic} >> 4 4 4 >> or like >> Kpoints >> 72 >> 0 0 0 >> 0 0.1 0 >> / >> / >> 0.5 0 0 >> >> and i asked that if i have to give k points in second option, How to >> give that ???? can i give these 72 k points generated by the scf >> output ???? > > Second: > > Many people has already replied you, giving different possible > solutions to your problem. > I'm quite astonished that after a dozen of mails we have not been > able to close this > thread. Maybe there is some problem in communication/understanding > which I cannot capture. > I summarize what are the possible solutions if you still want to use > epsilon.x and don't > want to try Yambo or SaX and co. > In the nscf run, either use > > nosym=.TRUE. > noinv=.TRUE. > > K_POINTS AUTOMATIC > 4 4 4 0 0 0 > > Such that the code will NOT use symmetry to reduce the 4x4x4 k-point > mesh. > > Or use: > > nosym=.TRUE. > > K_POINTS CRYSTAL > > > and then append a list of k-points corresponding to a 4x4x4 regular > grid of key points > that you can easily generate either using the piece of code given by > Layla, or adapting > the code in pwtools/kpoints.f or PW/kpoint_grid.f90. > > Obviously you must test for convergence with respect to the BZ > sampling (i.e. number of > k-points) > > HTH > > > GS > > > > > > > >> Thnaks >> Dev Sharma, >> Unoveristy of Delhi >> >> On Mon, Oct 12, 2009 at 8:21 PM, Andrea Ferretti > > wrote: >> >> >> >> Hi, >> >>> Sir, i want to calculate dielectric constant as a function of >> frequency. I >>> have read the manual And it is written there that >>> Epsilon.x doesn't support the reduction of the k-points grid into >>> the >>> unreducible Brillouin zone, so the previous PW runs must be >> performed with a >>> uniform k-points grid and all k-points weights must be equal to >> each other, >>> i.e. in the k-points card the k-points coordinates must be given >> manually in >>> \emph{crystal} or \emph{alat} or \emph{bohr}, but not with the >>> \emph{automatic} option. >>> Thanks >>> >> >> I think this is correct.. >> a good procedure could be first to run a scf calculation using >> whatever >> symmetrized kpt-mesh. Then you can run a nscf calculation with a >> suitable >> number of empty bands (according to your needs) and a regular mesh >> of kpts >> over the whole BZ. >> >> andrea >> >> DMSE, MIT >> Massachusetts >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > 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/20091013/8b744cd5/attachment-0001.htm From noreply at netlogmail.com Wed Oct 14 02:18:49 2009 From: noreply at netlogmail.com (=?UTF-8?B?6buE5a6d6Zm1?=) Date: Wed, 14 Oct 2009 02:18:49 +0200 (CEST) Subject: [Pw_forum] =?utf-8?b?6K6/6Zeu5oiR55qETmV0bG9n5Liq5Lq65Li76aG1?= Message-ID: <20091014001857.07ED215610F@democritos.sissa.it> ?, ?????????????????????????????????????????????????????????????????????????????????????????????????? ???: http://zh.netlog.com/go/mailurl/type=invite_1&mailid=69201817&id=1&url=-L2dvL3JlZ2lzdGVyL2lkPTg3NjU0ODc5OSZpPXQ5MQ__ ??, ??? ---------------------------------------------------------------- ?????????????????????? http://zh.netlog.com/go/mailurl/type=invite_1&mailid=69201817&id=2&url=-L2dvL25vbWFpbHMvaW52aXRlL2VtYWlsPS1jSGRmWm05eWRXMUFjSGR6WTJZdWIzSm4mY29kZT0wMDA4NjUxMCZpZD04NzY1NDg3OTkmaT10OTI_ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/e5a1ab98/attachment.htm From mtt_turion at hotmail.com Wed Oct 14 05:23:22 2009 From: mtt_turion at hotmail.com (mtt physics) Date: Wed, 14 Oct 2009 10:23:22 +0700 Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 32 In-Reply-To: References: Message-ID: Dear Prof. Stefano Baroni My name is Abdulmutta I study in Department of Physics Facalty of science Prince of Songkla University Thailand I used Quantum-ESPRESSO to compute Mateials science is PZT(PbZrxTi1-x)O3 Lead zerconate titanate I have a problem to find Optimized Structure of tetragonal phase (c/a), rhombohedral phase in metal oxide alloy of PZT Can you help me please > From: pw_forum-request at pwscf.org > Subject: Pw_forum Digest, Vol 28, Issue 32 > To: pw_forum at pwscf.org > Date: Tue, 13 Oct 2009 21:14:18 +0200 > > Send Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-owner at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. Wavefunction projections (Lex Kemper) > 2. Re: Wavefunction projections (Stefano de Gironcoli) > 3. Re: k points in epsilon.x (Stefano Baroni) > 4. Re: k points in epsilon.x (Stefano Baroni) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 13 Oct 2009 12:52:00 -0400 > From: Lex Kemper > Subject: [Pw_forum] Wavefunction projections > To: pw_forum at pwscf.org > Message-ID: <4AD4B030.9070206 at phys.ufl.edu> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hello PWSCF community, > > I have a brief question. projwfc.x projects the charge density onto > atomic orbitals, which are Ylms time a radial coordinate. The brief > question is what direction the Ylms are oriented with respect to. > Suppose I have a crystal structure described by: > > CELL_PARAMETERS (alat) > 5.614610000 -5.57421000 0.000000000 > 5.614610000 5.574210000 0.000000000 > 0.000000000 0.000000000 12.94533 > > > then if I look at the d orbital, and the projection onto d_{xz}, is that > along a1 (5.61,-5.57,0) or some other direction? > > Thanks, > > Lex Kemper > Department of Physics > University of Florida > > > ------------------------------ > > Message: 2 > Date: Tue, 13 Oct 2009 20:16:48 +0200 > From: Stefano de Gironcoli > Subject: Re: [Pw_forum] Wavefunction projections > To: PWSCF Forum > Message-ID: <4AD4C410.1010402 at sissa.it> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > In your example > a1 = 5.614610000 x -5.57421000 y + 0.000000000 z > a2 = 5.614610000 x +5.574210000 y + 0.000000000 z > a3 = 0.000000000 x + 0.000000000 y + 12.94533 z > ans d_xy is x*y ... > that is it has large positive lobes for x=y (ROUGHLY in the a2 > direction) and large negative lobes for x=-y (ROUGHLY in the a1 direction). > > stefano > > > Lex Kemper wrote: > > Hello PWSCF community, > > > > I have a brief question. projwfc.x projects the charge density onto > > atomic orbitals, which are Ylms time a radial coordinate. The brief > > question is what direction the Ylms are oriented with respect to. > > Suppose I have a crystal structure described by: > > > > CELL_PARAMETERS (alat) > > 5.614610000 -5.57421000 0.000000000 > > 5.614610000 5.574210000 0.000000000 > > 0.000000000 0.000000000 12.94533 > > > > > > then if I look at the d orbital, and the projection onto d_{xz}, is that > > along a1 (5.61,-5.57,0) or some other direction? > > > > Thanks, > > > > Lex Kemper > > Department of Physics > > University of Florida > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > ------------------------------ > > Message: 3 > Date: Tue, 13 Oct 2009 21:12:34 +0200 > From: Stefano Baroni > Subject: Re: [Pw_forum] k points in epsilon.x > To: PWSCF Forum > Message-ID: <2D5BCDEF-D387-4A23-A31E-D0E9661985E5 at sissa.it> > Content-Type: text/plain; charset="iso-8859-1" > > > On Oct 13, 2009, at 8:57 AM, dev sharma wrote: > > > Dear Madam, > > Just as a note of lore, Andrea Ferretti (as well as Nicola Marzari, > for that matter) is Italian. In Italian "Andrea" (as well as "Nicola") > are masculine names (the feminine counterparts being "Andreina" and > "Nicoletta"). > > This is being said, I myselfe I am totally unable to guess the gender > of many (virtually, all, but a few) foreign languages. > > Stefano B > > ... > > > > > On Mon, Oct 12, 2009 at 8:21 PM, Andrea Ferretti > > wrote: > > ... > > > > > > > andrea > > > > DMSE, MIT > > Massachusetts > > > > _______________________________________________ > > Pw_forum mailing l > > --- > 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/20091013/7347e94d/attachment-0001.htm > > ------------------------------ > > Message: 4 > Date: Tue, 13 Oct 2009 21:14:07 +0200 > From: Stefano Baroni > Subject: Re: [Pw_forum] k points in epsilon.x > To: PWSCF Forum > Message-ID: <4664FA4B-1F30-4955-BD0C-C8E792A023CD at sissa.it> > Content-Type: text/plain; charset="iso-8859-1" > > oops ... > many other picky Italian speaking members of this list ... > > S. > > On Oct 13, 2009, at 9:30 AM, Gabriele Sclauzero wrote: > > > First: > > > > Andrea Ferretti is Sir, Layla Martin Samos is Madam, not the > > opposite... > > > > dev sharma wrote: > >> Dear Madam, > >> > >> May be my question was not clear to all. originally my question was > >> the > >> same . That as u suggested do a scf and then a nscf with the nbnd > >> and > >> regular mesh of k points. What i want to know , that in nscf > >> calculation > >> i have to give > >> kpoints {automatic} > >> 4 4 4 > >> or like > >> Kpoints > >> 72 > >> 0 0 0 > >> 0 0.1 0 > >> / > >> / > >> 0.5 0 0 > >> > >> and i asked that if i have to give k points in second option, How to > >> give that ???? can i give these 72 k points generated by the scf > >> output ???? > > > > Second: > > > > Many people has already replied you, giving different possible > > solutions to your problem. > > I'm quite astonished that after a dozen of mails we have not been > > able to close this > > thread. Maybe there is some problem in communication/understanding > > which I cannot capture. > > I summarize what are the possible solutions if you still want to use > > epsilon.x and don't > > want to try Yambo or SaX and co. > > In the nscf run, either use > > > > nosym=.TRUE. > > noinv=.TRUE. > > > > K_POINTS AUTOMATIC > > 4 4 4 0 0 0 > > > > Such that the code will NOT use symmetry to reduce the 4x4x4 k-point > > mesh. > > > > Or use: > > > > nosym=.TRUE. > > > > K_POINTS CRYSTAL > > > > > > and then append a list of k-points corresponding to a 4x4x4 regular > > grid of key points > > that you can easily generate either using the piece of code given by > > Layla, or adapting > > the code in pwtools/kpoints.f or PW/kpoint_grid.f90. > > > > Obviously you must test for convergence with respect to the BZ > > sampling (i.e. number of > > k-points) > > > > HTH > > > > > > GS > > > > > > > > > > > > > > > >> Thnaks > >> Dev Sharma, > >> Unoveristy of Delhi > >> > >> On Mon, Oct 12, 2009 at 8:21 PM, Andrea Ferretti >> > wrote: > >> > >> > >> > >> Hi, > >> > >>> Sir, i want to calculate dielectric constant as a function of > >> frequency. I > >>> have read the manual And it is written there that > >>> Epsilon.x doesn't support the reduction of the k-points grid into > >>> the > >>> unreducible Brillouin zone, so the previous PW runs must be > >> performed with a > >>> uniform k-points grid and all k-points weights must be equal to > >> each other, > >>> i.e. in the k-points card the k-points coordinates must be given > >> manually in > >>> \emph{crystal} or \emph{alat} or \emph{bohr}, but not with the > >>> \emph{automatic} option. > >>> Thanks > >>> > >> > >> I think this is correct.. > >> a good procedure could be first to run a scf calculation using > >> whatever > >> symmetrized kpt-mesh. Then you can run a nscf calculation with a > >> suitable > >> number of empty bands (according to your needs) and a regular mesh > >> of kpts > >> over the whole BZ. > >> > >> andrea > >> > >> DMSE, MIT > >> Massachusetts > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> http://www.democritos.it/mailman/listinfo/pw_forum > >> > >> > >> > >> ------------------------------------------------------------------------ > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > > > > o ------------------------------------------------ o > > | Gabriele Sclauzero, PhD Student | > > | c/o: SISSA & CNR-INFM Democritos, | > > | via Beirut 2-4, 34014 Trieste (Italy) | > > | email: sclauzer at sissa.it | > > | phone: +39 040 3787 511 | > > | skype: gurlonotturno | > > o ------------------------------------------------ o > > _______________________________________________ > > 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/20091013/8b744cd5/attachment.htm > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 28, Issue 32 > **************************************** _________________________________________________________________ Windows Live: ????????????????????????????????????????????????????? Facebook http://www.microsoft.com/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:th-th:SI_SB_2:092009 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/86a1965c/attachment-0001.htm From wangqj1 at 126.com Wed Oct 14 05:04:33 2009 From: wangqj1 at 126.com (Q.J.Wang) Date: Wed, 14 Oct 2009 11:04:33 +0800 (CST) Subject: [Pw_forum] Lowdin charge Message-ID: <5453217.792101255489473630.JavaMail.coremail@bj126app26.126.com> Dear PWSCF users I want to do Lowdin population analysis .But After I get the pdos,the outfile of pdos only gave the total charge of s ,p ,d orbital respectively . e.g s=2.8945, p=6.8221,d= 5.6231,but the s orbital include 3s and 4s in the USPP.So how can I get the charge of 3s ,4s respectively ? -- Best regards Q.J.Wang XiangTan University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/2dc7192a/attachment.htm From sclauzer at sissa.it Wed Oct 14 08:44:21 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Wed, 14 Oct 2009 08:44:21 +0200 Subject: [Pw_forum] k points in epsilon.x In-Reply-To: <4664FA4B-1F30-4955-BD0C-C8E792A023CD@sissa.it> References: <4AD42C9F.2010705@sissa.it> <4664FA4B-1F30-4955-BD0C-C8E792A023CD@sissa.it> Message-ID: <4AD57345.4060808@sissa.it> I apologize if my reply looked too much aggressive... anyway I perfectly agree with what Lorenzo said: ... > every language I'm aware of. You can avoid this kind of mistakes by > calling people directly with their names, instead of using Sir/Madam. This looks to me more appropriate for the informal and friendly style of this forum, but I understand that perhaps in other countries there may be different habits/mores of addressing people. Cheers Gabriele Stefano Baroni wrote: > oops ... > many other picky Italian speaking members of this list ... > > S. > > On Oct 13, 2009, at 9:30 AM, Gabriele Sclauzero wrote: > >> First: >> >> Andrea Ferretti is Sir, Layla Martin Samos is Madam, not the >> opposite... >> -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From giannozz at democritos.it Wed Oct 14 16:34:39 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 14 Oct 2009 16:34:39 +0200 Subject: [Pw_forum] Lowdin charge In-Reply-To: <5453217.792101255489473630.JavaMail.coremail@bj126app26.126.com> References: <5453217.792101255489473630.JavaMail.coremail@bj126app26.126.com> Message-ID: <2F8AC1D6-04E4-4DF2-98C5-ED794C8B11EA@democritos.it> On Oct 14, 2009, at 5:04 , Q.J.Wang wrote: > I want to do Lowdin population analysis .But After I get the > pdos,the outfile of pdos only gave the total charge of s ,p ,d > orbital respectively . e.g s=2.8945, p=6.8221,d= 5.6231,but the s > orbital include 3s and 4s in the USPP.So how can I get the charge > of 3s ,4s respectively ? the code writes a bunch of files in addition to the output file. I think that all data is there. In any case, 3s should be close to a full occupancy (2). I guess you also have 3p, also close to full occupancy (6) P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From ferretti at MIT.EDU Wed Oct 14 17:28:54 2009 From: ferretti at MIT.EDU (Andrea Ferretti) Date: Wed, 14 Oct 2009 11:28:54 -0400 (EDT) Subject: [Pw_forum] Lowdin charge In-Reply-To: <2F8AC1D6-04E4-4DF2-98C5-ED794C8B11EA@democritos.it> Message-ID: Dear Q.J.Wang, I don't know if Paolo refers to that, but the projections of atomic (orthogonalized) orbitals and eigenfunctions needed to build up the pDOS's are written together with the eigenvalues (plus other small data and dimensions) in the file $prefix.save/atomic_proj.xml Andrea > > the code writes a bunch of files in addition to the output file. I > think that all data > is there. In any case, 3s should be close to a full occupancy (2). I > guess you > also have 3p, also close to full occupancy (6) > > P. > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Andrea Ferretti MIT, Dept Material Science & Engineering bldg 13-4078, 77, Massachusetts Ave, Cambridge, MA Tel: +1 617-452-2455; Skype: andrea_ferretti URL: http://quasiamore.mit.edu Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From sh.shapt at gmail.com Wed Oct 14 18:36:49 2009 From: sh.shapt at gmail.com (Shaptrishi Sharma) Date: Wed, 14 Oct 2009 17:36:49 +0100 Subject: [Pw_forum] Calculating projected density of states Message-ID: Hi QE Users, I want to calculate the projected density of states for my system in quantum espresso. I have already calculate band structure . However honestly speaking I ahve not understood the concept of projectd density of states. Can anyone please give me some links and ideas about PDOS. Also how to calculate PDOS from band structure. I tried by running projwfc.x still I did not get my result. Thanks SS -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/4f93b199/attachment.htm From baroni at sissa.it Wed Oct 14 18:46:59 2009 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 14 Oct 2009 18:46:59 +0200 Subject: [Pw_forum] Calculating projected density of states In-Reply-To: References: Message-ID: <914FF9B3-B9C7-44D0-B206-2501705F38CF@sissa.it> The PDOS is the imaginary part of the diagonal matrix element of the one-electron Green function, calculated with respect to a given (usually, but not necessarily, localized) orbital. In a formula: \begin{eqnarray*} g[\phi](\epsilon) &=& \frac{1}{2\pi} \mathrm{Im} \langle \phi | (\epsilon-H)^{-1}|\phi \rangle \\ &=& \sum_n|\langle \phi|\psi_n \rangle |^2 \delta(\epsilon-E_n) \end{eqnarray*} (see attachement) SB On Oct 14, 2009, at 6:36 PM, Shaptrishi Sharma wrote: > Hi QE Users, > > I want to calculate the projected density of states for my system in > quantum espresso. I have already calculate band structure . However > honestly speaking I ahve not understood the concept of projectd > density of states. Can anyone please give me some links and ideas > about PDOS. Also how to calculate PDOS from band structure. I tried > by running projwfc.x still I did not get my result. > > Thanks > > SS > _______________________________________________ > 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/20091014/577e06e6/attachment-0002.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: latex-image-1.pdf Type: application/pdf Size: 63381 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20091014/577e06e6/attachment-0001.pdf -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/577e06e6/attachment-0003.htm From decboy9 at gmail.com Wed Oct 14 19:02:33 2009 From: decboy9 at gmail.com (dev sharma) Date: Wed, 14 Oct 2009 22:32:33 +0530 Subject: [Pw_forum] Calculating projected density of states In-Reply-To: References: Message-ID: Dear Saptrishi, Have u done nscf ?? for PDOS you have to do the three things first do scf second do nscf (with nbnd option ) third run the projwfc.x and for the plottings of files like *pdos_atm#1(Zn)_wfc#2(s)* u can use gnuplot etc. for the analysis u can read the forum and i am pasting one of the explanation for you On Gio, Settembre 4, 2008 05:31, wangqj1 wrote: >* *>* Dear all *>* After I used projwfc.x ,I got the pdos of every atom ,but I don't *>* know the signification of the following characters of *>* pdos_atm#1(Zn)_wfc#1(d) and pdos_atm#1(Zn)_wfc#2(s). * >* pdos_atm#1(Zn)_wfc#1(d) *projected dos of atom number 1 (Zinc) wavefunction number 1 (d-orbital) >* pdos_atm#1(Zn)_wfc#2(s) *projected dos of atom number 1 (Zinc) wavefunction number 2 (s-orbital) Each file contains 3 + (2l+1) columns, where l is the angular momentum of its specific wavefunction. The first column is the energy. The second and third are the total spin up and spin down dos; where "total" means its summed over all values of m; than you find the projected dos for each value of m, up and down, one at a time. e.g. for a p orbital you have p_x, p_y and p_z orbitals: # E(eV) ldosup(E) ldosdw(E) pdosup(E) pdosdw(E) pdosup(E)... energy [sum_xyz p up] [sum_xyz p down] [p_z up] [p_z down] [p_x up]... I hope I was clear, regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ =================================== for more help. please provide your input file. In that case , people may be able to help you more take care, Dev, Univeristy of Delhi . On Wed, Oct 14, 2009 at 10:06 PM, Shaptrishi Sharma wrote: > Hi QE Users, > > I want to calculate the projected density of states for my system in > quantum espresso. I have already calculate band structure . However honestly > speaking I ahve not understood the concept of projectd density of states. > Can anyone please give me some links and ideas about PDOS. Also how to > calculate PDOS from band structure. I tried by running projwfc.x still I > did not get my result. > > Thanks > > SS > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/d0ee6f25/attachment.htm From paulatto at sissa.it Wed Oct 14 19:05:00 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 14 Oct 2009 19:05:00 +0200 Subject: [Pw_forum] Calculating projected density of states In-Reply-To: References: Message-ID: In data 14 ottobre 2009 alle ore 19:02:33, dev sharma ha scritto: > second do nscf (with nbnd option ) This step is not necessary, unless you want to include more bands in the projection. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From degironc at sissa.it Wed Oct 14 21:14:38 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Wed, 14 Oct 2009 21:14:38 +0200 Subject: [Pw_forum] Reference for Fock Exchange Minimization Scheme? In-Reply-To: References: <053EE3E5-F11C-4787-A540-10C260077FCE@sissa.it> Message-ID: <4AD6231E.5020109@sissa.it> No, there isn't much more. There is a paper in J. of Comp. Chem. 29, 2098 (2008) by Gillan, Alfe`, de Gironcoli and Manby and one on JACS 129, 10402 (200& by Wang, deGirocnoli, Hush, and Reimers but they hardly discuss the implementation.... which is however "standard", except for the detail discussaed in the QE paper. see for instance Chawla and Voth, JCP 108, 4697 (1998); Sorouri, Foulkes and Hine JCP 124, 064105 (2006), Spencer and Alavi, PRB77, 193110 (2008); and many others that before and after claim to have solved the problem and basically all do the same: build auxiliary densities rho_{-q}=phi^{*}_{k+q}*psi_k in real space, FFT to reciprocal space, solve Poisson equation, back FFT to real space, multiply by phi_{k+q} and accumulate the result... the only tricky point is what you do for q->0 and there Gygi-Baldereschi is still the reference point. stefano stefano William Parker wrote: > That is the "reference paper appendix" I was referring to, but it's worth > pointing out that a little bit is there. Anywhere else? > > --William > > ====================================================================== > William Parker Tel : +1 (614) 292-2887 > Graduate Research Associate Fax : +1 (614) 292-7557 > Ohio State University > Department of Physics > 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu > Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker > U.S.A. > > Office: 2025 Physics Research Building > ====================================================================== > > On Tue, 13 Oct 2009, Stefano Baroni wrote: > > >> On Oct 12, 2009, at 5:34 PM, William Parker wrote: >> >> >>> Hello everyone, >>> >>> Is there a publication that contains a more detailed description of >>> the >>> Fock exchange minimization scheme in QE than the reference paper >>> appendix >>> or the README in examples/EXX_example? >>> >> something (not much) is here: http://stacks.iop.org/0953-8984/21/395502 >> >> have you tried it out? >> >> SB >> >> --- >> 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 >> From anvari_meh at physics.iust.ac.ir Wed Oct 14 21:16:25 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Wed, 14 Oct 2009 22:46:25 +0330 Subject: [Pw_forum] forum searching Message-ID: Hello I have a question about pwscf.org site. It seems that it has a problem so how could I search previous questions in pwscf forum? Unfortunately I couldn't find any way in quantum espresso site . Best Mehrnaz Anvari Iran University of Science & Technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/ca6b8fb0/attachment.htm From jjr19 at zips.uakron.edu Wed Oct 14 21:37:04 2009 From: jjr19 at zips.uakron.edu (James J Ramsey) Date: Wed, 14 Oct 2009 15:37:04 -0400 Subject: [Pw_forum] forum searching In-Reply-To: References: Message-ID: <46e014700910141237s1935a774uaf9cb18b778920e2@mail.gmail.com> 2009/10/14 Mehrnaz Anvari : > Hello > I have a question about pwscf.org site.?It seems that it has?a problem > so?how could?I search previous questions in pwscf forum? Unfortunately I > couldn't find any way in quantum espresso site?. See this page: http://www.quantum-espresso.org/tools.php From luotengf at msu.edu Wed Oct 14 21:39:40 2009 From: luotengf at msu.edu (Tengfei Luo) Date: Wed, 14 Oct 2009 15:39:40 -0400 Subject: [Pw_forum] forum searching In-Reply-To: References: Message-ID: <6B09F4C6A56F41CCAD37294C2F6E38EE@mit.edu> You may want to bookmark this page: http://www.democritos.it/cgi-bin/htsearch?words=pwscf Tengfei ----- Original Message ----- From: Mehrnaz Anvari To: pw_forum at pwscf.org Sent: Wednesday, October 14, 2009 3:16 PM Subject: [Pw_forum] forum searching Hello I have a question about pwscf.org site. It seems that it has a problem so how could I search previous questions in pwscf forum? Unfortunately I couldn't find any way in quantum espresso site . Best Mehrnaz Anvari Iran University of Science & Technology ------------------------------------------------------------------------------ _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/bb805cbe/attachment.htm From wparker at mps.ohio-state.edu Wed Oct 14 23:17:27 2009 From: wparker at mps.ohio-state.edu (William Parker) Date: Wed, 14 Oct 2009 17:17:27 -0400 (EDT) Subject: [Pw_forum] Reference for Fock Exchange Minimization Scheme? In-Reply-To: <4AD6231E.5020109@sissa.it> References: <053EE3E5-F11C-4787-A540-10C260077FCE@sissa.it> <4AD6231E.5020109@sissa.it> Message-ID: Thank you for the references, Stefano. It is always a little frustrating trying to understand a method from a smattering of oblique references and reading the source code as one's primary explanation. --William On Wed, 14 Oct 2009, Stefano de Gironcoli wrote: > No, there isn't much more. > There is a paper in J. of Comp. Chem. 29, 2098 (2008) by Gillan, > Alfe`, de Gironcoli and Manby and one on JACS 129, 10402 (200& by Wang, > deGirocnoli, Hush, and Reimers > but they hardly discuss the implementation.... which is however > "standard", except for the detail discussaed in the QE paper. > see for instance Chawla and Voth, JCP 108, 4697 (1998); Sorouri, Foulkes > and Hine JCP 124, 064105 (2006), Spencer and Alavi, PRB77, 193110 > (2008); and many others that before and after claim to have solved the > problem and basically all do the same: build auxiliary densities > rho_{-q}=phi^{*}_{k+q}*psi_k in real space, FFT to reciprocal space, > solve Poisson equation, back FFT to real space, multiply by phi_{k+q} > and accumulate the result... the only tricky point is what you do for > q->0 and there Gygi-Baldereschi is still the reference point. > > stefano > > > > stefano > > William Parker wrote: > > That is the "reference paper appendix" I was referring to, but it's worth > > pointing out that a little bit is there. Anywhere else? > > > > --William > > > > ====================================================================== > > William Parker Tel : +1 (614) 292-2887 > > Graduate Research Associate Fax : +1 (614) 292-7557 > > Ohio State University > > Department of Physics > > 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu > > Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker > > U.S.A. > > > > Office: 2025 Physics Research Building > > ====================================================================== > > > > On Tue, 13 Oct 2009, Stefano Baroni wrote: > > > > > >> On Oct 12, 2009, at 5:34 PM, William Parker wrote: > >> > >> > >>> Hello everyone, > >>> > >>> Is there a publication that contains a more detailed description of > >>> the > >>> Fock exchange minimization scheme in QE than the reference paper > >>> appendix > >>> or the README in examples/EXX_example? > >>> > >> something (not much) is here: http://stacks.iop.org/0953-8984/21/395502 > >> > >> have you tried it out? > >> > >> SB > >> > >> --- > >> 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From vtmtrinh at caltech.edu Thu Oct 15 02:16:44 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Wed, 14 Oct 2009 17:16:44 -0700 (PDT) Subject: [Pw_forum] Questions on phonon calculation in Example02 and 06 Message-ID: <4237.137.78.73.28.1255565804.squirrel@webmail.caltech.edu> Dear PWSCF Users, Since I would like to compare the phonon calculations by using 2 different ways as explained in Example02 and Example06, I rerun the whole calculation in Example06. Everything was fine. I then rerun the calculation of one specific q-point (-0.25, 0.25, -0.25) from Example06, but using the way shown in Example02, ie. performing scf, calculation at Gamma, nscf at q, and phonon at q. I compared the 2 results from the 2 methods, I saw that both ways gave the same phonon frequencies, but the displacement vectors are totally are different. I thought both should have yielded the same results. I apprciate if you could you help me to explain this. Below are the results: Following Example06: q = ( -0.250000000 0.250000000 -0.250000000 ) ************************************************************************** omega( 1) = 1.765425 [THz] = 58.888627 [cm-1] ( -0.141032 0.460061 0.040238 0.471692 0.181270 0.011631 ) ( -0.113839 0.478895 0.071975 0.478895 0.185814 0.000000 ) omega( 2) = 1.765425 [THz] = 58.888627 [cm-1] ( 0.253069 0.173719 -0.286906 0.139072 -0.539975 -0.034647 ) ( 0.269727 0.160766 -0.283781 0.160766 -0.553509 0.000000 ) omega( 3) = 4.536195 [THz] = 151.312190 [cm-1] ( -0.204646 -0.290110 0.204646 0.290110 -0.204646 -0.290110 ) ( -0.455290 0.000000 0.455290 0.000000 -0.455290 0.000000 ) omega( 4) = 11.004563 [THz] = 367.075159 [cm-1] ( -0.081755 -0.466090 0.446212 -0.432214 0.527968 0.033877 ) ( 0.039228 0.161901 -0.147017 0.161901 -0.186245 0.000000 ) omega( 5) = 11.004563 [THz] = 367.075159 [cm-1] ( -0.451526 0.406642 0.107022 0.442481 0.558547 0.035839 ) ( 0.149460 -0.153037 -0.047572 -0.153037 -0.197033 0.000000 ) omega( 6) = 12.136012 [THz] = 404.816481 [cm-1] ( -0.320406 -0.454213 0.320406 0.454213 -0.320406 -0.454213 ) ( 0.156090 0.000000 -0.156090 0.000000 0.156090 0.000000 ) ************************************************************************** Following Example02 q = ( -0.250000000 0.250000000 -0.250000000 ) ************************************************************************** omega( 1) = 1.765442 [THz] = 58.889209 [cm-1] ( 0.318551 0.007983 -0.249525 -0.028467 -0.568077 -0.036450 ) ( 0.325719 -0.012716 -0.256595 -0.012716 -0.582315 0.000000 ) omega( 2) = 1.765442 [THz] = 58.889209 [cm-1] ( -0.451869 0.141577 -0.493353 0.138915 -0.041485 -0.002662 ) ( -0.452021 0.174129 -0.494545 0.174129 -0.042524 0.000000 ) omega( 3) = 4.536219 [THz] = 151.312986 [cm-1] ( -0.204633 -0.290122 0.204633 0.290122 -0.204633 -0.290122 ) ( -0.455289 0.000000 0.455289 0.000000 -0.455289 0.000000 ) omega( 4) = 11.004565 [THz] = 367.075220 [cm-1] ( -0.184054 0.641087 -0.184054 0.641087 0.000000 0.000000 ) ( 0.050209 -0.229371 0.050209 -0.229371 0.000000 0.000000 ) omega( 5) = 11.004565 [THz] = 367.075220 [cm-1] ( 0.384293 0.024658 -0.384293 -0.024658 -0.768587 -0.049316 ) ( -0.135563 0.000000 0.135563 0.000000 0.271126 0.000000 ) omega( 6) = 12.136049 [THz] = 404.817695 [cm-1] ( -0.320383 -0.454229 0.320383 0.454229 -0.320383 -0.454229 ) ( 0.156091 0.000000 -0.156091 0.000000 0.156091 0.000000 ) ************************************************************************** Best, MyTrinh Vo From giannozz at democritos.it Thu Oct 15 08:24:41 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 15 Oct 2009 08:24:41 +0200 Subject: [Pw_forum] Questions on phonon calculation in Example02 and 06 In-Reply-To: <4237.137.78.73.28.1255565804.squirrel@webmail.caltech.edu> References: <4237.137.78.73.28.1255565804.squirrel@webmail.caltech.edu> Message-ID: On Oct 15, 2009, at 2:16 , vtmtrinh at caltech.edu wrote: > I compared the 2 results from the 2 methods, I saw that both ways gave > the same phonon frequencies, but the displacement vectors are totally > different. no they aren't. Displacement vectors for non-degenerate modes (number 3 and 6) are the same (within numerical accuracy). For degenerate modes, you get different linear combinations (because of the way matrix diagonalization works), but the subspace spanned by the two sets of degenerate eigenvectors is the same (I haven't verified, but I am sure it is: plase verify) P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From wangqj1 at 126.com Thu Oct 15 15:38:39 2009 From: wangqj1 at 126.com (Q.J.Wang) Date: Thu, 15 Oct 2009 21:38:39 +0800 (CST) Subject: [Pw_forum] total force ascend in the course of VC-relax Message-ID: <27317892.572221255613919203.JavaMail.coremail@bj126app106.126.com> Dear all I did geometry optimization with vc-relax in a supercell. When I increase the concentration of doped atoms ,the value of total force ascend instead of declining.Before I did not increase the concentration of doped atoms ,It converges well .Anyone who encounter the same problem as me ? The second problem is when I increase the relative distance of the two doped atoms in the supercell and did geometry optimization with vc-relax,and the error turn up as like : from c_bands : error # 1 too many bands are not converged So I am puzzled why before I increase relative position of two doped atoms the vc-relax converged and the latter turns up error in the same parameter . Any advice will be appreciated . -- Best regards Q.J.Wang XiangTan University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091015/3423199a/attachment.htm From paulatto at sissa.it Thu Oct 15 15:42:34 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 15 Oct 2009 15:42:34 +0200 Subject: [Pw_forum] total force ascend in the course of VC-relax In-Reply-To: <27317892.572221255613919203.JavaMail.coremail@bj126app106.126.com> References: <27317892.572221255613919203.JavaMail.coremail@bj126app106.126.com> Message-ID: In data 15 ottobre 2009 alle ore 15:38:39, Q.J.Wang ha scritto: > Dear all > I did geometry optimization with vc-relax in a supercell. When I > increase the concentration of doped atoms ,the value of total force > ascend instead of declining.Before I did not increase the concentration > of doped atoms ,It converges well .Anyone who encounter the same problem > as me ? I sounds normal to me, when you perturb the system you will have to redo the relaxation... why do you find it puzzling? > The second problem is when I increase the relative distance of the > two doped atoms in the supercell and did geometry optimization with > vc-relax,and the error turn up as like : from c_bands : error # 1 > too many bands are not converged > So I am puzzled why before I increase relative position of two doped > atoms the vc-relax converged and the latter turns up error in the same > parameter . It happens with peculiar configurations, the easiest way to get rid of it is normally to increase the number of bands. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From marzari at MIT.EDU Thu Oct 15 15:47:54 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 15 Oct 2009 09:47:54 -0400 Subject: [Pw_forum] Everything you always wanted to know about QE, Message-ID: <4AD7280A.60002@mit.edu> but were afraid to ask... Dear All, now all (or most of your questions) have been answered ! As you might recall, this summer NSF/ICMR and Democritos organized a summer school on materials modeling using QE at the University of Santa Barbara. Sadly, we had only 48 slots (half US-based, half from outside) and 300 applicants. So, we decided to put all the lectures (slides, videos !, and tutorials) on the web. You can find all of these linked from http://media.quantum-espresso.org . A few videos are corrupted (but not the audio) and a few of the keynote lectures are missing, but overall this is a very large and articulate amount of material that can act for everyone on this mailing list as the first resource to learn about all the different features of QE. So, if you have a question on the basics, or on some advanced techniques, go first through the relevant part of this material - most questions are answered there. Many thanks to everyone that made this possible: all the lecturers, keynote speakers, NSF and Democritos, but especially the ICTP Science Dissemination Unit (Enrique Canessa, Marco Zennaro and Carlo Fonda) for setting up the recording system OpenEyA, Gabriele Sclauzero (SISSA and Democritos) and Paolo Umari (Democritos) for recording the lectures, and Alberto Campagnari (Democritos) for preparing them online. Enjoy ! nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From lanhaiping at gmail.com Thu Oct 15 16:10:05 2009 From: lanhaiping at gmail.com (lan haiping) Date: Thu, 15 Oct 2009 22:10:05 +0800 Subject: [Pw_forum] Everything you always wanted to know about QE, In-Reply-To: <4AD7280A.60002@mit.edu> References: <4AD7280A.60002@mit.edu> Message-ID: Thanks for the efforts. Regards, Hai-Ping On Thu, Oct 15, 2009 at 9:47 PM, Nicola Marzari wrote: > > > but were afraid to ask... > > > Dear All, > > > now all (or most of your questions) have been answered ! As you might > recall, > this summer NSF/ICMR and Democritos organized a summer school on materials > modeling using QE at the University of Santa Barbara. Sadly, we had only > 48 slots > (half US-based, half from outside) and 300 applicants. So, we decided to > put all > the lectures (slides, videos !, and tutorials) on the web. > > You can find all of these linked from http://media.quantum-espresso.org . > > A few videos are corrupted (but not the audio) and a few of the keynote > lectures > are missing, but overall this is a very large and articulate amount of > material that > can act for everyone on this mailing list as the first resource to learn > about all the > different features of QE. So, if you have a question on the basics, or > on some advanced > techniques, go first through the relevant part of this material - most > questions are > answered there. > > Many thanks to everyone that made this possible: all the lecturers, > keynote speakers, NSF and Democritos, but especially the ICTP Science > Dissemination Unit (Enrique Canessa, Marco Zennaro and Carlo Fonda) for > setting up the recording system OpenEyA, Gabriele Sclauzero (SISSA and > Democritos) and > Paolo Umari (Democritos) for recording the lectures, and Alberto > Campagnari (Democritos) for > preparing them online. > > Enjoy ! > > nicola > > -- > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091015/b8d1be45/attachment.htm From mtt_turion at hotmail.com Thu Oct 15 17:17:36 2009 From: mtt_turion at hotmail.com (mtt physics) Date: Thu, 15 Oct 2009 22:17:36 +0700 Subject: [Pw_forum] how i can find tetragonal phase with PZT structure In-Reply-To: References: Message-ID: Dear all with PZT project I have a problem with calculate c/a for tetragonal phase My structure is perovskite help me please (mtt_turion at hotmail.com) > From: pw_forum-request at pwscf.org > Subject: Pw_forum Digest, Vol 28, Issue 36 > To: pw_forum at pwscf.org > Date: Thu, 15 Oct 2009 02:16:57 +0200 > > Send Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-owner at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. Re: Calculating projected density of states (dev sharma) > 2. Re: Calculating projected density of states (Lorenzo Paulatto) > 3. Re: Reference for Fock Exchange Minimization Scheme? > (Stefano de Gironcoli) > 4. forum searching (Mehrnaz Anvari) > 5. Re: forum searching (James J Ramsey) > 6. Re: forum searching (Tengfei Luo) > 7. Re: Reference for Fock Exchange Minimization Scheme? > (William Parker) > 8. Questions on phonon calculation in Example02 and 06 > (vtmtrinh at caltech.edu) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 14 Oct 2009 22:32:33 +0530 > From: dev sharma > Subject: Re: [Pw_forum] Calculating projected density of states > To: PWSCF Forum > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > Dear Saptrishi, > > Have u done nscf ?? > > for PDOS you have to do the three things > > first do scf > second do nscf (with nbnd option ) > third run the projwfc.x > > and for the plottings of files like > > *pdos_atm#1(Zn)_wfc#2(s)* > > u can use gnuplot etc. > > for the analysis u can read the forum and i am pasting one of the > explanation for you > > On Gio, Settembre 4, 2008 05:31, wangqj1 wrote: > >* > *>* Dear all > *>* After I used projwfc.x ,I got the pdos of every atom ,but I don't > *>* know the signification of the following characters of > *>* pdos_atm#1(Zn)_wfc#1(d) and pdos_atm#1(Zn)_wfc#2(s). > * > >* pdos_atm#1(Zn)_wfc#1(d) > *projected dos of atom number 1 (Zinc) wavefunction number 1 (d-orbital) > > >* pdos_atm#1(Zn)_wfc#2(s) > *projected dos of atom number 1 (Zinc) wavefunction number 2 (s-orbital) > > Each file contains 3 + (2l+1) columns, where l is the angular momentum of > its specific wavefunction. The first column is the energy. The second and > third are the total spin up and spin down dos; where "total" means its > summed over all values of m; than you find the projected dos for each > value of m, up and down, one at a time. > > e.g. for a p orbital you have p_x, p_y and p_z orbitals: > > # E(eV) ldosup(E) ldosdw(E) pdosup(E) pdosdw(E) pdosup(E)... > energy [sum_xyz p up] [sum_xyz p down] [p_z up] [p_z down] [p_x up]... > > I hope I was clear, regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > =================================== > > > > > for more help. please provide your input file. In that case , people may be > able to help you more > > take care, > Dev, > Univeristy of Delhi . > > On Wed, Oct 14, 2009 at 10:06 PM, Shaptrishi Sharma wrote: > > > Hi QE Users, > > > > I want to calculate the projected density of states for my system in > > quantum espresso. I have already calculate band structure . However honestly > > speaking I ahve not understood the concept of projectd density of states. > > Can anyone please give me some links and ideas about PDOS. Also how to > > calculate PDOS from band structure. I tried by running projwfc.x still I > > did not get my result. > > > > Thanks > > > > SS > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/d0ee6f25/attachment-0001.htm > > ------------------------------ > > Message: 2 > Date: Wed, 14 Oct 2009 19:05:00 +0200 > From: "Lorenzo Paulatto" > Subject: Re: [Pw_forum] Calculating projected density of states > To: "PWSCF Forum" > Message-ID: > Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes > > In data 14 ottobre 2009 alle ore 19:02:33, dev sharma > ha scritto: > > second do nscf (with nbnd option ) > > This step is not necessary, unless you want to include more bands in the > projection. > > cheers > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > > > ------------------------------ > > Message: 3 > Date: Wed, 14 Oct 2009 21:14:38 +0200 > From: Stefano de Gironcoli > Subject: Re: [Pw_forum] Reference for Fock Exchange Minimization > Scheme? > To: PWSCF Forum > Message-ID: <4AD6231E.5020109 at sissa.it> > Content-Type: text/plain; charset=UTF-8; format=flowed > > No, there isn't much more. > There is a paper in J. of Comp. Chem. 29, 2098 (2008) by Gillan, > Alfe`, de Gironcoli and Manby and one on JACS 129, 10402 (200& by Wang, > deGirocnoli, Hush, and Reimers > but they hardly discuss the implementation.... which is however > "standard", except for the detail discussaed in the QE paper. > see for instance Chawla and Voth, JCP 108, 4697 (1998); Sorouri, Foulkes > and Hine JCP 124, 064105 (2006), Spencer and Alavi, PRB77, 193110 > (2008); and many others that before and after claim to have solved the > problem and basically all do the same: build auxiliary densities > rho_{-q}=phi^{*}_{k+q}*psi_k in real space, FFT to reciprocal space, > solve Poisson equation, back FFT to real space, multiply by phi_{k+q} > and accumulate the result... the only tricky point is what you do for > q->0 and there Gygi-Baldereschi is still the reference point. > > stefano > > > > stefano > > William Parker wrote: > > That is the "reference paper appendix" I was referring to, but it's worth > > pointing out that a little bit is there. Anywhere else? > > > > --William > > > > ====================================================================== > > William Parker Tel : +1 (614) 292-2887 > > Graduate Research Associate Fax : +1 (614) 292-7557 > > Ohio State University > > Department of Physics > > 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu > > Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker > > U.S.A. > > > > Office: 2025 Physics Research Building > > ====================================================================== > > > > On Tue, 13 Oct 2009, Stefano Baroni wrote: > > > > > >> On Oct 12, 2009, at 5:34 PM, William Parker wrote: > >> > >> > >>> Hello everyone, > >>> > >>> Is there a publication that contains a more detailed description of > >>> the > >>> Fock exchange minimization scheme in QE than the reference paper > >>> appendix > >>> or the README in examples/EXX_example? > >>> > >> something (not much) is here: http://stacks.iop.org/0953-8984/21/395502 > >> > >> have you tried it out? > >> > >> SB > >> > >> --- > >> 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 > >> > > > > ------------------------------ > > Message: 4 > Date: Wed, 14 Oct 2009 22:46:25 +0330 > From: "Mehrnaz Anvari" > Subject: [Pw_forum] forum searching > To: pw_forum at pwscf.org > Message-ID: > Content-Type: text/plain; charset="windows-1256" > > Hello > I have a question about pwscf.org site. It seems that it has a problem so > how could I search previous questions in pwscf forum? Unfortunately I > couldn't find any way in quantum espresso site . > Best > Mehrnaz Anvari > Iran University of Science & Technology > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/ca6b8fb0/attachment-0001.htm > > ------------------------------ > > Message: 5 > Date: Wed, 14 Oct 2009 15:37:04 -0400 > From: James J Ramsey > Subject: Re: [Pw_forum] forum searching > To: PWSCF Forum > Message-ID: > <46e014700910141237s1935a774uaf9cb18b778920e2 at mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > 2009/10/14 Mehrnaz Anvari : > > Hello > > I have a question about pwscf.org site.?It seems that it has?a problem > > so?how could?I search previous questions in pwscf forum? Unfortunately I > > couldn't find any way in quantum espresso site?. > > See this page: http://www.quantum-espresso.org/tools.php > > > ------------------------------ > > Message: 6 > Date: Wed, 14 Oct 2009 15:39:40 -0400 > From: "Tengfei Luo" > Subject: Re: [Pw_forum] forum searching > To: "PWSCF Forum" > Message-ID: <6B09F4C6A56F41CCAD37294C2F6E38EE at mit.edu> > Content-Type: text/plain; charset="windows-1256" > > You may want to bookmark this page: > http://www.democritos.it/cgi-bin/htsearch?words=pwscf > > Tengfei > > ----- Original Message ----- > From: Mehrnaz Anvari > To: pw_forum at pwscf.org > Sent: Wednesday, October 14, 2009 3:16 PM > Subject: [Pw_forum] forum searching > > > Hello > I have a question about pwscf.org site. It seems that it has a problem so how could I search previous questions in pwscf forum? Unfortunately I couldn't find any way in quantum espresso site . > Best > Mehrnaz Anvari > Iran University of Science & Technology > > > > ------------------------------------------------------------------------------ > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091014/bb805cbe/attachment-0001.htm > > ------------------------------ > > Message: 7 > Date: Wed, 14 Oct 2009 17:17:27 -0400 (EDT) > From: William Parker > Subject: Re: [Pw_forum] Reference for Fock Exchange Minimization > Scheme? > To: PWSCF Forum > Message-ID: > Content-Type: TEXT/PLAIN; charset=utf-8 > > Thank you for the references, Stefano. It is always a little frustrating > trying to understand a method from a smattering of oblique references and > reading the source code as one's primary explanation. > > --William > > On Wed, 14 Oct 2009, Stefano de Gironcoli wrote: > > > No, there isn't much more. > > There is a paper in J. of Comp. Chem. 29, 2098 (2008) by Gillan, > > Alfe`, de Gironcoli and Manby and one on JACS 129, 10402 (200& by Wang, > > deGirocnoli, Hush, and Reimers > > but they hardly discuss the implementation.... which is however > > "standard", except for the detail discussaed in the QE paper. > > see for instance Chawla and Voth, JCP 108, 4697 (1998); Sorouri, Foulkes > > and Hine JCP 124, 064105 (2006), Spencer and Alavi, PRB77, 193110 > > (2008); and many others that before and after claim to have solved the > > problem and basically all do the same: build auxiliary densities > > rho_{-q}=phi^{*}_{k+q}*psi_k in real space, FFT to reciprocal space, > > solve Poisson equation, back FFT to real space, multiply by phi_{k+q} > > and accumulate the result... the only tricky point is what you do for > > q->0 and there Gygi-Baldereschi is still the reference point. > > > > stefano > > > > > > > > stefano > > > > William Parker wrote: > > > That is the "reference paper appendix" I was referring to, but it's worth > > > pointing out that a little bit is there. Anywhere else? > > > > > > --William > > > > > > ====================================================================== > > > William Parker Tel : +1 (614) 292-2887 > > > Graduate Research Associate Fax : +1 (614) 292-7557 > > > Ohio State University > > > Department of Physics > > > 191 West Woodruff Avenue email: wparker at mps.ohio-state.edu > > > Columbus, OH 43210-1117 http://www.physics.ohio-state.edu/~wparker > > > U.S.A. > > > > > > Office: 2025 Physics Research Building > > > ====================================================================== > > > > > > On Tue, 13 Oct 2009, Stefano Baroni wrote: > > > > > > > > >> On Oct 12, 2009, at 5:34 PM, William Parker wrote: > > >> > > >> > > >>> Hello everyone, > > >>> > > >>> Is there a publication that contains a more detailed description of > > >>> the > > >>> Fock exchange minimization scheme in QE than the reference paper > > >>> appendix > > >>> or the README in examples/EXX_example? > > >>> > > >> something (not much) is here: http://stacks.iop.org/0953-8984/21/395502 > > >> > > >> have you tried it out? > > >> > > >> SB > > >> > > >> --- > > >> 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 mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > ------------------------------ > > Message: 8 > Date: Wed, 14 Oct 2009 17:16:44 -0700 (PDT) > From: vtmtrinh at caltech.edu > Subject: [Pw_forum] Questions on phonon calculation in Example02 and > 06 > To: pw_forum at pwscf.org > Message-ID: > <4237.137.78.73.28.1255565804.squirrel at webmail.caltech.edu> > Content-Type: text/plain;charset=iso-8859-1 > > Dear PWSCF Users, > > Since I would like to compare the phonon calculations by using 2 > different ways as explained in Example02 and Example06, I rerun the > whole calculation in Example06. Everything was fine. I then rerun the > calculation of one specific q-point (-0.25, 0.25, -0.25) from Example06, > but using the way shown in Example02, ie. performing scf, calculation at > Gamma, nscf at q, and phonon at q. I compared the 2 results from the 2 > methods, I saw that both ways gave the same phonon frequencies, but the > displacement vectors are totally are different. I thought both should > have yielded the same results. I apprciate if you could you help me to > explain this. > > Below are the results: > > Following Example06: > > q = ( -0.250000000 0.250000000 -0.250000000 ) > ************************************************************************** > omega( 1) = 1.765425 [THz] = 58.888627 [cm-1] > ( -0.141032 0.460061 0.040238 0.471692 0.181270 0.011631 ) > ( -0.113839 0.478895 0.071975 0.478895 0.185814 0.000000 ) > omega( 2) = 1.765425 [THz] = 58.888627 [cm-1] > ( 0.253069 0.173719 -0.286906 0.139072 -0.539975 -0.034647 ) > ( 0.269727 0.160766 -0.283781 0.160766 -0.553509 0.000000 ) > omega( 3) = 4.536195 [THz] = 151.312190 [cm-1] > ( -0.204646 -0.290110 0.204646 0.290110 -0.204646 -0.290110 ) > ( -0.455290 0.000000 0.455290 0.000000 -0.455290 0.000000 ) > omega( 4) = 11.004563 [THz] = 367.075159 [cm-1] > ( -0.081755 -0.466090 0.446212 -0.432214 0.527968 0.033877 ) > ( 0.039228 0.161901 -0.147017 0.161901 -0.186245 0.000000 ) > omega( 5) = 11.004563 [THz] = 367.075159 [cm-1] > ( -0.451526 0.406642 0.107022 0.442481 0.558547 0.035839 ) > ( 0.149460 -0.153037 -0.047572 -0.153037 -0.197033 0.000000 ) > omega( 6) = 12.136012 [THz] = 404.816481 [cm-1] > ( -0.320406 -0.454213 0.320406 0.454213 -0.320406 -0.454213 ) > ( 0.156090 0.000000 -0.156090 0.000000 0.156090 0.000000 ) > ************************************************************************** > > > > Following Example02 > > q = ( -0.250000000 0.250000000 -0.250000000 ) > ************************************************************************** > omega( 1) = 1.765442 [THz] = 58.889209 [cm-1] > ( 0.318551 0.007983 -0.249525 -0.028467 -0.568077 -0.036450 ) > ( 0.325719 -0.012716 -0.256595 -0.012716 -0.582315 0.000000 ) > omega( 2) = 1.765442 [THz] = 58.889209 [cm-1] > ( -0.451869 0.141577 -0.493353 0.138915 -0.041485 -0.002662 ) > ( -0.452021 0.174129 -0.494545 0.174129 -0.042524 0.000000 ) > omega( 3) = 4.536219 [THz] = 151.312986 [cm-1] > ( -0.204633 -0.290122 0.204633 0.290122 -0.204633 -0.290122 ) > ( -0.455289 0.000000 0.455289 0.000000 -0.455289 0.000000 ) > omega( 4) = 11.004565 [THz] = 367.075220 [cm-1] > ( -0.184054 0.641087 -0.184054 0.641087 0.000000 0.000000 ) > ( 0.050209 -0.229371 0.050209 -0.229371 0.000000 0.000000 ) > omega( 5) = 11.004565 [THz] = 367.075220 [cm-1] > ( 0.384293 0.024658 -0.384293 -0.024658 -0.768587 -0.049316 ) > ( -0.135563 0.000000 0.135563 0.000000 0.271126 0.000000 ) > omega( 6) = 12.136049 [THz] = 404.817695 [cm-1] > ( -0.320383 -0.454229 0.320383 0.454229 -0.320383 -0.454229 ) > ( 0.156091 0.000000 -0.156091 0.000000 0.156091 0.000000 ) > ************************************************************************** > > > Best, > > MyTrinh Vo > > > > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 28, Issue 36 > **************************************** _________________________________________________________________ Windows Live: ?????????????????????????????????????????????????????????????????? http://www.microsoft.com/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:th-th:SI_SB_1:092010 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091015/13bd5bc3/attachment-0001.htm From paulatto at sissa.it Thu Oct 15 18:02:47 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 15 Oct 2009 18:02:47 +0200 Subject: [Pw_forum] how i can find tetragonal phase with PZT structure In-Reply-To: References: Message-ID: In data 15 ottobre 2009 alle ore 17:17:36, mtt physics ha scritto: > I have a problem with calculate c/a for tetragonal phase Deat mtt, it is actually very easy: just take c and divide it by a, that's the number you are looking for! Now, seriously; if you want some helpful advice you should provide more information about your problem. E.g. the input/output of your calculation, the kind of system, the pseudopotential you are using. Have you had a look at the example files? They used to be bundled together with the code, but they became too big, you can download them from here: Finally, the videos from a 2-week series of lectures on how to use quantum-espresso has recently been published on-line: you can find them here: . Please take some time watching at least the first few (by Baroni, Narasimhan, de Gironcoli and Dal Corso),they provide the very basic knowledge you really need to have before starting with electronic structure calculations. It is boring, I know, but it pays back in saved time. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From rfaccio at fq.edu.uy Fri Oct 16 00:19:35 2009 From: rfaccio at fq.edu.uy (Ricardo Faccio) Date: Thu, 15 Oct 2009 20:19:35 -0200 (UYST) Subject: [Pw_forum] how to use startingpot for different ecutwfc? In-Reply-To: References: Message-ID: <1419.190.188.81.16.1255645175.squirrel@webmail.fq.edu.uy> Dear QE users I want to perform a convergence study on k-points and ecut for any selected material. Let?s take Si for example. I used ?startingwfc? and ?startingpot? in order to use the last convergence stuff as a starting point for the next one. In the case of k-points there is no problems, we can use the charge-density and wfc without problems. But, when we increase the Ecut, pw.x stops with this error: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from read_rho_xml : error # 1 dimensions do not match %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Is there exist any way to use the previous charge-density in order to generate the starting potential? In particular looking for convergence in Ecutwfc I spent lot of time starting from the superposition of atoms. Thanks in advance. Ricardo -- ------------------------------------------------------------------------- ----- Dr. Ricardo Faccio Mail: Cryssmat-Lab., C?tedra de F?sica, DETEMA Facultad de Qu?mica, Universidad de la Rep?blica Av. Gral. Flores 2124, C.C. 1157 C.P. 11800, Montevideo, Uruguay. E-mail: rfaccio at fq.edu.uy Phone: 598 2 9241860 Int. 109 598 2 9290705 Fax: 598 2 9241906 Web: http://cryssmat.fq.edu.uy/ricardo/ricardo.htm From vtmtrinh at caltech.edu Fri Oct 16 01:39:50 2009 From: vtmtrinh at caltech.edu (vtmtrinh at caltech.edu) Date: Thu, 15 Oct 2009 16:39:50 -0700 (PDT) Subject: [Pw_forum] Questions on phonon calculation in Example02 and 06 In-Reply-To: References: <4237.137.78.73.28.1255565804.squirrel@webmail.caltech.edu> Message-ID: <1511.137.78.73.28.1255649990.squirrel@webmail.caltech.edu> Thank you very much, Paolo. I see what you said. I'll verify the second part. If I see anything differnt, I'll let you know. MyTrinh > > On Oct 15, 2009, at 2:16 , vtmtrinh at caltech.edu wrote: > >> I compared the 2 results from the 2 methods, I saw that both ways gave >> the same phonon frequencies, but the displacement vectors are totally >> different. > > no they aren't. Displacement vectors for non-degenerate modes (number 3 > and 6) are the same (within numerical accuracy). For degenerate > modes, you > get different linear combinations (because of the way matrix > diagonalization > works), but the subspace spanned by the two sets of degenerate > eigenvectors > is the same (I haven't verified, but I am sure it is: plase verify) > > P. > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From ttduyle at gmail.com Fri Oct 16 02:47:20 2009 From: ttduyle at gmail.com (Duy Le) Date: Thu, 15 Oct 2009 20:47:20 -0400 Subject: [Pw_forum] how to use startingpot for different ecutwfc? In-Reply-To: <1419.190.188.81.16.1255645175.squirrel@webmail.fq.edu.uy> References: <1419.190.188.81.16.1255645175.squirrel@webmail.fq.edu.uy> Message-ID: <8974d3b20910151747j76323493g8292185d65d474ea@mail.gmail.com> Hi,First of all, I am not sure why you changed number of k-points and/or ecutwfc, the wave function was read successfully. As I remembers, the wave functions expansion has k and G as indexes. So, when you changed number of k-points or cutoff energy, the wavefunction must not be read successfully. Anyway, this is not the problem you are asking. For "dimensions do not match": The dimensions of charge density usually depends on the size of unit cell and the energy cutoff for charge density. You probably changed the energy cutoff for charge density. If you do want to change the energy cutoff for charge density, you should consider to fix nr1, nr2, and nr3. D. -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. On Thu, Oct 15, 2009 at 6:19 PM, Ricardo Faccio wrote: > Dear QE users > I want to perform a convergence study on k-points and ecut for any > selected material. Let?s take Si for example. > I used ?startingwfc? and ?startingpot? in order to use the last > convergence stuff as a starting point for the next one. > In the case of k-points there is no problems, we can use the > charge-density and wfc without problems. But, when we increase the Ecut, > pw.x stops with this error: > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from read_rho_xml : error # 1 > dimensions do not match > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > Is there exist any way to use the previous charge-density in order to > generate the starting potential? In particular looking for convergence in > Ecutwfc > I spent lot of time starting from the superposition of atoms. > Thanks in advance. > Ricardo > > -- > ------------------------------------------------------------------------- > ----- Dr. Ricardo Faccio > > Mail: Cryssmat-Lab., C?tedra de F?sica, DETEMA > Facultad de Qu?mica, Universidad de la Rep?blica > Av. Gral. Flores 2124, C.C. 1157 > C.P. 11800, Montevideo, Uruguay. > E-mail: rfaccio at fq.edu.uy > Phone: 598 2 9241860 Int. 109 > 598 2 9290705 > Fax: 598 2 9241906 > Web: http://cryssmat.fq.edu.uy/ricardo/ricardo.htm > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091015/77414a01/attachment.htm From wangqj1 at 126.com Fri Oct 16 03:24:44 2009 From: wangqj1 at 126.com (Q.J.Wang) Date: Fri, 16 Oct 2009 09:24:44 +0800 (CST) Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 38 In-Reply-To: References: Message-ID: <20221367.676951255656284343.JavaMail.coremail@bj126app82.126.com> >Message: 2 >Date: Thu, 15 Oct 2009 15:42:34 +0200 >From: "Lorenzo Paulatto" >Subject: Re: [Pw_forum] total force ascend in the course of VC-relax >To: "PWSCF Forum" >Message-ID: >Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes > >In data 15 ottobre 2009 alle ore 15:38:39, Q.J.Wang ha >scritto: > >> Dear all >> I did geometry optimization with vc-relax in a supercell. When I >> increase the concentration of doped atoms ,the value of total force >> ascend instead of declining.Before I did not increase the concentration >> of doped atoms ,It converges well .Anyone who encounter the same problem >> as me ? > >I sounds normal to me, when you perturb the system you will have to redo >the relaxation... why do you find it puzzling? The problem is when I redo the relaxation ,the value of 'Total force'and the 'P' become large in every bfgs cycle .And at last shows error : from bfgs : error # 1 bfgs history already reset at previous step The input file as following : &CONTROL calculation = 'vc-relax' , restart_mode = 'from_scratch' , outdir = '/root/work3/' , pseudo_dir = '/root/pseudo/' , prefix = 'GaAs' , nstep = 250 , tstress = .true. , tprnfor = .true. , / &SYSTEM nat = 48, ntyp = 4, ecutwfc = 30 , ecutrho = 300 , nosym = .true. , occupations = 'smearing' , degauss = 0.005 , smearing = 'gaussian' , nspin = 2 , starting_magnetization(1) = 0.5, starting_magnetization(2) = 0.5, starting_magnetization(3) = 0.5, starting_magnetization(4) = 0.5, lda_plus_u = .false. , / &ELECTRONS conv_thr = 1.D-7 , mixing_mode = 'plain' , mixing_beta = 0.3 , / &IONS ion_dynamics = 'bfgs' , phase_space = 'full' , pot_extrapolation = 'second_order' , wfc_extrapolation = 'second_order' , upscale = 100 , / &CELL cell_dynamics = 'bfgs' , / ATOMIC_SPECIES Ga ........... ATOMIC_POSITIONS angstrom ............... K_POINTS automatic 3 3 3 1 1 1 > > >> The second problem is when I increase the relative distance of the >> two doped atoms in the supercell and did geometry optimization with >> vc-relax,and the error turn up as like : from c_bands : error # 1 >> too many bands are not converged >> So I am puzzled why before I increase relative position of two doped >> atoms the vc-relax converged and the latter turns up error in the same >> parameter . > >It happens with peculiar configurations, the easiest way to get rid of it >is normally to increase the number of bands. > >cheers >-- Best regards Q.J.Wang XiangTan University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091016/a21b3671/attachment-0001.htm From mambom1902 at yahoo.com Fri Oct 16 05:03:19 2009 From: mambom1902 at yahoo.com (loc duong ding) Date: Thu, 15 Oct 2009 20:03:19 -0700 (PDT) Subject: [Pw_forum] How to calculate the isotope effect Message-ID: <544606.65923.qm@web38804.mail.mud.yahoo.com> Dear PWSCF users, I am trying to consider effect of isotope.?How can I?get the PP of isotope (ex.?O(16) and?O(17))? Best,?----------------------------------------------- Loc Duong Dinh Ms-Ph.D Student Sungkyunkwan Advanced Institute of Nanotechnology, Sungkyunkwan University, Suwon, 440-746, Korea Email: mambom1902 at yahoo.com, ddloc at skku.edu From zqwu00 at gmail.com Fri Oct 16 05:52:17 2009 From: zqwu00 at gmail.com (zq wu) Date: Thu, 15 Oct 2009 20:52:17 -0700 Subject: [Pw_forum] partial phonon density of state Message-ID: <3ebf88c60910152052p29e1137bj91fe0906b2358881@mail.gmail.com> Dear all, I need to calculate the partial phonon density of state. Has the Eyvaz's code, which can calculate the partial phonon density of state, been incorporated in quantum expresso? Thanks Zhongqing -- CACS University of Southern California -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091015/2f0d293d/attachment.htm From lanhaiping at gmail.com Fri Oct 16 06:10:54 2009 From: lanhaiping at gmail.com (lan haiping) Date: Fri, 16 Oct 2009 12:10:54 +0800 Subject: [Pw_forum] How to calculate the isotope effect In-Reply-To: <544606.65923.qm@web38804.mail.mud.yahoo.com> References: <544606.65923.qm@web38804.mail.mud.yahoo.com> Message-ID: Hi, The isotope effect only has impact on PHONON calculation. In this situation , you need to specify different atomic mass for the different isotopes. Please have a look at INPUT_PH. Hopt it helps Regards, Hai-Ping On Fri, Oct 16, 2009 at 11:03 AM, loc duong ding wrote: > Dear PWSCF users, > > I am trying to consider effect of isotope. How can I get the PP of isotope > (ex. O(16) and O(17))? > > Best, ----------------------------------------------- > Loc Duong Dinh > Ms-Ph.D Student > Sungkyunkwan Advanced Institute of Nanotechnology, > Sungkyunkwan University, > Suwon, 440-746, Korea > Email: mambom1902 at yahoo.com, ddloc at skku.edu > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091016/da5bb099/attachment.htm From paulatto at sissa.it Fri Oct 16 08:10:08 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 16 Oct 2009 08:10:08 +0200 (CEST) Subject: [Pw_forum] How to calculate the isotope effect In-Reply-To: <544606.65923.qm@web38804.mail.mud.yahoo.com> References: <544606.65923.qm@web38804.mail.mud.yahoo.com> Message-ID: <57590.78.12.170.175.1255673408.squirrel@webmail.sissa.it> On Fri, October 16, 2009 05:03, loc duong ding wrote: > I am trying to consider effect of isotope.?How can I?get the PP of > isotope (ex.?O(16) and?O(17))? Dear Ioc Duong, different isotopes share the same electronic wavefunctions, well there is an effect because of non point-like nucleus, but you can ignore it safely. On of the most isotope effect is breaking the symmetry, to achieve this you can define 2 different atomic species for Oxygen: ATOMIC_SPECIES O16 16.00 O.pbe.blah.UPF O17 17.00 O.pbe.blah.UPF and then use them appropriately in the structure. The number between the species identifier and the pseudopotential file is the nuclear mass! regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From paulatto at sissa.it Fri Oct 16 08:14:51 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 16 Oct 2009 08:14:51 +0200 (CEST) Subject: [Pw_forum] how to use startingpot for different ecutwfc? In-Reply-To: <1419.190.188.81.16.1255645175.squirrel@webmail.fq.edu.uy> References: <1419.190.188.81.16.1255645175.squirrel@webmail.fq.edu.uy> Message-ID: <45941.78.12.170.175.1255673691.squirrel@webmail.sissa.it> On Fri, October 16, 2009 00:19, Ricardo Faccio wrote: > Is there exist any way to use the previous charge-density in order to > generate the starting potential? In particular looking for convergence in > Ecutwfc I spent lot of time starting from the superposition of atoms. Dear Riccardo, it is not possible in the current implementation; it can be implemented but, as far as I know, nobody is working on it. If want it you can try to implement it yourself. More specifically, convergence of ecut depends only on the pseudopotentials, why don't you study it on some simple system (pure bulk, isolated atom, small molecule) and then transfer to your system? regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From paulatto at sissa.it Fri Oct 16 08:18:35 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 16 Oct 2009 08:18:35 +0200 (CEST) Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 38 In-Reply-To: <20221367.676951255656284343.JavaMail.coremail@bj126app82.126.com> References: <20221367.676951255656284343.JavaMail.coremail@bj126app82.126.com> Message-ID: <48411.78.12.170.175.1255673915.squirrel@webmail.sissa.it> On Fri, October 16, 2009 03:24, Q.J.Wang wrote: > The problem is when I redo the relaxation ,the value of 'Total force'and > the 'P' become large in every bfgs cycle .And at last shows error : I see. Apart from a wacky initial configuration (you should check with xcrysden..) I suspect you might have used the same outdir and prefix for the two calculations, this usually produces some funny problems, can you check? regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From giannozz at democritos.it Fri Oct 16 08:30:47 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 16 Oct 2009 08:30:47 +0200 Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 38 In-Reply-To: <20221367.676951255656284343.JavaMail.coremail@bj126app82.126.com> References: <20221367.676951255656284343.JavaMail.coremail@bj126app82.126.com> Message-ID: <12DAC9A4-4846-4DBC-926B-6855066F67D5@democritos.it> On Oct 16, 2009, at 3:24 , Q.J.Wang wrote: > pot_extrapolation = 'second_order' , > wfc_extrapolation = 'second_order' these are useless for structural optimizaton --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Fri Oct 16 08:31:33 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 16 Oct 2009 08:31:33 +0200 Subject: [Pw_forum] How to calculate the isotope effect In-Reply-To: <544606.65923.qm@web38804.mail.mud.yahoo.com> References: <544606.65923.qm@web38804.mail.mud.yahoo.com> Message-ID: On Oct 16, 2009, at 5:03 , loc duong ding wrote: > I am trying to consider effect of isotope. on what? phonons? electronic structure? --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From sclauzer at sissa.it Fri Oct 16 08:37:28 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 16 Oct 2009 08:37:28 +0200 Subject: [Pw_forum] partial phonon density of state In-Reply-To: <3ebf88c60910152052p29e1137bj91fe0906b2358881@mail.gmail.com> References: <3ebf88c60910152052p29e1137bj91fe0906b2358881@mail.gmail.com> Message-ID: <4AD814A8.10805@sissa.it> Dear Zhongqing, I think you can download the code from the qe-forge.org website: https://qe-forge.org/frs/?group_id=28 Regards, Gabriele zq wu wrote: > > Dear all, > > I need to calculate the partial phonon density of state. Has the Eyvaz's > code, which can calculate the partial phonon density of state, been > incorporated in quantum expresso? > > Thanks > > Zhongqing > -- > > CACS University of Southern California > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From mambom1902 at yahoo.com Fri Oct 16 11:07:52 2009 From: mambom1902 at yahoo.com (loc duong ding) Date: Fri, 16 Oct 2009 02:07:52 -0700 (PDT) Subject: [Pw_forum] How to calculate the isotope effect Message-ID: <256586.33453.qm@web38801.mail.mud.yahoo.com> Thank you for your reply. ? By the way, I consider the isofect effect to electronic and phonon of carbon nanotube. Sincerely,----------------------------------------------- Loc Duong Dinh Ms-Ph.D Student Sungkyunkwan Advanced Institute of Nanotechnology, Sungkyunkwan University, Suwon, 440-746, Korea Email: mambom1902 at yahoo.com, ddloc at skku.edu From baroni at sissa.it Fri Oct 16 11:28:17 2009 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 16 Oct 2009 11:28:17 +0200 Subject: [Pw_forum] How to calculate the isotope effect In-Reply-To: <544606.65923.qm@web38804.mail.mud.yahoo.com> References: <544606.65923.qm@web38804.mail.mud.yahoo.com> Message-ID: peseudopotentials do not depend on the nuclear mass the electronic structure may depend on isotopic effect through the crystal geometry (which depend _very slightly_ on nuclear masses because of zero-point vibrations) and through electron-phonon effects. SB On Oct 16, 2009, at 5:03 AM, loc duong ding wrote: > Dear PWSCF users, > > I am trying to consider effect of isotope. How can I get the PP of > isotope (ex. O(16) and O(17))? > > Best, ----------------------------------------------- > Loc Duong Dinh > Ms-Ph.D Student > Sungkyunkwan Advanced Institute of Nanotechnology, > Sungkyunkwan University, > Suwon, 440-746, Korea > Email: mambom1902 at yahoo.com, ddloc at skku.edu > > > > _______________________________________________ > 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/20091016/9c8ac524/attachment.htm From baroni at sissa.it Fri Oct 16 11:29:04 2009 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 16 Oct 2009 11:29:04 +0200 Subject: [Pw_forum] partial phonon density of state In-Reply-To: <3ebf88c60910152052p29e1137bj91fe0906b2358881@mail.gmail.com> References: <3ebf88c60910152052p29e1137bj91fe0906b2358881@mail.gmail.com> Message-ID: I think it should be available on qe-forge.org - aint it, Eyvaz? Stefano On Oct 16, 2009, at 5:52 AM, zq wu wrote: > > Dear all, > > I need to calculate the partial phonon density of state. Has the > Eyvaz's code, which can calculate the partial phonon density of > state, been incorporated in quantum expresso? > > Thanks > > Zhongqing > -- > > CACS University of Southern California > _______________________________________________ > 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/20091016/3fb65643/attachment.htm From paulatto at sissa.it Fri Oct 16 13:41:10 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 16 Oct 2009 13:41:10 +0200 Subject: [Pw_forum] How to calculate the isotope effect In-Reply-To: <57590.78.12.170.175.1255673408.squirrel@webmail.sissa.it> References: <544606.65923.qm@web38804.mail.mud.yahoo.com> <57590.78.12.170.175.1255673408.squirrel@webmail.sissa.it> Message-ID: In data 16 ottobre 2009 alle ore 08:10:08, Lorenzo Paulatto ha scritto: > regards Sorry for the bad English, I should not write emails that early in the morning. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From wangqj1 at 126.com Fri Oct 16 14:48:08 2009 From: wangqj1 at 126.com (Q.J.Wang) Date: Fri, 16 Oct 2009 20:48:08 +0800 (CST) Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 40 In-Reply-To: References: Message-ID: <31196709.1174431255697288729.JavaMail.coremail@bj126app27.126.com> >Message: 6 >Date: Fri, 16 Oct 2009 08:18:35 +0200 (CEST) >From: "Lorenzo Paulatto" >Subject: Re: [Pw_forum] Pw_forum Digest, Vol 28, Issue 38 >To: "PWSCF Forum" >Message-ID: <48411.78.12.170.175.1255673915.squirrel at webmail.sissa.it> >Content-Type: text/plain;charset=iso-8859-1 > > >On Fri, October 16, 2009 03:24, Q.J.Wang wrote: >> The problem is when I redo the relaxation ,the value of 'Total force'and >> the 'P' become large in every bfgs cycle .And at last shows error : > >I see. Apart from a wacky initial configuration (you should check with >xcrysden..) I suspect you might have used the same outdir and prefix for >the two calculations, this usually produces some funny problems, can you >check? According to your advice ,I checked the initial configuration with xcrysden ,there is no problem about the initia configuration .The outdir and prefix I used also not the same in the two caculations . >regards > > >-- >Lorenzo Paulatto >SISSA & DEMOCRITOS (Trieste) >phone: +39 040 3787 511 >skype: paulatz >www: http://people.sissa.it/~paulatto/ > > > >---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > > >------------------------------ > >Message: 7 >Date: Fri, 16 Oct 2009 08:30:47 +0200 >From: Paolo Giannozzi >Subject: Re: [Pw_forum] Pw_forum Digest, Vol 28, Issue 38 >To: PWSCF Forum >Message-ID: <12DAC9A4-4846-4DBC-926B-6855066F67D5 at democritos.it> >Content-Type: text/plain; charset=US-ASCII; format=flowed > > >On Oct 16, 2009, at 3:24 , Q.J.Wang wrote: > >> pot_extrapolation = 'second_order' , >> wfc_extrapolation = 'second_order' > >these are useless for structural optimizaton OK ,Let me have a try . >--- >Paolo Giannozzi, Dept of Physics, University of Udine >via delle Scienze 208, 33100 Udine, Italy >Phone +39-0432-558216, fax +39-0432-558222 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091016/80eba0c4/attachment-0001.htm From Matthias.Timmer at uni-due.de Fri Oct 16 15:20:32 2009 From: Matthias.Timmer at uni-due.de (Matthias Timmer) Date: Fri, 16 Oct 2009 15:20:32 +0200 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? Message-ID: <4AD87320.7000702@uni-due.de> Dear readers, I have two points: 1) I am trying to install Espresso (v 4.0.5) on a Xeon Cluster, trying to use MKL 10.2.1.017 (with ifort 11.1). Unfortunately, the configure-script does not recognize the MKL libraries (the dgemm-tests fail), although the correct path is known and tested - probably due to a rather drastic change in the structure of the library (e.g., there is no file named 'libmkl_em64t.a' any more, which seems to have been changed for MKL 10.2). There is no change in this behavior when I try to compile Espresso v 4.1. Can anyone comment whether Espresso is at the moment compatible with MKL 10.2.1.017? And if so, is there a need to change the configure-script, or is this just a local problem I have? 2) When the configure-script runs, it always tells me, that the "x86_64"-architecture of the Xeons we have is "amd64" (see line 1443 in the configure-script). I think this is highly irritating. Must something be changed, or can I safely ignore this message? I will provide any files (especially config.log) you might ask for, if they turn out to be necessary. Kind regards, Matthias Timmer University of Duisburg-Essen -------------- next part -------------- A non-text attachment was scrubbed... Name: Matthias_Timmer.vcf Type: text/x-vcard Size: 318 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20091016/7d8e2fa2/attachment.vcf From giannozz at democritos.it Fri Oct 16 15:55:19 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 16 Oct 2009 15:55:19 +0200 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: <4AD87320.7000702@uni-due.de> References: <4AD87320.7000702@uni-due.de> Message-ID: <0B2AFA97-DA84-47D1-B3BD-C36632A25FA2@democritos.it> On Oct 16, 2009, at 15:20 , Matthias Timmer wrote: > Unfortunately, the configure-script does not recognize the MKL > libraries > (the dgemm-tests fail), although the correct path is known and > tested - > probably due to a rather drastic change in the structure of the > library nothing surprising. Each new version of Intel software is orthogonal to all previous versions, like in Lanczos chains. > Can anyone comment whether Espresso is at the moment compatible > with MKL 10.2.1.017? of course it is (unless they changed the way blas and lapack routines are called). Carlo Nervi told me a few days ago that he managed to have everything working with ifort 11.1.056 and MKL 10.2.2.025. > And if so, is there a need to change the configure-script if you want mkl to be recognized by configure, yes, configure.ac has to be modified. > 2) When the configure-script runs, it always tells me, that the > "x86_64"-architecture of the Xeons we have is "amd64" (see line > 1443 in the configure-script). I think this is highly irritating. > Must something be changed, or can I safely ignore this message? configure reports what your operating system says. If it works, it is fine. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From jjr19 at zips.uakron.edu Fri Oct 16 16:02:22 2009 From: jjr19 at zips.uakron.edu (James J Ramsey) Date: Fri, 16 Oct 2009 10:02:22 -0400 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: <4AD87320.7000702@uni-due.de> References: <4AD87320.7000702@uni-due.de> Message-ID: <46e014700910160702r68e814f5nb3b7a2156000bc20@mail.gmail.com> On Fri, Oct 16, 2009 at 9:20 AM, Matthias Timmer wrote: > 2) When the configure-script runs, it always tells me, that the > "x86_64"-architecture of the Xeons we have is "amd64" (see line 1443 in the > configure-script). I think this is highly irritating. On Debian systems, x86_64 is often called amd64: http://alioth.debian.org/docman/view.php/30192/21/debian-amd64-howto.html#id250846 Not sure if you have a Debian (or Ubuntu) system or not, but that might help explain what you are seeing. From gbilalbegovic at gmail.com Fri Oct 16 16:11:12 2009 From: gbilalbegovic at gmail.com (Goranka Bilalbegovic) Date: Fri, 16 Oct 2009 16:11:12 +0200 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: <4AD87320.7000702@uni-due.de> References: <4AD87320.7000702@uni-due.de> Message-ID: <7332fead0910160711u1667f02by9417a371a172869f@mail.gmail.com> On Fri, Oct 16, 2009 at 3:20 PM, Matthias Timmer wrote 1) I am trying to install Espresso (v 4.0.5) on a Xeon Cluster, trying to use MKL 10.2.1.017 (with ifort 11.1). Hello Matthias, On our Linux cluster, I have recently compiled espresso-4.1 using MKL 10.2.1.017 and ifort 11.1 without any problems. Best wishes, Goranka Bilalbegovic Department of Physics, University of Zagreb Croatia -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091016/37046b46/attachment.htm From carlo.nervi at unito.it Fri Oct 16 16:39:19 2009 From: carlo.nervi at unito.it (Carlo Nervi) Date: Fri, 16 Oct 2009 16:39:19 +0200 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: <0B2AFA97-DA84-47D1-B3BD-C36632A25FA2@democritos.it> References: <4AD87320.7000702@uni-due.de> <0B2AFA97-DA84-47D1-B3BD-C36632A25FA2@democritos.it> Message-ID: <4AD88597.4010003@unito.it> I confirm that I was able to compile QE with intel ifort 11.1.056 and mkl 10.2.2.025. It added manually the libraries. Of course, you cannot use -lmkl_em64t, but: BLAS_LIBS = -lmkl_blas95_lp64 -lmkl_intel_lp64 -lmkl_intel_sequential -lmkl_core LAPACK_LIBS = -lfftw3xf_intel -lmkl_lapack95_lp64 -lmkl_intel_lp64 -lmkl_intel_sequential -lmkl_core hope this helps, Carlo -- ------------------------------------------------------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ From zqwu00 at gmail.com Fri Oct 16 19:32:29 2009 From: zqwu00 at gmail.com (zq wu) Date: Fri, 16 Oct 2009 10:32:29 -0700 Subject: [Pw_forum] partial phonon density of state In-Reply-To: References: <3ebf88c60910152052p29e1137bj91fe0906b2358881@mail.gmail.com> Message-ID: <3ebf88c60910161032t2cc078eco4a6d66d0c8d9e2c0@mail.gmail.com> Thanks all of you. I get it in Quasiharmonic Approximation section Zhongqing On Fri, Oct 16, 2009 at 2:29 AM, Stefano Baroni wrote: > I think it should be available on qe-forge.org - aint it, Eyvaz?Stefano > > On Oct 16, 2009, at 5:52 AM, zq wu wrote: > > > Dear all, > > I need to calculate the partial phonon density of state. Has the Eyvaz's > code, which can calculate the partial phonon density of state, been > incorporated in quantum expresso? > > Thanks > > Zhongqing > -- > > CACS University of Southern California > _______________________________________________ > 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 > > -- CACS University of Southern California -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091016/835b2d1c/attachment.htm From eyvaz_isaev at yahoo.com Fri Oct 16 20:09:45 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 16 Oct 2009 11:09:45 -0700 (PDT) Subject: [Pw_forum] partial phonon density of state In-Reply-To: <3ebf88c60910161032t2cc078eco4a6d66d0c8d9e2c0@mail.gmail.com> Message-ID: <309106.73656.qm@web65715.mail.ac4.yahoo.com> Hi, --- On Fri, 10/16/09, zq wu wrote: > Thanks all of you. I get it in > Quasiharmonic Approximation section Veeeery Good! Any question, welcome. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Fri, 10/16/09, zq wu wrote: > From: zq wu > Subject: Re: [Pw_forum] partial phonon density of state > To: "PWSCF Forum" > Date: Friday, October 16, 2009, 9:32 PM > Thanks all of you. I get it in > Quasiharmonic Approximation section > > Zhongqing > > On Fri, Oct 16, 2009 at 2:29 AM, > Stefano Baroni > wrote: > > I think it should be > available on qe-forge.org - aint it, > Eyvaz? > Stefano > On Oct 16, 2009, > at 5:52 AM, zq wu wrote: > > Dear all, > > I need to calculate the partial phonon density of state. > Has the Eyvaz's code, which can calculate the partial > phonon density of state, been incorporated in quantum > expresso? > > > Thanks > > Zhongqing > -- > > CACS University of Southern California > > _______________________________________________ > 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 > > > > > > -- > > CACS University of Southern California > > > -----Inline Attachment Follows----- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From baris.malcioglu at gmail.com Fri Oct 16 22:19:59 2009 From: baris.malcioglu at gmail.com (O. Baris Malcioglu) Date: Fri, 16 Oct 2009 23:19:59 +0300 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: <4AD88597.4010003@unito.it> References: <4AD87320.7000702@uni-due.de> <0B2AFA97-DA84-47D1-B3BD-C36632A25FA2@democritos.it> <4AD88597.4010003@unito.it> Message-ID: Dear all, For manually compiling QE using intel mkl, you may also refer to http://software.intel.com/en-us/articles/intel-mkl-link-line-advisor/ to obtain the "link line" corresponding to system at hand. I have successfully tested this with current version of QE at CVS. Best, Baris 2009/10/16 Carlo Nervi : > I confirm that I was able to compile QE with intel ifort 11.1.056 and > mkl 10.2.2.025. > It added manually the libraries. Of course, you cannot use -lmkl_em64t, but: > > BLAS_LIBS ? ? ?= -lmkl_blas95_lp64 -lmkl_intel_lp64 > -lmkl_intel_sequential -lmkl_core > > LAPACK_LIBS ? ?= ?-lfftw3xf_intel -lmkl_lapack95_lp64 -lmkl_intel_lp64 > -lmkl_intel_sequential -lmkl_core > > hope this helps, > ? ? ? ?Carlo > > -- > ------------------------------------------------------ > Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 > Fax: +39 011 6707855 ? - ? Dipartimento di Chimica IFM > via P. Giuria 7, 10125 Torino, Italy > http://lem.ch.unito.it/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From quantumdft at gmail.com Sat Oct 17 07:54:29 2009 From: quantumdft at gmail.com (vega lew) Date: Sat, 17 Oct 2009 13:54:29 +0800 Subject: [Pw_forum] Could Q-E benefit a lot from GPU computation. Message-ID: <412f6c680910162254k3847e73esb3ae9ab06e3ebbab@mail.gmail.com> Dear all, In china, there is a huge computer center of GPU computation located in Beijing. Now there is a pretty good opportunity of collaboration for us. But the most important thing is that the computer center only has GPU with CUDA for calculation instead of CPU with Fortran. Is there a possibility now to transplant the current Q-E code easily to the GPU computer with single precession? Do you think Double precession is so important for Q-E codes? thank you for reading vega -- ================================================================================== Vega Lew ( weijia liu) PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China ****************************************************************************************************************** Email: vegalew at gmail.com Office: Room A705, Technical Innovation Building, Xinmofan Road 5#, Nanjing, Jiangsu, China ****************************************************************************************************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091017/60b50101/attachment.htm From giannozz at democritos.it Sat Oct 17 10:14:52 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 17 Oct 2009 10:14:52 +0200 Subject: [Pw_forum] Could Q-E benefit a lot from GPU computation. In-Reply-To: <412f6c680910162254k3847e73esb3ae9ab06e3ebbab@mail.gmail.com> References: <412f6c680910162254k3847e73esb3ae9ab06e3ebbab@mail.gmail.com> Message-ID: <005005C9-59BD-4CA3-B98C-A682D64B73BC@democritos.it> On Oct 17, 2009, at 7:54 , vega lew wrote: > But the most important thing is that the computer center only has GPU > with CUDA for calculation instead of CPU with Fortran. Is there a > possibility > now to transplant the current Q-E code easily to the GPU computer with > single precession? Do you think Double precession is so important > for Q-E codes? you mean: single/double "precision" ? I think double precision is important. Anyway, there is some ongoing work on porting Q-E to GPUs. P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From udayagiri3 at gmail.com Sat Oct 17 11:31:14 2009 From: udayagiri3 at gmail.com (udayagiri sai babu) Date: Sat, 17 Oct 2009 15:01:14 +0530 Subject: [Pw_forum] Cr GGA potential is underestimating the cohesive energy. Message-ID: Dear all I am calculated the cohesive energy of Cr using the pseudopotential Cr.pbe-sp-van.UPF. I got is *3.63* *ev/atom* as the cohesive energy where as the experimental value is *4.1 ev/atom*. As i know the GGA pseudopotential should overestimate the cohesive energy value but here it is underestimating. What is the problem? can somebody please help me out. The magnetic state of the Cr in the bulk is antiferromagnetic at zero kelvin (Neel temperature is ~312 K). For the sake of complete information below are my input files. *Bulk* &control calculation = 'scf', restart_mode='from_scratch', outdir='/home/sai/softwares/espresso-4.0.1/workspace/cr_paramagnetic/temp', prefix='chromium' pseudo_dir='/home/sai/softwares/espresso-4.0.1/pseudo/', / &system ibrav=1 celldm(1)=5.250000, nat=2, ntyp=2, ecutwfc=60, ecutrho=600, occupations='smearing', smearing='mp', degauss=0.010000 nspin=2 starting_magnetization(1)=0.5 starting_magnetization(2)=-0.5 / &electrons conv_thr=1.D-8, diagonalization='david', mixing_mode='plain', mixing_beta= 0.7 / ATOMIC_SPECIES Cr1 51.9961 Cr.pbe-sp-van.UPF Cr2 51.9961 Cr.pbe-sp-van.UPF ATOMIC_POSITIONS (crystal) Cr1 0.00 0.00 0.00 Cr2 0.5 0.5 0.5 K_POINTS {automatic} 10 10 10 0 0 0 *Isolated* &control calculation = 'scf', restart_mode='from_scratch', outdir='/home/sai/softwares/espresso-4.0.1/workspace/cr_paramagnetic/temp', prefix='chromium' pseudo_dir='/home/sai/softwares/espresso-4.0.1/pseudo/', / &system ibrav=1 celldm(1)=15, nat=1, ntyp=1, ecutwfc=60, ecutrho=600, occupations='fixed', tot_magnetization=6 nspin=2 nosym=.true., / &electrons conv_thr=1.D-8, diagonalization='david', mixing_mode='plain', mixing_beta= 0.7 / ATOMIC_SPECIES Cr 51.9961 Cr.pbe-sp-van.UPF ATOMIC_POSITIONS (crystal) Cr 0.00 0.00 0.00 K_POINTS {automatic} 1 1 1 0 0 0 -- U.Saibabu PhD student, Deformation mechanisms modeling group, Materials engineering department, IISc Bangalore, India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091017/c586484a/attachment.htm From giannozz at democritos.it Sat Oct 17 12:37:01 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 17 Oct 2009 12:37:01 +0200 Subject: [Pw_forum] Cr GGA potential is underestimating the cohesive energy. In-Reply-To: References: Message-ID: <156151B9-0A65-465A-B6F6-B8ED03F3C9E0@democritos.it> On Oct 17, 2009, at 11:31 , udayagiri sai babu wrote: > I am calculated the cohesive energy of Cr using the pseudopotential > Cr.pbe-sp-van.UPF. I got is 3.63 ev/atom as the cohesive energy > where as the experimental value is 4.1 ev/atom. As i know the GGA > pseudopotential should overestimate the cohesive energy value should it? LDA overestimates, GGA not necessarily so. The cohesive energy of Cr has been already calculated by many authors, for sure. You may want to have a look at what other people have done before you. For instance, here: http://dare.ubvu.vu.nl/bitstream/1871/9949/1/B4.pdf (found in 5' of google search). They get 3.58eV for Becke-Perdew, 3.80eV for Perdew-Wang. So 3.63eV for PBE after all is not that bad. P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From xinhuolin at gmail.com Sat Oct 17 14:09:19 2009 From: xinhuolin at gmail.com (Huolin Xin) Date: Sat, 17 Oct 2009 08:09:19 -0400 Subject: [Pw_forum] Could Q-E benefit a lot from GPU computation. In-Reply-To: <005005C9-59BD-4CA3-B98C-A682D64B73BC@democritos.it> References: <412f6c680910162254k3847e73esb3ae9ab06e3ebbab@mail.gmail.com> <005005C9-59BD-4CA3-B98C-A682D64B73BC@democritos.it> Message-ID: Tesla C1060 and C1070 supports double precision. The next architecture "Fermi" (product will come out next year) has more efficient support of double precision. If I recall correctly, CUDA released CUDO_blas, but I am not sure about lapack. However, for FFT, it is a big boost. Huolin On Sat, Oct 17, 2009 at 4:14 AM, Paolo Giannozzi wrote: > > On Oct 17, 2009, at 7:54 , vega lew wrote: > >> But the most important thing is that the computer center only has GPU >> with CUDA for calculation instead of CPU with Fortran. Is there a >> possibility >> now to transplant the current Q-E code easily to the GPU computer with >> single precession? Do you think Double precession is so important >> for Q-E codes? > > you mean: single/double "precision" ? I think double precision is > important. > Anyway, there is some ongoing work on porting Q-E to GPUs. > > P. > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From quantumdft at gmail.com Sat Oct 17 18:00:11 2009 From: quantumdft at gmail.com (vega lew) Date: Sun, 18 Oct 2009 00:00:11 +0800 Subject: [Pw_forum] Could Q-E benefit a lot from GPU computation. In-Reply-To: <005005C9-59BD-4CA3-B98C-A682D64B73BC@democritos.it> References: <412f6c680910162254k3847e73esb3ae9ab06e3ebbab@mail.gmail.com> <005005C9-59BD-4CA3-B98C-A682D64B73BC@democritos.it> Message-ID: <412f6c680910170900u66fdaf82x99d422ef0cd0b00e@mail.gmail.com> Dear Sir, Thank you for your reply. I wonder whether Q-E has any stratagem to on porting Q-E to GPU computers with CUDA? Is there a schedule for this? Thank you again for you time. best, vega On Sat, Oct 17, 2009 at 4:14 PM, Paolo Giannozzi wrote: > > On Oct 17, 2009, at 7:54 , vega lew wrote: > > > But the most important thing is that the computer center only has GPU > > with CUDA for calculation instead of CPU with Fortran. Is there a > > possibility > > now to transplant the current Q-E code easily to the GPU computer with > > single precession? Do you think Double precession is so important > > for Q-E codes? > > you mean: single/double "precision" ? I think double precision is > important. > Anyway, there is some ongoing work on porting Q-E to GPUs. > > P. > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- ================================================================================== Vega Lew ( weijia liu) PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China ****************************************************************************************************************** Email: vegalew at gmail.com Office: Room A705, Technical Innovation Building, Xinmofan Road 5#, Nanjing, Jiangsu, China ****************************************************************************************************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091018/5659aa02/attachment.htm From ZChen at cau.edu Sat Oct 17 20:15:07 2009 From: ZChen at cau.edu (Chen, Zhifan) Date: Sat, 17 Oct 2009 14:15:07 -0400 Subject: [Pw_forum] c60 Message-ID: Dear all has anyone calculated the C60 total energy? I used DMol (all electron) and PWscf (pseudopotential) to calculated the C60 total energy. I found DMol result is five times biger than PWscf result. Do I suppose to get the almost same data from the calculations of all electron and pseudopotential representation? Zhifan Chen Clark Atlanta University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091017/e1bda0e3/attachment.htm From decboy9 at gmail.com Sat Oct 17 20:15:40 2009 From: decboy9 at gmail.com (dev sharma) Date: Sat, 17 Oct 2009 23:45:40 +0530 Subject: [Pw_forum] lapack and BLAS Message-ID: hi, can anybody help in installing LAPACK and BLIS libraries. I downloaded the lapack-3.2.1 and BLAS libraries. I want to use EXE code for which BLAS is required. I dont know, how to install or use these libraries. Please help and guide me. when i tried to make BLAS and LAPACK i got [dev at dev BLAS]$ make g77 -O3 -c isamax.f -o isamax.o make: g77: Command not found make: *** [isamax.o] Error 127 [dev at dev BLAS]$ Thanks in advance. Dev Sharma, Univeristy of Delhi, India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091017/88de63e8/attachment.htm From lanhaiping at gmail.com Sat Oct 17 20:18:05 2009 From: lanhaiping at gmail.com (lan haiping) Date: Sun, 18 Oct 2009 02:18:05 +0800 Subject: [Pw_forum] c60 In-Reply-To: References: Message-ID: Hi, The total energy doesnot make sense. And Dmol code adopts all-electron calculation for Carbon, while PWSCF is a pseudopotential scheme. So the total electrons in two calculations are not the same . Regards, On Sun, Oct 18, 2009 at 2:15 AM, Chen, Zhifan wrote: > Dear all > has anyone calculated the C60 total energy? I used DMol (all electron) > and PWscf (pseudopotential) to calculated the C60 total energy. I found DMol > result is five times biger than PWscf result. Do I suppose to get the almost > same data from the calculations of all electron and pseudopotential > representation? > > > Zhifan Chen > Clark Atlanta University > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091018/bae07af1/attachment-0001.htm From lanhaiping at gmail.com Sat Oct 17 20:19:34 2009 From: lanhaiping at gmail.com (lan haiping) Date: Sun, 18 Oct 2009 02:19:34 +0800 Subject: [Pw_forum] lapack and BLAS In-Reply-To: References: Message-ID: you donot have compiler g77, you may try gfortran or ifort , et al. Just edit the makefile for the packages, and check which compiler you have regards, hai-ping On Sun, Oct 18, 2009 at 2:15 AM, dev sharma wrote: > hi, > can anybody help in installing LAPACK and BLIS libraries. I downloaded the > lapack-3.2.1 and BLAS libraries. I want to use EXE code for which BLAS is > required. I dont know, how to install or use these libraries. Please help > and guide me. when i tried to make BLAS and LAPACK i got > > [dev at dev BLAS]$ make > g77 -O3 -c isamax.f -o isamax.o > make: g77: Command not found > make: *** [isamax.o] Error 127 > [dev at dev BLAS]$ > > Thanks in advance. > Dev Sharma, > Univeristy of Delhi, > India > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091018/d1c20653/attachment.htm From degironc at sissa.it Sat Oct 17 21:19:34 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sat, 17 Oct 2009 21:19:34 +0200 Subject: [Pw_forum] Could Q-E benefit a lot from GPU computation. In-Reply-To: <412f6c680910170900u66fdaf82x99d422ef0cd0b00e@mail.gmail.com> References: <412f6c680910162254k3847e73esb3ae9ab06e3ebbab@mail.gmail.com> <005005C9-59BD-4CA3-B98C-A682D64B73BC@democritos.it> <412f6c680910170900u66fdaf82x99d422ef0cd0b00e@mail.gmail.com> Message-ID: <20091017211934.d5k7f44mm8880ckk@webmail.sissa.it> there is interest and we are in contact with someone that has started making some experiment. the problem as usual is human time and expertise. stefano Quoting vega lew : > Dear Sir, > > Thank you for your reply. > > I wonder whether Q-E has any stratagem to on porting Q-E to GPU computers > with CUDA? Is there a schedule for this? > > Thank you again for you time. > > best, > > vega > > On Sat, Oct 17, 2009 at 4:14 PM, Paolo Giannozzi > wrote: > >> >> On Oct 17, 2009, at 7:54 , vega lew wrote: >> >> > But the most important thing is that the computer center only has GPU >> > with CUDA for calculation instead of CPU with Fortran. Is there a >> > possibility >> > now to transplant the current Q-E code easily to the GPU computer with >> > single precession? Do you think Double precession is so important >> > for Q-E codes? >> >> you mean: single/double "precision" ? I think double precision is >> important. >> Anyway, there is some ongoing work on porting Q-E to GPUs. >> >> P. >> --- >> 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 mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > > -- > ================================================================================== > Vega Lew ( weijia liu) > PH.D Candidate in Chemical Engineering > State Key Laboratory of Materials-oriented Chemical Engineering > College of Chemistry and Chemical Engineering > Nanjing University of Technology, 210009, Nanjing, Jiangsu, China > ****************************************************************************************************************** > Email: vegalew at gmail.com > Office: Room A705, Technical Innovation Building, Xinmofan Road 5#, Nanjing, > Jiangsu, China > ****************************************************************************************************************** > ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by Horde http://www.horde.org/ From decboy9 at gmail.com Sun Oct 18 09:24:54 2009 From: decboy9 at gmail.com (dev sharma) Date: Sun, 18 Oct 2009 12:54:54 +0530 Subject: [Pw_forum] yambo installation Message-ID: hi 2 all, I tried to install the yambo but not succeed. Please help. I download the yambo and configure it with ./configure --with-netdf=/home/dev/netcdf-4.0.1/ --with-iotk=/home/dev/espresso-4.0.2/iotk/src/libiotk.a --with-p2y=/home/dev/espresso-4.0.2/ after that i got, [VER] 3.2.1 r.448 # # [SYS] linux at i686 # [SRC] /home/dev/yambo-3.2.1-r.448 # [BIN] /home/dev/yambo-3.2.1-r.448/bin # [FFT] Goedecker Fast Fourier transform with 0 cache [FFT] Goedecker Fast Fourier transform with 0 cache # # [ ] Double precision # [X] Redundant compilation # [ ] MPI # [ ] PW (3.2) support # [ ] ETSF I/O support # [ ] SCALAPACK # [ ] NETCDF/Large Files # [XX ] Built-in BLAS/LAPACK/LOCAL # # [ CPP ] gcc -E -P # [ C ] gcc -g -O2 -D_C_US -D_FORTRAN_US # [MPICC] # [ F90 ] f95 -O # [MPIF ] # [ F77 ] f95 -O # [Cmain] # [NoOpt] # # [ MAKE ] make # [EDITOR] vim then i did make yambo interfaces after which i got a2y yambo and ypp in the bin but not p2y [image: :(] [image: :(]. Please guide. thanks Dev sharma, University of Delhi, India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091018/559d7b7d/attachment.htm From iori at theory.polytechnique.fr Sun Oct 18 11:13:39 2009 From: iori at theory.polytechnique.fr (iori at theory.polytechnique.fr) Date: Sun, 18 Oct 2009 11:13:39 +0200 (CEST) Subject: [Pw_forum] yambo installation In-Reply-To: References: Message-ID: <156d7bf1c4864f6baeb9bf669f2f6bee.squirrel@theory.polytechnique.fr> Hi. I think you should specify indeed: --with-p2y=4.0 but have a look also to the yambo website (documentation) http://www.yambo-code.org/forum/viewforum.php?f=1&sid=f6a47f967aced5269868973dd1c8b23d and to the foruminparticular where you can find whatever you need: http://www.yambo-code.org/doc/installing.php i hope it works! ciao federico iori. > hi 2 all, > > I tried to install the yambo but not succeed. Please help. > I download the yambo and configure it with > ./configure --with-netdf=/home/dev/netcdf-4.0.1/ > --with-iotk=/home/dev/espresso-4.0.2/iotk/src/libiotk.a > --with-p2y=/home/dev/espresso-4.0.2/ > > after that i got, > [VER] 3.2.1 r.448 > # > # [SYS] linux at i686 > # [SRC] /home/dev/yambo-3.2.1-r.448 > # [BIN] /home/dev/yambo-3.2.1-r.448/bin > # [FFT] Goedecker Fast Fourier transform with 0 cache > > [FFT] Goedecker Fast Fourier transform with 0 cache > # > # [ ] Double precision > # [X] Redundant compilation > # [ ] MPI > # [ ] PW (3.2) support > # [ ] ETSF I/O support > # [ ] SCALAPACK > # [ ] NETCDF/Large Files > # [XX ] Built-in BLAS/LAPACK/LOCAL > # > # [ CPP ] gcc -E -P > # [ C ] gcc -g -O2 -D_C_US -D_FORTRAN_US > # [MPICC] > # [ F90 ] f95 -O > # [MPIF ] > # [ F77 ] f95 -O > # [Cmain] > # [NoOpt] > # > # [ MAKE ] make > # [EDITOR] vim > > then i did make yambo interfaces > after which i got a2y yambo and ypp in the bin but not p2y [image: :(] > [image: > :(]. > Please guide. > thanks > Dev sharma, > University of Delhi, > India > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > Dr. Federico Iori, PhD! Laboratoire des Solides Irradi?s, Ecole Polytechnique - CEA/DSM, CNRS 91128 Palaiseau, France www.etsf.polytechnique.fr phone: +33 (0) 1 69 33 44 85 emails: federico.iori @ polytechnique.edu iori.federico @ unimore.it http://www.pdmodena.it/ http://www.comune.mo.it Dr. Federico Iori, PhD! Laboratoire des Solides Irradi?s, Ecole Polytechnique - CEA/DSM, CNRS 91128 Palaiseau, France www.etsf.polytechnique.fr phone: +33 (0) 1 69 33 44 85 emails: federico.iori @ polytechnique.edu iori.federico @ unimore.it http://www.pdmodena.it/ http://www.comune.mo.it From decboy9 at gmail.com Sun Oct 18 13:40:11 2009 From: decboy9 at gmail.com (dev sharma) Date: Sun, 18 Oct 2009 17:10:11 +0530 Subject: [Pw_forum] yambo installation In-Reply-To: <156d7bf1c4864f6baeb9bf669f2f6bee.squirrel@theory.polytechnique.fr> References: <156d7bf1c4864f6baeb9bf669f2f6bee.squirrel@theory.polytechnique.fr> Message-ID: SIr, I tried with ./configure --with-netdf=/home/dev/netcdf-4.0.1/ --with-iotk=/home/dev/espresso-4.0.2/iotk/src/libiotk.a --with-p2y=4.0 but the problem is same :( Please guide. Thanks, Dev University of Delhi India On Sun, Oct 18, 2009 at 2:43 PM, wrote: > Hi. > I think you should specify indeed: > > --with-p2y=4.0 > > but have a look also to the yambo website (documentation) > > > http://www.yambo-code.org/forum/viewforum.php?f=1&sid=f6a47f967aced5269868973dd1c8b23d > > > and to the foruminparticular where you can find whatever you need: > > http://www.yambo-code.org/doc/installing.php > > i hope it works! > ciao > federico iori. > > > hi 2 all, > > > > I tried to install the yambo but not succeed. Please help. > > I download the yambo and configure it with > > ./configure --with-netdf=/home/dev/netcdf-4.0.1/ > > --with-iotk=/home/dev/espresso-4.0.2/iotk/src/libiotk.a > > --with-p2y=/home/dev/espresso-4.0.2/ > > > > after that i got, > > [VER] 3.2.1 r.448 > > # > > # [SYS] linux at i686 > > # [SRC] /home/dev/yambo-3.2.1-r.448 > > # [BIN] /home/dev/yambo-3.2.1-r.448/bin > > # [FFT] Goedecker Fast Fourier transform with 0 cache > > > > [FFT] Goedecker Fast Fourier transform with 0 cache > > # > > # [ ] Double precision > > # [X] Redundant compilation > > # [ ] MPI > > # [ ] PW (3.2) support > > # [ ] ETSF I/O support > > # [ ] SCALAPACK > > # [ ] NETCDF/Large Files > > # [XX ] Built-in BLAS/LAPACK/LOCAL > > # > > # [ CPP ] gcc -E -P > > # [ C ] gcc -g -O2 -D_C_US -D_FORTRAN_US > > # [MPICC] > > # [ F90 ] f95 -O > > # [MPIF ] > > # [ F77 ] f95 -O > > # [Cmain] > > # [NoOpt] > > # > > # [ MAKE ] make > > # [EDITOR] vim > > > > then i did make yambo interfaces > > after which i got a2y yambo and ypp in the bin but not p2y [image: :(] > > [image: > > :(]. > > Please guide. > > thanks > > Dev sharma, > > University of Delhi, > > India > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > Dr. Federico Iori, PhD! > > Laboratoire des Solides Irradi?s, > Ecole Polytechnique - CEA/DSM, CNRS > 91128 Palaiseau, France > > www.etsf.polytechnique.fr > > phone: +33 (0) 1 69 33 44 85 > emails: federico.iori @ polytechnique.edu > iori.federico @ unimore.it > http://www.pdmodena.it/ > http://www.comune.mo.it > > > > Dr. Federico Iori, PhD! > > Laboratoire des Solides Irradi?s, > Ecole Polytechnique - CEA/DSM, CNRS > 91128 Palaiseau, France > > www.etsf.polytechnique.fr > > phone: +33 (0) 1 69 33 44 85 > emails: federico.iori @ polytechnique.edu > iori.federico @ unimore.it > http://www.pdmodena.it/ > http://www.comune.mo.it > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091018/bb59a918/attachment.htm From iori at theory.polytechnique.fr Sun Oct 18 13:51:49 2009 From: iori at theory.polytechnique.fr (iori at theory.polytechnique.fr) Date: Sun, 18 Oct 2009 13:51:49 +0200 (CEST) Subject: [Pw_forum] yambo installation In-Reply-To: References: <156d7bf1c4864f6baeb9bf669f2f6bee.squirrel@theory.polytechnique.fr> Message-ID: <856ecb2bc813d62850b00455fd6ccd5d.squirrel@theory.polytechnique.fr> at a first immediate look: --with-netdf=/home/dev/netcdf-4.0.1/ WRONG! --with-netcdf=/home/dev/netcdf-4.0.1/ RIGHT netdf is not netcdf !! please look at the compiling option of the confugure tool by ./configure -h to see how are the correct flags. and if you still find some problem, i think the best choice is to write to YAMBO FORUM since here is not the right mailing list. good luck! ciau, ff > SIr, > I tried with > ./configure --with-netdf=/home/dev/netcdf-4.0.1/ > --with-iotk=/home/dev/espresso-4.0.2/iotk/src/libiotk.a --with-p2y=4.0 > > but the problem is same :( > Please guide. > Thanks, > Dev > University of Delhi > India > > On Sun, Oct 18, 2009 at 2:43 PM, wrote: > >> Hi. >> I think you should specify indeed: >> >> --with-p2y=4.0 >> >> but have a look also to the yambo website (documentation) >> >> >> http://www.yambo-code.org/forum/viewforum.php?f=1&sid=f6a47f967aced5269868973dd1c8b23d >> >> >> and to the foruminparticular where you can find whatever you need: >> >> http://www.yambo-code.org/doc/installing.php >> >> i hope it works! >> ciao >> federico iori. >> >> > hi 2 all, >> > >> > I tried to install the yambo but not succeed. Please help. >> > I download the yambo and configure it with >> > ./configure --with-netdf=/home/dev/netcdf-4.0.1/ >> > --with-iotk=/home/dev/espresso-4.0.2/iotk/src/libiotk.a >> > --with-p2y=/home/dev/espresso-4.0.2/ >> > >> > after that i got, >> > [VER] 3.2.1 r.448 >> > # >> > # [SYS] linux at i686 >> > # [SRC] /home/dev/yambo-3.2.1-r.448 >> > # [BIN] /home/dev/yambo-3.2.1-r.448/bin >> > # [FFT] Goedecker Fast Fourier transform with 0 cache >> > >> > [FFT] Goedecker Fast Fourier transform with 0 cache >> > # >> > # [ ] Double precision >> > # [X] Redundant compilation >> > # [ ] MPI >> > # [ ] PW (3.2) support >> > # [ ] ETSF I/O support >> > # [ ] SCALAPACK >> > # [ ] NETCDF/Large Files >> > # [XX ] Built-in BLAS/LAPACK/LOCAL >> > # >> > # [ CPP ] gcc -E -P >> > # [ C ] gcc -g -O2 -D_C_US -D_FORTRAN_US >> > # [MPICC] >> > # [ F90 ] f95 -O >> > # [MPIF ] >> > # [ F77 ] f95 -O >> > # [Cmain] >> > # [NoOpt] >> > # >> > # [ MAKE ] make >> > # [EDITOR] vim >> > >> > then i did make yambo interfaces >> > after which i got a2y yambo and ypp in the bin but not p2y [image: :(] >> > [image: >> > :(]. >> > Please guide. >> > thanks >> > Dev sharma, >> > University of Delhi, >> > India >> > _______________________________________________ >> > Pw_forum mailing list >> > Pw_forum at pwscf.org >> > http://www.democritos.it/mailman/listinfo/pw_forum >> > >> >> >> Dr. Federico Iori, PhD! >> >> Laboratoire des Solides Irradi?s, >> Ecole Polytechnique - CEA/DSM, CNRS >> 91128 Palaiseau, France >> >> www.etsf.polytechnique.fr >> >> phone: +33 (0) 1 69 33 44 85 >> emails: federico.iori @ polytechnique.edu >> iori.federico @ unimore.it >> http://www.pdmodena.it/ >> http://www.comune.mo.it >> >> >> >> Dr. Federico Iori, PhD! >> >> Laboratoire des Solides Irradi?s, >> Ecole Polytechnique - CEA/DSM, CNRS >> 91128 Palaiseau, France >> >> www.etsf.polytechnique.fr >> >> phone: +33 (0) 1 69 33 44 85 >> emails: federico.iori @ polytechnique.edu >> iori.federico @ unimore.it >> http://www.pdmodena.it/ >> http://www.comune.mo.it >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > Dr. Federico Iori, PhD! Laboratoire des Solides Irradi?s, Ecole Polytechnique - CEA/DSM, CNRS 91128 Palaiseau, France www.etsf.polytechnique.fr phone: +33 (0) 1 69 33 44 85 emails: federico.iori @ polytechnique.edu iori.federico @ unimore.it http://www.pdmodena.it/ http://www.comune.mo.it From siyouber at yahoo.fr Sun Oct 18 17:47:44 2009 From: siyouber at yahoo.fr (Bertrand SITAMTZE) Date: Sun, 18 Oct 2009 15:47:44 +0000 (GMT) Subject: [Pw_forum] Band structure calculation In-Reply-To: <4AD1D006.2090709@sissa.it> Message-ID: <808806.98347.qm@web26503.mail.ukl.yahoo.com> Dear Stephano Thanks very much for this? IMPORTANT REMARK. Now I understand. The Fermi level is an electronic issue, not related to the ionic conduction. Thanks very much. ****************************************** ?Bertrand SITAMTZE YOUMBI ?Laboratory of Material sciences ?Department of physics University of Yaound? I-Cameroon PO Box 812 Yaound?-Cameroon ?******************************************** --- En date de?: Dim 11.10.09, Stefano de Gironcoli a ?crit?: De: Stefano de Gironcoli Objet: Re: [Pw_forum] Band structure calculation ?: "PWSCF Forum" Date: Dimanche 11 Octobre 2009, 14h31 An ionic conductor is a material where the charge carriers are ions NOT electrons... Molten NaCl will be able to sustain an electrical current but it will still be an insulator from the electronic point of view.. stefano Bertrand SITAMTZE wrote: > Dear all, > I calculated the bands structure of an ionic conductor with bands.x. > But when I plot it with plotband.x, the Fermi level is exactly at the > middle of the forbidden gap. > I think as a conductor, the Fermi level should be in the conduction band. > Has something gone wrong? > > Thanks for your help > > > > ****************************************** > Bertrand SITAMTZE YOUMBI > Laboratory of Material sciences > Department of physics > University of Yaound? I-Cameroon > PO Box 812 Yaound?-Cameroon > ******************************************** > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum >??? _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091018/5fe275a1/attachment-0001.htm From jdai3 at mail.ustc.edu.cn Sun Oct 18 18:04:17 2009 From: jdai3 at mail.ustc.edu.cn (Jun Dai) Date: Mon, 19 Oct 2009 00:04:17 +0800 (CST) Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <808806.98347.qm@web26503.mail.ukl.yahoo.com> References: <808806.98347.qm@web26503.mail.ukl.yahoo.com> Message-ID: <28060397.42421255881857407.JavaMail.coremail@mailweb> Dear Q-E users; I have a system has 38 oxygen atoms whoes occupation is 3/4, so i want to genrate a new O pp which is 3/4 of itself, can someone give some suggestions on how to do it, thanks in advance. Jun DAI ============================================ Ph.D. Candidate, Hefei National Laboratory For Physical Sciences at the Microscale, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 E-mail: jdai3 at mail.ustc.edu.cn ============================================ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091019/56f55c49/attachment.htm From Andrea.Marini at roma2.infn.it Sun Oct 18 21:56:17 2009 From: Andrea.Marini at roma2.infn.it (Andrea Marini) Date: Sun, 18 Oct 2009 21:56:17 +0200 (CEST) Subject: [Pw_forum] yambo installation In-Reply-To: <856ecb2bc813d62850b00455fd6ccd5d.squirrel@theory.polytechnique.fr> References: <156d7bf1c4864f6baeb9bf669f2f6bee.squirrel@theory.polytechnique.fr> <856ecb2bc813d62850b00455fd6ccd5d.squirrel@theory.polytechnique.fr> Message-ID: On Sun, 18 Oct 2009, iori at theory.polytechnique.fr wrote: > and if you still find some problem, i think the best choice is to write to > YAMBO FORUM since here is not the right mailing list. Dear Federico, many thanks ! We are indeed continuing the discussion with dev sharma on the Yambo forum. Cheers Andrea -------------------------------------------------------------------------- Andrea MARINI Physics Department, University of Rome "Tor Vergata" (Italy) - phone: +39-0672594894 - fax: +39-062023507 - - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - From baroni at sissa.it Sun Oct 18 22:43:05 2009 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 18 Oct 2009 22:43:05 +0200 Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <28060397.42421255881857407.JavaMail.coremail@mailweb> References: <808806.98347.qm@web26503.mail.ukl.yahoo.com> <28060397.42421255881857407.JavaMail.coremail@mailweb> Message-ID: I do not think this is a good idea. SB On Oct 18, 2009, at 6:04 PM, Jun Dai wrote: > Dear Q-E users; > > I have a system has 38 oxygen atoms whoes occupation is 3/4, so i > want to genrate a new O pp which is 3/4 of itself, > > can someone give some suggestions on how to do it, thanks in advance. > > > Jun DAI > > ============================================ > > Ph.D. Candidate, > Hefei National Laboratory For Physical Sciences at the Microscale, > University of Science and Technology of China, > Hefei, Anhui, 230026, > People's Republic of China > Tel.: 86-551-3606428 > Fax.: 86-551-3602969 > E-mail: jdai3 at mail.ustc.edu.cn > ============================================ > > _______________________________________________ > 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/20091018/61cfb9ef/attachment.htm From eyvaz_isaev at yahoo.com Mon Oct 19 00:11:02 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 18 Oct 2009 15:11:02 -0700 (PDT) Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <28060397.42421255881857407.JavaMail.coremail@mailweb> Message-ID: <215582.92018.qm@web65715.mail.ac4.yahoo.com> Dear Jun Dai, --- On Sun, 10/18/09, Jun Dai wrote: > I have a system has 38 oxygen atoms whoes occupation is 3/4, so i want > to genrate a new O pp which is 3/4 of itself, > can someone give some suggestions on how to do it, I suggest, the problem has no solution in ab initio calculations. What you mean generating of a new PsP for O (3/4 of O) means to generate a PsP ... for Carbon. Using virtual.x you can mix PsPs for elements with similar properties or neighbouring elements in the Periodic Table. But the problem you post is different: an O position is occupied by the probability of 0.75. I think the problem can not be solved also in the framework of alloy's theory where a site can be occupied in accordance with atomic concentration (but the site occupancy is 1!), say in a substitutional alloy A(1-x)B(x)C, where atoms A and B occupy the same site with weight 1-x and x, respectively. Hope this helps. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From jdai3 at mail.ustc.edu.cn Mon Oct 19 04:57:12 2009 From: jdai3 at mail.ustc.edu.cn (Jun Dai) Date: Mon, 19 Oct 2009 10:57:12 +0800 (CST) Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <215582.92018.qm@web65715.mail.ac4.yahoo.com> References: <215582.92018.qm@web65715.mail.ac4.yahoo.com> Message-ID: <12059541.64231255921032052.JavaMail.coremail@mailweb> Prof. Baroni and Prof Isaev, Thanks for your replies, my original idea is something like using virtual.x to mix a O pp with a ghost pp to generate a new O pp which has 3/4 of O to stand for of the 3/4 occupation. Like those in alloys AxB1-xs, if I take a ghost pp as A, and the original O pp as B. Is it possible or reasonable? Bests Jun DAI ============================================ Ph.D. Candidate, Hefei National Laboratory For Physical Sciences at the Microscale, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 E-mail: jdai3 at mail.ustc.edu.cn ============================================ > -----Original E-mail----- > From: "Eyvaz Isaev" > Sent Time: 2009-10-19 6:11:02 > To: "PWSCF Forum" > Cc: > Subject: Re: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x > > Dear Jun Dai, > > --- On Sun, 10/18/09, Jun Dai wrote: > > > I have a system has 38 oxygen atoms whoes occupation is 3/4, so i want > > to genrate a new O pp which is 3/4 of itself, > > can someone give some suggestions on how to do it, > > I suggest, the problem has no solution in ab initio calculations. What you mean generating of a new PsP for O (3/4 of O) means to generate a PsP ... for Carbon. Using virtual.x you can mix PsPs for elements with similar properties or neighbouring elements in the Periodic Table. > > But the problem you post is different: an O position is occupied by the probability of 0.75. I think the problem can not be solved also in the framework of alloy's theory where a site can be occupied in accordance with atomic concentration (but the site occupancy is 1!), say in a substitutional alloy A(1-x)B(x)C, where atoms A and B occupy the same site with weight 1-x and x, respectively. > > Hope this helps. > > Bests, > Eyvaz. > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, > Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at democritos.it Mon Oct 19 08:35:14 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 19 Oct 2009 08:35:14 +0200 Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <12059541.64231255921032052.JavaMail.coremail@mailweb> References: <215582.92018.qm@web65715.mail.ac4.yahoo.com> <12059541.64231255921032052.JavaMail.coremail@mailweb> Message-ID: <81E9B102-9E7A-442D-A72B-91FA5152CCD5@democritos.it> On Oct 19, 2009, at 4:57 , Jun Dai wrote: > Is it possible or reasonable? it is possible, but hardly reasonable. Virtual crystal approximation works for cases like A(x) B(1-x) substitutional alloys, when A and B are not too different. In your case A=Oxygen and B=nothing. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From sclauzer at sissa.it Mon Oct 19 08:53:12 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 19 Oct 2009 08:53:12 +0200 Subject: [Pw_forum] c60 In-Reply-To: References: Message-ID: <4ADC0CD8.4080206@sissa.it> lan haiping wrote: > Hi, > The total energy doesnot make sense. But total energy differences do! So that you can try to compare cohesive energies, defect energies, ... > > And Dmol code adopts all-electron calculation for Carbon, while PWSCF is > a pseudopotential scheme. > So the total electrons in two calculations are not the same . That's right. Anyway with the new PAW implementation you should be able to reconstruct the all-electron total energy (in the frozen core approximation) from the pseudo-wavefunctions . I think this has already been implemented, you simply need to use a PAW pseudopotential in place of the ultrasoft or normconserving one that you've been using. For instance in espresso-4.1.1/pseudo there is one: C.lda-paw_kj.UPF. Regards, GS > > Regards, > > On Sun, Oct 18, 2009 at 2:15 AM, Chen, Zhifan > wrote: > > Dear all > has anyone calculated the C60 total energy? I used DMol (all > electron) and PWscf (pseudopotential) to calculated the C60 total > energy. I found DMol result is five times biger than PWscf result. > Do I suppose to get the almost same data from the calculations of > all electron and pseudopotential representation? > > > Zhifan Chen > Clark Atlanta University > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > Hai-Ping Lan > Department of Electronics , > Peking University , Bejing, 100871 > lanhaiping at gmail.com , hplan at pku.edu.cn > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From baroni at sissa.it Mon Oct 19 09:00:33 2009 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 19 Oct 2009 09:00:33 +0200 Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <81E9B102-9E7A-442D-A72B-91FA5152CCD5@democritos.it> References: <215582.92018.qm@web65715.mail.ac4.yahoo.com> <12059541.64231255921032052.JavaMail.coremail@mailweb> <81E9B102-9E7A-442D-A72B-91FA5152CCD5@democritos.it> Message-ID: <52474134-B412-4277-ADE0-1F20F2525E60@sissa.it> may I add the recommendation that one should try to understand methods and approximations (if, when, and how they work) _BEFORE_ using them? cheers - SB On Oct 19, 2009, at 8:35 AM, Paolo Giannozzi wrote: > > On Oct 19, 2009, at 4:57 , Jun Dai wrote: > >> Is it possible or reasonable? > > it is possible, but hardly reasonable. Virtual crystal > approximation works for cases like A(x) B(1-x) > substitutional alloys, when A and B are not too > different. In your case A=Oxygen and B=nothing. > > Paolo > --- > 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 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/20091019/d74a78f8/attachment.htm From jdai3 at mail.ustc.edu.cn Mon Oct 19 09:08:09 2009 From: jdai3 at mail.ustc.edu.cn (Jun Dai) Date: Mon, 19 Oct 2009 15:08:09 +0800 (CST) Subject: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x In-Reply-To: <52474134-B412-4277-ADE0-1F20F2525E60@sissa.it> References: <52474134-B412-4277-ADE0-1F20F2525E60@sissa.it> <215582.92018.qm@web65715.mail.ac4.yahoo.com> <12059541.64231255921032052.JavaMail.coremail@mailweb> <81E9B102-9E7A-442D-A72B-91FA5152CCD5@democritos.it> Message-ID: <6074747.83171255936089557.JavaMail.coremail@mailweb> Dear Prof. Baroni & Giannozzi thanks, your replies help a lot. Bests DAI -----Original E-mail----- From: "Stefano Baroni" Sent Time: 2009-10-19 15:00:33 To: "PWSCF Forum" Cc: Subject: Re: [Pw_forum] How to generate a pp which is 3/4 of itself using virtual.x may I add the recommendation that one should try to understand methods and approximations (if, when, and how they work) _BEFORE_ using them? cheers - SB On Oct 19, 2009, at 8:35 AM, Paolo Giannozzi wrote: On Oct 19, 2009, at 4:57 , Jun Dai wrote: Is it possible or reasonable? it is possible, but hardly reasonable. Virtual crystal approximation works for cases like A(x) B(1-x) substitutional alloys, when A and B are not too different. In your case A=Oxygen and B=nothing. Paolo --- 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 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/20091019/901d70e8/attachment.htm From kazempoor2000 at yahoo.com Mon Oct 19 11:09:23 2009 From: kazempoor2000 at yahoo.com (ali kazempour) Date: Mon, 19 Oct 2009 02:09:23 -0700 (PDT) Subject: [Pw_forum] GWW-problem Message-ID: <925582.53126.qm@web112511.mail.gq1.yahoo.com> &inputpw4gww prefix='tio2' outdir='/p5/batch/kazempou/' lwannier=.true., cutoff_wpr_vc = 0.1d0 num_nbnd_first =50 num_nbndv=24 num_nbnds=34 l_truncated_coulomb=.false. remainder=-1 restart_gww=-1 numw_prod=1 n_gauss=79 omega_gauss=20.d0 tau_gauss=10.d0 l_zero=.true. l_wing=.true. cprim_first=1 cprim_last=34 cutoff_products=0.01d0 l_polarization_analysis=.true. cutoff_polarization=0.01d0 / &inputgww ggwin%n=79, ggwin%n_fit=120, ggwin%tau=10., ggwin%max_i=34, ggwin%prefix='tio2' ggwin%num_rows=200 ggwin%starting_point=1 ggwin%ending_point=7 ggwin%lnonorthogonal=.false. ggwin%lconduction=.true. ggwin%grid_time=2 ggwin%grid_freq=2 ggwin%omega=20 ggwin%omega_fit=20 ggwin%n_grid_fit=240 ggwin%i_min=1 ggwin%i_max=34 ggwin%l_head_epsilon=.true. ggwin%w_divergence=2 ggwin%l_wing_epsilon=.true. / &inputpp outdir='/p5/batch/kazempou/' prefix='tio2' lgww=.true. Emin=-30.0, Emax=20.0, DeltaE=0.05, ngauss=0, degauss=0.015 / thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091019/0b30b487/attachment-0001.htm From paulatto at sissa.it Mon Oct 19 11:14:49 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 19 Oct 2009 11:14:49 +0200 (CEST) Subject: [Pw_forum] GWW-problem In-Reply-To: <925582.53126.qm@web112511.mail.gq1.yahoo.com> References: <925582.53126.qm@web112511.mail.gq1.yahoo.com> Message-ID: <39447.78.12.172.175.1255943689.squirrel@webmail.sissa.it> On Mon, October 19, 2009 11:09, ali kazempour wrote: > ... > thanks a lot Which kind of problem do you have? regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From kazempoor2000 at yahoo.com Mon Oct 19 11:24:01 2009 From: kazempoor2000 at yahoo.com (ali kazempour) Date: Mon, 19 Oct 2009 02:24:01 -0700 (PDT) Subject: [Pw_forum] elctron-phonon Message-ID: <347042.83929.qm@web112513.mail.gq1.yahoo.com> Hi all I am studying the defect induced level in Ionic semiconductor like TiO2 and ZnO. In this material the coupling between electron and phonon is very strong specially when a defect forms in these materials. But by LDA and GGA normally we predict wrong defect level for them. I am in initial steps and Don't know whether the electron-phonon calculation help to predict and obtain correct defect level or not? I saw exampl07 is about electron-phonon coupling that at last prepare the lambda Vs. broadening , But I Don't know how to connect them to defect level? could anyone help me how can I use el-ph calculation to improve the levels? thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091019/b249a58f/attachment.htm From kazempoor2000 at yahoo.com Mon Oct 19 11:26:35 2009 From: kazempoor2000 at yahoo.com (ali kazempour) Date: Mon, 19 Oct 2009 02:26:35 -0700 (PDT) Subject: [Pw_forum] GWW-problem In-Reply-To: <39447.78.12.172.175.1255943689.squirrel@webmail.sissa.it> References: <925582.53126.qm@web112511.mail.gq1.yahoo.com> <39447.78.12.172.175.1255943689.squirrel@webmail.sissa.it> Message-ID: <312269.79335.qm@web112520.mail.gq1.yahoo.com> Dear Paolo I determine the range that include Valence and conduction but after doing GW calculation I dont see the conduction band in Dos. thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 ________________________________ From: Lorenzo Paulatto To: PWSCF Forum Sent: Mon, October 19, 2009 11:14:49 AM Subject: Re: [Pw_forum] GWW-problem On Mon, October 19, 2009 11:09, ali kazempour wrote: > ... > thanks a lot Which kind of problem do you have? regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091019/e02dbc11/attachment.htm From sxu2 at ncsu.edu Mon Oct 19 21:30:08 2009 From: sxu2 at ncsu.edu (shu xu) Date: Mon, 19 Oct 2009 15:30:08 -0400 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> Message-ID: <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> Dear users: When I submitted for the second time and restarted the ph calculation, it failed due to the following reason. How can I solve this problem? Electric Fields Calculation iter # 8 total cpu time : 539.9 secs av.it.: 2.6 thresh= 0.423E-05 alpha_mix = 0.700 |ddv_scf|^2 = 0.313E-08 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from ph_writefile : error # 37 cannot open xml_recover file for writing %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... my input of ph is as follows. When I submited it for the 1st time, I didn't add recover=.true. When time was up , I set recover = .true and submited again, then crashed. &inputph tr2_ph=1.0d-14, prefix='SC', amass(1)=1.00794, amass(2)=12.0107, amass(3)=28.0855, outdir='./', recover =.true., epsil=.true., lraman=.true., trans=.true., asr=.true. fildyn='sc.dyn' / 0.0 0.0 0.0 Shu Xu Dept. of Physics NCSU USA -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091019/acb531b4/attachment.htm From lfhuang at theory.issp.ac.cn Tue Oct 20 10:06:26 2009 From: lfhuang at theory.issp.ac.cn (=?utf-8?B?bGZodWFuZw==?=) Date: Tue, 20 Oct 2009 16:06:26 +0800 Subject: [Pw_forum] =?utf-8?q?elctron-phonon?= Message-ID: <20091020080626.26537.qmail@ms.hfcas.ac.cn> Dear ali kazempour: Now you have got the band structure and el-ph coupling constant from your DFT calculations, MAYBE self-energy of the electrons due to el-ph coupling can lead the band structure closer to realistics. But these things are so complicated for me to give you detailed descriptions derectly, and you can refer to some books on solid state theory for related solutions(e.g. semi-classical approximation). Best Wishes! Yours Sincerely L.F.Huang > Date: Mon, 19 Oct 2009 02:24:01 -0700 (PDT) > From: ali kazempour > Subject: [Pw_forum] elctron-phonon > To: pw > Message-ID: > Content-Type: text/plain; charset="us-ascii" > > Hi all > I am studying the defect induced level in Ionic semiconductor like TiO2 and ZnO. In this material the coupling between electron and phonon is very strong specially when a defect forms in these materials. > But by LDA and GGA normally we predict wrong defect level for them. I am in initial steps and Don't know whether the electron-phonon calculation help to predict and obtain correct defect level or not? I saw exampl07 is about electron-phonon coupling that at last prepare the lambda Vs. broadening , But I Don't know how to connect them to defect level? could anyone help me how can I use el-ph calculation to improve the levels? > thanks a lot > > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 ------ ====================================================================== L.F.Huang(???) ph.D candidate ====================================================================== Add: Research Laboratory for Computational Materials Sciences, Instutue of Solid State Physics,the Chinese Academy of Sciences, P.O.Box 1129, Hefei 230031, P.R.China Tel: 86-551-5591464-328(office) Fax: 86-551-5591434 Web: http://theory.issp.ac.cn (website of our theory group) http://www.issp.ac.cn (website of our institute) ====================================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091020/0d4b8901/attachment.htm From paulatto at sissa.it Tue Oct 20 10:31:42 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 20 Oct 2009 10:31:42 +0200 Subject: [Pw_forum] lapack and BLAS In-Reply-To: References: Message-ID: In data 17 ottobre 2009 alle ore 20:15:40, dev sharma ha scritto: > can anybody help in installing LAPACK and BLIS libraries. I downloaded > the lapack-3.2.1 and BLAS libraries. I want to use EXE code for which > BLASis required. I dont know, how to install or use these libraries. > Please help > and guide me. when i tried to make BLAS and LAPACK i got Dear Dev, BLAS and LAPACK are already included in the Quantum-ESPRESO distribution, you don't need to install them separately. Nevertheless it is better to use platform-specific high-performance libraries such as mkl (intel), acml (amd), essl (ibm power), sun performance libraries (sun sparc). Or, at least, the automatically-tuned open-source ATLAS libraries, which are a bit difficult to compile, but can often perform as good as commercial ones. In general, if you have problems with a specific library you should contact the developers of the library itself, which are definitely more experienced than us on the matter. Finally, this kind of libraries are normally well known and widely used, hence a simple google search will give you hundreds of help pages, addressing you very issue. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From giannozz at democritos.it Tue Oct 20 10:32:57 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 20 Oct 2009 10:32:57 +0200 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> Message-ID: <4ADD75B9.6070102@democritos.it> shu xu wrote: > from ph_writefile : error # 37 > cannot open xml_recover file for writing it is hard to say what happened, since restarting from interrupted calculations may go wrong in many ways and for many reasons (in addition to possible bugs). The message itself means that either the file wasn't there, or it wasn't writable. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From paulatto at sissa.it Tue Oct 20 10:40:01 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 20 Oct 2009 10:40:01 +0200 Subject: [Pw_forum] c60 In-Reply-To: <4ADC0CD8.4080206@sissa.it> References: <4ADC0CD8.4080206@sissa.it> Message-ID: In data 19 ottobre 2009 alle ore 08:53:12, Gabriele Sclauzero ha scritto: > That's right. Anyway with the new PAW implementation you should be able > to reconstruct the all-electron total energy (in the frozen core > approximation) from the pseudo-wavefunctions . I think this has already > been implemented, you simply need to use a PAW pseudopotential in place > of the ultrasoft or normconserving one that you've been using. > For instance in espresso-4.1.1/pseudo there is one: C.lda-paw_kj.UPF. You indeed can, i.e. if you only use PAW datasets in a calculation the code will spit out the "total all-electron energy", aside with the total valence energy. Nevertheless, I would like to stress that total energy has no meaning for isolated systems, i.e. the zero of energy is arbitrary and only energy differences matter. What you should do is compute the binding energy: E_bind = E_c60 - 60 x E_c where E_c comes from a calculation of an isolated C atom with the same cutoffs as the C60. This not only give a meaningful binding energy, but can also reduce possible truncation errors thanks to error cancellation. Computing the ground state of a single C atom is a bit tricky, but is a matter that have been discussed countless times in this mailing list: best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From lfhuang at theory.issp.ac.cn Tue Oct 20 10:26:58 2009 From: lfhuang at theory.issp.ac.cn (=?utf-8?B?bGZodWFuZw==?=) Date: Tue, 20 Oct 2009 16:26:58 +0800 Subject: [Pw_forum] =?utf-8?q?elctron-phonon?= Message-ID: <20091020082658.31315.qmail@ms.hfcas.ac.cn> Dear ali kazempour: Now you have got the band structure and el-ph coupling constant from your DFT calculations, MAYBE self-energy of the electrons due to el-ph coupling can lead the band structure closer to realistics. But these things are so complicated for me to give you detailed descriptions derectly, and you can refer to some books on solid state theory for related solutions(e.g. semi-classical approximation). Best Wishes! Yours Sincerely L.F.Huang > Date: Mon, 19 Oct 2009 02:24:01 -0700 (PDT) > From: ali kazempour > Subject: [Pw_forum] elctron-phonon > To: pw > Message-ID: > Content-Type: text/plain; charset="us-ascii" > > Hi all > I am studying the defect induced level in Ionic semiconductor like TiO2 and ZnO. In this material the coupling between electron and phonon is very strong specially when a defect forms in these materials. > But by LDA and GGA normally we predict wrong defect level for them. I am in initial steps and Don't know whether the electron-phonon calculation help to predict and obtain correct defect level or not? I saw exampl07 is about electron-phonon coupling that at last prepare the lambda Vs. broadening , But I Don't know how to connect them to defect level? could anyone help me how can I use el-ph calculation to improve the levels? > thanks a lot > > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 ------ ====================================================================== L.F.Huang(???) ph.D candidate ====================================================================== Add: Research Laboratory for Computational Materials Sciences, Instutue of Solid State Physics,the Chinese Academy of Sciences, P.O.Box 1129, Hefei 230031, P.R.China Tel: 86-551-5591464-328(office) Fax: 86-551-5591434 Web: http://theory.issp.ac.cn (website of our theory group) http://www.issp.ac.cn (website of our institute) ====================================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091020/0e445abc/attachment.htm From lanhaiping at gmail.com Tue Oct 20 10:42:55 2009 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 20 Oct 2009 16:42:55 +0800 Subject: [Pw_forum] c60 In-Reply-To: <4ADC0CD8.4080206@sissa.it> References: <4ADC0CD8.4080206@sissa.it> Message-ID: Thanks, Gabriele. However, Dmol3 also is a LCAO package. regards, H.P On Mon, Oct 19, 2009 at 2:53 PM, Gabriele Sclauzero wrote: > > > lan haiping wrote: > > Hi, > > The total energy doesnot make sense. > > But total energy differences do! So that you can try to compare cohesive > energies, defect > energies, ... > > > > > And Dmol code adopts all-electron calculation for Carbon, while PWSCF is > > a pseudopotential scheme. > > So the total electrons in two calculations are not the same . > > That's right. Anyway with the new PAW implementation you should be able to > reconstruct the > all-electron total energy (in the frozen core approximation) from the > pseudo-wavefunctions > . I think this has already been implemented, you simply need to use a PAW > pseudopotential > in place of the ultrasoft or normconserving one that you've been using. > For instance in espresso-4.1.1/pseudo there is one: C.lda-paw_kj.UPF. > > Regards, > > GS > > > > > Regards, > > > > On Sun, Oct 18, 2009 at 2:15 AM, Chen, Zhifan > > wrote: > > > > Dear all > > has anyone calculated the C60 total energy? I used DMol (all > > electron) and PWscf (pseudopotential) to calculated the C60 total > > energy. I found DMol result is five times biger than PWscf result. > > Do I suppose to get the almost same data from the calculations of > > all electron and pseudopotential representation? > > > > > > Zhifan Chen > > Clark Atlanta University > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > -- > > Hai-Ping Lan > > Department of Electronics , > > Peking University , Bejing, 100871 > > lanhaiping at gmail.com , hplan at pku.edu.cn > > > > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091020/69550b9a/attachment-0001.htm From lfhuang at theory.issp.ac.cn Tue Oct 20 10:34:55 2009 From: lfhuang at theory.issp.ac.cn (=?utf-8?B?bGZodWFuZw==?=) Date: Tue, 20 Oct 2009 16:34:55 +0800 Subject: [Pw_forum] =?utf-8?q?I_am_sorry_for_sending_two_same_mails_in_the?= =?utf-8?q?_previous_volume!?= Message-ID: <20091020083455.878.qmail@ms.hfcas.ac.cn> Dear folks: I am sorry for sending two same mails in the previous volume! Because there must be a problem with my email system. Best Wishes! Yours Sincerely L.F.Huang ------ ====================================================================== L.F.Huang(???) ph.D candidate ====================================================================== Add: Research Laboratory for Computational Materials Sciences, Instutue of Solid State Physics,the Chinese Academy of Sciences, P.O.Box 1129, Hefei 230031, P.R.China Tel: 86-551-5591464-328(office) Fax: 86-551-5591434 Web: http://theory.issp.ac.cn (website of our theory group) http://www.issp.ac.cn (website of our institute) ====================================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091020/be06c343/attachment.htm From jameslipd at gmail.com Tue Oct 20 16:02:04 2009 From: jameslipd at gmail.com (JAY) Date: Tue, 20 Oct 2009 10:02:04 -0400 Subject: [Pw_forum] Problem on compiling pwscf Message-ID: Dear All I tried to compile pwscf-4.1.1 on my FC10. mpi-1.2.7, fc=pgf90, cc=pgcc, lapack-3.2.1(blas), fftw-3.2.2, When i type in make all, the problem is: if test -d iotk ; then \ ( cd iotk ; if test "make" = "" ; then make TLDEPS= lib+util ; \ else make TLDEPS= lib+util ; fi ) ; fi make[1]: Entering directory `/home/jy/espresso/espresso-4.1.1/iotk' cd src ; make lib+util make[2]: Entering directory `/home/jy/espresso/espresso-4.1.1/iotk/src' mpif90 -O2 -fast -r8 -Mfree -Mx,119,0x200000 -D__GFORTRAN -D__FFTW -I../include -I./ -I../Modules -I../iotk/src -I../PW -I../PH -I../EE -I../GIPAW -c iotk_base.f90 PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 20) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 30) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 56) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 74) PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: 72) PGF90-S-0087-Non-constant expression where constant expression required (iotk_base.spp: 73) PGF90-F-0008-Error limit exceeded (iotk_base.spp: 73) PGF90/x86-64 Linux 7.0-5: compilation aborted make[2]: *** [iotk_base.o] Error 2 make[2]: Leaving directory `/home/jy/espresso/espresso-4.1.1/iotk/src' make[1]: *** [lib+util] Error 2 make[1]: Leaving directory `/home/jy/espresso/espresso-4.1.1/iotk' make: *** [libiotk] Error 2 and my make.sys is: # make.sys. Generated from make.sys.in by configure. # compilation rules .SUFFIXES : .SUFFIXES : .o .c .f .f90 # most fortran compilers can directly preprocess c-like directives: use # $(MPIF90) $(F90FLAGS) -c $< # if explicit preprocessing by the C preprocessor is needed, use: # $(CPP) $(CPPFLAGS) $< -o $*.F90 # $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o # remember the tabulator in the first column !!! .f90.o: $(MPIF90) $(F90FLAGS) -c $< # .f.o and .c.o: do not modify .f.o: $(F77) $(FFLAGS) -c $< .c.o: $(CC) $(CFLAGS) -c $< # DFLAGS = precompilation options (possible arguments to -D and -U) # used by the C compiler and preprocessor # FDFLAGS = as DFLAGS, for the f90 compiler # See include/defs.h.README for a list of options and their meaning # With the exception of IBM xlf, FDFLAGS = $(DFLAGS) # For IBM xlf, FDFLAGS is the same as DFLAGS with separating commas DFLAGS = -D__GFORTRAN -D__FFTW FDFLAGS = $(DFLAGS) # IFLAGS = how to locate directories where files to be included are # In most cases, IFLAGS = -I../include IFLAGS = -I../include # MODFLAGS = flag used by f90 compiler to locate modules # You need to search for modules in ./, in ../iotk/src, in ../Modules # Some applications also need modules in ../PW and ../PH MODFLAGS = -I./ -I../Modules -I../iotk/src \ -I../PW -I../PH -I../EE -I../GIPAW # Compilers: fortran-90, fortran-77, C # If a parallel compilation is desired, MPIF90 should be a fortran-90 # compiler that produces executables for parallel execution using MPI # (such as for instance mpif90, mpf90, mpxlf90,...); # otherwise, an ordinary fortran-90 compiler (f90, g95, xlf90, ifort,...) # If you have a parallel machine but no suitable candidate for MPIF90, # try to specify the directory containing "mpif.h" in IFLAGS # and to specify the location of MPI libraries in MPI_LIBS MPIF90 = mpif90 #F90 = gfortran CC = pgcc F77 = pgf77 # C preprocessor and preprocessing flags - for explicit preprocessing, # if needed (see the compilation rules above) # preprocessing flags must include DFLAGS and IFLAGS CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) # compiler flags: C, F90, F77 # C flags must include DFLAGS and IFLAGS # F90 flags must include MODFLAGS, IFLAGS, and FDFLAGS with appropriate syntax CFLAGS = -O3 $(DFLAGS) $(IFLAGS) #F90FLAGS = $(FFLAGS) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) F90FLAGS = $(FFLAGS) -Mfree -Mx,119,0x200000 $(FDFLAGS) $(IFLAGS) $(MODFLAGS) #F90FLAGS = $(FFLAGS) -x f95-cpp-input $(FDFLAGS) $(IFLAGS) $(MODFLAGS) #FFLAGS = -O3 FFLAGS = -O2 -fast -r8 # compiler flags without optimization for fortran-77 # the latter is NEEDED to properly compile dlamch.f, used by lapack FFLAGS_NOOPT = -O0 -Mx,119,0x200000 #FFLAGS_NOOPT = -O0 # Linker, linker-specific flags (if any) # Typically LD coincides with F90 or MPIF90, LD_LIBS is empty LD = mpif90 LDFLAGS = LD_LIBS = # External Libraries (if any) : blas, lapack, fft, MPI # If you have nothing better, use the local copy : ../flib/blas.a #BLAS_LIBS =/home/jy/espresso/lapack-3.2.1 BLAS_LIBS = /home/jy/espresso/libs/libfftw3.a # The following lapack libraries will be available in flib/ : # ../flib/lapack.a : contains all needed routines # ../flib/lapack_atlas.a: only routines not present in the Atlas library # For IBM machines with essl (-D__ESSL): load essl BEFORE lapack ! # remember that LAPACK_LIBS precedes BLAS_LIBS in loading order LAPACK_LIBS = /home/jy/espresso/lapack-3.2.1/lapack_LINUX.a # nothing needed here if the the internal copy of FFTW is compiled # (needs -D__FFTW in DFLAGS) #FFT_LIBS =/home/jy/espresso/fftw-3.2.2/lib FFT_LIBS = /home/jy/espresso/libs/libfftw3.a # For parallel execution, the correct path to MPI libraries must # be specified in MPI_LIBS (except for IBM if you use mpxlf) MPI_LIBS =/opt/mpich-1.2.5.2/lib -lmpich #MPI_LIBS = /opt/mpich1.2.5.2/lib/libpmpich.a -lmpich IBM-specific: MASS libraries, if available and if -D__MASS is defined in FDFLAGS MASS_LIBS = # pgplot libraries (used by some post-processing tools) PGPLOT_LIBS = # ar command and flags - for most architectures: AR = ar, ARFLAGS = ruv # ARFLAGS_DYNAMIC is used in iotk to produce a dynamical library, # for Mac OS-X with PowerPC and xlf compiler. In all other cases # ARFLAGS_DYNAMIC = $(ARFLAGS) AR = ar ARFLAGS = ruv ARFLAGS_DYNAMIC= ruv # ranlib command. If ranlib is not needed (it isn't in most cases) use # RANLIB = echo RANLIB = ranlib # all internal and external libraries - do not modify LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a ../iotk/src/libiotk.a ../Multigrid/mglib.a LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS) $(MASS_LIBS) $(PGPLOT_LIBS) $(LD_LIBS) Could anybody help me? Thank you in advance. By the way, how to choose 'F90FLAGS'? -- Sincerely Jiaye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091020/9c8d009a/attachment.htm From giannozz at democritos.it Tue Oct 20 16:08:35 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 20 Oct 2009 16:08:35 +0200 Subject: [Pw_forum] Problem on compiling pwscf In-Reply-To: References: Message-ID: <4ADDC463.1030200@democritos.it> JAY wrote: > I tried to compile pwscf-4.1.1 on my FC10. mpi-1.2.7, fc=pgf90 and why then your make.sys is for gfortran? P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From Matthias.Timmer at uni-due.de Tue Oct 20 16:09:32 2009 From: Matthias.Timmer at uni-due.de (Matthias Timmer) Date: Tue, 20 Oct 2009 16:09:32 +0200 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: References: Message-ID: <4ADDC49C.7020909@uni-due.de> Dear readers, thank you for your replies and great help. Just to answer and comment on a few things: >> 2) When the configure-script runs, it always tells me, that the >> "x86_64"-architecture of the Xeons we have is "amd64" (see line 1443 in the >> configure-script). I think this is highly irritating. >> > > On Debian systems, x86_64 is often called amd64: > http://alioth.debian.org/docman/view.php/30192/21/debian-amd64-howto.html#id250846 > > Not sure if you have a Debian (or Ubuntu) system or not, but that > might help explain what you are seeing. > According to my last information, we have scientific linux. > > BLAS_LIBS = -lmkl_blas95_lp64 -lmkl_intel_lp64 > -lmkl_intel_sequential -lmkl_core > > LAPACK_LIBS = -lfftw3xf_intel -lmkl_lapack95_lp64 -lmkl_intel_lp64 > -lmkl_intel_sequential -lmkl_core Since the linker does not recognize -lmkl_intel_sequential I needed to change this towards -lmkl_sequential. I also added -lpthread for BLAS_LIBS, following http://software.intel.com/en-us/articles/intel-mkl-link-line-advisor/ pointed out by Baris. For the moment it works perfectly, and I get a very considerable reduction in calculation time (seems to be >50% after first tests on our Xeons, just because of the MKL library instead of PWSCF internal - can that be true?). Compared to our Opteron cluster (ifort 9.1 / mkl 9.0), the Xeons (ifort 11.1 / mkl 10.2.1.017) save > 50% calculation time. Many thanks to all of you who tried to help! Kind regards, Matthias Timmer, University of Duisburg-Essen -------------- next part -------------- A non-text attachment was scrubbed... Name: Matthias_Timmer.vcf Type: text/x-vcard Size: 318 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20091020/87cdae12/attachment-0001.vcf From paulatto at sissa.it Tue Oct 20 16:13:26 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 20 Oct 2009 16:13:26 +0200 Subject: [Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64? In-Reply-To: <4ADDC49C.7020909@uni-due.de> References: <4ADDC49C.7020909@uni-due.de> Message-ID: In data 20 ottobre 2009 alle ore 16:09:32, Matthias Timmer ha scritto: > For the moment it works perfectly, and I get a very considerable > reduction in calculation time (seems to be >50% after first tests on our > Xeons, just because of the MKL library instead of PWSCF internal - can > that be true?). Yes, especially for systems with many electrons, according to the performance improvement they boast here: http://software.intel.com/sites/products/collateral/hpc/mkl/mkl_indepth.pdf > Compared to our Opteron cluster (ifort 9.1 / mkl 9.0), the Xeons (ifort > 11.1 / mkl 10.2.1.017) save > 50% calculation time. Yes, for the same reason as above. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From marsamos at democritos.it Tue Oct 20 16:47:54 2009 From: marsamos at democritos.it (marsamos at democritos.it) Date: Tue, 20 Oct 2009 16:47:54 +0200 Subject: [Pw_forum] Problem on compiling pwscf In-Reply-To: References: Message-ID: <20091020164754.cevyntr2m8kc4sg0@mail.democritos.it> Dear Jay, sometimes this time of error is related to the prepocessor. Maybe try to uncomment the line related to prepocessor from make.sys (and comment the older one). # most fortran compilers can directly preprocess c-like directives: use # $(MPIF90) $(F90FLAGS) -c $< # if explicit preprocessing by the C preprocessor is needed, use: # $(CPP) $(CPPFLAGS) $< -o $*.F90 # $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o # remember the tabulator in the first column !!! bests Layla Quoting JAY : > Dear All > > I tried to compile pwscf-4.1.1 on my FC10. mpi-1.2.7, fc=pgf90, cc=pgcc, > lapack-3.2.1(blas), fftw-3.2.2, > > When i type in make all, the problem is: > if test -d iotk ; then \ > ( cd iotk ; if test "make" = "" ; then make TLDEPS= lib+util ; \ > else make TLDEPS= lib+util ; fi ) ; fi > make[1]: Entering directory `/home/jy/espresso/espresso-4.1.1/iotk' > cd src ; make lib+util > make[2]: Entering directory `/home/jy/espresso/espresso-4.1.1/iotk/src' > mpif90 -O2 -fast -r8 -Mfree -Mx,119,0x200000 -D__GFORTRAN -D__FFTW > -I../include -I./ -I../Modules -I../iotk/src -I../PW -I../PH -I../EE > -I../GIPAW -c iotk_base.f90 > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 20) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 30) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 56) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 74) > PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: > 72) > PGF90-S-0087-Non-constant expression where constant expression required > (iotk_base.spp: 73) > PGF90-F-0008-Error limit exceeded (iotk_base.spp: 73) > PGF90/x86-64 Linux 7.0-5: compilation aborted > make[2]: *** [iotk_base.o] Error 2 > make[2]: Leaving directory `/home/jy/espresso/espresso-4.1.1/iotk/src' > make[1]: *** [lib+util] Error 2 > make[1]: Leaving directory `/home/jy/espresso/espresso-4.1.1/iotk' > make: *** [libiotk] Error 2 > > > and my make.sys is: > # make.sys. Generated from make.sys.in by configure. > > # compilation rules > > .SUFFIXES : > .SUFFIXES : .o .c .f .f90 > > # most fortran compilers can directly preprocess c-like directives: use > # $(MPIF90) $(F90FLAGS) -c $< > # if explicit preprocessing by the C preprocessor is needed, use: > # $(CPP) $(CPPFLAGS) $< -o $*.F90 > # $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o > # remember the tabulator in the first column !!! > > .f90.o: > $(MPIF90) $(F90FLAGS) -c $< > > # .f.o and .c.o: do not modify > > .f.o: > $(F77) $(FFLAGS) -c $< > > .c.o: > $(CC) $(CFLAGS) -c $< > > > # DFLAGS = precompilation options (possible arguments to -D and -U) > # used by the C compiler and preprocessor > # FDFLAGS = as DFLAGS, for the f90 compiler > # See include/defs.h.README for a list of options and their meaning > # With the exception of IBM xlf, FDFLAGS = $(DFLAGS) > # For IBM xlf, FDFLAGS is the same as DFLAGS with separating commas > > DFLAGS = -D__GFORTRAN -D__FFTW > FDFLAGS = $(DFLAGS) > > # IFLAGS = how to locate directories where files to be included are > # In most cases, IFLAGS = -I../include > > IFLAGS = -I../include > > # MODFLAGS = flag used by f90 compiler to locate modules > # You need to search for modules in ./, in ../iotk/src, in ../Modules > # Some applications also need modules in ../PW and ../PH > > MODFLAGS = -I./ -I../Modules -I../iotk/src \ > -I../PW -I../PH -I../EE -I../GIPAW > > # Compilers: fortran-90, fortran-77, C > # If a parallel compilation is desired, MPIF90 should be a fortran-90 > # compiler that produces executables for parallel execution using MPI > # (such as for instance mpif90, mpf90, mpxlf90,...); > # otherwise, an ordinary fortran-90 compiler (f90, g95, xlf90, ifort,...) > # If you have a parallel machine but no suitable candidate for MPIF90, > # try to specify the directory containing "mpif.h" in IFLAGS > # and to specify the location of MPI libraries in MPI_LIBS > > MPIF90 = mpif90 > #F90 = gfortran > CC = pgcc > F77 = pgf77 > > # C preprocessor and preprocessing flags - for explicit preprocessing, > # if needed (see the compilation rules above) > # preprocessing flags must include DFLAGS and IFLAGS > > CPP = cpp > CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) > > # compiler flags: C, F90, F77 > # C flags must include DFLAGS and IFLAGS > # F90 flags must include MODFLAGS, IFLAGS, and FDFLAGS with appropriate > syntax > > CFLAGS = -O3 $(DFLAGS) $(IFLAGS) > #F90FLAGS = $(FFLAGS) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) > F90FLAGS = $(FFLAGS) -Mfree -Mx,119,0x200000 $(FDFLAGS) $(IFLAGS) > $(MODFLAGS) > #F90FLAGS = $(FFLAGS) -x f95-cpp-input $(FDFLAGS) $(IFLAGS) > $(MODFLAGS) > #FFLAGS = -O3 > FFLAGS = -O2 -fast -r8 > > # compiler flags without optimization for fortran-77 > # the latter is NEEDED to properly compile dlamch.f, used by lapack > > FFLAGS_NOOPT = -O0 -Mx,119,0x200000 > #FFLAGS_NOOPT = -O0 > > # Linker, linker-specific flags (if any) > # Typically LD coincides with F90 or MPIF90, LD_LIBS is empty > > LD = mpif90 > LDFLAGS = > LD_LIBS = > > # External Libraries (if any) : blas, lapack, fft, MPI > > # If you have nothing better, use the local copy : ../flib/blas.a > > #BLAS_LIBS =/home/jy/espresso/lapack-3.2.1 > BLAS_LIBS = /home/jy/espresso/libs/libfftw3.a > > # The following lapack libraries will be available in flib/ : > # ../flib/lapack.a : contains all needed routines > # ../flib/lapack_atlas.a: only routines not present in the Atlas library > # For IBM machines with essl (-D__ESSL): load essl BEFORE lapack ! > # remember that LAPACK_LIBS precedes BLAS_LIBS in loading order > > LAPACK_LIBS = /home/jy/espresso/lapack-3.2.1/lapack_LINUX.a > > # nothing needed here if the the internal copy of FFTW is compiled > # (needs -D__FFTW in DFLAGS) > > #FFT_LIBS =/home/jy/espresso/fftw-3.2.2/lib > FFT_LIBS = /home/jy/espresso/libs/libfftw3.a > > # For parallel execution, the correct path to MPI libraries must > # be specified in MPI_LIBS (except for IBM if you use mpxlf) > > MPI_LIBS =/opt/mpich-1.2.5.2/lib -lmpich > #MPI_LIBS = /opt/mpich1.2.5.2/lib/libpmpich.a -lmpich > > IBM-specific: MASS libraries, if available and if -D__MASS is defined in > FDFLAGS > > MASS_LIBS = > > # pgplot libraries (used by some post-processing tools) > > PGPLOT_LIBS = > > # ar command and flags - for most architectures: AR = ar, ARFLAGS = ruv > # ARFLAGS_DYNAMIC is used in iotk to produce a dynamical library, > # for Mac OS-X with PowerPC and xlf compiler. In all other cases > # ARFLAGS_DYNAMIC = $(ARFLAGS) > > AR = ar > ARFLAGS = ruv > ARFLAGS_DYNAMIC= ruv > > # ranlib command. If ranlib is not needed (it isn't in most cases) use > # RANLIB = echo > > RANLIB = ranlib > > # all internal and external libraries - do not modify > > LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a > ../iotk/src/libiotk.a ../Multigrid/mglib.a > LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS) > $(MASS_LIBS) $(PGPLOT_LIBS) $(LD_LIBS) > > > Could anybody help me? Thank you in advance. > > By the way, how to choose 'F90FLAGS'? > > > > -- > Sincerely > > Jiaye > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From Peter.Pallister at nrc-cnrc.gc.ca Tue Oct 20 18:29:49 2009 From: Peter.Pallister at nrc-cnrc.gc.ca (Pallister, Peter) Date: Tue, 20 Oct 2009 12:29:49 -0400 Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: Message-ID: Hello again, After using your script Dr. Seitsonen to generate the GIPAW part of the .UPF file it seems that there is always a strange discontinuity in the projections above 11 bohr. However, examining another GIPAW pseudopotential on your website, http://www-int.impmc.upmc.fr/~software/gipaw/, these discontinuities are normal or expected? If that is the case then the pseudopotential generated from the following input into the ld1.x module should work? &input title = 'Mg' prefix = 'Mg' zed = 12.0 rel = 0 nld = 3 rlderiv = 5.0 eminld = -6.0 emaxld = -1.0 deld = 0.01 config = '1s2 2s2 2p6 3s2 3p-1 3d-1 4s-1' iswitch = 3 dft = 'PBE' / &inputp pseudotype = 1 lloc = 2 file_pseudopw = 'Mg.pbe-gipaw-oct09-02.UPF' file_recon = 'Mg.pbe-gipaw-oct09-02.recon' lgipaw_reconstruction = .true. tm = .true. / 2 3S 1 0 2.00 0.00 1.800 1.800 3D 3 2 0.00 -0.30 1.800 1.800 &test / 2 3S 1 0 2.00 0.00 1.800 1.800 4S 2 0 0.00 -0.40 1.800 1.800 As you can see the reference energy for the unbound states is certainly set as non-zero. However, the pseudopotential generated from this input, when used in a simple lattice, MgO, will produce an error in the scf calculation: - two of the Mg atoms overlap I am sure the crystal parameters used are accurate so it must be a problem with the pseudopotential? I greatly appreciate any help or assistance anyone might have to offer! Peter Pallister -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Ari P Seitsonen Sent: October 13, 2009 11:31 AM To: PWSCF Forum Subject: Re: [Pw_forum] GIPAW pseudopotentials Dear Peter et al, Sorry for the long time it took me to answer!! :( The *.wfc files are generated by default, as far as I know. There is a special issue concerning the wave functions for the GIPAW reconstruction: They are _not_ generated in the normal procedure how the ld1.x/atomic code works, but for GIPAW they are calculated _after_ the *.wfc files have been written: the standard part of the code calculates them once, with some value (possibly zero) as the reference energy; for GIPAW one then recalculates them, but with the reference energies taken from the '&test' section. This is why the wave functions for the GIPAW reconstruction can be different from the ones in the *.wfc files. And this is also why one _has to_ define a reference energy different from zero, at least for unbound states (ie. whose eigenvalue would be positive) in the section '&test'. So, in practise it doesn't matter if the wave functions in the *.wfc corresponding to empty orbitals do not make sense, but you can check that they look ok in the GIPAW section at the end of the UPF file. I attach an unofficial script for writing out the GIPAW data in the UPF file (it is a very small script file). You can probably modify the output to show the properties you want to have. If you still have problems with the wave functions coming out from this script please check that you use a non-zero reference energy for unbound states, otherwise please contact me again. Greetings from Zurich, apsi -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Physikalisch-Chemisches Institut der Universit?t Z?rich Tel: +41 44 63 55 44 97 / Mobile: +41 79 71 90 935 On Tue, 13 Oct 2009, Pallister, Peter wrote: > Hi, > Actually, one of the tutorials on the QE site has a step-by-step > pseudopotential generation tutorial showing how to generate the AE > calculation, examine the logarithmic derivatives, etc. This tutorial > suggests looking at the file (well, in my case) 'Mg-oct09ps.wfc' to look > at the pseudo-generated orbitals. Is this incorrect? What do the > individual file/s mean in the "file_wfcncgen=" input? Thanks! > > Peter Pallister > > -----Original Message----- > From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On > Behalf Of Lorenzo Paulatto > Sent: October 9, 2009 4:20 AM > To: PWSCF Forum > Subject: Re: [Pw_forum] GIPAW pseudopotentials > > > On Wed, October 7, 2009 17:10, Pallister, Peter wrote: >> Hi all, >> >> I am attempting to generate a gipaw pseudopotentials, in this case > for >> Mg, but I am having tremendous difficulty. The wavefunctions produced >> with the input below (or minute variations thereof) continuously show >> discontinuities, specifically at a radius of 15bohrs. > > Dear Peter, > the wavefunctions are generated on a logarithmic radial grid, around 15 > bohr the points are very coarse, btw how are you examining the > wavfunctions? (there is not file_wfcncgen='..' in your input). > > best regards > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From Ari.P.Seitsonen at iki.fi Wed Oct 21 09:31:01 2009 From: Ari.P.Seitsonen at iki.fi (Ari P Seitsonen) Date: Wed, 21 Oct 2009 09:31:01 +0200 (CEST) Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: <4ADD75B9.6070102@democritos.it> References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> Message-ID: Dear Shu Xu, Paolo et al, I have had exactly the same problem in my own system, with the message %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from ph_writefile : error # 25 cannot open xml_recover file for writing %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% when using v4.1 or v4.1.1, and a core dump when using the CVS version of Monday. The machine I'm running on is an IBM Power6, with xlf compiler (I tried both versions 12.1 and 11.1 of xlf), and if I read the correct location in the reference manual of XLF Fortran, the error code 25 would correspond to 25 RECL= specifier omitted on an OPEN statement for a direct file. But I don't see where RECL's would be used for the XML files... Maybe some one can help (us) more. The job worked fine if it was given more time so that it could finish without a need for 'recover'. Greetings, apsi -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Physikalisch-Chemisches Institut der Universitaet Zuerich Tel: +41 44 63 54 497 / Mobile: +41 79 71 90 935 On Tue, 20 Oct 2009, Paolo Giannozzi wrote: > shu xu wrote: > >> from ph_writefile : error # 37 >> cannot open xml_recover file for writing > > it is hard to say what happened, since restarting from > interrupted calculations may go wrong in many ways and for > many reasons (in addition to possible bugs). The message > itself means that either the file wasn't there, or it > wasn't writable. > > P. > -- > Paolo Giannozzi, Democritos and University of Udine, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From paulatto at sissa.it Wed Oct 21 09:48:06 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 21 Oct 2009 09:48:06 +0200 Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: References: Message-ID: In data 20 ottobre 2009 alle ore 18:29:49, Pallister, Peter ha scritto: > http://www-int.impmc.upmc.fr/~software/gipaw/, these discontinuities are > normal or expected? They are totally normal. Atomic wavefunction computed at an energy different from the eigenvalue are unbound, unnormalizable states. The point at about 15 bohr where you see the discontinuity is just the end of the radial grid. Taking the grid further away introduces huge rounding errors when computing integrals. > - two of the Mg atoms overlap > I am sure the crystal parameters used are accurate so it must be a > problem with the pseudopotential? No, it is indeed a problem with your atomic positions! If you are using a recent enough version of QE (and you should) it does not only tell you that two atoms overlaps, but also which ones, e.g. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from check_atoms : error # 1 atoms # 1 and # 2 overlap! %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% For the rest, your ld1 input looks ok to me, I think you may have to add more projectors to have meaningful GIPAW data, but I'm no expert in that specific field. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From jhouska at kfy.zcu.cz Wed Oct 21 10:17:34 2009 From: jhouska at kfy.zcu.cz (Jiri Houska) Date: Wed, 21 Oct 2009 10:17:34 +0200 Subject: [Pw_forum] lelfield Message-ID: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> Dear users, (1) I successfully performed a relax calculation (gamma point only) on a 64-atom amorphous system. (2) After introducing an electric field: lelfield=.true., gdir=1, nppstr=1, nberrycyc=1, ... efield=0.0001, (same directory, from scratch), I experienced the famous problem " from davcio : error # 10 error while reading from file". Yes, I now that it has been posted many times but for my life I do not see a solution in the answers: the problem is reproducible, I have permissions and disc quota, the only thing which changed are the 5 lines above. When trying " disk_io='low' ", the error changed from davcio to " from h_epsi_her : error #256 error in zgefa " Thanks for an advice, Jiri Houska University of West Bohemia, Czech Republic From paulatto at sissa.it Wed Oct 21 10:21:33 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 21 Oct 2009 10:21:33 +0200 Subject: [Pw_forum] lelfield In-Reply-To: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> References: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> Message-ID: In data 21 ottobre 2009 alle ore 10:17:34, Jiri Houska ha scritto: > I have > permissions and disc quota, the only thing which changed are the 5 > lines above. Dear Jiri, the information you provide is clearly not enough to give you any advice... but I can ask you a couple of questions: 1) after adding the electric field via Berry-phase did you restarted the calculation from scratch? 2) which hardware are you using? 3) which version of QE are you using? 4) are you using any kind of smearing for occupations? cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From giannozz at democritos.it Wed Oct 21 10:23:34 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 21 Oct 2009 10:23:34 +0200 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> Message-ID: <4ADEC506.2030807@democritos.it> Ari P Seitsonen wrote: > Maybe some one can help (us) more. unlikely. As I mentioned yesterday (and many more times), problems in restart are difficult to pinpoint, unless you have access to the specific case and the specific machine where this happens. As a starting point, one should try to answer these two basic questions: - is the problem reproducible on another machine ? - is the problem reproducible with different data (if possible, leading to a quick execution) ? Then one can try to answer some less basic questions: - where in the code does the problem arise? - under which exact circumstances does the problem arise? Then it becomes maybe not easy, but at least possible, to find a solution. By the way: there is a bug-tracking facility (very buggy, by the way) on qe-forge. It can be used to file bug reports. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From jhouska at kfy.zcu.cz Wed Oct 21 10:41:33 2009 From: jhouska at kfy.zcu.cz (Jiri Houska) Date: Wed, 21 Oct 2009 10:41:33 +0200 Subject: [Pw_forum] lelfield In-Reply-To: References: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> Message-ID: <20091021104133.rlhoy810sgw40wk4@webmail.zcu.cz> Dear Lorenzo, thanks for the answer. The relaxation is from scratch (not restarted), QE 4.0.3, running on cluster consisting of quadcore Xeon E5472 (3 GHz, 12 MB cache), happened for both (1) fixed occupations [used originally for material with a band gap] and (2) cold smearing. Thanks for now, Jiri Quoting Lorenzo Paulatto : > In data 21 ottobre 2009 alle ore 10:17:34, Jiri Houska > ha scritto: >> I have >> permissions and disc quota, the only thing which changed are the 5 >> lines above. > > Dear Jiri, > the information you provide is clearly not enough to give you any > advice... but I can ask you a couple of questions: > 1) after adding the electric field via Berry-phase did you restarted the > calculation from scratch? > 2) which hardware are you using? > 3) which version of QE are you using? > 4) are you using any kind of smearing for occupations? > > cheers > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > From giannozz at democritos.it Wed Oct 21 10:45:27 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 21 Oct 2009 10:45:27 +0200 Subject: [Pw_forum] lelfield In-Reply-To: <20091021104133.rlhoy810sgw40wk4@webmail.zcu.cz> References: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> <20091021104133.rlhoy810sgw40wk4@webmail.zcu.cz> Message-ID: <4ADECA27.3020404@democritos.it> Did you read examples/example31/README ? P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From jameslipd at gmail.com Wed Oct 21 13:45:30 2009 From: jameslipd at gmail.com (JAY) Date: Wed, 21 Oct 2009 19:45:30 +0800 Subject: [Pw_forum] Problem on compiling pwscf In-Reply-To: <20091020164754.cevyntr2m8kc4sg0@mail.democritos.it> References: <20091020164754.cevyntr2m8kc4sg0@mail.democritos.it> Message-ID: Dear Layla Thank you for the reply. I tried to understand the makefile of pwscf, but due to the lack of my knowledge, I am not able to find what should do. I typed ./configure in the source directory, it reports that the Fortran 77 compiler cannot create executables. I will check what's wrong with the Fortran compiler. On Tue, Oct 20, 2009 at 10:47 PM, wrote: > Dear Jay, sometimes this time of error is related to the prepocessor. Maybe > try to uncomment the line related to prepocessor from make.sys (and comment > the older one). > > # most fortran compilers can directly preprocess c-like directives: use > # $(MPIF90) $(F90FLAGS) -c $< > # if explicit preprocessing by the C preprocessor is needed, use: > # $(CPP) $(CPPFLAGS) $< -o $*.F90 > # $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o > # remember the tabulator in the first column !!! > > bests > > Layla > > > Quoting JAY : > > Dear All >> >> I tried to compile pwscf-4.1.1 on my FC10. mpi-1.2.7, fc=pgf90, cc=pgcc, >> lapack-3.2.1(blas), fftw-3.2.2, >> >> When i type in make all, the problem is: >> if test -d iotk ; then \ >> ( cd iotk ; if test "make" = "" ; then make TLDEPS= lib+util ; \ >> else make TLDEPS= lib+util ; fi ) ; fi >> make[1]: Entering directory `/home/jy/espresso/espresso-4.1.1/iotk' >> cd src ; make lib+util >> make[2]: Entering directory `/home/jy/espresso/espresso-4.1.1/iotk/src' >> mpif90 -O2 -fast -r8 -Mfree -Mx,119,0x200000 -D__GFORTRAN -D__FFTW >> -I../include -I./ -I../Modules -I../iotk/src -I../PW -I../PH -I../EE >> -I../GIPAW -c iotk_base.f90 >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 20) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 30) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 56) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 74) >> PGF90-S-0021-Label field of continuation line is not blank (iotk_base.spp: >> 72) >> PGF90-S-0087-Non-constant expression where constant expression required >> (iotk_base.spp: 73) >> PGF90-F-0008-Error limit exceeded (iotk_base.spp: 73) >> PGF90/x86-64 Linux 7.0-5: compilation aborted >> make[2]: *** [iotk_base.o] Error 2 >> make[2]: Leaving directory `/home/jy/espresso/espresso-4.1.1/iotk/src' >> make[1]: *** [lib+util] Error 2 >> make[1]: Leaving directory `/home/jy/espresso/espresso-4.1.1/iotk' >> make: *** [libiotk] Error 2 >> >> >> and my make.sys is: >> # make.sys. Generated from make.sys.in by configure. >> >> # compilation rules >> >> .SUFFIXES : >> .SUFFIXES : .o .c .f .f90 >> >> # most fortran compilers can directly preprocess c-like directives: use >> # $(MPIF90) $(F90FLAGS) -c $< >> # if explicit preprocessing by the C preprocessor is needed, use: >> # $(CPP) $(CPPFLAGS) $< -o $*.F90 >> # $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o >> # remember the tabulator in the first column !!! >> >> .f90.o: >> $(MPIF90) $(F90FLAGS) -c $< >> >> # .f.o and .c.o: do not modify >> >> .f.o: >> $(F77) $(FFLAGS) -c $< >> >> .c.o: >> $(CC) $(CFLAGS) -c $< >> >> >> # DFLAGS = precompilation options (possible arguments to -D and -U) >> # used by the C compiler and preprocessor >> # FDFLAGS = as DFLAGS, for the f90 compiler >> # See include/defs.h.README for a list of options and their meaning >> # With the exception of IBM xlf, FDFLAGS = $(DFLAGS) >> # For IBM xlf, FDFLAGS is the same as DFLAGS with separating commas >> >> DFLAGS = -D__GFORTRAN -D__FFTW >> FDFLAGS = $(DFLAGS) >> >> # IFLAGS = how to locate directories where files to be included are >> # In most cases, IFLAGS = -I../include >> >> IFLAGS = -I../include >> >> # MODFLAGS = flag used by f90 compiler to locate modules >> # You need to search for modules in ./, in ../iotk/src, in ../Modules >> # Some applications also need modules in ../PW and ../PH >> >> MODFLAGS = -I./ -I../Modules -I../iotk/src \ >> -I../PW -I../PH -I../EE -I../GIPAW >> >> # Compilers: fortran-90, fortran-77, C >> # If a parallel compilation is desired, MPIF90 should be a fortran-90 >> # compiler that produces executables for parallel execution using MPI >> # (such as for instance mpif90, mpf90, mpxlf90,...); >> # otherwise, an ordinary fortran-90 compiler (f90, g95, xlf90, ifort,...) >> # If you have a parallel machine but no suitable candidate for MPIF90, >> # try to specify the directory containing "mpif.h" in IFLAGS >> # and to specify the location of MPI libraries in MPI_LIBS >> >> MPIF90 = mpif90 >> #F90 = gfortran >> CC = pgcc >> F77 = pgf77 >> >> # C preprocessor and preprocessing flags - for explicit preprocessing, >> # if needed (see the compilation rules above) >> # preprocessing flags must include DFLAGS and IFLAGS >> >> CPP = cpp >> CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) >> >> # compiler flags: C, F90, F77 >> # C flags must include DFLAGS and IFLAGS >> # F90 flags must include MODFLAGS, IFLAGS, and FDFLAGS with appropriate >> syntax >> >> CFLAGS = -O3 $(DFLAGS) $(IFLAGS) >> #F90FLAGS = $(FFLAGS) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) >> F90FLAGS = $(FFLAGS) -Mfree -Mx,119,0x200000 $(FDFLAGS) $(IFLAGS) >> $(MODFLAGS) >> #F90FLAGS = $(FFLAGS) -x f95-cpp-input $(FDFLAGS) $(IFLAGS) >> $(MODFLAGS) >> #FFLAGS = -O3 >> FFLAGS = -O2 -fast -r8 >> >> # compiler flags without optimization for fortran-77 >> # the latter is NEEDED to properly compile dlamch.f, used by lapack >> >> FFLAGS_NOOPT = -O0 -Mx,119,0x200000 >> #FFLAGS_NOOPT = -O0 >> >> # Linker, linker-specific flags (if any) >> # Typically LD coincides with F90 or MPIF90, LD_LIBS is empty >> >> LD = mpif90 >> LDFLAGS = >> LD_LIBS = >> >> # External Libraries (if any) : blas, lapack, fft, MPI >> >> # If you have nothing better, use the local copy : ../flib/blas.a >> >> #BLAS_LIBS =/home/jy/espresso/lapack-3.2.1 >> BLAS_LIBS = /home/jy/espresso/libs/libfftw3.a >> >> # The following lapack libraries will be available in flib/ : >> # ../flib/lapack.a : contains all needed routines >> # ../flib/lapack_atlas.a: only routines not present in the Atlas library >> # For IBM machines with essl (-D__ESSL): load essl BEFORE lapack ! >> # remember that LAPACK_LIBS precedes BLAS_LIBS in loading order >> >> LAPACK_LIBS = /home/jy/espresso/lapack-3.2.1/lapack_LINUX.a >> >> # nothing needed here if the the internal copy of FFTW is compiled >> # (needs -D__FFTW in DFLAGS) >> >> #FFT_LIBS =/home/jy/espresso/fftw-3.2.2/lib >> FFT_LIBS = /home/jy/espresso/libs/libfftw3.a >> >> # For parallel execution, the correct path to MPI libraries must >> # be specified in MPI_LIBS (except for IBM if you use mpxlf) >> >> MPI_LIBS =/opt/mpich-1.2.5.2/lib -lmpich >> #MPI_LIBS = /opt/mpich1.2.5.2/lib/libpmpich.a -lmpich >> >> IBM-specific: MASS libraries, if available and if -D__MASS is defined in >> FDFLAGS >> >> MASS_LIBS = >> >> # pgplot libraries (used by some post-processing tools) >> >> PGPLOT_LIBS = >> >> # ar command and flags - for most architectures: AR = ar, ARFLAGS = ruv >> # ARFLAGS_DYNAMIC is used in iotk to produce a dynamical library, >> # for Mac OS-X with PowerPC and xlf compiler. In all other cases >> # ARFLAGS_DYNAMIC = $(ARFLAGS) >> >> AR = ar >> ARFLAGS = ruv >> ARFLAGS_DYNAMIC= ruv >> >> # ranlib command. If ranlib is not needed (it isn't in most cases) use >> # RANLIB = echo >> >> RANLIB = ranlib >> >> # all internal and external libraries - do not modify >> >> LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a >> ../iotk/src/libiotk.a ../Multigrid/mglib.a >> LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS) >> $(MASS_LIBS) $(PGPLOT_LIBS) $(LD_LIBS) >> >> >> Could anybody help me? Thank you in advance. >> >> By the way, how to choose 'F90FLAGS'? >> >> >> >> -- >> Sincerely >> >> Jiaye >> >> > > > ---------------------------------------------------------------- > This message was sent using IMP, the Internet Messaging Program. > > -- Sincerely Jiaye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091021/169fa34a/attachment-0001.htm From paulatto at sissa.it Wed Oct 21 14:33:23 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 21 Oct 2009 14:33:23 +0200 Subject: [Pw_forum] Problem on compiling pwscf In-Reply-To: References: <20091020164754.cevyntr2m8kc4sg0@mail.democritos.it> Message-ID: In data 21 ottobre 2009 alle ore 13:45:30, JAY ha scritto: > Thank you for the reply. I tried to understand the makefile of pwscf, but > due to the lack of my knowledge, I am not able to find what should do. I > typed ./configure in the source directory, it reports that the Fortran 77 > compiler cannot create executables. I will check what's wrong with the > Fortran compiler. We are not able to understand what you should do either: your are sending contradicting emails. Let's try to recap what you've tried so far: 1. you run ./configure and it said it could not find a working f77 compiler, which means you could not create a make.sys whatsoever 2. you have a make.sys prepared for gfortran (GNU fortran compiler) in which you have simply replaced gfortran with mpif90, and added the location of mpich to MPI_LIBS. This make.sys will never work, as several additional switches are required to compile the parallel code. 3. where was the original make.sys coming from? 4. are you sure that mpif90 is actually belonging to the installation of mpich that is situated in /opt/mpich-1.2.5.2/? System administrators love to install many different versions of the same or equivalent libraries, often only one of them is actually working. 5. maybe the system you are using is module-based, you should really read the system documentation. Or, at the *very* least, type the command "module avail" and report the result... cheers P.S. please provide your full name+surname and affiliation, this is an informal mailing list, but we like to know who we're talking to. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From paulatto at sissa.it Wed Oct 21 14:39:16 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 21 Oct 2009 14:39:16 +0200 Subject: [Pw_forum] lelfield In-Reply-To: <20091021104133.rlhoy810sgw40wk4@webmail.zcu.cz> References: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> <20091021104133.rlhoy810sgw40wk4@webmail.zcu.cz> Message-ID: In data 21 ottobre 2009 alle ore 10:41:33, Jiri Houska ha scritto: > Dear Lorenzo, > thanks for the answer. The relaxation is from scratch (not restarted), > QE 4.0.3, running on cluster consisting of quadcore Xeon E5472 (3 GHz, > 12 MB cache), happened for both (1) fixed occupations [used originally > for material with a band gap] and (2) cold smearing. Dear Jiri, what happened in both cases? The i/o error or the algebra error? As far as I know, the Berry-phase method does not work with pools (for implementation reasons), also I'm not quite sure it works with smearing, and with metals in general (I think the theory only applies to insulators). The davcio error you have experienced may happen if you restart a calculation with a different number of pools. On the other hand, the zgefa error may be caused by dipfield+smearing or dipfield+pools. You could check if one of the aforementioned conditions is verified in your setup. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From paulatto at sissa.it Wed Oct 21 14:44:43 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 21 Oct 2009 14:44:43 +0200 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> Message-ID: In data 21 ottobre 2009 alle ore 09:31:01, Ari P Seitsonen ha scritto: > 25 RECL= specifier omitted on an OPEN statement for a direct file. > > But I don't see where RECL's would be used for the XML files... It would go in iotk/src/iotk_tool.spp. I'm having tons of i/o errors on sp6 too, I suspect it may have something to do with the specific filesystem used, but couldn't get reproducible fails so far. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From prasenjit.jnc at gmail.com Wed Oct 21 16:45:58 2009 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Wed, 21 Oct 2009 16:45:58 +0200 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> Message-ID: <627e0ffa0910210745x65b5eb5dq5b85a0236f8d0b14@mail.gmail.com> > > I'm having tons of i/o errors on sp6 too, I suspect it may have something > to do with the specific filesystem used, but couldn't get reproducible > fails so far. > > Lorenzo, the i/o errors on sp6 is most probably due to the problem in the $CINECA_SCRATCH area. In fact I'm also getting input output errors there and I think many other people are also getting it. Prasenjit -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091021/cef54fa9/attachment.htm From paulatto at sissa.it Wed Oct 21 16:53:53 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 21 Oct 2009 16:53:53 +0200 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: <627e0ffa0910210745x65b5eb5dq5b85a0236f8d0b14@mail.gmail.com> References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> <627e0ffa0910210745x65b5eb5dq5b85a0236f8d0b14@mail.gmail.com> Message-ID: In data 21 ottobre 2009 alle ore 16:45:58, Prasenjit Ghosh ha scritto: > the i/o errors on sp6 is most probably due to the problem in the > $CINECA_SCRATCH area. > In fact I'm also getting input output errors there and I think many other > people are also getting it. Yep, I suspected it. They started after some maintenance last month, so I'm hoping they will disappear sooner or later. We're really going off-topic now, but I wished to post this, to reassure other users who may be clueless and frustrated. Thank you for confirming. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From udayagiri3 at gmail.com Wed Oct 21 17:05:13 2009 From: udayagiri3 at gmail.com (udayagiri sai babu) Date: Wed, 21 Oct 2009 20:35:13 +0530 Subject: [Pw_forum] Cr GGA potential is underestimating the cohesive energy. In-Reply-To: <156151B9-0A65-465A-B6F6-B8ED03F3C9E0@democritos.it> References: <156151B9-0A65-465A-B6F6-B8ED03F3C9E0@democritos.it> Message-ID: Thank you dear Paolo for the reference you provided me. On Sat, Oct 17, 2009 at 4:07 PM, Paolo Giannozzi wrote: > > On Oct 17, 2009, at 11:31 , udayagiri sai babu wrote: > > > I am calculated the cohesive energy of Cr using the pseudopotential > > Cr.pbe-sp-van.UPF. I got is 3.63 ev/atom as the cohesive energy > > where as the experimental value is 4.1 ev/atom. As i know the GGA > > pseudopotential should overestimate the cohesive energy value > > should it? LDA overestimates, GGA not necessarily so. > > The cohesive energy of Cr has been already calculated by many authors, > for sure. You may want to have a look at what other people have done > before you. For instance, here: > http://dare.ubvu.vu.nl/bitstream/1871/9949/1/B4.pdf > (found in 5' of google search). They get 3.58eV for Becke-Perdew, > 3.80eV for Perdew-Wang. So 3.63eV for PBE after all is not that bad. > > P. > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- U.Saibabu PhD student, Deformation mechanisms modeling group, Materials engineering department, IISc Bangalore, India. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091021/7dbda4e4/attachment.htm From Ari.P.Seitsonen at iki.fi Wed Oct 21 17:30:52 2009 From: Ari.P.Seitsonen at iki.fi (Ari P Seitsonen) Date: Wed, 21 Oct 2009 17:30:52 +0200 (CEST) Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> <627e0ffa0910210745x65b5eb5dq5b85a0236f8d0b14@mail.gmail.com> Message-ID: Dear Lorenzo et al, I can confirm that I also manage to run exactly the same system on a Linux cluster (ifort 10.*, MKL) that caused the problem on the IBM Power6. Unfortunately the latter is the only machine where I could run the biiiiiig system... Well, I'll think of something. Thanks to all for the support!! Greetings, apsi -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Physikalisch-Chemisches Institut der Universitaet Zuerich Tel: +41 44 63 54 497 / Mobile: +41 79 71 90 935 On Wed, 21 Oct 2009, Lorenzo Paulatto wrote: > In data 21 ottobre 2009 alle ore 16:45:58, Prasenjit Ghosh > ha scritto: >> the i/o errors on sp6 is most probably due to the problem in the >> $CINECA_SCRATCH area. >> In fact I'm also getting input output errors there and I think many other >> people are also getting it. > > Yep, I suspected it. They started after some maintenance last month, so > I'm hoping they will disappear sooner or later. > > We're really going off-topic now, but I wished to post this, to reassure > other users who may be clueless and frustrated. Thank you for confirming. > > cheers > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From marsamos at democritos.it Wed Oct 21 17:57:55 2009 From: marsamos at democritos.it (marsamos at democritos.it) Date: Wed, 21 Oct 2009 17:57:55 +0200 Subject: [Pw_forum] "cannot open xml_recover file for writing" In-Reply-To: References: <36d476df0910191228l32a9f74dube8f94be4be5602d@mail.gmail.com> <36d476df0910191230x2840c5eeq701cbab56a2100b1@mail.gmail.com> <4ADD75B9.6070102@democritos.it> <627e0ffa0910210745x65b5eb5dq5b85a0236f8d0b14@mail.gmail.com> Message-ID: <20091021175755.x7sl43e3kgso8csw@mail.democritos.it> Dear all, I also find big i-o problems on sp6 (running GW) in particular when handling big files ... sometimes it works sometimes it does not work!!!!! HPC ;-) Layla Quoting Ari P Seitsonen : > > Dear Lorenzo et al, > > I can confirm that I also manage to run exactly the same system on a > Linux cluster (ifort 10.*, MKL) that caused the problem on the IBM Power6. > Unfortunately the latter is the only machine where I could run the > biiiiiig system... Well, I'll think of something. Thanks to all for the > support!! > > Greetings, > > apsi > > -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- > Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ > Physikalisch-Chemisches Institut der Universitaet Zuerich > Tel: +41 44 63 54 497 / Mobile: +41 79 71 90 935 > > On Wed, 21 Oct 2009, Lorenzo Paulatto wrote: > >> In data 21 ottobre 2009 alle ore 16:45:58, Prasenjit Ghosh >> ha scritto: >>> the i/o errors on sp6 is most probably due to the problem in the >>> $CINECA_SCRATCH area. >>> In fact I'm also getting input output errors there and I think many other >>> people are also getting it. >> >> Yep, I suspected it. They started after some maintenance last month, so >> I'm hoping they will disappear sooner or later. >> >> We're really going off-topic now, but I wished to post this, to reassure >> other users who may be clueless and frustrated. Thank you for confirming. >> >> cheers >> >> >> -- >> Lorenzo Paulatto >> SISSA & DEMOCRITOS (Trieste) >> phone: +39 040 3787 511 >> skype: paulatz >> www: http://people.sissa.it/~paulatto/ >> >> *** save italian brains *** >> http://saveitalianbrains.wordpress.com/ >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From manoj at phys.ufl.edu Wed Oct 21 20:13:31 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Wed, 21 Oct 2009 14:13:31 -0400 (EDT) Subject: [Pw_forum] Left and right going Bloch's states from PWCOND Message-ID: Dear All, I am trying to figure out the left and right going Bloch's states in the lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only prints out Bloch's state moving in one direction. eg. in one of the calculation- k//=(0.375,-0.375) Nchannels of the left tip = 1 k1(2pi/a) k2(2pi/a) E-Ef (eV) 0.3157801 0.0000000 0.0000000 Now if I want Bloch's state moving in right as well as left direction, I can go to kbloch.f90 subroutine, and print out all the eigen values of AX=exp(ikd)BX, and out of those the ones with real solution would be our Bloch's state, so I get for each channel two solutions- kval (-0.275409421993275,1.823688001395235E-010) kval (0.315780119742506,-3.611201785292708E-012) To figure out the direction, I can calculate current associated with these Bloch's sate and if the current is +ive it is right moving , and if '-'ive its left moving Bloch's state. I can print out current from jbloch.f90 subroutine which are - current eigenvalue -1.86502143831863 1.59149029314457 So, clearly the first state with kval=-0.2754094 is left moving and the other one right moving. Upto here its clear to me how to identify left and right moving states. I get confused when for a given (kx,ky,E), I have more than one Bloch' state. In another calculation where i get multiple Bloch's state- Nchannels of the left tip = 5 k1(2pi/a) k2(2pi/a) E-Ef (eV) -0.0746301 0.0000000 0.0000000 0.1205527 0.0000000 0.0000000 0.3112908 0.0000000 0.0000000 0.4200218 0.0000000 0.0000000 -0.4935150 0.0000000 0.0000000 so i did the same trick i did above to first print out kz and then current, which gives me - kval (-0.420023481074359,1.979595081419732E-010) (call it a) kval (0.420023367986768,2.500979698670295E-011) (b) kval (-0.306507431678779,-1.236804629184431E-011) (c) kval (-0.125376071175573,-6.134512510438736E-011) (d) kval (-7.945001124706894E-002,6.683546930037856E-011)(e) kval (0.106554601758169,-6.427946951285107E-011) (f) kval (8.866867725358024E-002,8.342250371574646E-011) (g) kval (0.325333314672671,1.260810749228185E-011) (h) kval (-0.488725859521576,1.769197678346003E-010) (i) kval (0.479509832763231,1.765499400037283E-010) (j) current eigenvalue -9.31389492882581 -1.24296522993488 -1.21324078359658 -1.11950286753963 -1.08166842367443 1.08187482164864 1.11973146584263 1.21295295042188 1.24280031534940 9.313897787790 So, the first 5 are left moving and rest are right moving. But I dont know the pairs. for example for left moving state a, what is the corresponding right moving state whether its f or g ... j ? Any help would be appreciated. Regards, Manoj Srivastava University of Florida, Gainesville. From jameslipd at gmail.com Thu Oct 22 03:07:35 2009 From: jameslipd at gmail.com (JAY) Date: Thu, 22 Oct 2009 09:07:35 +0800 Subject: [Pw_forum] Problem on compiling pwscf Message-ID: Dear Lorenzo Paulatto This is Jay Lee. The affiliate is Cug. I downloaded the tar file which contains the make.sys, however, I cannot succeed compiling it anyway. Then I tried to type ./configure to see what's wrong with my system, as reported the fortran 77 seems to be unavailable. I probably found where the bug is, because I updated the gcc to 4.4.1 version and the glibc to 2.9.3 version. The config.log reported that option '-cxxlib-gcc' not supported with the current installed GCC. I guess this is the problem that caused the error above. By the way, I want to learn more about how to switch the options in make.sys, but I cannot find any document that can help. Is there some? The system is FC10, the kernel is 2.6.27.35-170.2.94.fc10.x86_64 On Thu, Oct 22, 2009 at 2:13 AM, wrote: > Send Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-owner at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. Re: Problem on compiling pwscf (Lorenzo Paulatto) > 2. Re: lelfield (Lorenzo Paulatto) > 3. Re: "cannot open xml_recover file for writing" (Lorenzo Paulatto) > 4. Re: "cannot open xml_recover file for writing" (Prasenjit Ghosh) > 5. Re: "cannot open xml_recover file for writing" (Lorenzo Paulatto) > 6. Re: Cr GGA potential is underestimating the cohesive energy. > (udayagiri sai babu) > 7. Re: "cannot open xml_recover file for writing" (Ari P Seitsonen) > 8. Re: "cannot open xml_recover file for writing" > (marsamos at democritos.it) > 9. Left and right going Bloch's states from PWCOND (Manoj Srivastava) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 21 Oct 2009 14:33:23 +0200 > From: "Lorenzo Paulatto" > Subject: Re: [Pw_forum] Problem on compiling pwscf > To: "PWSCF Forum" > Message-ID: > Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes > > In data 21 ottobre 2009 alle ore 13:45:30, JAY ha > scritto: > > Thank you for the reply. I tried to understand the makefile of pwscf, but > > due to the lack of my knowledge, I am not able to find what should do. I > > typed ./configure in the source directory, it reports that the Fortran 77 > > compiler cannot create executables. I will check what's wrong with the > > Fortran compiler. > > We are not able to understand what you should do either: your are sending > contradicting emails. Let's try to recap what you've tried so far: > > 1. you run ./configure and it said it could not find a working f77 > compiler, which means you could not create a make.sys whatsoever > > 2. you have a make.sys prepared for gfortran (GNU fortran compiler) in > which you have simply replaced gfortran with mpif90, and added the > location of mpich to MPI_LIBS. > This make.sys will never work, as several additional switches are required > to compile the parallel code. > > 3. where was the original make.sys coming from? > > 4. are you sure that mpif90 is actually belonging to the installation of > mpich that is situated in /opt/mpich-1.2.5.2/? System administrators love > to install many different versions of the same or equivalent libraries, > often only one of them is actually working. > > 5. maybe the system you are using is module-based, you should really read > the system documentation. Or, at the *very* least, type the command > "module avail" and report the result... > > > cheers > > > P.S. please provide your full name+surname and affiliation, this is an > informal mailing list, but we like to know who we're talking to. > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > > > ------------------------------ > > Message: 2 > Date: Wed, 21 Oct 2009 14:39:16 +0200 > From: "Lorenzo Paulatto" > Subject: Re: [Pw_forum] lelfield > To: "PWSCF Forum" > Message-ID: > Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes > > In data 21 ottobre 2009 alle ore 10:41:33, Jiri Houska > ha scritto: > > > Dear Lorenzo, > > thanks for the answer. The relaxation is from scratch (not restarted), > > QE 4.0.3, running on cluster consisting of quadcore Xeon E5472 (3 GHz, > > 12 MB cache), happened for both (1) fixed occupations [used originally > > for material with a band gap] and (2) cold smearing. > > > Dear Jiri, > what happened in both cases? The i/o error or the algebra error? > As far as I know, the Berry-phase method does not work with pools (for > implementation reasons), also I'm not quite sure it works with smearing, > and with metals in general (I think the theory only applies to insulators). > > The davcio error you have experienced may happen if you restart a > calculation with a different number of pools. On the other hand, the zgefa > error may be caused by dipfield+smearing or dipfield+pools. > > You could check if one of the aforementioned conditions is verified in > your setup. > > cheers > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > > > ------------------------------ > > Message: 3 > Date: Wed, 21 Oct 2009 14:44:43 +0200 > From: "Lorenzo Paulatto" > Subject: Re: [Pw_forum] "cannot open xml_recover file for writing" > To: "Ari Paavo Seitsonen" , "PWSCF Forum" > > Message-ID: > Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes > > In data 21 ottobre 2009 alle ore 09:31:01, Ari P Seitsonen > ha scritto: > > 25 RECL= specifier omitted on an OPEN statement for a direct file. > > > > But I don't see where RECL's would be used for the XML files... > > It would go in iotk/src/iotk_tool.spp. > I'm having tons of i/o errors on sp6 too, I suspect it may have something > to do with the specific filesystem used, but couldn't get reproducible > fails so far. > > cheers > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > > > ------------------------------ > > Message: 4 > Date: Wed, 21 Oct 2009 16:45:58 +0200 > From: Prasenjit Ghosh > Subject: Re: [Pw_forum] "cannot open xml_recover file for writing" > To: PWSCF Forum > Message-ID: > <627e0ffa0910210745x65b5eb5dq5b85a0236f8d0b14 at mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > > > > I'm having tons of i/o errors on sp6 too, I suspect it may have something > > to do with the specific filesystem used, but couldn't get reproducible > > fails so far. > > > > > Lorenzo, > > the i/o errors on sp6 is most probably due to the problem in the > $CINECA_SCRATCH area. > In fact I'm also getting input output errors there and I think many other > people are also getting it. > > Prasenjit > > -- > PRASENJIT GHOSH, > POST-DOC, > ROOM NO: 265, MAIN BUILDING, > CM SECTION, ICTP, > STRADA COSTERIA 11, > TRIESTE, 34104, > ITALY > PHONE: +39 040 2240 369 (O) > +39 3807528672 (M) > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > http://www.democritos.it/pipermail/pw_forum/attachments/20091021/cef54fa9/attachment-0001.htm > > ------------------------------ > > Message: 5 > Date: Wed, 21 Oct 2009 16:53:53 +0200 > From: "Lorenzo Paulatto" > Subject: Re: [Pw_forum] "cannot open xml_recover file for writing" > To: "PWSCF Forum" > Message-ID: > Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes > > In data 21 ottobre 2009 alle ore 16:45:58, Prasenjit Ghosh > ha scritto: > > the i/o errors on sp6 is most probably due to the problem in the > > $CINECA_SCRATCH area. > > In fact I'm also getting input output errors there and I think many other > > people are also getting it. > > Yep, I suspected it. They started after some maintenance last month, so > I'm hoping they will disappear sooner or later. > > We're really going off-topic now, but I wished to post this, to reassure > other users who may be clueless and frustrated. Thank you for confirming. > > cheers > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > > > ------------------------------ > > Message: 6 > Date: Wed, 21 Oct 2009 20:35:13 +0530 > From: udayagiri sai babu > Subject: Re: [Pw_forum] Cr GGA potential is underestimating the > cohesive energy. > To: PWSCF Forum > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > Thank you dear Paolo for the reference you provided me. > > On Sat, Oct 17, 2009 at 4:07 PM, Paolo Giannozzi >wrote: > > > > > On Oct 17, 2009, at 11:31 , udayagiri sai babu wrote: > > > > > I am calculated the cohesive energy of Cr using the pseudopotential > > > Cr.pbe-sp-van.UPF. I got is 3.63 ev/atom as the cohesive energy > > > where as the experimental value is 4.1 ev/atom. As i know the GGA > > > pseudopotential should overestimate the cohesive energy value > > > > should it? LDA overestimates, GGA not necessarily so. > > > > The cohesive energy of Cr has been already calculated by many authors, > > for sure. You may want to have a look at what other people have done > > before you. For instance, here: > > http://dare.ubvu.vu.nl/bitstream/1871/9949/1/B4.pdf > > (found in 5' of google search). They get 3.58eV for Becke-Perdew, > > 3.80eV for Perdew-Wang. So 3.63eV for PBE after all is not that bad. > > > > P. > > --- > > 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 mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > -- > U.Saibabu > PhD student, > Deformation mechanisms modeling group, > Materials engineering department, > IISc Bangalore, > India. > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > http://www.democritos.it/pipermail/pw_forum/attachments/20091021/7dbda4e4/attachment-0001.htm > > ------------------------------ > > Message: 7 > Date: Wed, 21 Oct 2009 17:30:52 +0200 (CEST) > From: Ari P Seitsonen > Subject: Re: [Pw_forum] "cannot open xml_recover file for writing" > To: PWSCF Forum > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed > > > Dear Lorenzo et al, > > I can confirm that I also manage to run exactly the same system on a > Linux cluster (ifort 10.*, MKL) that caused the problem on the IBM Power6. > Unfortunately the latter is the only machine where I could run the > biiiiiig system... Well, I'll think of something. Thanks to all for the > support!! > > Greetings, > > apsi > > > -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- > Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ > Physikalisch-Chemisches Institut der Universitaet Zuerich > Tel: +41 44 63 54 497 / Mobile: +41 79 71 90 935 > > On Wed, 21 Oct 2009, Lorenzo Paulatto wrote: > > > In data 21 ottobre 2009 alle ore 16:45:58, Prasenjit Ghosh > > ha scritto: > >> the i/o errors on sp6 is most probably due to the problem in the > >> $CINECA_SCRATCH area. > >> In fact I'm also getting input output errors there and I think many > other > >> people are also getting it. > > > > Yep, I suspected it. They started after some maintenance last month, so > > I'm hoping they will disappear sooner or later. > > > > We're really going off-topic now, but I wished to post this, to reassure > > other users who may be clueless and frustrated. Thank you for confirming. > > > > cheers > > > > > > -- > > Lorenzo Paulatto > > SISSA & DEMOCRITOS (Trieste) > > phone: +39 040 3787 511 > > skype: paulatz > > www: http://people.sissa.it/~paulatto/ > > > > *** save italian brains *** > > http://saveitalianbrains.wordpress.com/ > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > ------------------------------ > > Message: 8 > Date: Wed, 21 Oct 2009 17:57:55 +0200 > From: marsamos at democritos.it > Subject: Re: [Pw_forum] "cannot open xml_recover file for writing" > To: pw_forum at pwscf.org > Message-ID: <20091021175755.x7sl43e3kgso8csw at mail.democritos.it> > Content-Type: text/plain; charset=ISO-8859-1; format="flowed" > > Dear all, I also find big i-o problems on sp6 (running GW) in > particular when handling big files ... sometimes it works sometimes it > does not work!!!!! HPC ;-) > > Layla > > Quoting Ari P Seitsonen : > > > > > Dear Lorenzo et al, > > > > I can confirm that I also manage to run exactly the same system on a > > Linux cluster (ifort 10.*, MKL) that caused the problem on the IBM > Power6. > > Unfortunately the latter is the only machine where I could run the > > biiiiiig system... Well, I'll think of something. Thanks to all for the > > support!! > > > > Greetings, > > > > apsi > > > > > -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- > > Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ > > Physikalisch-Chemisches Institut der Universitaet Zuerich > > Tel: +41 44 63 54 497 / Mobile: +41 79 71 90 935 > > > > On Wed, 21 Oct 2009, Lorenzo Paulatto wrote: > > > >> In data 21 ottobre 2009 alle ore 16:45:58, Prasenjit Ghosh > >> ha scritto: > >>> the i/o errors on sp6 is most probably due to the problem in the > >>> $CINECA_SCRATCH area. > >>> In fact I'm also getting input output errors there and I think many > other > >>> people are also getting it. > >> > >> Yep, I suspected it. They started after some maintenance last month, so > >> I'm hoping they will disappear sooner or later. > >> > >> We're really going off-topic now, but I wished to post this, to reassure > >> other users who may be clueless and frustrated. Thank you for > confirming. > >> > >> cheers > >> > >> > >> -- > >> Lorenzo Paulatto > >> SISSA & DEMOCRITOS (Trieste) > >> phone: +39 040 3787 511 > >> skype: paulatz > >> www: http://people.sissa.it/~paulatto/ > >> > >> *** save italian brains *** > >> http://saveitalianbrains.wordpress.com/ > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> http://www.democritos.it/mailman/listinfo/pw_forum > >> > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > ---------------------------------------------------------------- > This message was sent using IMP, the Internet Messaging Program. > > > > ------------------------------ > > Message: 9 > Date: Wed, 21 Oct 2009 14:13:31 -0400 (EDT) > From: Manoj Srivastava > Subject: [Pw_forum] Left and right going Bloch's states from PWCOND > To: pw_forum at pwscf.org > Message-ID: > > Content-Type: TEXT/PLAIN; charset=US-ASCII > > Dear All, > I am trying to figure out the left and right going Bloch's states in the > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > prints out Bloch's state moving in one direction. eg. in one of the > calculation- > k//=(0.375,-0.375) > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > 0.3157801 0.0000000 0.0000000 > > Now if I want Bloch's state moving in right as well as left direction, I > can go to kbloch.f90 subroutine, and print out all the eigen values of > AX=exp(ikd)BX, and out of those the ones with real solution would be our > Bloch's state, so I get for each channel two solutions- > kval (-0.275409421993275,1.823688001395235E-010) > kval (0.315780119742506,-3.611201785292708E-012) > > To figure out the direction, I can calculate current associated with these > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > its left moving Bloch's state. I can print out current from jbloch.f90 > subroutine which are - > current eigenvalue -1.86502143831863 1.59149029314457 > > So, clearly the first state with kval=-0.2754094 is left moving and the > other one right moving. Upto here its clear to me how to identify left and > right moving states. > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > state. In another calculation where i get multiple Bloch's state- > Nchannels of the left tip = 5 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > -0.0746301 0.0000000 0.0000000 > 0.1205527 0.0000000 0.0000000 > 0.3112908 0.0000000 0.0000000 > 0.4200218 0.0000000 0.0000000 > -0.4935150 0.0000000 0.0000000 > > so i did the same trick i did above to first print out kz and then > current, which gives me - > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > kval (0.420023367986768,2.500979698670295E-011) (b) > kval (-0.306507431678779,-1.236804629184431E-011) (c) > kval (-0.125376071175573,-6.134512510438736E-011) (d) > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > kval (0.106554601758169,-6.427946951285107E-011) (f) > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > kval (0.325333314672671,1.260810749228185E-011) (h) > kval (-0.488725859521576,1.769197678346003E-010) (i) > kval (0.479509832763231,1.765499400037283E-010) (j) > > current eigenvalue -9.31389492882581 -1.24296522993488 > -1.21324078359658 -1.11950286753963 -1.08166842367443 > 1.08187482164864 1.11973146584263 1.21295295042188 > 1.24280031534940 9.313897787790 > > So, the first 5 are left moving and rest are right moving. But I dont know > the pairs. for example for left moving state a, what is the corresponding > right moving state whether its f or g ... j ? > > Any help would be appreciated. > > Regards, > Manoj Srivastava > University of Florida, Gainesville. > > > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 28, Issue 54 > **************************************** > -- Sincerely Jay -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/b8117b01/attachment-0001.htm From decboy9 at gmail.com Thu Oct 22 06:06:38 2009 From: decboy9 at gmail.com (dev sharma) Date: Thu, 22 Oct 2009 09:36:38 +0530 Subject: [Pw_forum] GW calculations Message-ID: hi 2 all, Can anybody suggest the any example for GW calculations. thanks taker care Dev University of Delhi India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/b80fcbb6/attachment.htm From jhouska at kfy.zcu.cz Thu Oct 22 10:14:51 2009 From: jhouska at kfy.zcu.cz (Jiri Houska) Date: Thu, 22 Oct 2009 10:14:51 +0200 Subject: [Pw_forum] lelfield In-Reply-To: <4ADECA27.3020404@democritos.it> References: <20091021101734.bzsgp4et6sg484kc@webmail.zcu.cz> <20091021104133.rlhoy810sgw40wk4@webmail.zcu.cz> <4ADECA27.3020404@democritos.it> Message-ID: <20091022101451.vsmpy1esgg000swk@webmail.zcu.cz> Thanks - so it is necessary to optimize the wavefunction before relax WITH LELFIELD. Works now. Thanks again, Jiri Quoting Paolo Giannozzi : > Did you read examples/example31/README ? > > P. > -- > Paolo Giannozzi, Democritos and University of Udine, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From pc229 at kent.ac.uk Thu Oct 22 11:01:54 2009 From: pc229 at kent.ac.uk (Pieremanuele Canepa) Date: Thu, 22 Oct 2009 10:01:54 +0100 Subject: [Pw_forum] How dose the restart work? Message-ID: <8b443ba10910220201x4df3da47le45fcbf7e08ef772@mail.gmail.com> Dear all , my question for today is: How does the restart work in PWscf when in a vc-relax job? Once activated the restart, as restart_mode = 'restart', in the control section, will PWscf read either the last wave function and the last geometry or just the last density? Would it possible to restart a job from the previous geometry instead of the the wave function? Can I do both of them? All the best, Piero -- Pieremanuele Canepa Room 104 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom ----------------------------------------------------------- From smogunov at sissa.it Thu Oct 22 11:08:26 2009 From: smogunov at sissa.it (Alexander Smogunov) Date: Thu, 22 Oct 2009 11:08:26 +0200 Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: References: Message-ID: <1256202506.3829.17.camel@alex-laptop> Dear Manoj. The output of complex k vectors is performed in summary_band.f90 routine. If you want to see all the complex k vectors, not only propagating ones, you can change at the end of this routine: ------------------- do i = 1, nchanl WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev enddo ------------------- to ------------------- do i = 1, 2*nstl WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev enddo ------------------- Altogether there are 2*nstl (or 2*nstr) Bloch states in the left (or right) lead. First half, [1,nstl], are propagating or decaying to the right states, another half, [nstl+1,2*nstl], - propagating or decaying to the left. In each group, first nchanl states are propagating states. The propagating states are normalized by the current and are arranged in the above order at the end of jbloch.f90 routine, after the following lines: ! ! Right ordering (+, >, -, <) ! Notice, that in the last versions the code gives in output both propagating to the right and to the left states. Hope this helps, Alexander On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > Dear All, > I am trying to figure out the left and right going Bloch's states in the > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > prints out Bloch's state moving in one direction. eg. in one of the > calculation- > k//=(0.375,-0.375) > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > 0.3157801 0.0000000 0.0000000 > > Now if I want Bloch's state moving in right as well as left direction, I > can go to kbloch.f90 subroutine, and print out all the eigen values of > AX=exp(ikd)BX, and out of those the ones with real solution would be our > Bloch's state, so I get for each channel two solutions- > kval (-0.275409421993275,1.823688001395235E-010) > kval (0.315780119742506,-3.611201785292708E-012) > > To figure out the direction, I can calculate current associated with these > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > its left moving Bloch's state. I can print out current from jbloch.f90 > subroutine which are - > current eigenvalue -1.86502143831863 1.59149029314457 > > So, clearly the first state with kval=-0.2754094 is left moving and the > other one right moving. Upto here its clear to me how to identify left and > right moving states. > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > state. In another calculation where i get multiple Bloch's state- > Nchannels of the left tip = 5 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > -0.0746301 0.0000000 0.0000000 > 0.1205527 0.0000000 0.0000000 > 0.3112908 0.0000000 0.0000000 > 0.4200218 0.0000000 0.0000000 > -0.4935150 0.0000000 0.0000000 > > so i did the same trick i did above to first print out kz and then > current, which gives me - > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > kval (0.420023367986768,2.500979698670295E-011) (b) > kval (-0.306507431678779,-1.236804629184431E-011) (c) > kval (-0.125376071175573,-6.134512510438736E-011) (d) > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > kval (0.106554601758169,-6.427946951285107E-011) (f) > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > kval (0.325333314672671,1.260810749228185E-011) (h) > kval (-0.488725859521576,1.769197678346003E-010) (i) > kval (0.479509832763231,1.765499400037283E-010) (j) > > current eigenvalue -9.31389492882581 -1.24296522993488 > -1.21324078359658 -1.11950286753963 -1.08166842367443 > 1.08187482164864 1.11973146584263 1.21295295042188 > 1.24280031534940 9.313897787790 > > So, the first 5 are left moving and rest are right moving. But I dont know > the pairs. for example for left moving state a, what is the corresponding > right moving state whether its f or g ... j ? > > Any help would be appreciated. > > Regards, > Manoj Srivastava > University of Florida, Gainesville. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- e-mail: smogunov at sissa.it home-page: http://people.sissa.it/~smogunov From paulatto at sissa.it Thu Oct 22 11:25:15 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 22 Oct 2009 11:25:15 +0200 Subject: [Pw_forum] How dose the restart work? In-Reply-To: <8b443ba10910220201x4df3da47le45fcbf7e08ef772@mail.gmail.com> References: <8b443ba10910220201x4df3da47le45fcbf7e08ef772@mail.gmail.com> Message-ID: In data 22 ottobre 2009 alle ore 11:01:54, Pieremanuele Canepa ha scritto: > Once activated the restart, as restart_mode = 'restart', in the > control section, will PWscf read either the last wave function and the > last geometry or just the last density? It always reads the geometry, and the geometry history too for bfgs algorithm. It also reads the last charge density (which may be from an half completed scf cycle) and, if available, the last wavefunctions. > Would it possible to restart a job from the previous geometry instead > of the the wave function? Yes, you have to set startingwfc='whatever', remember to unset it afterward. Or you can copy the last geometry by hand and change the prefix or clean the tmp. > Can I do both of them? I'm not sure I understand what you mean here, if you mean "can I restart from both the last geometry and the last wfcs," then it is the default behaviour. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From sclauzer at sissa.it Thu Oct 22 11:26:17 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 22 Oct 2009 11:26:17 +0200 Subject: [Pw_forum] How dose the restart work? In-Reply-To: <8b443ba10910220201x4df3da47le45fcbf7e08ef772@mail.gmail.com> References: <8b443ba10910220201x4df3da47le45fcbf7e08ef772@mail.gmail.com> Message-ID: <4AE02539.10205@sissa.it> Dear Pieremanuele Pieremanuele Canepa wrote: > Dear all , > my question for today is: How does the restart work in PWscf when in a > vc-relax job? > Once activated the restart, as restart_mode = 'restart', in the > control section, will PWscf read either the last wave function and the > last geometry or just the last density? I don't know much about specificity of vc-relax, but in general when restarting pwscf the atomic configuration will be read from file (see in prefix.save/data_file.xml) as well as the wavefuntions and charge density from the last iteration of the previous run (if the files are availalable). If using wfc_extrapolation and pot_extrapolation, perhaps also older wfcs and older rho from previous run will be read from file. > Would it possible to restart a job from the previous geometry instead > of the the wave function? You can avoid the wavefunction to be read specifying startingwfc='atomic' (or simply removing/renaming the file). The code will use a superposition of atomic wavefunctions to build the initial guess (as for the first iteration of a run "from_scratch"), but this will require more time to perform the first iteration of the scf cycle. > Can I do both of them? Both what? HTH GS > > All the best, Piero > -- > Pieremanuele Canepa > Room 104 > School of Physical Sciences, Ingram Building, > University of Kent, Canterbury, Kent, > CT2 7NH > United Kingdom > ----------------------------------------------------------- > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From pc229 at kent.ac.uk Thu Oct 22 11:46:36 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Thu, 22 Oct 2009 10:46:36 +0100 Subject: [Pw_forum] How dose the restart work? In-Reply-To: <4AE02539.10205@sissa.it> References: <8b443ba10910220201x4df3da47le45fcbf7e08ef772@mail.gmail.com> <4AE02539.10205@sissa.it> Message-ID: thank you, both of you were very clear ! Best regards ----- Original Message ----- From: Gabriele Sclauzero Date: Thursday, October 22, 2009 10:26 am Subject: Re: [Pw_forum] How dose the restart work? To: PWSCF Forum > Dear Pieremanuele > > Pieremanuele Canepa wrote: > > Dear all , > > my question for today is: How does the restart work in PWscf > when in a > > vc-relax job? > > Once activated the restart, as restart_mode = 'restart', in the > > control section, will PWscf read either the last wave function > and the > > last geometry or just the last density? > > I don't know much about specificity of vc-relax, but in general > when restarting pwscf the > atomic configuration will be read from file (see in > prefix.save/data_file.xml) as well as > the wavefuntions and charge density from the last iteration of > the previous run (if the > files are availalable). > If using wfc_extrapolation and pot_extrapolation, perhaps also > older wfcs and older rho > from previous run will be read from file. > > > Would it possible to restart a job from the previous geometry > instead> of the the wave function? > > You can avoid the wavefunction to be read specifying > startingwfc='atomic' (or simply > removing/renaming the file). The code will use a superposition > of atomic wavefunctions to > build the initial guess (as for the first iteration of a run > "from_scratch"), but this > will require more time to perform the first iteration of the scf > cycle. > > Can I do both of them? > > Both what? > > > HTH > > GS > > > > > All the best, Piero > > -- > > Pieremanuele Canepa > > Room 104 > > School of Physical Sciences, Ingram Building, > > University of Kent, Canterbury, Kent, > > CT2 7NH > > United Kingdom > > ----------------------------------------------------------- > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD > Student????????????????? | > | c/o:?? SISSA & CNR-INFM > Democritos,????????????? | > |??????? via Beirut 2-4, > 34014 Trieste (Italy)???? | > | email: > sclauzer at sissa.it???????????????????????? | > | phone: +39 040 3787 > 511????????????????????????? | > | skype: > gurlonotturno???????????????????????????? | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/272cbff5/attachment.htm From guido.roma at cea.fr Thu Oct 22 14:20:37 2009 From: guido.roma at cea.fr (Guido Roma) Date: Thu, 22 Oct 2009 14:20:37 +0200 Subject: [Pw_forum] check on starting_magnetization In-Reply-To: References: Message-ID: <4AE04E15.1040006@cea.fr> Dear all, by error I submitted a job with starting_magnetization(3)=0.7 and nspin=2, but I had only two atomic species in my job. I realized that there are no warnings and the program went on doing a non-magnetic calculation while I wanted a magnetic one. This is because the array starting_magnetization is initialized with dimension nsx (=ntypx=10), and a check is made (in PW/input.f90) on all the elements of the array. Wouldn't it be better to put, in PW/input.f90, instead of the line : ALL(starting_magnetization == sm_not_set) ) THEN the following : ALL(starting_magnetization(0:nt) == sm_not_set) ) THEN ? Guido -------------- next part -------------- A non-text attachment was scrubbed... Name: guido_roma.vcf Type: text/x-vcard Size: 260 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20091022/cabc8daf/attachment-0001.vcf From miguel.martinez at ehu.es Thu Oct 22 15:43:21 2009 From: miguel.martinez at ehu.es (Miguel =?UTF-8?B?TWFydMOtbmV6?=) Date: Thu, 22 Oct 2009 15:43:21 +0200 Subject: [Pw_forum] Missing Wannier90 in Q-E 4.1.1? Message-ID: <20091022154321.4af18593@lcpybm> Hello everybody, I noticed yesterday that there was a new bugfix release of q-e on the website, although it yielded a 403 error. It's fine, q-e forge was working OK and today 4.1.1 was available from the Q-E website. When trying to build wannier90, however, I noticed that the W90/src folder is actually empty. I feel like I'm missed something, since espresso-4.1 did contain those F90 files and I've seen no mention of removing W90 in either the release notes or the changelog. Is there any reason why W90 has been left out? In any case, the sources from q-e 4.1 seem to work fine. Regards, Miguel -- ---------------------------------------- Miguel Mart?nez Canales Condensed Matter Physics Dpt. UPV/EHU Faculty of Science and Technology Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5326 ---------------------------------------- "If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas." George Bernard Shaw From giannozz at democritos.it Thu Oct 22 15:52:31 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 22 Oct 2009 15:52:31 +0200 Subject: [Pw_forum] Problem on compiling pwscf In-Reply-To: References: Message-ID: <5D58BB98-D044-4D18-85F1-BDEA4CC0211A@democritos.it> On Oct 22, 2009, at 3:07 , JAY wrote: > The affiliate is Cug. ??? > By the way, I want to learn more about how to switch the options in > make.sys, > but I cannot find any document that can help. Is there some? there are pages and pages in the user guide (in Doc/) about compilation. See also the make.sys file itself. > The system is FC10, the kernel is 2.6.27.35-170.2.94.fc10.x86_64 It is not the kernel or the operating system that compiles a code: it is the compiler. Do you have working compilers? apparently not, since configure says so. Problem solved: you cannot compile anything, until you fix your compiler(s) P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Thu Oct 22 15:54:02 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 22 Oct 2009 15:54:02 +0200 Subject: [Pw_forum] check on starting_magnetization In-Reply-To: <4AE04E15.1040006@cea.fr> References: <4AE04E15.1040006@cea.fr> Message-ID: <79C742CD-086B-4E23-B5C5-FE62387849D4@democritos.it> On Oct 22, 2009, at 14:20 , Guido Roma wrote: > Wouldn't it be better to put, in PW/input.f90, instead of the line : > > ALL(starting_magnetization == sm_not_set) ) THEN > > the following : > > ALL(starting_magnetization(0:nt) == sm_not_set) ) THEN definitely so (with a small correction: 1:nt, not 0:nt) - P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Thu Oct 22 15:58:02 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 22 Oct 2009 15:58:02 +0200 Subject: [Pw_forum] Missing Wannier90 in Q-E 4.1.1? In-Reply-To: <20091022154321.4af18593@lcpybm> References: <20091022154321.4af18593@lcpybm> Message-ID: On Oct 22, 2009, at 15:43 , Miguel Mart?nez wrote: > Is there any reason why W90 has been left out? it is a bug in the script that produces the pakcages. Sources of W90 are *.F90, in all other cases are *.f90. Thank you for reporting this. It will be fixed soon(er or later) > In any case, the sources from q-e 4.1 seem to work fine. they are the same: nothing has changed in W90 recently Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From manoj at phys.ufl.edu Thu Oct 22 17:34:46 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Thu, 22 Oct 2009 11:34:46 -0400 (EDT) Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: <1256202506.3829.17.camel@alex-laptop> Message-ID: Dear Alexander, Thanks for your answer. I just want to make sure. Imagine we have total number of channels in the left lead 2, so total number of Bloch's state are 4. 2 of them left going say a and b, and 2 right going say c and d. So, are you saying that for left going state a, the corresponding right going state is c? Are they ordered this way? Regards, Manoj On Thu, 22 Oct 2009, Alexander Smogunov wrote: > Dear Manoj. > > The output of complex k vectors is performed in > summary_band.f90 routine. If you want to see all > the complex k vectors, not only propagating ones, > you can change at the end of this routine: > > ------------------- > do i = 1, nchanl > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > enddo > ------------------- > > to > ------------------- > > do i = 1, 2*nstl > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > enddo > ------------------- > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > (or right) lead. First half, [1,nstl], are propagating or decaying to > the right states, another half, [nstl+1,2*nstl], - propagating or > decaying to the left. In each group, first nchanl states are propagating > states. > > The propagating states are normalized by the current and are arranged in > the above order at the end of jbloch.f90 routine, after the following > lines: > > ! > ! Right ordering (+, >, -, <) > ! > > > > Notice, that in the last versions the code gives in output > both propagating to the right and to the left states. > > Hope this helps, > Alexander > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > Dear All, > > I am trying to figure out the left and right going Bloch's states in the > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > > prints out Bloch's state moving in one direction. eg. in one of the > > calculation- > > k//=(0.375,-0.375) > > Nchannels of the left tip = 1 > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > 0.3157801 0.0000000 0.0000000 > > > > Now if I want Bloch's state moving in right as well as left direction, I > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > AX=exp(ikd)BX, and out of those the ones with real solution would be our > > Bloch's state, so I get for each channel two solutions- > > kval (-0.275409421993275,1.823688001395235E-010) > > kval (0.315780119742506,-3.611201785292708E-012) > > > > To figure out the direction, I can calculate current associated with these > > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > > its left moving Bloch's state. I can print out current from jbloch.f90 > > subroutine which are - > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > other one right moving. Upto here its clear to me how to identify left and > > right moving states. > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > state. In another calculation where i get multiple Bloch's state- > > Nchannels of the left tip = 5 > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > -0.0746301 0.0000000 0.0000000 > > 0.1205527 0.0000000 0.0000000 > > 0.3112908 0.0000000 0.0000000 > > 0.4200218 0.0000000 0.0000000 > > -0.4935150 0.0000000 0.0000000 > > > > so i did the same trick i did above to first print out kz and then > > current, which gives me - > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > kval (0.420023367986768,2.500979698670295E-011) (b) > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > kval (0.325333314672671,1.260810749228185E-011) (h) > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > 1.08187482164864 1.11973146584263 1.21295295042188 > > 1.24280031534940 9.313897787790 > > > > So, the first 5 are left moving and rest are right moving. But I dont know > > the pairs. for example for left moving state a, what is the corresponding > > right moving state whether its f or g ... j ? > > > > Any help would be appreciated. > > > > Regards, > > Manoj Srivastava > > University of Florida, Gainesville. > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > -- > e-mail: smogunov at sissa.it > home-page: http://people.sissa.it/~smogunov > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From Andrea.Marini at roma2.infn.it Thu Oct 22 17:35:36 2009 From: Andrea.Marini at roma2.infn.it (Andrea Marini) Date: Thu, 22 Oct 2009 17:35:36 +0200 (CEST) Subject: [Pw_forum] GW calculations In-Reply-To: References: Message-ID: On Thu, 22 Oct 2009, dev sharma wrote: > hi 2 all, > ?Can anybody suggest the any example for GW calculations. Dear Dev, which code are you talking about ? Of course there is no unique input file for a GW calculation. It depends on the code and, more importantly, on the system you want to study. So, please, be more precise and tell us something more ;) Cheers Andrea -------------------------------------------------------------------------- Andrea MARINI Physics Department, University of Rome "Tor Vergata" (Italy) - phone: +39-0672594894 - fax: +39-062023507 - - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - From smogunov at sissa.it Thu Oct 22 17:47:05 2009 From: smogunov at sissa.it (Alexander Smogunov) Date: Thu, 22 Oct 2009 17:47:05 +0200 Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: References: Message-ID: <1256226425.3829.27.camel@alex-laptop> On Thu, 2009-10-22 at 11:34 -0400, Manoj Srivastava wrote: > Dear Alexander, > Thanks for your answer. I just want to make sure. Imagine we have total > number of channels in the left lead 2, so total number of Bloch's state > are 4. 2 of them left going say a and b, and 2 right going say c and d. > So, are you saying that for left going state a, the corresponding right > going state is c? Are they ordered this way? what do you mean by corresponding? Left and right moving Bloch states are in general not related one to another, you can even have different number of them ... Only if you have some symmetry S which brings kz to -kz conserving k_parallel, then the state with \psi_{-kz} will be S \psi{kz}. This is true for example at 2D G point when you have time reversal operation. Now the code simply arranges the propagating states in the order of increasing |k_z|... Regards, Alexander > > Regards, > Manoj > > > On Thu, 22 Oct 2009, Alexander > Smogunov wrote: > > > Dear Manoj. > > > > The output of complex k vectors is performed in > > summary_band.f90 routine. If you want to see all > > the complex k vectors, not only propagating ones, > > you can change at the end of this routine: > > > > ------------------- > > do i = 1, nchanl > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > enddo > > ------------------- > > > > to > > ------------------- > > > > do i = 1, 2*nstl > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > enddo > > ------------------- > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > the right states, another half, [nstl+1,2*nstl], - propagating or > > decaying to the left. In each group, first nchanl states are propagating > > states. > > > > The propagating states are normalized by the current and are arranged in > > the above order at the end of jbloch.f90 routine, after the following > > lines: > > > > ! > > ! Right ordering (+, >, -, <) > > ! > > > > > > > > Notice, that in the last versions the code gives in output > > both propagating to the right and to the left states. > > > > Hope this helps, > > Alexander > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > Dear All, > > > I am trying to figure out the left and right going Bloch's states in the > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > > > prints out Bloch's state moving in one direction. eg. in one of the > > > calculation- > > > k//=(0.375,-0.375) > > > Nchannels of the left tip = 1 > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > Now if I want Bloch's state moving in right as well as left direction, I > > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > > AX=exp(ikd)BX, and out of those the ones with real solution would be our > > > Bloch's state, so I get for each channel two solutions- > > > kval (-0.275409421993275,1.823688001395235E-010) > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > To figure out the direction, I can calculate current associated with these > > > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > > > its left moving Bloch's state. I can print out current from jbloch.f90 > > > subroutine which are - > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > > other one right moving. Upto here its clear to me how to identify left and > > > right moving states. > > > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > > state. In another calculation where i get multiple Bloch's state- > > > Nchannels of the left tip = 5 > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > -0.0746301 0.0000000 0.0000000 > > > 0.1205527 0.0000000 0.0000000 > > > 0.3112908 0.0000000 0.0000000 > > > 0.4200218 0.0000000 0.0000000 > > > -0.4935150 0.0000000 0.0000000 > > > > > > so i did the same trick i did above to first print out kz and then > > > current, which gives me - > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > 1.24280031534940 9.313897787790 > > > > > > So, the first 5 are left moving and rest are right moving. But I dont know > > > the pairs. for example for left moving state a, what is the corresponding > > > right moving state whether its f or g ... j ? > > > > > > Any help would be appreciated. > > > > > > Regards, > > > Manoj Srivastava > > > University of Florida, Gainesville. > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > e-mail: smogunov at sissa.it > > home-page: http://people.sissa.it/~smogunov > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- e-mail: smogunov at sissa.it home-page: http://people.sissa.it/~smogunov From stewart at cnf.cornell.edu Thu Oct 22 17:47:36 2009 From: stewart at cnf.cornell.edu (Derek Stewart) Date: Thu, 22 Oct 2009 11:47:36 -0400 Subject: [Pw_forum] hard drive becomes read only during parallel QE 4.1.1 run with openmpi-1.3.3 intelv11 compilers Message-ID: Hi everyone, I have recently been trying out the new version of QE (4.1.1) with different MPI libraries (mpich, openmpi) etc. I recently came across a strange problem where a QE run for nickel test case (large k-mesh 48x48x48) would crash when I ran it in parallel across 5 nodes (10 processors) with openmpi-1.3.3 and Intel v11 compiles/MKL 10.2. It appears that the hard drive on the last node becomes read-only and QE can no longer write to the local wavefunction files. After the run, the local scratch drive remains read only and I end up having to reboot the system to eliminate this problem. I have been able to reproduce this problem on the same node. However, when I remove this node from the list, QE runs fine. Also, running QE with mpich2 doesn't have a problem on that node. I suspect that it could be a hardware issue (harddrive close to dying perhaps) or an issue with openmpi, but I wanted to check to see anyone else has run into this problem while using QE. For additional technical info, this is on a system with Redhat Enterprise 4, 2 Xeon processors (3 GHz), 2GB ram. 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 From decboy9 at gmail.com Thu Oct 22 18:54:26 2009 From: decboy9 at gmail.com (dev sharma) Date: Thu, 22 Oct 2009 22:24:26 +0530 Subject: [Pw_forum] GW calculations In-Reply-To: References: Message-ID: Dear Andrea, I want to calculate the reflection, absorption and EEELS for the YVO4. I earlier tried the epsilon.x but all the guides on this forum suggested me to use YAMBO. And for YAMBO , i think i have to produce result with GW approximation. Please guide, Thanks With regards, Dev On Thu, Oct 22, 2009 at 9:05 PM, Andrea Marini wrote: > On Thu, 22 Oct 2009, dev sharma wrote: > > hi 2 all, >> Can anybody suggest the any example for GW calculations. >> > > Dear Dev, > > which code are you talking about ? Of course there is no unique > input file for a GW calculation. It depends on the code and, more > importantly, on the system you want to study. > > So, please, be more precise and tell us something more ;) > > Cheers > > Andrea > > -------------------------------------------------------------------------- > Andrea MARINI > > Physics Department, University of Rome "Tor Vergata" (Italy) > - phone: +39-0672594894 - fax: +39-062023507 - > > - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/e2b4d0af/attachment.htm From Andrea.Marini at roma2.infn.it Thu Oct 22 21:21:36 2009 From: Andrea.Marini at roma2.infn.it (Andrea Marini) Date: Thu, 22 Oct 2009 21:21:36 +0200 (CEST) Subject: [Pw_forum] GW calculations In-Reply-To: References: Message-ID: On Thu, 22 Oct 2009, dev sharma wrote: > I want to calculate the reflection, absorption and EEELS for the YVO4. I > earlier tried the epsilon.x but all the guides on this forum suggested me to > use YAMBO. And for YAMBO , i think i have to produce result with GW > approximation. Ok Dev, step by step. Reflection, absorption and EELS can be computed without performing any self-energy calculations. Moreover I strongly suggest you to 1. post questions and doubts regarding any aspect of Yambo in the dedicated forum on the yambo web page (www.yambo-code.org/forum). There are several sections covering all features of the code. This is a PW mailing list and I am not sure people will be interested in all the technicalities of yambo. 2. start using the code on simple systems, trying to reproduce some well known properties (like the excitonic optical absorption and/or the quasiparticle gap correction). So, I am looking forward to see on our forum. Cheers Andrea > Please guide, > Thanks > With regards, > Dev > > On Thu, Oct 22, 2009 at 9:05 PM, Andrea Marini > wrote: > >> On Thu, 22 Oct 2009, dev sharma wrote: >> >> hi 2 all, >>> Can anybody suggest the any example for GW calculations. >>> >> >> Dear Dev, >> >> which code are you talking about ? Of course there is no unique >> input file for a GW calculation. It depends on the code and, more >> importantly, on the system you want to study. >> >> So, please, be more precise and tell us something more ;) >> >> Cheers >> >> Andrea >> >> -------------------------------------------------------------------------- >> Andrea MARINI >> >> Physics Department, University of Rome "Tor Vergata" (Italy) >> - phone: +39-0672594894 - fax: +39-062023507 - >> >> - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > -------------------------------------------------------------------------- Andrea MARINI Physics Department, University of Rome "Tor Vergata" (Italy) - phone: +39-0672594894 - fax: +39-062023507 - - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - From bmalone at civet.berkeley.edu Thu Oct 22 21:29:52 2009 From: bmalone at civet.berkeley.edu (Brad Malone) Date: Thu, 22 Oct 2009 12:29:52 -0700 Subject: [Pw_forum] problems calculating large numbers of empty states Message-ID: Hi, I'm trying to calculate a large number of conduction band states for cubic Si (to be used in a subsequent GW calculation). However, when doing this I sometimes run into errors like "problems computing cholesky decomposition" or "too many bands are unconverged". An example file of the nscf calculation that generates an error for me is below: The problems can sometimes be avoided by going to a larger wavefunction cutoff, but I'm not sure if this is for good reason or if changing the wavefunction cutoff just happens to avoid the error in the cases I've tried. Also, for a chosen grid (not the one listed below) I get errors when trying to calculate 400,500, and 700 total bands, but the calculation works fine for 600 bands. Any insight to why this might happen or suggestions as to how to avoid it? Thanks, Brad Malone UC Berkeley ------------------------------------------------------------- &control prefix = 'si' calculation = 'nscf' restart_mode = 'from_scratch' wf_collect = .true. tstress = .false. tprnfor = .false. outdir = './' wfcdir = './' pseudo_dir = './' / &system ibrav = 0 celldm(1) = 10.2612 nat = 2 ntyp = 1 nbnd = 300 ecutwfc = 40.0 / &electrons conv_thr = 1.0d-10 diagonalization = 'cg' diago_full_acc = .true. / CELL_PARAMETERS cubic +0.000000000 +0.500000000 +0.500000000 +0.500000000 +0.000000000 +0.500000000 +0.500000000 +0.500000000 +0.000000000 ATOMIC_SPECIES Si 28.086 Si.UPF ATOMIC_POSITIONS crystal Si -0.125000000 -0.125000000 -0.125000000 Si +0.125000000 +0.125000000 +0.125000000 K_POINTS crystal 8 0.000000000 0.000000000 0.000000000 1.0 0.000000000 0.000000000 0.250000000 8.0 0.000000000 0.000000000 0.500000000 4.0 0.000000000 0.250000000 0.250000000 6.0 0.000000000 0.250000000 0.500000000 24.0 0.000000000 0.250000000 0.750000000 12.0 0.000000000 0.500000000 0.500000000 3.0 0.250000000 0.500000000 0.750000000 6.0 ---------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/92715f32/attachment.htm From jameslipd at gmail.com Fri Oct 23 02:58:19 2009 From: jameslipd at gmail.com (JAY) Date: Thu, 22 Oct 2009 20:58:19 -0400 Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 58Re: Problem on compiling pwscf Message-ID: Dear Paolo Thank you for answering me. Cug is China University of Geosciences. I don't have the working compilers. I tried to update the gcclib and gcc, it seems work well for other packages, like BLAS, fftw, however, when compile lapack-3.2.1, it reports make[1]: Entering directory `/home/jy/espresso/lapack-3.2.1/SRC' mpif90 -O2 -c ../INSTALL/second_INT_ETIME.f -o ../INSTALL/second_INT_ETIME.o PGF90-S-0126-Name etime is not an intrinsic function (../INSTALL/second_INT_ETIME.f: 23) 0 inform, 0 warnings, 1 severes, 0 fatal for second make[1]: *** [../INSTALL/second_INT_ETIME.o] Error 2 make[1]: Leaving directory `/home/jy/espresso/lapack-3.2.1/SRC' make: *** [lapacklib] Error 2 so, I have to comment the lines in second_INT_ETIME.f and dsecnd_INT_ETIME.f. Then the compiling of lapack gives no error, but compile of make.sys gives error: ../Modules/fft_scalar.o: In function `fft_scalar_cft_1z_': fft_scalar.F90:(.text+0x1a6): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x231): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x32f): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x3dd): undefined reference to `zfft1mx_' ../Modules/fft_scalar.o: In function `fft_scalar_cft_2xy_': fft_scalar.F90:(.text+0x696): undefined reference to `zfft1mx_' ../Modules/fft_scalar.o:fft_scalar.F90:(.text+0x701): more undefined references to `zfft1mx_' follow ../Modules/fft_scalar.o: In function `fft_scalar_cfft3d_': fft_scalar.F90:(.text+0xeea): undefined reference to `zfft3dy_' fft_scalar.F90:(.text+0xfa5): undefined reference to `zfft3dy_' fft_scalar.F90:(.text+0x10e3): undefined reference to `zfft3dy_' fft_scalar.F90:(.text+0x11cf): undefined reference to `zfft3dy_' ../Modules/fft_scalar.o: In function `fft_scalar_cfft3ds_': fft_scalar.F90:(.text+0x142c): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x14a8): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x1551): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x15cd): undefined reference to `zfft1mx_' fft_scalar.F90:(.text+0x1682): undefined reference to `zfft1mx_' ../Modules/fft_scalar.o:fft_scalar.F90:(.text+0x16fe): more undefined references to `zfft1mx_' follow ../Multigrid/mglib.a(maind.o): In function `mgmain_': maind.F90:(.text+0x13c): undefined reference to `tstart_' maind.F90:(.text+0x1ef): undefined reference to `readit_' maind.F90:(.text+0x269): undefined reference to `mgsize_' maind.F90:(.text+0x652): undefined reference to `readit_' maind.F90:(.text+0x7a1): undefined reference to `fillco_' maind.F90:(.text+0x7be): undefined reference to `tstop_' maind.F90:(.text+0x7d3): undefined reference to `tstart_' maind.F90:(.text+0x8a3): undefined reference to `cgmgdriv_' maind.F90:(.text+0x91b): undefined reference to `tstop_' /opt/mpich-1.2.5.2/lib/libmpich.a(farg.o): In function `mpir_iargc_': farg.f:(.text+0xe): undefined reference to `for_iargc' make[1]: *** [pw.x] Error 2 make[1]: Leaving directory `/home/jy/espresso/espresso-4.1.1/PW' make: *** [pw] Error 2 I guess there must be something wrong with the compiler, (maybe ifort). On Thu, Oct 22, 2009 at 11:57 PM, wrote: > Send Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-owner at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. Missing Wannier90 in Q-E 4.1.1? (Miguel Mart?nez) > 2. Re: Problem on compiling pwscf (Paolo Giannozzi) > 3. Re: check on starting_magnetization (Paolo Giannozzi) > 4. Re: Missing Wannier90 in Q-E 4.1.1? (Paolo Giannozzi) > 5. Re: Left and right going Bloch's states from PWCOND > (Manoj Srivastava) > 6. Re: GW calculations (Andrea Marini) > 7. Re: Left and right going Bloch's states from PWCOND > (Alexander Smogunov) > 8. hard drive becomes read only during parallel QE 4.1.1 run > with openmpi-1.3.3 intelv11 compilers (Derek Stewart) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 22 Oct 2009 15:43:21 +0200 > From: Miguel Mart?nez > Subject: [Pw_forum] Missing Wannier90 in Q-E 4.1.1? > To: PWscf Forum > Message-ID: <20091022154321.4af18593 at lcpybm> > Content-Type: text/plain; charset=UTF-8 > > Hello everybody, > > I noticed yesterday that there was a new bugfix release of q-e on the > website, although it yielded a 403 error. It's fine, q-e forge was > working OK and today 4.1.1 was available from the Q-E website. When > trying to build wannier90, however, I noticed that the W90/src folder > is actually empty. > > I feel like I'm missed something, since espresso-4.1 did contain those > F90 files and I've seen no mention of removing W90 in either the > release notes or the changelog. Is there any reason why W90 has been > left out? In any case, the sources from q-e 4.1 seem to work fine. > > Regards, > > Miguel > > -- > ---------------------------------------- > Miguel Mart?nez Canales > Condensed Matter Physics Dpt. > UPV/EHU > Faculty of Science and Technology > Apdo. 644 > 48080 Bilbao (Spain) > Fax: +34 94 601 3500 > Tlf: +34 94 601 5326 > ---------------------------------------- > > "If you have an apple and I have an apple and > we exchange these apples then you and I will > still each have one apple. But if you have an > idea and I have an idea and we exchange these > ideas, then each of us will have two ideas." > > George Bernard Shaw > > > > ------------------------------ > > Message: 2 > Date: Thu, 22 Oct 2009 15:52:31 +0200 > From: Paolo Giannozzi > Subject: Re: [Pw_forum] Problem on compiling pwscf > To: PWSCF Forum > Message-ID: <5D58BB98-D044-4D18-85F1-BDEA4CC0211A at democritos.it> > Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed > > > On Oct 22, 2009, at 3:07 , JAY wrote: > > > The affiliate is Cug. > > ??? > > > By the way, I want to learn more about how to switch the options in > > make.sys, > > but I cannot find any document that can help. Is there some? > > there are pages and pages in the user guide (in Doc/) about compilation. > See also the make.sys file itself. > > > The system is FC10, the kernel is 2.6.27.35-170.2.94.fc10.x86_64 > > It is not the kernel or the operating system that compiles a code: it > is the compiler. > Do you have working compilers? apparently not, since configure says so. > Problem solved: you cannot compile anything, until you fix your > compiler(s) > > P. > --- > Paolo Giannozzi, Dept of Physics, University of Udine > via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > > > > > ------------------------------ > > Message: 3 > Date: Thu, 22 Oct 2009 15:54:02 +0200 > From: Paolo Giannozzi > Subject: Re: [Pw_forum] check on starting_magnetization > To: PWSCF Forum > Message-ID: <79C742CD-086B-4E23-B5C5-FE62387849D4 at democritos.it> > Content-Type: text/plain; charset=US-ASCII; format=flowed > > > On Oct 22, 2009, at 14:20 , Guido Roma wrote: > > > Wouldn't it be better to put, in PW/input.f90, instead of the line : > > > > ALL(starting_magnetization == sm_not_set) ) THEN > > > > the following : > > > > ALL(starting_magnetization(0:nt) == sm_not_set) ) THEN > > definitely so (with a small correction: 1:nt, not 0:nt) - P. > --- > Paolo Giannozzi, Dept of Physics, University of Udine > via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > > > > > ------------------------------ > > Message: 4 > Date: Thu, 22 Oct 2009 15:58:02 +0200 > From: Paolo Giannozzi > Subject: Re: [Pw_forum] Missing Wannier90 in Q-E 4.1.1? > To: PWSCF Forum > Message-ID: > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > > On Oct 22, 2009, at 15:43 , Miguel Mart?nez wrote: > > > Is there any reason why W90 has been left out? > > it is a bug in the script that produces the pakcages. > Sources of W90 are *.F90, in all other cases are *.f90. > Thank you for reporting this. It will be fixed soon(er or later) > > > In any case, the sources from q-e 4.1 seem to work fine. > > they are the same: nothing has changed in W90 recently > > Paolo > --- > Paolo Giannozzi, Dept of Physics, University of Udine > via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > > > > > ------------------------------ > > Message: 5 > Date: Thu, 22 Oct 2009 11:34:46 -0400 (EDT) > From: Manoj Srivastava > Subject: Re: [Pw_forum] Left and right going Bloch's states from > PWCOND > To: PWSCF Forum > Message-ID: > > Content-Type: TEXT/PLAIN; charset=US-ASCII > > Dear Alexander, > Thanks for your answer. I just want to make sure. Imagine we have total > number of channels in the left lead 2, so total number of Bloch's state > are 4. 2 of them left going say a and b, and 2 right going say c and d. > So, are you saying that for left going state a, the corresponding right > going state is c? Are they ordered this way? > > Regards, > Manoj > > > On Thu, 22 Oct 2009, Alexander > Smogunov wrote: > > > Dear Manoj. > > > > The output of complex k vectors is performed in > > summary_band.f90 routine. If you want to see all > > the complex k vectors, not only propagating ones, > > you can change at the end of this routine: > > > > ------------------- > > do i = 1, nchanl > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > enddo > > ------------------- > > > > to > > ------------------- > > > > do i = 1, 2*nstl > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > enddo > > ------------------- > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > the right states, another half, [nstl+1,2*nstl], - propagating or > > decaying to the left. In each group, first nchanl states are propagating > > states. > > > > The propagating states are normalized by the current and are arranged in > > the above order at the end of jbloch.f90 routine, after the following > > lines: > > > > ! > > ! Right ordering (+, >, -, <) > > ! > > > > > > > > Notice, that in the last versions the code gives in output > > both propagating to the right and to the left states. > > > > Hope this helps, > > Alexander > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > Dear All, > > > I am trying to figure out the left and right going Bloch's states in > the > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code > only > > > prints out Bloch's state moving in one direction. eg. in one of the > > > calculation- > > > k//=(0.375,-0.375) > > > Nchannels of the left tip = 1 > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > Now if I want Bloch's state moving in right as well as left direction, > I > > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > > AX=exp(ikd)BX, and out of those the ones with real solution would be > our > > > Bloch's state, so I get for each channel two solutions- > > > kval (-0.275409421993275,1.823688001395235E-010) > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > To figure out the direction, I can calculate current associated with > these > > > Bloch's sate and if the current is +ive it is right moving , and if > '-'ive > > > its left moving Bloch's state. I can print out current from jbloch.f90 > > > subroutine which are - > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > > other one right moving. Upto here its clear to me how to identify left > and > > > right moving states. > > > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > > state. In another calculation where i get multiple Bloch's state- > > > Nchannels of the left tip = 5 > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > -0.0746301 0.0000000 0.0000000 > > > 0.1205527 0.0000000 0.0000000 > > > 0.3112908 0.0000000 0.0000000 > > > 0.4200218 0.0000000 0.0000000 > > > -0.4935150 0.0000000 0.0000000 > > > > > > so i did the same trick i did above to first print out kz and then > > > current, which gives me - > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > 1.24280031534940 9.313897787790 > > > > > > So, the first 5 are left moving and rest are right moving. But I dont > know > > > the pairs. for example for left moving state a, what is the > corresponding > > > right moving state whether its f or g ... j ? > > > > > > Any help would be appreciated. > > > > > > Regards, > > > Manoj Srivastava > > > University of Florida, Gainesville. > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > e-mail: smogunov at sissa.it > > home-page: http://people.sissa.it/~smogunov > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > ------------------------------ > > Message: 6 > Date: Thu, 22 Oct 2009 17:35:36 +0200 (CEST) > From: Andrea Marini > Subject: Re: [Pw_forum] GW calculations > To: PWSCF Forum > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > On Thu, 22 Oct 2009, dev sharma wrote: > > > hi 2 all, > > ?Can anybody suggest the any example for GW calculations. > > Dear Dev, > > which code are you talking about ? Of course there is no unique > input file for a GW calculation. It depends on the code and, more > importantly, on the system you want to study. > > So, please, be more precise and tell us something more ;) > > Cheers > > Andrea > > -------------------------------------------------------------------------- > Andrea MARINI > > Physics Department, University of Rome "Tor Vergata" (Italy) > - phone: +39-0672594894 - fax: +39-062023507 - > > - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - > > ------------------------------ > > Message: 7 > Date: Thu, 22 Oct 2009 17:47:05 +0200 > From: Alexander Smogunov > Subject: Re: [Pw_forum] Left and right going Bloch's states from > PWCOND > To: PWSCF Forum > Message-ID: <1256226425.3829.27.camel at alex-laptop> > Content-Type: text/plain > > On Thu, 2009-10-22 at 11:34 -0400, Manoj Srivastava wrote: > > Dear Alexander, > > Thanks for your answer. I just want to make sure. Imagine we have total > > number of channels in the left lead 2, so total number of Bloch's state > > are 4. 2 of them left going say a and b, and 2 right going say c and d. > > So, are you saying that for left going state a, the corresponding right > > going state is c? Are they ordered this way? > > what do you mean by corresponding? Left and right moving Bloch > states are in general not related one to another, you can even have > different number of them ... Only if you have some symmetry S which > brings kz to -kz conserving k_parallel, then the state with > \psi_{-kz} will be S \psi{kz}. This is true for example at 2D G point > when you have time reversal operation. > > Now the code simply arranges the propagating states in the order of > increasing |k_z|... > > Regards, Alexander > > > > > > Regards, > > Manoj > > > > > > On Thu, 22 Oct 2009, Alexander > > Smogunov wrote: > > > > > Dear Manoj. > > > > > > The output of complex k vectors is performed in > > > summary_band.f90 routine. If you want to see all > > > the complex k vectors, not only propagating ones, > > > you can change at the end of this routine: > > > > > > ------------------- > > > do i = 1, nchanl > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > enddo > > > ------------------- > > > > > > to > > > ------------------- > > > > > > do i = 1, 2*nstl > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > enddo > > > ------------------- > > > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > > the right states, another half, [nstl+1,2*nstl], - propagating or > > > decaying to the left. In each group, first nchanl states are > propagating > > > states. > > > > > > The propagating states are normalized by the current and are arranged > in > > > the above order at the end of jbloch.f90 routine, after the following > > > lines: > > > > > > ! > > > ! Right ordering (+, >, -, <) > > > ! > > > > > > > > > > > > Notice, that in the last versions the code gives in output > > > both propagating to the right and to the left states. > > > > > > Hope this helps, > > > Alexander > > > > > > > > > > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > > Dear All, > > > > I am trying to figure out the left and right going Bloch's states in > the > > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code > only > > > > prints out Bloch's state moving in one direction. eg. in one of the > > > > calculation- > > > > k//=(0.375,-0.375) > > > > Nchannels of the left tip = 1 > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > > > Now if I want Bloch's state moving in right as well as left > direction, I > > > > can go to kbloch.f90 subroutine, and print out all the eigen values > of > > > > AX=exp(ikd)BX, and out of those the ones with real solution would be > our > > > > Bloch's state, so I get for each channel two solutions- > > > > kval (-0.275409421993275,1.823688001395235E-010) > > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > > > To figure out the direction, I can calculate current associated with > these > > > > Bloch's sate and if the current is +ive it is right moving , and if > '-'ive > > > > its left moving Bloch's state. I can print out current from > jbloch.f90 > > > > subroutine which are - > > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and > the > > > > other one right moving. Upto here its clear to me how to identify > left and > > > > right moving states. > > > > > > > > I get confused when for a given (kx,ky,E), I have more than one > Bloch' > > > > state. In another calculation where i get multiple Bloch's state- > > > > Nchannels of the left tip = 5 > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > -0.0746301 0.0000000 0.0000000 > > > > 0.1205527 0.0000000 0.0000000 > > > > 0.3112908 0.0000000 0.0000000 > > > > 0.4200218 0.0000000 0.0000000 > > > > -0.4935150 0.0000000 0.0000000 > > > > > > > > so i did the same trick i did above to first print out kz and then > > > > current, which gives me - > > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > > 1.24280031534940 9.313897787790 > > > > > > > > So, the first 5 are left moving and rest are right moving. But I dont > know > > > > the pairs. for example for left moving state a, what is the > corresponding > > > > right moving state whether its f or g ... j ? > > > > > > > > Any help would be appreciated. > > > > > > > > Regards, > > > > Manoj Srivastava > > > > University of Florida, Gainesville. > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > -- > > > e-mail: smogunov at sissa.it > > > home-page: http://people.sissa.it/~smogunov > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > -- > e-mail: smogunov at sissa.it > home-page: http://people.sissa.it/~smogunov > > > > ------------------------------ > > Message: 8 > Date: Thu, 22 Oct 2009 11:47:36 -0400 > From: "Derek Stewart" > Subject: [Pw_forum] hard drive becomes read only during parallel QE > 4.1.1 run with openmpi-1.3.3 intelv11 compilers > To: > Message-ID: > Content-Type: text/plain; charset="iso-8859-1" > > Hi everyone, > > I have recently been trying out the new version of QE (4.1.1) with > different MPI libraries (mpich, openmpi) etc. I recently came across a > strange problem where a QE run for nickel test case (large k-mesh 48x48x48) > would crash when I ran it in parallel across 5 nodes (10 processors) with > openmpi-1.3.3 and Intel v11 compiles/MKL 10.2. It appears that the hard > drive on the last node becomes read-only and QE can no longer write to the > local wavefunction files. > > After the run, the local scratch drive remains read only and I end up > having to reboot the system to eliminate this problem. I have been able to > reproduce this problem on the same node. However, when I remove this node > from the list, QE runs fine. Also, running QE with mpich2 doesn't have a > problem on that node. > > I suspect that it could be a hardware issue (harddrive close to dying > perhaps) or an issue with openmpi, but I wanted to check to see anyone else > has run into this problem while using QE. > > For additional technical info, this is on a system with Redhat Enterprise > 4, 2 Xeon processors (3 GHz), 2GB ram. > > 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 > > > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 28, Issue 58 > **************************************** > -- Sincerely yours Jay -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/7b9d9800/attachment-0001.htm From ttduyle at gmail.com Fri Oct 23 03:28:14 2009 From: ttduyle at gmail.com (Duy Le) Date: Thu, 22 Oct 2009 21:28:14 -0400 Subject: [Pw_forum] Missing Wannier90 in Q-E 4.1.1? In-Reply-To: References: <20091022154321.4af18593@lcpybm> Message-ID: <8974d3b20910221828h722c20bu507e776a4c9632c6@mail.gmail.com> I asked the instructor (sorry I forgot his name) about the missing in the wannier handson section (in Santa Barbara), he told me to download it from wannier.org :) -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. "Men don't need hand to do things" On Thu, Oct 22, 2009 at 9:58 AM, Paolo Giannozzi wrote: > > On Oct 22, 2009, at 15:43 , Miguel Mart?nez wrote: > > > Is there any reason why W90 has been left out? > > it is a bug in the script that produces the pakcages. > Sources of W90 are *.F90, in all other cases are *.f90. > Thank you for reporting this. It will be fixed soon(er or later) > > > In any case, the sources from q-e 4.1 seem to work fine. > > they are the same: nothing has changed in W90 recently > > Paolo > --- > 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 mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091022/f583eefc/attachment.htm From giannozz at democritos.it Fri Oct 23 09:33:15 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 23 Oct 2009 09:33:15 +0200 Subject: [Pw_forum] problems calculating large numbers of empty states In-Reply-To: References: Message-ID: On Oct 22, 2009, at 21:29 , Brad Malone wrote: > Hi, I'm trying to calculate a large number of conduction band > states for cubic Si (to be used in a subsequent GW calculation). > However, when doing this I sometimes run into errors like > "problems computing cholesky decomposition" or "too many > bands are unconverged". the algorithm used in PWscf is based on the assumption that the number of Kohn-Sham states (aka bands) is much smaller than the number of basis functions (i.e. plane waves). If this assumption doesn't hold (and it doesn't if you calculate 500 bands in fcc Si unless you use a very large cutoff) you may run into trouble. Moreover iterative diagonalization becomes slower and more memory-consuming than conventional diagonalization. If you really need to do that, you need to modify the code Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From yuminqian at gmail.com Fri Oct 23 12:35:57 2009 From: yuminqian at gmail.com (Y.M.Qian) Date: Fri, 23 Oct 2009 18:35:57 +0800 Subject: [Pw_forum] electronic-phonon coupling error References: Message-ID: <200910231835547506316@gmail.com> Hello, PWSCF users I meet a prolem during the calculation of electronic-phonon coupling coefficient First ,I do the 'scf' calculation with dense mesh 16X16X16,and switch on the la2F='true' to get the **.a2Fsave file Second, I do the 'scf' calulation with spare mesh 8x8x8. Third , I do the phonon calculation with a q-mesh 4x4x4 this is the input file for scf &control calculation='scf' restart_mode='from_scratch', pseudo_dir = '~/software/espresso-4test/pseudo', outdir='./outscf' nstep=60 prefix='nm-type' tstress = .true. tprnfor = .true. wf_collect=.true. / &system ibrav = 1, nat= 5, ntyp= 3, celldm(1)=7.20415879, ecutwfc = 45.0, ecutrho = 350.0, occupations='smearing', smearing='methfessel-paxton', degauss=0.02, / &electrons diagonalization='david', conv_thr = 1.0e-7, mixing_beta = 0.5, / ........... this is the input file for phonon and electron-phonon coupling &inputph tr2_ph=1.0d-12, amass(1)=87.62, amass(2)=51.9961, amass(3)=15.9994, prefix='nm-type', outdir='./outscf', fildyn='nm-type.dyn', alpha_mix(1)=0.4, fildvscf='scdv', elph=.true., trans=.true., ldisp=.true., nq1=4, nq2=4, nq3=4 / But the output give an errror Mode symmetry, O_h (m-3m) point group: omega( 1 - 3) = 73.3 [cm-1] --> T_1u G_15 G_4- I omega( 4 - 6) = 183.3 [cm-1] --> T_1u G_15 G_4- I omega( 7 - 9) = 302.3 [cm-1] --> T_1u G_15 G_4- I omega( 10 - 12) = 322.1 [cm-1] --> T_2u G_25 G_5- omega( 13 - 15) = 541.4 [cm-1] --> T_1u G_15 G_4- I ************************************************************************** electron-phonon interaction ... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from lint : error # 1 cannot locate k point xk %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... p0_30378: p4_error: : 0 p0_30378: (12095.527344) net_send: could not write to fd=4, errno = 32 Is anyone know what is wrong with me ? I use the version 4.1.1 package. many thanks / Y.M.Qian -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091023/0b74f69f/attachment.htm From jjr19 at zips.uakron.edu Fri Oct 23 14:25:42 2009 From: jjr19 at zips.uakron.edu (James J Ramsey) Date: Fri, 23 Oct 2009 08:25:42 -0400 Subject: [Pw_forum] Pw_forum Digest, Vol 28, Issue 58Re: Problem on compiling pwscf In-Reply-To: References: Message-ID: <46e014700910230525o18e2c1bex4b351bd7ad606aed@mail.gmail.com> On Thu, Oct 22, 2009 at 8:58 PM, JAY wrote:> Dear Paolo > I tried to update the gcclib and gcc, it > seems work well for other packages, like BLAS, fftw, however, when compile > lapack-3.2.1, it reports > > I guess there must be something wrong with the compiler, (maybe ifort). You are dealing with gcc, and thus presumably gfortran as well, yet you are blaming ifort, a proprietary Fortran compiler that isn't part of the GCC project at all? You aren't making much sense here. From marzari at MIT.EDU Fri Oct 23 15:31:59 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Fri, 23 Oct 2009 09:31:59 -0400 Subject: [Pw_forum] problems calculating large numbers of empty states In-Reply-To: References: Message-ID: <4AE1B04F.2050206@mit.edu> Paolo Giannozzi wrote: > On Oct 22, 2009, at 21:29 , Brad Malone wrote: > >> Hi, I'm trying to calculate a large number of conduction band >> states for cubic Si (to be used in a subsequent GW calculation). >> However, when doing this I sometimes run into errors like >> "problems computing cholesky decomposition" or "too many >> bands are unconverged". > > the algorithm used in PWscf is based on the assumption that > the number of Kohn-Sham states (aka bands) is much smaller > than the number of basis functions (i.e. plane waves). If this > assumption doesn't hold (and it doesn't if you calculate 500 > bands in fcc Si unless you use a very large cutoff) you may > run into trouble. Moreover iterative diagonalization becomes > slower and more memory-consuming than conventional > diagonalization. If you really need to do that, you need to > modify the code Hi Paolo, Brad, could one use CP instead - i.e. are the files cross-readable between the codes ? If they were, could one do a calculation with CP and 4 valence bands, all full, and then restart that calculations requesting 500 bands, frozen Hamiltonian, and all bands filled. Then, a damped dynamics or conjugate gradients should converge all those 500 filled bands, in the frozen/correct Hamiltonian, to the ground state ? nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From giannozz at democritos.it Fri Oct 23 15:54:49 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 23 Oct 2009 15:54:49 +0200 Subject: [Pw_forum] problems calculating large numbers of empty states In-Reply-To: <4AE1B04F.2050206@mit.edu> References: <4AE1B04F.2050206@mit.edu> Message-ID: <095514CF-FD86-485E-8606-7565155E36BA@democritos.it> On Oct 23, 2009, at 15:31 , Nicola Marzari wrote: > could one use CP instead - i.e. are the files cross-readable > between the codes ? files are cross-readable (or at least, they were the last time I tried), but 1) CP can only read Gamma-point wavefunctions, 2) CP doesn't write the charge density (required by PWscf) unless variable disk_io is set to 'high' > P. -- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Fri Oct 23 16:02:36 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 23 Oct 2009 16:02:36 +0200 Subject: [Pw_forum] electronic-phonon coupling error In-Reply-To: <200910231835547506316@gmail.com> References: <200910231835547506316@gmail.com> Message-ID: On Oct 23, 2009, at 12:35 , Y.M.Qian wrote: > Is anyone know what is wrong with me ? with you, nothing that I know. With your data: did you by any chance use "translated" k-points grid (e.g. 16 16 16 1 1 1) ? you should use grids centered in k- (e.g. 16 16 16 0 0 0) P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From rezaebraahimi at yahoo.com Sat Oct 24 14:39:20 2009 From: rezaebraahimi at yahoo.com (Reza Ebraahimi) Date: Sat, 24 Oct 2009 05:39:20 -0700 (PDT) Subject: [Pw_forum] gap in the phonon DOS Message-ID: <422079.29576.qm@web59408.mail.ac4.yahoo.com> Hi I have a question. Does the existence of a gap in the high-energy region of the phonon DOS means ionicity of the crystal (splitting of the TO and LO-like phonons)? The higher the gap, means the more ionicity? Thank You for any reply Reza Ebraahimi- Graduate Student Tehran University? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091024/26f9be6f/attachment.htm From nkxirainbow at gmail.com Sat Oct 24 15:45:16 2009 From: nkxirainbow at gmail.com (xirainbow) Date: Sat, 24 Oct 2009 13:45:16 +0000 Subject: [Pw_forum] gap in the phonon DOS In-Reply-To: <422079.29576.qm@web59408.mail.ac4.yahoo.com> References: <422079.29576.qm@web59408.mail.ac4.yahoo.com> Message-ID: <21fbc4790910240645g39250b05ld172497fe89f4eaa@mail.gmail.com> Dear Reza Ebraahimi: I have a question. Does the existence of a gap in the high-energy region of > the phonon DOS means ionicity of the crystal (splitting of the TO and > LO-like phonons)? The higher the gap, means the more ionicity? > I do not think so. You can see the "one-dimensional diatomic chain model" as given in written by Max Born and Kun Huang. ____________________________________ Hui Wang School of physics, Nankai University, Tianjin, China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091024/9f20ea2f/attachment.htm From eyvaz_isaev at yahoo.com Sat Oct 24 16:31:10 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sat, 24 Oct 2009 07:31:10 -0700 (PDT) Subject: [Pw_forum] gap in the phonon DOS In-Reply-To: <422079.29576.qm@web59408.mail.ac4.yahoo.com> Message-ID: <292710.96638.qm@web65707.mail.ac4.yahoo.com> Dear Reza, --- On Sat, 10/24/09, Reza Ebraahimi wrote: > Does the existence of a gap in the high-energy region of the phonon DOS > means ionicity of the crystal No, does not. It means just the mass difference, i.e. one of compound componenet is rather light, than other(s), as examlpe, NiAl, FeAl, FeH, PdH, etc. > (splitting of the TO and LO-like phonons)? This one is more relevant for the ionicity issue. This happens in polar semiconductors/insulatoras. Another excitnig example is gamma-boron phase where a cage of B-atoms (icosahedra, B12) is charged negatively, but dumbbell B-atoms B2 are charged positively giving rise for LO-TO splitting. > The higher the gap, means the more ionicity? The gap depends on both the Born effective charge and static dielectric constant (see any textbook, or Rev.Mod.Phys. 73, 515, 2002, see page 527). So, the gap is the result of the competition of these parameters. I suppose analyzing this relation for yourself will give you more insight. Hope this helps. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From forum.san at gmail.com Sun Oct 25 09:55:14 2009 From: forum.san at gmail.com (Sangamesh B) Date: Sun, 25 Oct 2009 14:25:14 +0530 Subject: [Pw_forum] Espresso-4.1 failing to compile Message-ID: Dear Espresso users, I'm trying to compile espresso-4.1.tar.gz with Intel compilers and Openmpi. But during make all, its giving following error: # make all test -d bin || mkdir bin if test -d iotk ; then \ ( cd iotk ; if test "make" = "" ; then make TLDEPS= lib+util ; \ else make TLDEPS= lib+util ; fi ) ; fi make[1]: Entering directory `/opt/apps/espresso-4.1/iotk' cd src ; make lib+util make[2]: Entering directory `/opt/apps/espresso-4.1/iotk/src' ( cd ../../bin ; ln -sf ../iotk/src/iotk_print_kinds.x . ) ( cd ../../bin ; ln -sf ../iotk/src/iotk.x . ) ( cd ../../bin ; ln -sf ../iotk/tools/iotk . ) make[2]: Leaving directory `/opt/apps/espresso-4.1/iotk/src' make[1]: Leaving directory `/opt/apps/espresso-4.1/iotk' ( cd Modules ; if test "make" = "" ; then make TLDEPS= all ; \ else make TLDEPS= all ; fi ) make[1]: Entering directory `/opt/apps/espresso-4.1/Modules' make[1]: Nothing to be done for `all'. make[1]: Leaving directory `/opt/apps/espresso-4.1/Modules' ( cd clib ; if test "make" = "" ; then make TLDEPS= all ; \ else make TLDEPS= all ; fi ) make[1]: Entering directory `/opt/apps/espresso-4.1/clib' icc -O3 -xT -D__INTEL -D__FFTW -D__MPI -D__PARA -I../include -c stack.c stack.c(16): error: function returning function is not allowed void F77_FUNC_(remove_stack_limit,REMOVE_STACK_LIMIT) (void) { ^ stack.c(16): warning #1193: standard requires that parameter "remove_stack_limit" be given a type by a subsequent declaration ("int" assumed) void F77_FUNC_(remove_stack_limit,REMOVE_STACK_LIMIT) (void) { ^ stack.c(16): warning #1193: standard requires that parameter "REMOVE_STACK_LIMIT" be given a type by a subsequent declaration ("int" assumed) void F77_FUNC_(remove_stack_limit,REMOVE_STACK_LIMIT) (void) { ^ compilation aborted for stack.c (code 2) make[1]: *** [stack.o] Error 2 make[1]: Leaving directory `/opt/apps/espresso-4.1/clib' make: *** [libs] Error 2 This looks like a coding problem. Let me know if any of you faced this error and how this can be fixed? Thanks, Sangamesh -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091025/17a21e4b/attachment.htm From yuminqian at gmail.com Sun Oct 25 13:14:05 2009 From: yuminqian at gmail.com (Y.M.Qian) Date: Sun, 25 Oct 2009 20:14:05 +0800 Subject: [Pw_forum] electronic-phonon coupling error References: , <200910231835547506316@gmail.com>, Message-ID: <200910252014024844598@gmail.com> Thanks for your reply . Yes , I used the k-point grid (16,16,16,1,1,1) So I changed it following your suggestion But there is still mistake and the output is Atomic displacements: There are 10 irreducible representations Representation 1 2 modes - To be done Representation 2 2 modes - To be done Representation 3 2 modes - To be done Representation 4 1 modes - To be done Representation 5 2 modes - To be done Representation 6 1 modes - To be done Representation 7 1 modes - To be done Representation 8 1 modes - To be done Representation 9 1 modes - To be done Representation 10 2 modes - To be done %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from openfilq : error # 1 file nm-type.wfc not found %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 2009-10-25 Y.M.Qian ???? Paolo Giannozzi ????? 2009-10-23 22:03:01 ???? PWSCF Forum ??? ??? Re: [Pw_forum] electronic-phonon coupling error On Oct 23, 2009, at 12:35 , Y.M.Qian wrote: > Is anyone know what is wrong with me ? with you, nothing that I know. With your data: did you by any chance use "translated" k-points grid (e.g. 16 16 16 1 1 1) ? you should use grids centered in k- (e.g. 16 16 16 0 0 0) P. --- 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 mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091025/1928661f/attachment.htm From anvari_meh at physics.iust.ac.ir Sun Oct 25 14:10:44 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Sun, 25 Oct 2009 16:40:44 +0330 Subject: [Pw_forum] about elph calculations Message-ID: Hi I have a question about elph calculations in graphene. well, in graphene the electronic density of state at fermi surface is zero, so it could be a problem during elph calculations. However according to Lazzeri's paper(PRL,vol93,2004)these equations could be restated to solve this problem. Now I want to know whether this correction has been implemented in PWscf package or not.I've done elph calculations for graphene and apperantly no problems exist, I mean up to now I've not got any problem, but I want to become sure. Best Mehrnaz Anvari Iran university of science & technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091025/bbafd1b4/attachment.htm From rezaebraahimi at yahoo.com Sun Oct 25 15:43:43 2009 From: rezaebraahimi at yahoo.com (Reza Ebraahimi) Date: Sun, 25 Oct 2009 07:43:43 -0700 (PDT) Subject: [Pw_forum] gap in the phonon DOS In-Reply-To: <292710.96638.qm@web65707.mail.ac4.yahoo.com> References: <292710.96638.qm@web65707.mail.ac4.yahoo.com> Message-ID: <840587.62388.qm@web59415.mail.ac4.yahoo.com> Hi Thanks for your reply What i meant to say was, does not the ionicity (Born charges and ..) of the crystal, which splits the TO-LO phonons, show itself as a gap in high-energy region of the phonon DOS? If not, what is the sign of this splitting?in the phonon dos diagram? Thanks again R. Ebraahimi Grad. Student- Tehran Uni. ________________________________ > Does the existence of a gap in the high-energy region of the phonon DOS > means ionicity of the? crystal No, does not. It means just the mass difference, i.e. one of compound componenet is rather light, than other(s), as examlpe, NiAl, FeAl, FeH, PdH, etc. > (splitting of the TO and LO-like phonons)? This one is more relevant for the ionicity issue. This happens in polar semiconductors/insulatoras. Another excitnig example is gamma-boron phase where a cage of B-atoms (icosahedra, B12) is charged negatively, but dumbbell B-atoms B2 are charged positively giving rise for LO-TO splitting. > The higher the gap, means the more ionicity? The gap depends on both? the Born effective charge and static dielectric constant (see any textbook, or Rev.Mod.Phys. 73, 515, 2002, see page 527). So, the gap is the result of the competition of these parameters.? I suppose analyzing this relation for yourself will give you more insight.? Hope this helps. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com ? ? ? _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091025/3df80cac/attachment-0001.htm From baroni at sissa.it Sun Oct 25 15:51:32 2009 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 25 Oct 2009 15:51:32 +0100 Subject: [Pw_forum] gap in the phonon DOS In-Reply-To: <840587.62388.qm@web59415.mail.ac4.yahoo.com> References: <292710.96638.qm@web65707.mail.ac4.yahoo.com> <840587.62388.qm@web59415.mail.ac4.yahoo.com> Message-ID: On Oct 25, 2009, at 3:43 PM, Reza Ebraahimi wrote: > > Hi > Thanks for your reply > What i meant to say was, does not the ionicity (Born charges and ..) > of the crystal, which splits the TO-LO phonons, show itself as a gap > in high-energy region of the phonon DOS? it may, or may not, according to the relative magnitude of the LO-TO splitting vs. band-width > If not, what is the sign of this splitting in the phonon dos diagram? LO phonons have a higher frequency Notice, however, that in low-simmetry crystals the distinction between transferse and longitudinal modes is not as meaningful as, say, in cubic crystals, because long-wavelengths phonons are neither purely transverse, nor longitudinal Stefano B > Thanks again > > R. Ebraahimi > Grad. Student- Tehran Uni. > > Does the existence of a gap in the high-energy region of the > phonon DOS > means ionicity of the crystal > > No, does not. It means just the mass difference, i.e. one of > compound componenet is rather light, than other(s), as examlpe, > NiAl, FeAl, FeH, PdH, etc. > > > (splitting of the TO and LO-like phonons)? > > This one is more relevant for the ionicity issue. This happens in > polar semiconductors/insulatoras. Another excitnig example is gamma- > boron phase where a cage of B-atoms (icosahedra, B12) is charged > negatively, but dumbbell B-atoms B2 are charged positively giving > rise for LO-TO splitting. > > > The higher the gap, means the more ionicity? > > The gap depends on both the Born effective charge and static > dielectric constant (see any textbook, or Rev.Mod.Phys. 73, 515, > 2002, see page 527). So, the gap is the result of the competition of > these parameters. > I suppose analyzing this relation for yourself will give you more > insight. > > Hope this helps. > > Bests, > Eyvaz. > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & > Alloys, Russia, > Department of Physics, Chemistry, and Biology (IFM), Linkoping > University, Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > _______________________________________________ > 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/20091025/03601ca9/attachment.htm From marzari at MIT.EDU Sun Oct 25 15:55:54 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Sun, 25 Oct 2009 10:55:54 -0400 Subject: [Pw_forum] gap in the phonon DOS In-Reply-To: <840587.62388.qm@web59415.mail.ac4.yahoo.com> References: <292710.96638.qm@web65707.mail.ac4.yahoo.com> <840587.62388.qm@web59415.mail.ac4.yahoo.com> Message-ID: <4AE466FA.9020602@mit.edu> Reza Ebraahimi wrote: > > If not, what is the sign of this > splitting in the phonon dos diagram? > Thanks again You could also calculate the DOS with and without the non-analytic contribution (not sure what's the simplest way - worse case scenario, just put the effective charges to zero in the gamma file). nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From eyvaz_isaev at yahoo.com Sun Oct 25 22:17:12 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 25 Oct 2009 14:17:12 -0700 (PDT) Subject: [Pw_forum] gap in the phonon DOS In-Reply-To: <840587.62388.qm@web59415.mail.ac4.yahoo.com> Message-ID: <663097.85462.qm@web65709.mail.ac4.yahoo.com> Hi, --- On Sun, 10/25/09, Reza Ebraahimi wrote: > Thanks for your reply Welcome! > What i meant to say was, does not the ionicity (Born charges and ..) of > the crystal, which splits the TO-LO phonons, show itself as a gap in > high-energy region of the phonon DOS? Sorry, I am not sure I understood this sentence properly. And I suppose you mixup two quanities: gap and splitting. 1. A gap for high-energy optical modes is related to the mass difference. It usually occures along high symmetry directions. In fact, in some cases it might be related to specific atomic vibrations (see below). 2. LO-TO splitting occurs ONLY!!! at the Gamma point. It is always connected to dipols in the crystal (and respectively, to Born effective charges and macroscopic dielectric constant) For both cases, again, Gamma-Boron is a good example. Because it has three optical modes which are separaed from others, though, all B atoms have the same mass. Another example (for mass difference case) is MeH. Besides, LO-TO splitting takes place for B28 at low frequency range (but optical modes again). For MgO LO-TO takes place, but there is no gap in the phonon dispersion relations. > If not, what is the sign of this splitting?in the phonon dos diagram? This one is quite difficult to predict apriori, as it might be different for different systems. In fact, there will be some re-distribution of phonon DOS. For a system of interest just follow Nicola's advice. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From giannozz at democritos.it Mon Oct 26 08:21:44 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 26 Oct 2009 08:21:44 +0100 Subject: [Pw_forum] Espresso-4.1 failing to compile In-Reply-To: References: Message-ID: On Oct 25, 2009, at 9:55 , Sangamesh B wrote: > icc -O3 -xT -D__INTEL -D__FFTW -D__MPI -D__PARA -I../include -c > stack.c > stack.c(16): error: function returning function is not allowed > void F77_FUNC_(remove_stack_limit,REMOVE_STACK_LIMIT) (void) { > in include/c_defs.h, there should be two lines (produced by "configure") that look like #define F77_FUNC(name,NAME) name ## _ #define F77_FUNC_(name,NAME) name ## _ P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Mon Oct 26 09:32:42 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 26 Oct 2009 09:32:42 +0100 Subject: [Pw_forum] electronic-phonon coupling error In-Reply-To: <200910252014024844598@gmail.com> References: , <200910231835547506316@gmail.com>, <200910252014024844598@gmail.com> Message-ID: <4AE55EAA.8020200@democritos.it> Y.M.Qian wrote: > But there is still mistake > from openfilq : error # 1 > file nm-type.wfc not found it is not there. Check your input data. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From mohnish.iitk at gmail.com Mon Oct 26 14:53:36 2009 From: mohnish.iitk at gmail.com (mohnish pandey) Date: Mon, 26 Oct 2009 19:23:36 +0530 Subject: [Pw_forum] calculating equilibrium lattice constant Message-ID: I am beginner of quantum espresso. I am trying to calculate the lattice constant for zinc oxide graphitic and wurtzitic structure . I am using a script for that. But I have to see each and every file to check the energy and to decide which one is minimum. Can you guys please help me to get a shortcut for this I mean after running a loop I get a minimum energy file instead of doing it manually and then deciding which lattice parammeters fits to minimum energy.. Thanks in advance. Mohnish Pandey -- Mohnish Pandey Y6262,4th Year Undergraduate, Department of Chemical Engineering, IIT KANPUR -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091026/d40da90d/attachment.htm From marzari at MIT.EDU Mon Oct 26 15:00:25 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 26 Oct 2009 10:00:25 -0400 Subject: [Pw_forum] calculating equilibrium lattice constant In-Reply-To: References: Message-ID: <4AE5AB79.5080308@mit.edu> mohnish pandey wrote: > I am beginner of quantum espresso. I am trying to calculate the > lattice constant for zinc oxide graphitic and wurtzitic structure . I > am using a script for that. But I have to see each and every file to > check the energy and to decide which one is minimum. Can you guys > please help me to get a shortcut for this I mean after running a loop > I get a minimum energy file instead of doing it manually and then > deciding which lattice parammeters fits to minimum energy.. > > Thanks in advance. > Mohnish Pandey Dear Mohnish, Use the linux command 'grep' to extract from a file keywords like 'total energy'; in particular the converged one has an exclamation mark '!'. so, grep '!' ZnO_out* would work, where ZnO_out* includes all the files whose name starts with ZnO_out. Familiarize yourself with the 'awk' command, with the program 'xmgrace', and have a close look at all examples in media.quantum-espresso.org and ocw.mit.edu/OcwWeb/Materials-Science-and-Engineering/3-320Spring-2005/CourseHome/ (before posting again). Best luck ! nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From hande at newton.physics.metu.edu.tr Mon Oct 26 15:00:25 2009 From: hande at newton.physics.metu.edu.tr (Hande Ustunel) Date: Mon, 26 Oct 2009 16:00:25 +0200 (EET) Subject: [Pw_forum] calculating equilibrium lattice constant In-Reply-To: Message-ID: Dear Mohnish If it is a shell script that you are running you might try something like for file in out.* do E=`grep ! $file | tail -1 | awk '{print $5}'` echo E >> energies-file done This is just a snippet you can place in your script as required. Please do be careful about the way the quotation marks are pointing. Best wishes, Hande On Mon, 26 Oct 2009, mohnish pandey wrote: > I am beginner of quantum espresso. I am trying to calculate the lattice > constant for zinc oxide graphitic and wurtzitic structure . I am using a > script for that. But I have to see each and every file to check the energy > and to decide which one is minimum. Can you guys please help me to get a > shortcut for this I mean after running a loop I get a minimum energy file > instead of doing it manually and then deciding which lattice parammeters > fits to minimum energy.. > > Thanks in advance. > Mohnish Pandey > > -- Hande Toffoli Department of Physics Office 439 Middle East Technical University Ankara 06531, Turkey Tel : +90 312 210 3264 http://www.physics.metu.edu.tr/~hande From giovanni.cantele at na.infn.it Mon Oct 26 15:11:09 2009 From: giovanni.cantele at na.infn.it (Giovanni Cantele) Date: Mon, 26 Oct 2009 15:11:09 +0100 Subject: [Pw_forum] calculating equilibrium lattice constant In-Reply-To: References: Message-ID: On Oct 26, 2009, at 2:53 PM, mohnish pandey wrote: > I am beginner of quantum espresso. I am trying to calculate the > lattice constant for zinc oxide graphitic and wurtzitic structure . > I am using a script for that. But I have to see each and every file > to check the energy and to decide which one is minimum. Can you guys > please help me to get a shortcut for this I mean after running a > loop I get a minimum energy file instead of doing it manually and > then deciding which lattice parammeters fits to minimum energy.. > > Thanks in advance. > Mohnish Pandey - to extract the final energy of a scf loop just add, at the end of each step of your script loop grep ^! fileout | tail -1 | awk '{print $5}' or, if you want to store it into a variable etot etot=$(grep ^! fileout | tail -1 | awk '{print $5}') (the "tail -1" is maybe reduntant if you're using version >= 4.0) - to find the minimum, you can make parabolic fits of etot vs lattice_parameter or, maybe better, use and equation of state. Fort the latter, a simple utility is available within the QE package, read the header of pwtools/ev.f90 (useful for cubic / hexagonal systems). Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it Skype contact: giocan74 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091026/45175c90/attachment.htm From paulatto at sissa.it Mon Oct 26 15:28:29 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 26 Oct 2009 15:28:29 +0100 Subject: [Pw_forum] calculating equilibrium lattice constant In-Reply-To: References: Message-ID: In data 26 ottobre 2009 alle ore 15:00:25, Hande Ustunel ha scritto: > If it is a shell script that you are running you might try something like I don't want to be picky, but there was a missing "$", this is a correct script: for file in out.* do E=`grep ! $file | tail -1 | awk '{print $5}'` echo $E >> energies-file done If you only want to find the energy minimum you can just do a vc-relax calculation, which will do all the work for you. It will not give you the energy/volume curve though, and it may relax more than you expect (e.g. moving some parameter that you wanted to keep fixed). regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From hande at newton.physics.metu.edu.tr Mon Oct 26 16:15:47 2009 From: hande at newton.physics.metu.edu.tr (Hande Ustunel) Date: Mon, 26 Oct 2009 17:15:47 +0200 (EET) Subject: [Pw_forum] calculating equilibrium lattice constant In-Reply-To: Message-ID: > I don't want to be picky, but there was a missing "$", this is a correct > script: > > for file in out.* > do > E=`grep ! $file | tail -1 | awk '{print $5}'` > echo $E >> energies-file > done > Absolutely. Thanks, Lorenzo. > If you only want to find the energy minimum you can just do a vc-relax > calculation, which will do all the work for you. It will not give you the > energy/volume curve though, and it may relax more than you expect (e.g. > moving some parameter that you wanted to keep fixed). One last comment on this is that vc-relax is indeed very handy but it can also be a little bit tricky. If you want to try it out you might want to try many different starting conditions and parameters and make sure they all end up in the same final configuration. Best wishes, Hande -- Hande Toffoli Department of Physics Office 439 Middle East Technical University Ankara 06531, Turkey Tel : +90 312 210 3264 http://www.physics.metu.edu.tr/~hande From daijiayu at nudt.edu.cn Mon Oct 26 17:12:26 2009 From: daijiayu at nudt.edu.cn (Jiayu Dai) Date: Tue, 27 Oct 2009 00:12:26 +0800 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valence electrons) Message-ID: <456573546.16447@nudt.edu.cn> Dear Users, I met a problem when i tried to construct a NCPP of Al using ld1. In this PP, the 2s and 2p orbitals should be considered as valence orbitals. I used the input file below, but the generated PP did not work well according to my test for the crystal constal and bulk modulus for Al crystal. I found the energies of orbitals in PPs are not equal to the energies of all electron calculation. We can find the information in the output file of ld1.x: n l nl e AE (Ry) e PS (Ry) De AE-PS (Ry) 1 0 2S 1( 2.00) -7.94999 -6.17363 -1.77636 2 1 2P 1( 6.00) -5.11868 -3.82463 -1.29405 2 0 3S 1( 2.00) -0.56982 -0.49211 -0.07771 3 1 3P 1( 1.00) -0.19934 -0.16136 -0.03798 It is shown that the difference is so large. Lorenzo suggested to add more projectors of 3S and 3P, but the problem seems to be there. So, how can i solve this problem? Any suggestion is much appreciated. The input file: &input title='Al', zed=13.0, rel=1, config='[He] 2s2 2p6 3s2 3p1 3d-2' iswitch=3, dft='PBE', / &inputp lloc=2, pseudotype=2, nlcc = .true. tm = .true. , file_pseudopw='Al.pbe-tm-nc.UPF', / 5 2S 1 0 2.00 0.00 2.00 2.00 2P 2 1 6.00 0.00 2.00 2.00 3S 2 0 2.00 0.00 2.40 2.40 3P 3 1 1.00 0.00 2.70 2.70 3D 5 2 -2.00 0.25 2.30 2.30 ------------------------------ ------------------------------------------- Jiayu Dai Department of Physics National University of Defense Technology, Changsha, 410073, P R China ----------------------------------------- From siyouber at yahoo.fr Mon Oct 26 18:03:00 2009 From: siyouber at yahoo.fr (Bertrand SITAMTZE) Date: Mon, 26 Oct 2009 17:03:00 +0000 (GMT) Subject: [Pw_forum] System size Message-ID: <284142.43693.qm@web26503.mail.ukl.yahoo.com> Dears Users, I have a question that seems to be simple, but the answer is of great interest for me and I would like to have the correct: What is the size of the system in QE calculations (1 unit cell, 2, 3, ....)? Thanking you all for your help ********************************** Bertrand SITAMTZE YOUMBI Laboratory of Material Sciences Department of physics University of Yaounde I-Cameroon ************************************* -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091026/430594e0/attachment-0001.htm From paulatto at sissa.it Mon Oct 26 18:20:31 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 26 Oct 2009 18:20:31 +0100 Subject: [Pw_forum] System size In-Reply-To: <284142.43693.qm@web26503.mail.ukl.yahoo.com> References: <284142.43693.qm@web26503.mail.ukl.yahoo.com> Message-ID: In data 26 ottobre 2009 alle ore 18:03:00, Bertrand SITAMTZE ha scritto: > What is the size of the system in QE calculations (1 unit cell, 2, 3, > ....)? Dear Betrand, most of Quantum-ESPRESSO programs and utilities use a plane-wave basis set and periodically boundary conditions: the size of the system is always formally infinite. The ions are repeats infinite times whatever you put in the unit cell. The periodicity of electronic wavefunctions is a slightly different story which depends on the number and positions of the k-points. It is a bit long to explain. I can go in details, if necessary (although not tonight for lack of time), but it would be better for you to study it off some solid-state book first and ask for the necessary clarifications. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From marcello.rosini at unimore.it Mon Oct 26 18:24:54 2009 From: marcello.rosini at unimore.it (Marcello Rosini) Date: Mon, 26 Oct 2009 18:24:54 +0100 Subject: [Pw_forum] vc-relax bug? Message-ID: <4AE5DB66.4000704@unimore.it> Hallo everybody I am using pw.x version 4.1 and I noticed a very strange behaviour when performing vc-relax calculation. I have done many tests, and I have reduced to the simplest case possible wich I illustrate here: InAs fcc lattice: * the lattice parameter scan and the murnaghan fit lead me to 11.177au lattice constant with total energy -16.91675895 Ry *starting from 11.177 fcc lattice I perform a vc-relax, and I find a new unit cell with ~0.98% smaller lattice constant. All the symmetries are preseved and the InAs distance is still (1/4,1/4,1/4) of the new lattice parameter. Total energy is now -16.91886930 Ry *I run a new scf calculation with the new lattice parameters and now I get Etot=-16.91485951... obviously higher, as it was or the murnaghnan calculation. the question is: is there something wrong in the vc-relax procedure? (you can find my input in the following) thank you Marcello ----------------------- &control calculation = 'vc-relax' restart_mode='from_scratch', nstep=100, pseudo_dir = '~/AB_INITIO/PSEUDO/LDA', outdir='./' etot_conv_thr=1.0e-7, forc_conv_thr=1.0e-5, / &system ibrav=2, celldm(1)=11.177, nat=2, ntyp=2, ecutwfc =15, / &electrons electron_maxstep = 300, diagonalization='david' mixing_mode = 'plain' mixing_beta = 0.7 conv_thr = 1.0e-8, / &IONS ion_dynamics='bfgs', / &CELL cell_dynamics='bfgs', / ATOMIC_SPECIES In 114.818 in.cpi.UPF As 74.922 As.pz-bhs.UPF ATOMIC_POSITIONS In 0.00 0.00 0.00 As 0.25 0.25 0.25 K_POINTS automatic 6 6 6 1 1 1 -- ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Dipartimento di Fisica, Universit? di Modena e Reggio Emilia and S3 National Research Centre of CNR-INFM via campi 213/a - 41125 Modena - Italy tel: +39 059.205.5067 email: marcello.rosini at unimore.it www.nanomodelling.unimore.it ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ From degironc at sissa.it Mon Oct 26 18:45:30 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 26 Oct 2009 18:45:30 +0100 Subject: [Pw_forum] vc-relax bug? In-Reply-To: <4AE5DB66.4000704@unimore.it> References: <4AE5DB66.4000704@unimore.it> Message-ID: <4AE5E03A.8020502@sissa.it> Dear Marcello Rosini, vc-relax uses the information on forces and STRESS to search for the minimum configuration. The stress is calculated as the derivative of the total nergy w.r.t. cell dimension AT FIXED NUMBER PW and in the evolution of the system the same number of PW is kept. When you minimize by a muraghan fit you are instead comparing energies calculated at FIXED CUTOFF . This is why you obtain different results. Obviously for a well converged calculation derivatives at fixed number of PW or fixed cutoff should agree but 15 Ry is likely insufficient for converging stress calculation in InAs. stefano Marcello Rosini wrote: > Hallo everybody > I am using pw.x version 4.1 and I noticed a very strange behaviour when > performing vc-relax calculation. I have done many tests, and I have > reduced to the simplest case possible wich I illustrate here: > > InAs fcc lattice: > * the lattice parameter scan and the murnaghan fit lead me to 11.177au > lattice constant with total energy -16.91675895 Ry > *starting from 11.177 fcc lattice I perform a vc-relax, and I find a new > unit cell with ~0.98% smaller lattice constant. All the symmetries are > preseved and the InAs distance is still (1/4,1/4,1/4) of the new lattice > parameter. Total energy is now -16.91886930 Ry > *I run a new scf calculation with the new lattice parameters and now I > get Etot=-16.91485951... obviously higher, as it was or the murnaghnan > calculation. > > the question is: > is there something wrong in the vc-relax procedure? (you can find my > input in the following) > > thank you > Marcello > > ----------------------- > > &control > calculation = 'vc-relax' > restart_mode='from_scratch', > nstep=100, > pseudo_dir = '~/AB_INITIO/PSEUDO/LDA', > outdir='./' > etot_conv_thr=1.0e-7, > forc_conv_thr=1.0e-5, > / > &system > ibrav=2, > celldm(1)=11.177, > nat=2, > ntyp=2, > ecutwfc =15, > / > &electrons > electron_maxstep = 300, > diagonalization='david' > mixing_mode = 'plain' > mixing_beta = 0.7 > conv_thr = 1.0e-8, > / > &IONS > ion_dynamics='bfgs', > / > &CELL > cell_dynamics='bfgs', > / > ATOMIC_SPECIES > In 114.818 in.cpi.UPF > As 74.922 As.pz-bhs.UPF > ATOMIC_POSITIONS > In 0.00 0.00 0.00 > As 0.25 0.25 0.25 > K_POINTS automatic > 6 6 6 1 1 1 > > > From eariel99 at gmail.com Mon Oct 26 19:00:22 2009 From: eariel99 at gmail.com (Eduardo Ariel Menendez Proupin) Date: Mon, 26 Oct 2009 15:00:22 -0300 Subject: [Pw_forum] about potentials given by PP Message-ID: Hello, I have some doubts about the input of the postprocessing code PP Variable: plot_num Type: INTEGER Description: selects what to save in filplot: 0 = charge 1 = total potential V_bare+V_H + V_xc 2 = local ionic potential 11 = the V_bare + V_H potential 12 = the electric field potential Doubt one: "local ionic potential" = V_bare ? Doubt two: " the electric field potential" = V_H? Doubt three "Does V_bare" includes only the local part of the pseudopotentials? Thank you -- Eduardo Menendez Departamento de Fisica Facultad de Ciencias Universidad de Chile Phone: (56)(2)9787439 URL: http://fisica.ciencias.uchile.cl/~emenendez -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091026/74c3a159/attachment.htm From marzari at MIT.EDU Mon Oct 26 18:20:17 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 26 Oct 2009 13:20:17 -0400 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valence electrons) In-Reply-To: <456573546.16447@nudt.edu.cn> References: <456573546.16447@nudt.edu.cn> Message-ID: <4AE5DA51.3000303@mit.edu> Jiayu Dai wrote: > Dear Users, > > I met a problem when i tried to construct a NCPP of Al using ld1. In this PP, the > 2s and 2p orbitals should be considered as valence orbitals. I used the input file > below, but the generated PP did not work well according to my test for the crystal > constal and bulk modulus for Al crystal. I found the energies of orbitals in PPs > are not equal to the energies of all electron calculation. We can find the > information in the output file of ld1.x: > Dear Jiayu, we constructed a ultrasoft 11e pseudopotential for Al (attached) and tested it in Physical Review B 77, 172102 (2008) (if needed, maybe it could be included in the pseudo table, with reference to the paper above). I would be a bit careful with a NCPP and 11e, since you will have only one projector per angular momentum, but the energy of e.g. the 2s and 3s are very different. nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Al_USPBE_11e.van Url: http://www.democritos.it/pipermail/pw_forum/attachments/20091026/dd528f58/attachment-0001.txt From paulatto at sissa.it Mon Oct 26 20:17:55 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 26 Oct 2009 20:17:55 +0100 (CET) Subject: [Pw_forum] about potentials given by PP In-Reply-To: References: Message-ID: <49453.78.12.172.175.1256584675.squirrel@webmail.sissa.it> On Mon, October 26, 2009 19:00, Eduardo Ariel Menendez Proupin wrote: > Doubt one: "local ionic potential" = V_bare ? yes > Doubt two: " the electric field potential" = V_H? yes, but it will include also the sawtooth potential if you are using it (tefield=.true., dipole corrections) > Doubt three "Does V_bare" includes only the local part of the > pseudopotentials? yes, the projectors are non-local operators that cannot be represented as a 3D scalar field. > Thank you you're welcome -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From daijiayu at nudt.edu.cn Tue Oct 27 02:52:33 2009 From: daijiayu at nudt.edu.cn (Jiayu Dai) Date: Tue, 27 Oct 2009 09:52:33 +0800 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valenceelectrons) Message-ID: <456608353.10175@nudt.edu.cn> > >we constructed a ultrasoft 11e pseudopotential for Al (attached) and >tested it in Physical Review B 77, 172102 (2008) (if needed, maybe it >could be included in the pseudo table, with reference to the paper above). > >I would be a bit careful with a NCPP and 11e, since you will have only >one projector per angular momentum, but the energy of e.g. the 2s and >3s are very different. Thanks, nicola. For NCPP, there is really big difficult to generate PPs like this. I read your paper about Al at high pressure. There is something very similar to my job now. The 2s and 2p electrons should be considered in my system. Furthermore, would you please show some details about the PPs you sent? If i want to consider higher orbital such as 3D, i think i should generate new PP. If i understood correctly, the 3D orbital is not included in your PP. Is it? Thanks very much. Jiayu ------------------------------ ------------------------------------------- Jiayu Dai Department of Physics National University of Defense Technology, Changsha, 410073, P R China ----------------------------------------- From marzari at MIT.EDU Tue Oct 27 03:10:50 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 26 Oct 2009 22:10:50 -0400 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valenceelectrons) In-Reply-To: <456608353.10175@nudt.edu.cn> References: <456608353.10175@nudt.edu.cn> Message-ID: <4AE656AA.5060902@mit.edu> Jiayu Dai wrote: > Thanks, nicola. For NCPP, there is really big difficult to generate PPs like this. > I read your paper about Al at high pressure. There is something very similar to my > job now. The 2s and 2p electrons should be considered in my system. Furthermore, > would you please show some details about the PPs you sent? If i want to consider > higher orbital such as 3D, i think i should generate new PP. If i understood > correctly, the 3D orbital is not included in your PP. Is it? > > Thanks very much. > > Jiayu Dear Jiayu, I will send you separately the input files for Vanderbilt's pseudopotential code. The pseudo had 2 projectors for s and p, and one for d, so you should be ok. nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From siyouber at yahoo.fr Tue Oct 27 08:23:53 2009 From: siyouber at yahoo.fr (Bertrand SITAMTZE) Date: Tue, 27 Oct 2009 07:23:53 +0000 (GMT) Subject: [Pw_forum] System size In-Reply-To: Message-ID: <132303.75364.qm@web26507.mail.ukl.yahoo.com> Dear Lorenzo, Thanks very much for this answer. The informations about the ions system was the must important and I got it. Thanks once again ******************************* Bertrand SITAMTZE YOUMBI Laboratory of Material Sciences Department of Physics University of Yaounde I-Cameroon *********************************** --- En date de?: Lun 26.10.09, Lorenzo Paulatto a ?crit?: De: Lorenzo Paulatto Objet: Re: [Pw_forum] System size ?: "PWSCF Forum" Date: Lundi 26 Octobre 2009, 18h20 In data 26 ottobre 2009 alle ore 18:03:00, Bertrand SITAMTZE? ha scritto: > What is the size of the system in QE calculations (1 unit cell, 2, 3,? > ....)? Dear Betrand, most of Quantum-ESPRESSO programs and utilities use a plane-wave basis set? and periodically boundary conditions: the size of the system is always? formally infinite. The ions are repeats infinite times whatever you put in? the unit cell. The periodicity of electronic wavefunctions is a slightly different story? which depends on the number and positions of the k-points. It is a bit? long to explain. I can go in details, if necessary (although not tonight? for lack of time), but it would be better for you to study it off some? solid-state book first and ask for the necessary clarifications. best regards -- Lorenzo Paulatto SISSA? &? DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www:???http://people.sissa.it/~paulatto/ ? ???*** save italian brains *** ? http://saveitalianbrains.wordpress.com/ _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/4bc1892d/attachment.htm From paulatto at sissa.it Tue Oct 27 09:05:52 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 27 Oct 2009 09:05:52 +0100 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valenceelectrons) In-Reply-To: <456608353.10175@nudt.edu.cn> References: <456608353.10175@nudt.edu.cn> Message-ID: In data 27 ottobre 2009 alle ore 02:52:33, Jiayu Dai ha scritto: > If i understood > correctly, the 3D orbital is not included in your PP. Is it? Technically it isn't even in yours! You have specified lloc=2, which means that the empty 3D orbital will be used to generate the local potential, i.e: V(r) = frac{\epsilon \phi(r) - 1/(2m) \nabla^2 \phi(r)}{\phi(r)} where \phi is the 3D orbital and \epsilon its eigen-energy. Since you have specified an arbitrary energy (0.25Ry) for the eigenvalue you have not included eigenvector (provided it be bound) in the pseudopotential. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From giannozz at democritos.it Tue Oct 27 09:42:52 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 27 Oct 2009 09:42:52 +0100 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valence electrons) In-Reply-To: <456573546.16447@nudt.edu.cn> References: <456573546.16447@nudt.edu.cn> Message-ID: <5E4E7453-39DA-4976-BB1C-8B34051922BD@democritos.it> On Oct 26, 2009, at 17:12 , Jiayu Dai wrote: > 2S 1 0 2.00 0.00 2.00 2.00 > 2P 2 1 6.00 0.00 2.00 2.00 Rc=2a.u. for 2s and 2p is way too large. The peak of 2s and 2p orbitals is around 0.7 a.u. P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at democritos.it Tue Oct 27 09:43:07 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 27 Oct 2009 09:43:07 +0100 Subject: [Pw_forum] PPs of Aluminum using ld1 code(including 11 valence electrons) In-Reply-To: <4AE5DA51.3000303@mit.edu> References: <456573546.16447@nudt.edu.cn> <4AE5DA51.3000303@mit.edu> Message-ID: On Oct 26, 2009, at 18:20 , Nicola Marzari wrote: > I would be a bit careful with a NCPP and 11e, since you will have only > one projector per angular momentum not necessarily so: you can have more than one projector per angular momentum in NCPP as well. The problem with the "atomic" code in QE is that the pseudo-orbitalls have to be nodeless. This makes it impossible to choose for instance 2s and 3s as reference orbitals. P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From gregor.mali at ki.si Tue Oct 27 14:08:35 2009 From: gregor.mali at ki.si (Gregor Mali) Date: Tue, 27 Oct 2009 14:08:35 +0100 Subject: [Pw_forum] convergence problems using GIPAW In-Reply-To: <8C0B47760A05FB4AB504C9BCAD32D256012E52AE@postar.ki.si> References: <001901ca4044$efca0b30$cf5e2190$@ac.kr> <8C0B47760A05FB4AB504C9BCAD32D256012E52AE@postar.ki.si> Message-ID: <8C0B47760A05FB4AB504C9BCAD32D256013EB614@postar.ki.si> Dear PWscf/GIPAW users, When doing test calculations with GIPAW module on a Li-Mn-oxide, after relaxing the structure, checking the convergence with respect to ecutwfc (80 or more), convergence threshold (1d-10 to 1d-12), and number of k points (4 4 4 or more), I was able to calculate hyperfine and quadrupolar couplings. However, attempts to calculate NMR shifts and g tensors always resulted in messages that convergence could not be reached. In gipaw the convergence threshold is by default set to 1d-14. And the error or warning messages in my case always mentioned the accuracy of only 1d-5 or 1d-6: Computing the magnetic susceptibility isolve=0 ethr=0.1000E-13 k-point # 1 of 54 ik 1 ibnd 51 linter: root not converged 0.524E-06 ik 1 ibnd 51 linter: root not converged 0.210E-06 ik 1 ibnd 51 linter: root not converged 0.139E-05 ik 1 ibnd 51 linter: root not converged 0.369E-06 ik 1 ibnd 51 linter: root not converged 0.103E-06 ... Can anybody, please, give me some advice on how to circumvent the problem? Thak you very much. Best regards. Gregor Mali Gregor Mali Kemijski institut/National Institute of Chemistry Hajdrihova 19 SI-1001 Ljubljana From wangqj1 at 126.com Tue Oct 27 14:37:42 2009 From: wangqj1 at 126.com (Q.J.Wang) Date: Tue, 27 Oct 2009 21:37:42 +0800 (CST) Subject: [Pw_forum] AFM not equal to 0 Message-ID: <32482005.570431256650662940.JavaMail.coremail@bj126app21.126.com> Dear all When I use 'relax ' to optimize the geometry of a doped system ,but the total magnetization of AFM equal to 1 or -1 Bohr mag/cell ,but not to 0 .I think it should equal to 0. So I want to know whether the result credible or not ? If not credible ,what I should do to improve it ? -- Best regards Q.J.Wang XiangTan University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/3a06a628/attachment.htm From ttduyle at gmail.com Tue Oct 27 14:44:04 2009 From: ttduyle at gmail.com (Duy Le) Date: Tue, 27 Oct 2009 09:44:04 -0400 Subject: [Pw_forum] AFM not equal to 0 In-Reply-To: <32482005.570431256650662940.JavaMail.coremail@bj126app21.126.com> References: <32482005.570431256650662940.JavaMail.coremail@bj126app21.126.com> Message-ID: <8974d3b20910270644v516a7b7s4535d68b4eb16c35@mail.gmail.com> I think it depends on how you dope the system and the structure of the system. If you don't mind, could you pls share the input. -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. "Men don't need hand to do things" 2009/10/27 Q.J.Wang > Dear all > When I use 'relax ' to optimize the geometry of a doped system ,but > the total magnetization of AFM equal to 1 or -1 Bohr mag/cell ,but not to 0 > .I think it should equal to 0. So I want to know whether the result credible > or not ? If not credible ,what I should do to improve it ? > > > -- > Best regards > > Q.J.Wang > > XiangTan University > > > ------------------------------ > ?????????????????3D????????? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/64c338e4/attachment.htm From ceresoli at MIT.EDU Tue Oct 27 14:53:27 2009 From: ceresoli at MIT.EDU (Davide Ceresoli) Date: Tue, 27 Oct 2009 09:53:27 -0400 (EDT) Subject: [Pw_forum] convergence problems using GIPAW In-Reply-To: <8C0B47760A05FB4AB504C9BCAD32D256013EB614@postar.ki.si> References: <001901ca4044$efca0b30$cf5e2190$@ac.kr> <8C0B47760A05FB4AB504C9BCAD32D256012E52AE@postar.ki.si> <8C0B47760A05FB4AB504C9BCAD32D256013EB614@postar.ki.si> Message-ID: On Tue, 27 Oct 2009, Gregor Mali wrote: > Dear PWscf/GIPAW users, > > When doing test calculations with GIPAW module on a Li-Mn-oxide, after > relaxing the structure, checking the convergence with respect to ecutwfc > (80 or more), convergence threshold (1d-10 to 1d-12), and number of k > points (4 4 4 or more), I was able to calculate hyperfine and > quadrupolar couplings. However, attempts to calculate NMR shifts and g > tensors always resulted in messages that convergence could not be > reached. > In gipaw the convergence threshold is by default set to 1d-14. And the > error or warning messages in my case always mentioned the accuracy of > only 1d-5 or 1d-6: Hi Gregor, is your system metallic? Davide > Can anybody, please, give me some advice on how to circumvent the > problem? > > Thak you very much. > > Best regards. > Gregor Mali From mohnish.iitk at gmail.com Tue Oct 27 15:00:25 2009 From: mohnish.iitk at gmail.com (mohnish pandey) Date: Tue, 27 Oct 2009 19:30:25 +0530 Subject: [Pw_forum] geometry optimization Message-ID: Dear users! I am trying to calculate equilibrium lattice constant for wurtzitic and graphitic structure of zinc oxide by 'vc-relax' but this is breaking the symmetry means the final positions of Zn and O are different from where it should be in wurtzitic and graphitic structure. Can anybody please help me how I can fix this problem by preserving the symmetry and getting eqb. lattice constan.? PS: I have gone through many tutorials availble for pwscf, but unluckily i didnt found it. Heartly thanks in advance, Mohnish -- Mohnish Pandey Y6262,4th Year Undergraduate, Department of Chemical Engineering, IIT KANPUR -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/0bd42307/attachment.htm From mohnish.iitk at gmail.com Tue Oct 27 16:17:15 2009 From: mohnish.iitk at gmail.com (mohnish pandey) Date: Tue, 27 Oct 2009 20:47:15 +0530 Subject: [Pw_forum] calculating equilibrium lattice constant In-Reply-To: References: Message-ID: Thanks a lot Dear Hande and Cantele... On Mon, Oct 26, 2009 at 7:30 PM, Hande Ustunel < hande at newton.physics.metu.edu.tr> wrote: > Dear Mohnish > > If it is a shell script that you are running you might try something like > > for file in out.* > do > E=`grep ! $file | tail -1 | awk '{print $5}'` > echo E >> energies-file > done > > This is just a snippet you can place in your script as required. Please do > be careful about the way the quotation marks are pointing. > > Best wishes, > Hande > > On Mon, 26 Oct 2009, mohnish pandey wrote: > > > I am beginner of quantum espresso. I am trying to calculate the lattice > > constant for zinc oxide graphitic and wurtzitic structure . I am using a > > script for that. But I have to see each and every file to check the > energy > > and to decide which one is minimum. Can you guys please help me to get a > > shortcut for this I mean after running a loop I get a minimum energy file > > instead of doing it manually and then deciding which lattice parammeters > > fits to minimum energy.. > > > > Thanks in advance. > > Mohnish Pandey > > > > > > -- > Hande Toffoli > Department of Physics > Office 439 > Middle East Technical University > Ankara 06531, Turkey > Tel : +90 312 210 3264 > http://www.physics.metu.edu.tr/~hande > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Mohnish Pandey Y6262,4th Year Undergraduate, Department of Chemical Engineering, IIT KANPUR -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/7ca5ee36/attachment.htm From siyouber at yahoo.fr Tue Oct 27 16:49:13 2009 From: siyouber at yahoo.fr (Bertrand SITAMTZE) Date: Tue, 27 Oct 2009 15:49:13 +0000 (GMT) Subject: [Pw_forum] pair correlation function and the choice of fnosep Message-ID: <451793.84992.qm@web26505.mail.ukl.yahoo.com> Dear Users, I have two preccupations: 1) The first one is on how to calculate the pair-correlation functions from CP.x output? 2) The second is the following: when sampling the canonical ensemble in cp.x we should choice the parameter "fnosep". In all texts related to QE, it is written that fnosep "should be chosen to be comparable with the center of the vibrational spectrum of the system". My question is: Does it means that we have first to compute the phonon spectrum of the system? or what to do? Best regards *************************** Bertrand SITAMTZE YOUMBI Laboratory of Material Sciences Department of Physics University of Yaound? I-Cameroon tel: 237 99 16 53 57 ***************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/e28677ae/attachment-0001.htm From paulatto at sissa.it Tue Oct 27 17:02:54 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 27 Oct 2009 17:02:54 +0100 Subject: [Pw_forum] geometry optimization In-Reply-To: References: Message-ID: In data 27 ottobre 2009 alle ore 15:00:25, mohnish pandey ha scritto: > I can fix this problem by preserving the symmetry and > getting eqb. lattice constan.? Dear Mohnish, the code actually do preserve the symmetry, it is enforced actively when computing the forces and cannot be broken during the relax procedure. What's happening is that the system has a lower symmetry than you think since the very beginning. There are two possible reasons: 1. you have not specified the initial atomic positions with sufficient precision, the code consider an atom to be in a high symmetry position if it is less than 1.d-6 away from it (I think in units of alat=celldm(0), but I would have to check) 2. the symmetry of your system has a free parameter you where not aware of, e.g. in a Wurzite structure the distance between alternate planes can change in a specific way without breaking symmetry. These two problems have two corresponding solutions: 1. specify more digits, or (with QE 4.1 and next) specify the position as a fraction, e.g. Ga 1/3 2/3 0 note that xcrysden cannot yet interpret an input file written like this and it is still a bit experimental 2. lock the atoms in position, you can do that adding three zeroes after the atom's coordinate in the input, e.g.: Ga 0.3333333 0.6666667 0.000000 0 0 0 have a look at the manual (Doc/INPUT_PW.txt) for details on details on this notation best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From mohnish.iitk at gmail.com Tue Oct 27 17:26:22 2009 From: mohnish.iitk at gmail.com (mohnish pandey) Date: Tue, 27 Oct 2009 21:56:22 +0530 Subject: [Pw_forum] geometry optimization In-Reply-To: References: Message-ID: Thank you so much dear Lorenzo... On Tue, Oct 27, 2009 at 9:32 PM, Lorenzo Paulatto wrote: > In data 27 ottobre 2009 alle ore 15:00:25, mohnish pandey > ha scritto: > > I can fix this problem by preserving the symmetry and > > getting eqb. lattice constan.? > > Dear Mohnish, > the code actually do preserve the symmetry, it is enforced actively when > computing the forces and cannot be broken during the relax procedure. > What's happening is that the system has a lower symmetry than you think > since the very beginning. > > There are two possible reasons: > 1. you have not specified the initial atomic positions with sufficient > precision, the code consider an atom to be in a high symmetry position if > it is less than 1.d-6 away from it (I think in units of alat=celldm(0), > but I would have to check) > 2. the symmetry of your system has a free parameter you where not aware > of, e.g. in a Wurzite structure the distance between alternate planes can > change in a specific way without breaking symmetry. > > These two problems have two corresponding solutions: > 1. specify more digits, or (with QE 4.1 and next) specify the position as > a fraction, e.g. > Ga 1/3 2/3 0 > note that xcrysden cannot yet interpret an input file written like this > and it is still a bit experimental > 2. lock the atoms in position, you can do that adding three zeroes after > the atom's coordinate in the input, e.g.: > Ga 0.3333333 0.6666667 0.000000 0 0 0 > have a look at the manual (Doc/INPUT_PW.txt) for details on details on > this notation > > best regards > > > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Mohnish Pandey Y6262,4th Year Undergraduate, Department of Chemical Engineering, IIT KANPUR -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/6ecf85bf/attachment.htm From giannozz at democritos.it Tue Oct 27 17:31:40 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 27 Oct 2009 17:31:40 +0100 Subject: [Pw_forum] pair correlation function and the choice of fnosep In-Reply-To: <451793.84992.qm@web26505.mail.ukl.yahoo.com> References: <451793.84992.qm@web26505.mail.ukl.yahoo.com> Message-ID: <4AE7206C.3090806@democritos.it> Bertrand SITAMTZE wrote: > when sampling the canonical ensemble in cp.x we should choose the > parameter "fnosep" [...] "should be chosen to be comparable with the > center of the vibrational spectrum of the system" [...]. Does it mean > that we have first to compute the phonon spectrum of the system? no, it means that you need an idea of what the typical vibrational frequencies of the system are. It is a much simpler task that calculating the entire phonon spectrum P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From uccaati at ucl.ac.uk Tue Oct 27 19:01:02 2009 From: uccaati at ucl.ac.uk (Antonio Tilocca) Date: Tue, 27 Oct 2009 18:01:02 +0000 (GMT) Subject: [Pw_forum] pair correlation function and the choice of fnosep (fwd) Message-ID: Dear Bertrand: > Dear Users, > I have two preccupations: > 1) The first one is on how to calculate the pair-correlation functions > from CP.x output? you will need a simple code to post-process the atomic coordinates taken from the cp.pos trajectory file. Have a look for instance at the routines in the Allen-Tildesley book. > 2) The second is the following: when sampling the canonical ensemble in > cp.x we should choice the parameter "fnosep". In all texts related to > QE, it is written that fnosep "should be chosen to be comparable with the > center of the vibrational spectrum of the system". > My question is: Does it means that we have first to compute the phonon > spectrum of the system? or what to do? you do not need to calculate the phonon spectrum: you could pick fnosep based on experimental vibrational data, if available, for your system, or based on some general knowledge about its characteristic vibrational modes. Antonio Tilocca Chemistry - UCL ------------------------- Date: Tue, 27 Oct 2009 15:49:13 +0000 (GMT) From: Bertrand SITAMTZE Subject: [Pw_forum] pair correlation function and the choice of From mulazzi at spring8.or.jp Wed Oct 28 02:09:42 2009 From: mulazzi at spring8.or.jp (Mulazzi Mattia) Date: Wed, 28 Oct 2009 10:09:42 +0900 Subject: [Pw_forum] Charge ordering with QE Message-ID: Dear QE community, I am currently studying charge-density wave ordered materials form the theoretical point of view. In my system, VS2, the Vanadium planes order in an hexagonal 2D lattice and are sandwiched by two Sulfur planes. As a basic approach, an hexagonal unit cell with one V and two S atoms produces a band structure that is in fair agreement with the experimental data and is consistent with other all-electron calculations. But both theoretically and experimentally there are hints that the system is orbital-ordered, i.e. that the d electrons on the Vanadium sites occupy orbitals of different symmetry on different Vanadium sites. In order to calculate the electronic structure of an orbitally-ordered system I think I have to use a supercell and specify different occupation numbers for the different Vanadium sites. The problem is that I don't know how to do that. I tried specifying the starting_ns_eigenvalue variable on different sites, but the converged calculation is identical to the calculation with starting_ns_eigenvalue left unspecified. In the INPUT_PW it is written that the occupation matrices can be specified only when the LDA+U scheme is used. However, I cannot use the LDA+U on my system (Mac OS X 10.5.6) because every time I use the Vanadium pseudopotential (either the one I generated, ether the one on the QE database) I find a "segmentation fault" that I cannot solve. BTW, the segmentation fault error does not appear for Co or Fe or Ni pseudopotentials. Thanks everybody for the help, Mattia Mulazzi FPR Fellow of RIKEN at Spring8 Excitation Order Research Team 1-1-1 Sayo-cho Sayo-gun, Hyogo Japan From daijiayu at nudt.edu.cn Wed Oct 28 02:09:56 2009 From: daijiayu at nudt.edu.cn (Jiayu Dai) Date: Wed, 28 Oct 2009 09:09:56 +0800 Subject: [Pw_forum] mixing_mode: "local-TF" and "TF". Message-ID: <456692196.14141@nudt.edu.cn> Dear Users, Recently, when i did some calculations of molecular dynamics using pwscf code, i found that the mixing_mode of local-TF and TF can improve the convergence greatly. However, i did not understand how it works. Could anybody tell me? I know the Thomas-Fermi theory, which is the foundation of DFT. So, the TF here is the same as the Thomas-Fermi theory? Are there any references about it? And is there any limitation for the usage of this mixing_mode? Thanks a lot. Jiayu ------------------------------ ------------------------------------------- Jiayu Dai Department of Physics National University of Defense Technology, Changsha, 410073, P R China ----------------------------------------- From marzari at MIT.EDU Wed Oct 28 02:49:39 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Tue, 27 Oct 2009 21:49:39 -0400 Subject: [Pw_forum] Charge ordering with QE In-Reply-To: References: Message-ID: <4AE7A333.4050605@mit.edu> Dear Mattia, first step would be to have LDA+U working - not sure if the V problem is from your computer, or from the V in the pwscf database (if the former, ask a friend to run your calculation on a different machine, to see if it works. Also, if you plan to do comptuational research, you might want to try and get access to some machines bigger, if not better, than a Mac). Second - I don't think there is a canned solution ready, but what I would do is something akin to what we did a few years ago (Sit, Cococcioni, Marzari, PRL06) in adding a penalty functional to the total energy functional - to force or to gently push the electronic ground state towards occupying certain orbitals. The algorithms is not implemented in pwscf, but it is implemented in CP - Patrick, cc'd here, has a copy, and there was the hope/plan to put it in the CVS. Taking the subroutines from the CP code and putting them in PWscf would not be too difficult, but would require expertise on the two codes. The CP implementation does work, though (contact Patrick), albeit Gamma sampling only, forcing you to use large supercells to get decent BZ sampling. Some kind of workaround could be this, with PWscf: get LDA+U to work, define each vanadium atom as being different (V1, V2, V3...) but with the same pseudopotential, and modify the part of the code that calculates the occupation matrix n. By changing the n for V1 so that it has 1 only for the oribtals that you want occupied, and same for V2, V3, etc..., and by choosing "judiciously" the U, you might force the system to the orbital ordering that you want. Then, restart your calculation without any constraint, and see if the electronic-state is stable. Note, you might need a true U in this second case, applied equally to all V, so the steps are first to become an expert in LDA+U (Cococcioni's thesis and papers), and then move onf rom there. It's not easy, but very interesting. nicola Mulazzi Mattia wrote: > Dear QE community, > I am currently studying charge-density wave ordered materials form the > theoretical point of view. In my system, VS2, the Vanadium planes > order in an hexagonal 2D lattice and are sandwiched by two Sulfur > planes. As a basic approach, an hexagonal unit cell with one V and two > S atoms produces a band structure that is in fair agreement with the > experimental data and is consistent with other all-electron > calculations. But both theoretically and experimentally there are > hints that the system is orbital-ordered, i.e. that the d electrons on > the Vanadium sites occupy orbitals of different symmetry on different > Vanadium sites. > In order to calculate the electronic structure of an orbitally-ordered > system I think I have to use a supercell and specify different > occupation numbers for the different Vanadium sites. The problem is > that I don't know how to do that. I tried specifying the > starting_ns_eigenvalue variable on different sites, but the converged > calculation is identical to the calculation with > starting_ns_eigenvalue left unspecified. In the INPUT_PW it is written > that the occupation matrices can be specified only when the LDA+U > scheme is used. > However, I cannot use the LDA+U on my system (Mac OS X 10.5.6) because > every time I use the Vanadium pseudopotential (either the one I > generated, ether the one on the QE database) I find a "segmentation > fault" that I cannot solve. BTW, the segmentation fault error does > not appear for Co or Fe or Ni pseudopotentials. > > Thanks everybody for the help, > > Mattia Mulazzi > > > FPR Fellow of RIKEN at Spring8 > Excitation Order Research Team > 1-1-1 Sayo-cho Sayo-gun, Hyogo > Japan > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From lfhuang at theory.issp.ac.cn Wed Oct 28 03:17:30 2009 From: lfhuang at theory.issp.ac.cn (=?utf-8?B?bGZodWFuZw==?=) Date: Wed, 28 Oct 2009 10:17:30 +0800 Subject: [Pw_forum] =?utf-8?q?about_the_nelec_in_the_outputfile?= Message-ID: <20091028021730.15578.qmail@ms.hfcas.ac.cn> Dear PWSCF Developers: I am calculating the electronic structure of charged systems, and there is one thing that has confused me: Whatever how much "tot_charge"(in unit of 1 electron, am I right?) I set in the input files, the "nelec" or the " starting charge" in the output files always appears to be about 200.0. It seems this is not related to the problem of unit transformation, am I right? Then, what does the number "200.0" means? Any kind-hearted reply would be highly appreciated! Thanks very much! Best Wishes! Yours Sincerely L. F. Huang ------ ====================================================================== L.F.Huang(???) ph.D candidate, DFT and Phonon physics ====================================================================== Add: Research Laboratory for Computational Materials Sciences, Instutue of Solid State Physics,the Chinese Academy of Sciences, P.O.Box 1129, Hefei 230031, P.R.China Tel: 86-551-5591464-328(office) Fax: 86-551-5591434 Web: http://theory.issp.ac.cn (website of our theory group) http://www.issp.ac.cn (website of our institute) ====================================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091028/b1694f5a/attachment-0001.htm From ttduyle at gmail.com Wed Oct 28 04:17:54 2009 From: ttduyle at gmail.com (Duy Le) Date: Tue, 27 Oct 2009 23:17:54 -0400 Subject: [Pw_forum] about the nelec in the outputfile In-Reply-To: <20091028021730.15578.qmail@ms.hfcas.ac.cn> References: <20091028021730.15578.qmail@ms.hfcas.ac.cn> Message-ID: <8974d3b20910272017k6b68277bp2f3144e32e2f86c3@mail.gmail.com> Did you check carefully? If you change the tot_charge, the number of electron must be charged accordingly. The starting charge is generated to be a non-charge system but then will be renormalized accordingly. That number 200 is the total electron in your unit cell corresponding to the neutral system. -------------------------------------------------- Duy Le PhD Student Department of Physics University of Central Florida. "Men don't need hand to do things" On Tue, Oct 27, 2009 at 10:17 PM, lfhuang wrote: > Dear PWSCF Developers: > > I am calculating the electronic structure of charged systems, and there > is one thing that has confused me: Whatever how much "tot_charge"(in unit of > 1 electron, am I right?) I set in the input files, the "nelec" or the " > starting charge" in the output files always appears to be about 200.0. It > seems this is not related to the problem of unit transformation, am I > right? Then, what does the number "200.0" means? > > Any kind-hearted reply would be highly appreciated! Thanks very much! > > Best Wishes! > > Yours Sincerely > > L. F. Huang > > ------ > > ====================================================================== > L.F.Huang(???) ph.D candidate, DFT and Phonon physics > ====================================================================== > Add: Research Laboratory for Computational Materials Sciences, > Instutue of Solid State Physics,the Chinese Academy of Sciences, > P.O.Box 1129, Hefei 230031, P.R.China > Tel: 86-551-5591464-328(office) > Fax: 86-551-5591434 > Web: http://theory.issp.ac.cn (website of our theory group) > http://www.issp.ac.cn (website of our institute) > ====================================================================== > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091027/c1166f45/attachment.htm From helman at tandar.cnea.gov.ar Wed Oct 28 05:49:25 2009 From: helman at tandar.cnea.gov.ar (Christian Santiago Helman) Date: Wed, 28 Oct 2009 01:49:25 -0300 Subject: [Pw_forum] Looking for Manganse pseudopotential In-Reply-To: References: Message-ID: Hello world! I'm looking for a Manganse pseudopotencial. At the QE home page there is one but with semicore states in valence. Does anybody have a pseudo without semicore. Of course, Scalar-Relativistic Calculation and GGA ( the PBE one) and Ultrasoft are the main caracteristics Thanks everybody!! Lic. Christian Helman Gerencia de Investigaciones y Aplicaciones Centro At?mico Constituyentes-CNEA Argentina Tel:++54-11-6772-7102 From manoj at phys.ufl.edu Wed Oct 28 06:49:34 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Wed, 28 Oct 2009 01:49:34 -0400 (EDT) Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: <1256226425.3829.27.camel@alex-laptop> Message-ID: Dear Alexander, You are correct, left and right moving states are in general not related. I was wrong about trying to create a pair. Thanks for clarification. I have a few more questions. 1. I want to find out the reflection coefficient in the code. In transmit.f90, transmission coefficient is |tmat(ig,n)|^2, where tmat is vec1(ntran-n2d-npol*nocrosr+ig,n). (line 206, of version 4.0.5). Reflection is not explicitly calculated in the code, but it should be modulus square of vec1(2*n2d+npol*norbs+ig,n) (line 234). Am I correct? 2. In the code reflection and transmission coefficients are calculated for one direction lets say t++ and t+-, which is transmission and reflection coefficient for Bloch' state moving from - to + , or in other words, left to right. To calculate transmission and reflection for Bloch's state going from right to left, what should I do? Should i just rotate my system 180 degree, that way leads get interchanged and also scattering area has orientation 180 different from before. 3. In the subroutine sunitary.f90, is raux measure of conservation of flux? If I sum all the transmission and reflecion coefficient for a particular Bloch's state, I should get 1, and computationally a number close to 1. How large could it be for S matrix not to be unitary, i m getting raux more than one e.g. 1.0011 but code still considers smatrix to be unitary. Regards, Manoj Srivastava University of Florida, Gainesville, FL On Thu, 22 Oct 2009, Alexander Smogunov wrote: > On Thu, 2009-10-22 at 11:34 -0400, Manoj Srivastava wrote: > > Dear Alexander, > > Thanks for your answer. I just want to make sure. Imagine we have total > > number of channels in the left lead 2, so total number of Bloch's state > > are 4. 2 of them left going say a and b, and 2 right going say c and d. > > So, are you saying that for left going state a, the corresponding right > > going state is c? Are they ordered this way? > > what do you mean by corresponding? Left and right moving Bloch > states are in general not related one to another, you can even have > different number of them ... Only if you have some symmetry S which > brings kz to -kz conserving k_parallel, then the state with > \psi_{-kz} will be S \psi{kz}. This is true for example at 2D G point > when you have time reversal operation. > > Now the code simply arranges the propagating states in the order of > increasing |k_z|... > > Regards, Alexander > > > > > > Regards, > > Manoj > > > > > > On Thu, 22 Oct 2009, Alexander > > Smogunov wrote: > > > > > Dear Manoj. > > > > > > The output of complex k vectors is performed in > > > summary_band.f90 routine. If you want to see all > > > the complex k vectors, not only propagating ones, > > > you can change at the end of this routine: > > > > > > ------------------- > > > do i = 1, nchanl > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > enddo > > > ------------------- > > > > > > to > > > ------------------- > > > > > > do i = 1, 2*nstl > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > enddo > > > ------------------- > > > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > > the right states, another half, [nstl+1,2*nstl], - propagating or > > > decaying to the left. In each group, first nchanl states are propagating > > > states. > > > > > > The propagating states are normalized by the current and are arranged in > > > the above order at the end of jbloch.f90 routine, after the following > > > lines: > > > > > > ! > > > ! Right ordering (+, >, -, <) > > > ! > > > > > > > > > > > > Notice, that in the last versions the code gives in output > > > both propagating to the right and to the left states. > > > > > > Hope this helps, > > > Alexander > > > > > > > > > > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > > Dear All, > > > > I am trying to figure out the left and right going Bloch's states in the > > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > > > > prints out Bloch's state moving in one direction. eg. in one of the > > > > calculation- > > > > k//=(0.375,-0.375) > > > > Nchannels of the left tip = 1 > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > > > Now if I want Bloch's state moving in right as well as left direction, I > > > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > > > AX=exp(ikd)BX, and out of those the ones with real solution would be our > > > > Bloch's state, so I get for each channel two solutions- > > > > kval (-0.275409421993275,1.823688001395235E-010) > > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > > > To figure out the direction, I can calculate current associated with these > > > > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > > > > its left moving Bloch's state. I can print out current from jbloch.f90 > > > > subroutine which are - > > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > > > other one right moving. Upto here its clear to me how to identify left and > > > > right moving states. > > > > > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > > > state. In another calculation where i get multiple Bloch's state- > > > > Nchannels of the left tip = 5 > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > -0.0746301 0.0000000 0.0000000 > > > > 0.1205527 0.0000000 0.0000000 > > > > 0.3112908 0.0000000 0.0000000 > > > > 0.4200218 0.0000000 0.0000000 > > > > -0.4935150 0.0000000 0.0000000 > > > > > > > > so i did the same trick i did above to first print out kz and then > > > > current, which gives me - > > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > > 1.24280031534940 9.313897787790 > > > > > > > > So, the first 5 are left moving and rest are right moving. But I dont know > > > > the pairs. for example for left moving state a, what is the corresponding > > > > right moving state whether its f or g ... j ? > > > > > > > > Any help would be appreciated. > > > > > > > > Regards, > > > > Manoj Srivastava > > > > University of Florida, Gainesville. > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > -- > > > e-mail: smogunov at sissa.it > > > home-page: http://people.sissa.it/~smogunov > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > -- > e-mail: smogunov at sissa.it > home-page: http://people.sissa.it/~smogunov > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From yuminqian at gmail.com Wed Oct 28 08:00:51 2009 From: yuminqian at gmail.com (Y.M.Qian) Date: Wed, 28 Oct 2009 15:00:51 +0800 Subject: [Pw_forum] electronic-phonon coupling error References: , <200910231835547506316@gmail.com>, , <200910252014024844598@gmail.com>, <4AE55EAA.8020200@democritos.it> Message-ID: <200910281500472027808@gmail.com> Thanks for your reply, I find the reason of this problem. When I erase the following control parameter ,verything is ok . wf_collect=.true. Thank you very much . 2009-10-28 Y.M.Qian ???? Paolo Giannozzi ????? 2009-10-26 16:32:50 ???? PWSCF Forum ??? ??? Re: [Pw_forum] electronic-phonon coupling error Y.M.Qian wrote: > But there is still mistake > from openfilq : error # 1 > file nm-type.wfc not found it is not there. Check your input data. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091028/28a05bcd/attachment.htm From giannozz at democritos.it Wed Oct 28 08:55:22 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 28 Oct 2009 08:55:22 +0100 Subject: [Pw_forum] mixing_mode: "local-TF" and "TF". In-Reply-To: <456692196.14141@nudt.edu.cn> References: <456692196.14141@nudt.edu.cn> Message-ID: <4AE7F8EA.9070108@democritos.it> Jiayu Dai wrote: > i found that the mixing_mode of local-TF and TF can improve the > convergence greatly.[...] Are there any references about it? there is some explanation in the paper describing QE: J. Phys.: Condens. Matter 21, 395502 (2009). The relevant reference is D. Raczkowski, A. Canning, and L. W. Wang, Phys. Rev. B64, R121101 (2001) P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From lfhuang at theory.issp.ac.cn Wed Oct 28 08:53:54 2009 From: lfhuang at theory.issp.ac.cn (=?utf-8?B?bGZodWFuZw==?=) Date: Wed, 28 Oct 2009 15:53:54 +0800 Subject: [Pw_forum] =?utf-8?q?about_the_nelec_in_the_outputfile?= Message-ID: <20091028075354.14113.qmail@ms.hfcas.ac.cn> Dear Duy Le: Thanks very much for your kind reply! Because just a small amount of charges are added to my system, so that I did not noticed the small difference in the values. Thanks very much! Best Wishes! Yours Sincerely L.F.Huang > From: Duy Le > Subject: Re: [Pw_forum] about the nelec in the outputfile > To: PWSCF Forum > Message-ID: > > Content-Type: text/plain; charset="utf-8" > > Did you check carefully? If you change the tot_charge, the number of > electron must be charged accordingly. The starting charge is generated to be > a non-charge system but then will be renormalized accordingly. > > That number 200 is the total electron in your unit cell corresponding to the > neutral system. > -------------------------------------------------- > Duy Le > PhD Student > Department of Physics > University of Central Florida. > > "Men don't need hand to do things" ------ ====================================================================== L.F.Huang(???) ph.D candidate, DFT and phonon physics ====================================================================== Add: Research Laboratory for Computational Materials Sciences, Instutue of Solid State Physics,the Chinese Academy of Sciences, P.O.Box 1129, Hefei 230031, P.R.China Tel: 86-551-5591464-328(office) Fax: 86-551-5591434 Web: http://theory.issp.ac.cn (website of our theory group) http://www.issp.ac.cn (website of our institute) ====================================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091028/f4a96efe/attachment.htm From gregor.mali at ki.si Wed Oct 28 10:27:32 2009 From: gregor.mali at ki.si (Gregor Mali) Date: Wed, 28 Oct 2009 10:27:32 +0100 Subject: [Pw_forum] convergence problems using GIPAW In-Reply-To: References: <001901ca4044$efca0b30$cf5e2190$@ac.kr><8C0B47760A05FB4AB504C9BCAD32D256012E52AE@postar.ki.si><8C0B47760A05FB4AB504C9BCAD32D256013EB614@postar.ki.si> Message-ID: <8C0B47760A05FB4AB504C9BCAD32D256013EB6C5@postar.ki.si> Davide, > is your system metallic? > the material I am studying is Li2MnO3. It is an insulator, at low temperature there might exist a weak antiferromagnetic interaction between Mn ions. In this material I predominantley wanted to calculate the hyperfine interaction between Mn ion and Li nucleus. Therefore I performed spin-polarized calculation. The input file is given below. Nevertheless, I was also interested in the magnitude of Li chemical shifts. &CONTROL calculation = 'scf' , restart_mode = 'from_scratch' , outdir = './scratch/' , pseudo_dir = './pseudo/' , prefix = 'Li2MnO3' , tstress = .true. , tprnfor = .true. , verbosity = 'high' , / &SYSTEM ibrav = 14, A = 4.929200 , B = 8.531500 , C = 5.025100 , cosAB = 0 , cosAC = -0.3311 , cosBC = 0 , nat = 24, ntyp = 3, ecutwfc = 80.0 , ecutrho = 320.0 , occupations = 'smearing' , degauss = 0.02 , smearing = 'gaussian' , nspin = 2 , starting_magnetization(2) = 0.7, / &ELECTRONS conv_thr = 1.0d-10 , mixing_mode = 'plain' , mixing_beta = 0.7 , diagonalization = 'david' , / ATOMIC_SPECIES Li 6.94100 Li_pbe-20090916.UPF Mn 54.93800 Mn.pbe-rrkj-semi-gipaw.UPF O 15.99900 O.pbe-tm-gipaw.UPF ATOMIC_POSITIONS crystal Li 0.000000000 0.500000000 0.000000000 . . . K_POINTS automatic 4 3 4 1 1 1 Gregor From javaid at ipcms.u-strasbg.fr Wed Oct 28 16:02:45 2009 From: javaid at ipcms.u-strasbg.fr (saqib javaid) Date: Wed, 28 Oct 2009 16:02:45 +0100 Subject: [Pw_forum] About gaussian cubic format Message-ID: <4AE85D15.9060502@ipcms.u-strasbg.fr> Dear PWSCF users, I have a question regarding gaussian cubic format (output_format=6 in pp input file). I have used this format to get charge density. However, the xyz coordinates of the atoms given in the output file (which should be in bohr as per a previous post) do not match the coordinates used in SCF input file. I would appreciate if you can clarify this problem. Secondly, is there any way to increase the no. of mesh points in this format. Thanks in advance for your kind reply with best regards, saqib Javaid Unviersity of Strasbourg, France From giannozz at democritos.it Wed Oct 28 16:27:22 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 28 Oct 2009 16:27:22 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: <4AE85D15.9060502@ipcms.u-strasbg.fr> References: <4AE85D15.9060502@ipcms.u-strasbg.fr> Message-ID: <4AE862DA.4070707@democritos.it> saqib javaid wrote: > the xyz coordinates of the atoms given in the output file (...) > do not match the coordinates used in SCF input file. what do you mean by "do not match"? if you visualize the charge and the atoms, do they look misplaced? > Secondly, is there any way to increase the no. of mesh points in this > format. the number of mesh points is determined by the size of the FFT grid. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From javaid at ipcms.u-strasbg.fr Wed Oct 28 16:49:42 2009 From: javaid at ipcms.u-strasbg.fr (saqib javaid) Date: Wed, 28 Oct 2009 16:49:42 +0100 Subject: [Pw_forum] About gaussian cubic format Message-ID: <4AE86816.4000904@ipcms.u-strasbg.fr> Thanks a lot for your reply. What i mean by "do not match" is the fact that atomic coordinates are different from what i gave in SCF input file. When i load the guassina cubic file in the Xcrysden, i found that inter atomic distances are the same as in SCF input file, but origin for plotting has changed. I hope i am clear enough this time... with best regards saqib javaid University of strasbourg, France From giannozz at democritos.it Wed Oct 28 16:54:44 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 28 Oct 2009 16:54:44 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: <4AE86816.4000904@ipcms.u-strasbg.fr> References: <4AE86816.4000904@ipcms.u-strasbg.fr> Message-ID: <4AE86944.60709@democritos.it> saqib javaid wrote: > What i mean by "do not match" is the fact > that atomic coordinates are different from what i gave in SCF input atoms outside the cell are wrapped back into the cell P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From paulatto at sissa.it Wed Oct 28 16:55:03 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 28 Oct 2009 16:55:03 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: <4AE86816.4000904@ipcms.u-strasbg.fr> References: <4AE86816.4000904@ipcms.u-strasbg.fr> Message-ID: In data 28 ottobre 2009 alle ore 16:49:42, saqib javaid ha scritto: > i found that > inter atomic distances are the same as in SCF input file, but origin for > plotting has changed. I hope i am clear enough this time... Dear Saqib, it does not come from the Gaussian file, but from XCrysDen itself, which translates all the atoms inside the primitive unit cell. If you want all the atoms in they original positions follow the menu Displa -> Unit of repetition -> translational asymmetric unit. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From tone.kokalj at ijs.si Wed Oct 28 17:06:04 2009 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Wed, 28 Oct 2009 17:06:04 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: References: <4AE86816.4000904@ipcms.u-strasbg.fr> Message-ID: <1256745964.4653.854.camel@walk.ijs.si> On Wed, 2009-10-28 at 16:55 +0100, Lorenzo Paulatto wrote: > In data 28 ottobre 2009 alle ore 16:49:42, saqib javaid > ha scritto: > > i found that > > inter atomic distances are the same as in SCF input file, but origin for > > plotting has changed. I hope i am clear enough this time... > > Dear Saqib, > it does not come from the Gaussian file, but from XCrysDen itself, which > translates all the atoms inside the primitive unit cell. If you want all > the atoms in they original positions follow the menu Displa -> Unit of > repetition -> translational asymmetric unit. To add to this comment, which raises two issues: (1) if your system is periodic (i.e. crystal or surface), use XSF file instead of gaussian cube file as to have the possibility to use/generate periodic images (2) if your system is a molecule "in the box", put the molecule in the center of the box, as this will avoid the "translational" problem discussed above. Regards, Tone -- Anton Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) Please, if possible, avoid sending me Word or PowerPoint attachments. See: http://www.gnu.org/philosophy/no-word-attachments.html From javaid at ipcms.u-strasbg.fr Wed Oct 28 17:40:22 2009 From: javaid at ipcms.u-strasbg.fr (saqib javaid) Date: Wed, 28 Oct 2009 17:40:22 +0100 Subject: [Pw_forum] About gaussian cubic format Message-ID: <4AE873F6.3050009@ipcms.u-strasbg.fr> with regards to the Tone comments, infact i am using guassian cubic so that i can do bader population analysis. I am thankful to all for your timely help. with best regards saqib University of strasbourg, France. From eariel99 at gmail.com Wed Oct 28 19:50:50 2009 From: eariel99 at gmail.com (Eduardo Ariel Menendez Proupin) Date: Wed, 28 Oct 2009 15:50:50 -0300 Subject: [Pw_forum] plane averaged local potential Message-ID: Hello, I need to calculate plane averaged local potentials, and charges, for a heterostructure. I.e., V(z)=1/(Lx*Ly) \int\int V(x,y,z) dx dy I think I can do it using PP to get V(x,y,z) in the 3D FFT grid and a write a program that makes the average. Then I want to obtain the macroscopic average, as defined in a triestine paper by Baldereschi, Baroni and Resta PRL 61 ,734 (1988). Before spending one or two days writing the program and become sure that is works fine, I would like to confirm that this is not already implemented in PP or in other utility of Quantum-ESPRESSO. Thank you -- Eduardo Menendez Departamento de Fisica Facultad de Ciencias Universidad de Chile Phone: (56)(2)9787439 URL: http://fisica.ciencias.uchile.cl/~emenendez -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091028/6f36f848/attachment.htm From rezaebraahimi at yahoo.com Wed Oct 28 20:44:18 2009 From: rezaebraahimi at yahoo.com (Reza Ebraahimi) Date: Wed, 28 Oct 2009 12:44:18 -0700 (PDT) Subject: [Pw_forum] About Debye Temp. Message-ID: <337266.81273.qm@web59402.mail.ac4.yahoo.com> Hi I am trying to investigate the Debye Temp. of two polymorphs of a non-metal case, by calculating the heat capacity,?c_v?(using the phonon DOS) at very low temperatures. But the polymorph with higher Bulk Modulus, gets the lower Debye Temp. I think, as the acoustic frequencies have more important role in both stiffness and "c_v in low temps", the higher the Bulk Modulus, may get the higher Debye temp. Am i right? Is there anything wrong in my calculation? Thanks for any comment R. Ebraahimi Grad. Student Tehran University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091028/9855250f/attachment.htm From prasenjit.jnc at gmail.com Wed Oct 28 22:18:33 2009 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Wed, 28 Oct 2009 22:18:33 +0100 Subject: [Pw_forum] plane averaged local potential In-Reply-To: References: Message-ID: <627e0ffa0910281418tc41f397p9e87c0dff492b350@mail.gmail.com> 2009/10/28 Eduardo Ariel Menendez Proupin > Hello, > I need to calculate plane averaged local potentials, and charges, for a > heterostructure. I.e., > V(z)=1/(Lx*Ly) \int\int V(x,y,z) dx dy > > I think I can do it using PP to get V(x,y,z) in the 3D FFT grid and a > write a program that makes the average. Then I want to obtain the > macroscopic average, as defined in a triestine paper by Baldereschi, Baroni > and Resta PRL 61 ,734 (1988). > > Before spending one or two days writing the program and become sure that > is works fine, I would like to confirm that this is not already implemented > in PP or in other utility of Quantum-ESPRESSO. > > Hi Eduardo, The executable average.x can do the job........for the input file you can see the example on Work function calculation...... However, if you want to calculate the planar average of charge density, you need to multiply it with the area of the xy plane, assuming you want to plot along the z-direction. Also I think (I'm not sure) it works only if the z- is perpendicular to the xy-plane. With regards, Prasenjit. -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091028/b81cdc30/attachment.htm From kajalmh18 at gmail.com Thu Oct 29 02:52:20 2009 From: kajalmh18 at gmail.com (kajal jindal) Date: Thu, 29 Oct 2009 07:22:20 +0530 Subject: [Pw_forum] (no subject) Message-ID: Hii, I am working on ZnO doped with Nitrogen .I have used 2*2*2 ZnO supercell ..If i replace 1 Zn atom by 1 N atom in that supercell,doped N atom does not make any bonds with the surrounding atoms even after optimizing the lattice constants and atomic positions..It remains at that position as an isolated atom only...So can anybody plz suggest me what should i do in this case????Also i want to know that whether this will affect my band structure calculations??? sincerely, Miss Kajal (UTA)(University of Delhi) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091029/078954f1/attachment.htm From marzari at MIT.EDU Thu Oct 29 03:13:32 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 28 Oct 2009 22:13:32 -0400 Subject: [Pw_forum] (no subject) In-Reply-To: References: Message-ID: <4AE8FA4C.60606@mit.edu> kajal jindal wrote: > Hii, > > I am working on ZnO doped with Nitrogen .I have used 2*2*2 ZnO > supercell ..If i replace 1 Zn atom by 1 N atom in that supercell,doped > N atom does not make any bonds with the surrounding atoms even after > optimizing the lattice constants and atomic positions..It remains at > that position as an isolated atom only...So can anybody plz suggest me > what should i do in this case????Also i want to know that whether > this will affect my band structure calculations??? > > sincerely, > Miss Kajal > (UTA)(University of Delhi) Dear Miss Kajal, it probably made some bonds - simplest way to tell would be to look at the charge density, or the charge density difference between the two cases (use the unit cell of the doped system, but relax fully the atoms in both cases). Sure, the N will affect your band structures. Might be worth studying the media.quantum-espresso.org lectures. nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From sagarambavale at yahoo.co.in Thu Oct 29 05:43:25 2009 From: sagarambavale at yahoo.co.in (ambavale sagar) Date: Thu, 29 Oct 2009 10:13:25 +0530 (IST) Subject: [Pw_forum] mpi-pwcond error Message-ID: <22095.68694.qm@web94612.mail.in2.yahoo.com> Dear PwScf users, I am getting MPI error during PWCOND run at particular point. The standard output gives: MPI_Allreduce: invalid communicator: Unknown error 2064 (rank 0, comm 16) Rank (0, MPI_COMM_WORLD) : Call stack within LAM: Rank (0, MPI_COMM_WORLD) : -MPI_Allreduce() Rank (0, MPI_COMM_WORLD) : -main() ------------------------------------------------------ One of the processes started by mpirun has exited with a nonzero exit........ ... .. .. PID 16077 failed on node n1 (192.168.0.12) with exit status 1. ----------------------------- I am running espresso-4.0 installed with lam-mpi 7.1.4 and ifort 10.1 on dual cpu quad core xeon processors. There are 2 machines running in parallel. Thus 4 cpus or 16 cores are available. However, this calculation was ran on 4 cores only. The scf and relax calculations do not crash. One more thing to note is: recently I have enhanced RAM on both the machines. And after enhnacement did not do any reinstallation or recompilation of any software including lam-mpi. Note : Through search on forum archives I found a post by Derek : http://www.democritos.it/pipermail/pw_forum/2005-November/003219.html which indicates this might be a problem related to mpi or def.h...... Thank you. Regards Sagar Ambavale PhD Student The M.S. Uni. of Baroda India Yahoo! India has a new look. Take a sneak peek http://in.yahoo.com/trynew -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091029/ec9ec657/attachment.htm From tone.kokalj at ijs.si Thu Oct 29 10:48:08 2009 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Thu, 29 Oct 2009 10:48:08 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: <4AE873F6.3050009@ipcms.u-strasbg.fr> References: <4AE873F6.3050009@ipcms.u-strasbg.fr> Message-ID: <1256809688.12255.15.camel@walk.ijs.si> On Wed, 2009-10-28 at 17:40 +0100, saqib javaid wrote: > with regards to the Tone comments, infact i am using guassian cubic so > that i can do bader population analysis. Would you share the info of how you do the Bader analysis? Regards, Tone -- Anton Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) Please, if possible, avoid sending me Word or PowerPoint attachments. See: http://www.gnu.org/philosophy/no-word-attachments.html From giannozz at democritos.it Thu Oct 29 10:52:34 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 29 Oct 2009 10:52:34 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: <1256809688.12255.15.camel@walk.ijs.si> References: <4AE873F6.3050009@ipcms.u-strasbg.fr> <1256809688.12255.15.camel@walk.ijs.si> Message-ID: <4AE965E2.8080007@democritos.it> Tone Kokalj wrote: > Would you share the info of how you do the Bader analysis? maybe it is already shared .... I added the following paragraph a few days ago to the user guide: --- Note for people interested in Bader's analysis: there is a software (link here: \texttt{http://theory.cm.utexas.edu/bader/}) that performs Bader's analysis starting from charge on a regular grid. The required "cube" format can be produced by \qe\ using pp.x (info by G. Lapenna who has successfully used this technique). This code should perform decomposition into Voronoi polyhedra as well, in place of obsolete code voronoy.x (removed from distribution after v.4.1). --- Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From tone.kokalj at ijs.si Thu Oct 29 10:58:10 2009 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Thu, 29 Oct 2009 10:58:10 +0100 Subject: [Pw_forum] About gaussian cubic format In-Reply-To: <4AE965E2.8080007@democritos.it> References: <4AE873F6.3050009@ipcms.u-strasbg.fr> <1256809688.12255.15.camel@walk.ijs.si> <4AE965E2.8080007@democritos.it> Message-ID: <1256810290.12255.29.camel@walk.ijs.si> On Thu, 2009-10-29 at 10:52 +0100, Paolo Giannozzi wrote: > Tone Kokalj wrote: > > > Would you share the info of how you do the Bader analysis? > > maybe it is already shared .... I added the following paragraph > a few days ago to the user guide: Paolo - thank you for the info. Regards, Tone -- Anton Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) Please, if possible, avoid sending me Word or PowerPoint attachments. See: http://www.gnu.org/philosophy/no-word-attachments.html From smogunov at sissa.it Thu Oct 29 16:27:15 2009 From: smogunov at sissa.it (Alexander Smogunov) Date: Thu, 29 Oct 2009 16:27:15 +0100 Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: References: Message-ID: <1256830035.3467.13.camel@alex-laptop> Dear Manoj On Wed, 2009-10-28 at 01:49 -0400, Manoj Srivastava wrote: > Dear Alexander, > You are correct, left and right moving states are in general not > related. I was wrong about trying to create a pair. Thanks for > clarification. > I have a few more questions. > 1. I want to find out the reflection coefficient in the code. In > transmit.f90, transmission coefficient is |tmat(ig,n)|^2, where tmat is > vec1(ntran-n2d-npol*nocrosr+ig,n). (line 206, of version 4.0.5). > Reflection is not explicitly calculated in the code, but it should be > modulus square of vec1(2*n2d+npol*norbs+ig,n) (line 234). Am I correct? in the 4.1 version there is output of both t and r. > > 2. In the code reflection and transmission coefficients are calculated > for one direction lets say t++ and t+-, which is transmission and > reflection coefficient for Bloch' state moving from - to + , or in other > words, left to right. To calculate transmission and reflection for Bloch's > state going from right to left, what should I do? Should i just rotate my > system 180 degree, that way leads get interchanged and also scattering > area has orientation 180 different from before. > again in the 4.1 version you have possibility to calculate t and r for both directions, in the do_cond.f90 the transmit routine is called with an extra parameter which is true (false) if you want to calculate scattering states propagating from the left (from the right). > 3. In the subroutine sunitary.f90, is raux measure of conservation of > flux? If I sum all the transmission and reflecion coefficient for > a particular Bloch's state, I should get 1, and computationally a number > close to 1. How large could it be for S matrix not to be unitary, i m > getting raux more than one e.g. 1.0011 but code still considers smatrix > to be unitary. in the case when |R + T - 1| > 1.d-4 for some band, it should print out the value of R+T but continues to run anyway. All the best, Alexander > > Regards, > Manoj Srivastava > University of Florida, Gainesville, FL > > > > > > > On Thu, 22 > Oct 2009, Alexander Smogunov wrote: > > > On Thu, 2009-10-22 at 11:34 -0400, Manoj Srivastava wrote: > > > Dear Alexander, > > > Thanks for your answer. I just want to make sure. Imagine we have total > > > number of channels in the left lead 2, so total number of Bloch's state > > > are 4. 2 of them left going say a and b, and 2 right going say c and d. > > > So, are you saying that for left going state a, the corresponding right > > > going state is c? Are they ordered this way? > > > > what do you mean by corresponding? Left and right moving Bloch > > states are in general not related one to another, you can even have > > different number of them ... Only if you have some symmetry S which > > brings kz to -kz conserving k_parallel, then the state with > > \psi_{-kz} will be S \psi{kz}. This is true for example at 2D G point > > when you have time reversal operation. > > > > Now the code simply arranges the propagating states in the order of > > increasing |k_z|... > > > > Regards, Alexander > > > > > > > > > > Regards, > > > Manoj > > > > > > > > > On Thu, 22 Oct 2009, Alexander > > > Smogunov wrote: > > > > > > > Dear Manoj. > > > > > > > > The output of complex k vectors is performed in > > > > summary_band.f90 routine. If you want to see all > > > > the complex k vectors, not only propagating ones, > > > > you can change at the end of this routine: > > > > > > > > ------------------- > > > > do i = 1, nchanl > > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > > enddo > > > > ------------------- > > > > > > > > to > > > > ------------------- > > > > > > > > do i = 1, 2*nstl > > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > > enddo > > > > ------------------- > > > > > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > > > the right states, another half, [nstl+1,2*nstl], - propagating or > > > > decaying to the left. In each group, first nchanl states are propagating > > > > states. > > > > > > > > The propagating states are normalized by the current and are arranged in > > > > the above order at the end of jbloch.f90 routine, after the following > > > > lines: > > > > > > > > ! > > > > ! Right ordering (+, >, -, <) > > > > ! > > > > > > > > > > > > > > > > Notice, that in the last versions the code gives in output > > > > both propagating to the right and to the left states. > > > > > > > > Hope this helps, > > > > Alexander > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > > > Dear All, > > > > > I am trying to figure out the left and right going Bloch's states in the > > > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > > > > > prints out Bloch's state moving in one direction. eg. in one of the > > > > > calculation- > > > > > k//=(0.375,-0.375) > > > > > Nchannels of the left tip = 1 > > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > > > > > Now if I want Bloch's state moving in right as well as left direction, I > > > > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > > > > AX=exp(ikd)BX, and out of those the ones with real solution would be our > > > > > Bloch's state, so I get for each channel two solutions- > > > > > kval (-0.275409421993275,1.823688001395235E-010) > > > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > > > > > To figure out the direction, I can calculate current associated with these > > > > > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > > > > > its left moving Bloch's state. I can print out current from jbloch.f90 > > > > > subroutine which are - > > > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > > > > other one right moving. Upto here its clear to me how to identify left and > > > > > right moving states. > > > > > > > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > > > > state. In another calculation where i get multiple Bloch's state- > > > > > Nchannels of the left tip = 5 > > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > > > -0.0746301 0.0000000 0.0000000 > > > > > 0.1205527 0.0000000 0.0000000 > > > > > 0.3112908 0.0000000 0.0000000 > > > > > 0.4200218 0.0000000 0.0000000 > > > > > -0.4935150 0.0000000 0.0000000 > > > > > > > > > > so i did the same trick i did above to first print out kz and then > > > > > current, which gives me - > > > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > > > 1.24280031534940 9.313897787790 > > > > > > > > > > So, the first 5 are left moving and rest are right moving. But I dont know > > > > > the pairs. for example for left moving state a, what is the corresponding > > > > > right moving state whether its f or g ... j ? > > > > > > > > > > Any help would be appreciated. > > > > > > > > > > Regards, > > > > > Manoj Srivastava > > > > > University of Florida, Gainesville. > > > > > > > > > > _______________________________________________ > > > > > Pw_forum mailing list > > > > > Pw_forum at pwscf.org > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > > e-mail: smogunov at sissa.it > > > > home-page: http://people.sissa.it/~smogunov > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > e-mail: smogunov at sissa.it > > home-page: http://people.sissa.it/~smogunov > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- e-mail: smogunov at sissa.it home-page: http://people.sissa.it/~smogunov From paulatto at sissa.it Thu Oct 29 17:46:56 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 29 Oct 2009 17:46:56 +0100 Subject: [Pw_forum] mpi-pwcond error In-Reply-To: <22095.68694.qm@web94612.mail.in2.yahoo.com> References: <22095.68694.qm@web94612.mail.in2.yahoo.com> Message-ID: In data 29 ottobre 2009 alle ore 05:43:25, ambavale sagar ha scritto: > One of the processes started by mpirun has exited with a nonzero > exit........ One of the pwcond.x processes has terminated with an error. It may both be an internal error (e.g. bad input parameter or algebra problem) or an external error (not enough ram, not enough disk space). In either case we cannot tell with the little information you provide. best regards P.S. it is not necessary to recompile anything when you upgrade the RAM, unless the RAM is defective, in which case you gave to give it back for a refund. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From paulatto at sissa.it Fri Oct 30 09:22:30 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 30 Oct 2009 09:22:30 +0100 Subject: [Pw_forum] (no subject) In-Reply-To: References: Message-ID: In data 29 ottobre 2009 alle ore 02:52:20, kajal jindal ha scritto: > make any bonds with the surrounding atoms even after optimizing the > lattice constants and atomic positions..It remains at that position as > an isolated atom only...So can anybody plz suggest me what should i do > in this > case???? Dear Kajal, what do you mean precisely by it does not make any bond? If you just cannot see any bond in XCrysDen or similar graphic programs it does not mean it isn't bound. Also, if it is not feeling any force it may be due to symmetry, try to break it slightly. > Also i want to know that whether this will affect my band structure > calculations??? Yes, of course. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From giuseppe.mattioli at mlib.ism.cnr.it Fri Oct 30 11:18:29 2009 From: giuseppe.mattioli at mlib.ism.cnr.it (Giuseppe Mattioli) Date: Fri, 30 Oct 2009 12:18:29 +0200 Subject: [Pw_forum] (no subject) In-Reply-To: References: Message-ID: <200910301118.30226.giuseppe.mattioli@mlib.ism.cnr.it> Are you sure that N atoms inserted into the ZnO lattice substitute Zn atoms? I would put the N atom at the O site, but you know what you're doing... Yours Giuseppe Mattioli On Thursday 29 October 2009 02:52:20 kajal jindal wrote: > Hii, > > I am working on ZnO doped with Nitrogen .I have used 2*2*2 ZnO supercell > ..If i replace 1 Zn atom by 1 N atom in that supercell,doped N atom does > not make any bonds with the surrounding atoms even after optimizing the > lattice constants and atomic positions..It remains at that position as an > isolated atom only...So can anybody plz suggest me what should i do in this > case????Also i want to know that whether this will affect my band > structure calculations??? > > sincerely, > Miss Kajal > (UTA)(University of Delhi) -- ******************************************************** - Article premier - Les hommes naissent et demeurent libres et ?gaux en droits. Les distinctions sociales ne peuvent ?tre fond?es que sur l'utilit? commune - Article 2 - Le but de toute association politique est la conservation des droits naturels et imprescriptibles de l'homme. Ces droits sont la libert?, la propri?t?, la s?ret? et la r?sistance ? l'oppression. ******************************************************** Giuseppe Mattioli CNR - ISTITUTO DI STRUTTURA DELLA MATERIA v. Salaria Km 29,300 - C.P. 10 I 00016 - Monterotondo Stazione (RM) Tel + 39 06 90672836 - Fax +39 06 90672316 E-mail: -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091030/c2e528d4/attachment.htm From hongli at imr.ac.cn Fri Oct 30 12:20:41 2009 From: hongli at imr.ac.cn (Hong LI) Date: Fri, 30 Oct 2009 19:20:41 +0800 Subject: [Pw_forum] nothing is written to the output file References: <20091028075354.14113.qmail@ms.hfcas.ac.cn> Message-ID: Dear users, We are using espresso to relax the atomic structures of an alloy, there is a problem that after several days of running the size of *.out is still zero, nothing is written to the output file. If we kill the job then the output file is not empty. Some small jobs run fine, I mean the information is updated from time to time during the calculation. It seems this is just an occasional case. Does anyone here encounter similar problem before, any suggestions to solve this problem ? Many thanks. Best wishes. Hong -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091030/248dbb92/attachment.htm From kazempoor2000 at yahoo.com Fri Oct 30 12:52:33 2009 From: kazempoor2000 at yahoo.com (ali kazempour) Date: Fri, 30 Oct 2009 04:52:33 -0700 (PDT) Subject: [Pw_forum] el-ph for vacancy Message-ID: <247564.21335.qm@web112520.mail.gq1.yahoo.com> Hi all I want to study vacancy problem in ionic material. When a vacancy forms in an ionic materials like ZnO, there is stronger electron-lattice coupling because of electrostatic interaction with nearly atom cores. These interaction repels the surrounding nearest neighbors away from empty lattice site. So electronic structure pictures can be affected by this type of interaction. I want to know Does espresso help me to detect and calculate the effect of el-ph interaction on band renormalization. I saw in example07 for Al that is metal, but I dont know that whether or not it can be used for semiconductor ? thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091030/771bd59e/attachment.htm From arvifis at gmail.com Fri Oct 30 14:35:05 2009 From: arvifis at gmail.com (Arles V. Gil Rebaza) Date: Fri, 30 Oct 2009 11:35:05 -0200 Subject: [Pw_forum] Tetrahedra method Message-ID: Hi, QE user, I would like obtain the magnetic properties of gamma-Fe4N , using Tetrahedra method ( occupations='tetrahedra') and non-colineal magnetism. So, after the first scf the program report this error...!! WARNING: integrated charge= 66.17220285, expected= 69.00000000 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from electrons : error # 1 charge is wrong %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... ---------------------------------------------------------------------------------------------- And there is my input..!! &control calculation='scf' restart_mode='from_scratch', pseudo_dir = './', outdir='./', prefix='fe4n' / &system ibrav= 1, celldm(1) =7.46, nat= 5, ntyp= 2, ecutwfc = 45, ecutrho = 400.0, report=1, occupations='tetrahedra', #smearing='fd', degauss=0.001 starting_magnetization(1)=0.1, starting_magnetization(2)=0.1, starting_magnetization(3)=0.1, starting_magnetization(4)=0.1, starting_magnetization(5)=0.1, noncolin = .true. angle1(1) = 90.0 angle2(1) = 0.0 angle1(2) = 90.0 angle2(2) = 0.0 angle1(3) = 90.0 angle2(3) = 0.0 angle1(4) = 90.0 angle2(4) = 0.0 angle1(5) = 90.0 angle2(5) = 0.0 / &electrons diagonalization='david' conv_thr = 1.0e-6 mixing_beta = 0.1 / ATOMIC_SPECIES Fe 55.847 Fe.pbe-sp-van.UPF N 14.000 N.pbe-rrkjus.UPF ATOMIC_POSITIONS Fe 0.0 0.0 0.0 Fe 0.5 0.5 0.0 Fe 0.5 0.0 0.5 Fe 0.0 0.5 0.5 N 0.5 0.5 0.5 K_POINTS automatic 20 20 20 0 0 0 When i use occupations=smearing and smearing=fd,mp, etc.. this error not appears. Somebody can help me please.... what's is my mistake...!!??? Best..!! ----------------------------------------------------------------- Arles V. Gil Rebaza Instituto de F?sica de La Plata La Plata - Argentina. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091030/f32cb62c/attachment.htm From giannozz at democritos.it Fri Oct 30 14:58:20 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 30 Oct 2009 14:58:20 +0100 Subject: [Pw_forum] Tetrahedra method In-Reply-To: References: Message-ID: <8E2004EC-803D-4322-9468-085E375E3F36@democritos.it> I am not sure tetrahedron method works in the noncolinear case. In any event, be careful because its usage for self-consistency will lead to small errors on forces. The main usage of tetrahedra is for calculation of the DOS. P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From sclauzer at sissa.it Fri Oct 30 15:26:49 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 30 Oct 2009 15:26:49 +0100 Subject: [Pw_forum] nothing is written to the output file In-Reply-To: References: <20091028075354.14113.qmail@ms.hfcas.ac.cn> Message-ID: <4AEAF7A9.2030804@sissa.it> Dear Hong Li, Hong LI wrote: > Dear users, > > We are using espresso to relax the atomic structures of an alloy, there > is a problem that after several days of running the size of *.out is > still zero, nothing is written to the output file. If we kill the job > then the output file is not empty. Some small jobs run fine, I mean the I suppose you're running on a parallel machine with MPI. If this is the case, I think that the delay after which the stdout is available depends on the specific MPI implementation and/or on how the machine is configured. Anyway, during execution you may have access to the allocated nodes and at least check if the code is running (using top command for instance). > information is updated from time to time during the calculation. It > seems this is just an occasional case. Does anyone here encounter If it is an occasional situation and it cannot be reproduced, it is probably due to a temporary failure of one of the nodes on which the job was running. It may be caused by many things, you should look at stderr and stdout of the job for error messages from the operating system/compiler/libraries/... You may find that it has been caused by disk failures/too much memory required/network communication problems/... These are not related to QE and cannot be solved by the users of this forum. If you have evidence that the problem persists, please provide a (possibly small) sample input and output files Regards GS problem before, any suggestions to solve this problem ? Many thanks. > > Best wishes. > Hong > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Fri Oct 30 15:32:08 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 30 Oct 2009 15:32:08 +0100 Subject: [Pw_forum] mpi-pwcond error In-Reply-To: <22095.68694.qm@web94612.mail.in2.yahoo.com> References: <22095.68694.qm@web94612.mail.in2.yahoo.com> Message-ID: <4AEAF8E8.30103@sissa.it> Dear Sagar, ambavale sagar wrote: > Dear PwScf users, > I am getting MPI error during PWCOND run at particular point. The > standard output gives: > > MPI_Allreduce: invalid communicator: Unknown error 2064 (rank 0, comm 16) > Rank (0, MPI_COMM_WORLD) : Call stack within LAM: > Rank (0, MPI_COMM_WORLD) : -MPI_Allreduce() > Rank (0, MPI_COMM_WORLD) : -main() > ------------------------------------------------------ > One of the processes started by mpirun has exited with a nonzero > exit........ > ... > .. > .. > PID 16077 failed on node n1 (192.168.0.12) with exit status 1. > ----------------------------- Is the error reproducible? Have you managed to run in parallel pw.x+pwcond.x for a smaller system that, let's say, can be done on 2 cpus? > > > I am running espresso-4.0 installed with lam-mpi 7.1.4 and ifort 10.1 on > dual cpu quad core xeon processors. There are 2 machines running in > parallel. Thus 4 cpus or 16 cores are available. However, this > calculation was ran on 4 cores only. The scf and relax calculations do > not crash. One more thing to note is: recently I have enhanced RAM on > both the machines. And after enhnacement did not do any reinstallation > or recompilation of any software including lam-mpi. I don't think this is needed. You simply need to reboot the machine and see if the OS sees the new RAM banks, I suppose. > > Note : Through search on forum archives I found a post by Derek : > http://www.democritos.it/pipermail/pw_forum/2005-November/003219.html > which indicates this might be a problem related to mpi or def.h...... > > > Thank you. > > Regards > Sagar Ambavale > PhD Student > The M.S. Uni. of Baroda > India > > ------------------------------------------------------------------------ > Now, send attachments up to 25MB with Yahoo! India Mail. Learn how > . > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Fri Oct 30 15:37:30 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 30 Oct 2009 15:37:30 +0100 Subject: [Pw_forum] plane averaged local potential In-Reply-To: <627e0ffa0910281418tc41f397p9e87c0dff492b350@mail.gmail.com> References: <627e0ffa0910281418tc41f397p9e87c0dff492b350@mail.gmail.com> Message-ID: <4AEAFA2A.1020209@sissa.it> Dear Eduardo and Ghosh, Prasenjit Ghosh wrote: > > > 2009/10/28 Eduardo Ariel Menendez Proupin > > > Hello, > I need to calculate plane averaged local potentials, and charges, > for a heterostructure. I.e., > V(z)=1/(Lx*Ly) \int\int V(x,y,z) dx dy > > I think I can do it using PP to get V(x,y,z) in the 3D FFT grid and > a write a program that makes the average. Then I want to obtain the > macroscopic average, as defined in a triestine paper by Baldereschi, > Baroni and Resta PRL 61 ,734 (1988). > > Before spending one or two days writing the program and become sure > that is works fine, I would like to confirm that this is not already > implemented in PP or in other utility of Quantum-ESPRESSO. > > Hi Eduardo, > > The executable average.x can do the job........for the input file you > can see the example on Work function calculation...... > However, if you want to calculate the planar average of charge density, > you need to multiply it with the area of the xy plane, assuming you want > to plot along the z-direction. I can confirm. I've been using that and it should give exactly the integral you wrote above. > > Also I think (I'm not sure) it works only if the z- is perpendicular to > the xy-plane. There is a variable that let's you chose the direction along which to plot the quantity (the other two are integrated up). Please read the header of PP/average.f90. Cheers, GS > > With regards, > Prasenjit. > > > > -- > PRASENJIT GHOSH, > POST-DOC, > ROOM NO: 265, MAIN BUILDING, > CM SECTION, ICTP, > STRADA COSTERIA 11, > TRIESTE, 34104, > ITALY > PHONE: +39 040 2240 369 (O) > +39 3807528672 (M) > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Fri Oct 30 16:09:14 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 30 Oct 2009 16:09:14 +0100 Subject: [Pw_forum] Charge ordering with QE In-Reply-To: References: Message-ID: <4AEB019A.6090508@sissa.it> Dear Mattia, Mulazzi Mattia wrote: > In order to calculate the electronic structure of an orbitally-ordered > system I think I have to use a supercell and specify different > occupation numbers for the different Vanadium sites. The problem is > that I don't know how to do that. I tried specifying the > starting_ns_eigenvalue variable on different sites, but the converged > calculation is identical to the calculation with > starting_ns_eigenvalue left unspecified. In the INPUT_PW it is written > that the occupation matrices can be specified only when the LDA+U > scheme is used. I don't understand a thing: starting_ns_eigenvalue is used only on the species on which you put +U (i.e. Hubbard_U(?)>0), so if you cannot not use LDA+U for V with your QE installation, how could you say that using starting_ns_eigenvalue does not change the result? One other important point is that occupations are symmetrized according to the symmetry of the system. So that, perhaps, retrying after breaking some symmetry or using nosym=.TRUE. may change something. > However, I cannot use the LDA+U on my system (Mac OS X 10.5.6) because > every time I use the Vanadium pseudopotential (either the one I > generated, ether the one on the QE database) I find a "segmentation > fault" that I cannot solve. BTW, the segmentation fault error does > not appear for Co or Fe or Ni pseudopotentials. Which version of QE are you using? Please check if the Vanadium case has already been inserted in tabd.f90 and set_hubbard_l.f90 files. If not pw.x should stop with an error message like "pseudopotential not yet inserted" before causing any segmentation fault. From version 4.1 on, LDA+U should be working with any of the transition metal elements. If you have previous versions, you can easily get it working modifying those two files. Please report to us if the problem persists in v4.1 and if it is reproducible. Thanks, GS > > Thanks everybody for the help, > > Mattia Mulazzi > > > FPR Fellow of RIKEN at Spring8 > Excitation Order Research Team > 1-1-1 Sayo-cho Sayo-gun, Hyogo > Japan > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From manoj at phys.ufl.edu Fri Oct 30 17:57:12 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Fri, 30 Oct 2009 12:57:12 -0400 (EDT) Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: <1256830035.3467.13.camel@alex-laptop> Message-ID: Dear Alexander, Thank you very much! I read the code, it seems that there are quiet a few changes in 4.1.1 compared to 4.0.4 :) .So, just to make sure, if I want to calculate r and t for left going Bloch's state as well as right going Bloch's state, I need to run the code two times, one time with 'left_to_right' switch true, and another time false. I dont need to change scattering region or interchange left and right leads. Regards, Manoj Srivastava On Thu, 29 Oct 2009, Alexander Smogunov wrote: > Dear Manoj > > On Wed, 2009-10-28 at 01:49 -0400, Manoj Srivastava wrote: > > Dear Alexander, > > You are correct, left and right moving states are in general not > > related. I was wrong about trying to create a pair. Thanks for > > clarification. > > I have a few more questions. > > 1. I want to find out the reflection coefficient in the code. In > > transmit.f90, transmission coefficient is |tmat(ig,n)|^2, where tmat is > > vec1(ntran-n2d-npol*nocrosr+ig,n). (line 206, of version 4.0.5). > > Reflection is not explicitly calculated in the code, but it should be > > modulus square of vec1(2*n2d+npol*norbs+ig,n) (line 234). Am I correct? > > in the 4.1 version there is output of both t and r. > > > > > > 2. In the code reflection and transmission coefficients are calculated > > for one direction lets say t++ and t+-, which is transmission and > > reflection coefficient for Bloch' state moving from - to + , or in other > > words, left to right. To calculate transmission and reflection for Bloch's > > state going from right to left, what should I do? Should i just rotate my > > system 180 degree, that way leads get interchanged and also scattering > > area has orientation 180 different from before. > > > > again in the 4.1 version you have possibility to calculate t and r for > both directions, in the do_cond.f90 the transmit routine is called with > an extra parameter which is true (false) if you want to calculate > scattering states propagating from the left (from the right). > > > 3. In the subroutine sunitary.f90, is raux measure of conservation of > > flux? If I sum all the transmission and reflecion coefficient for > > a particular Bloch's state, I should get 1, and computationally a number > > close to 1. How large could it be for S matrix not to be unitary, i m > > getting raux more than one e.g. 1.0011 but code still considers smatrix > > to be unitary. > > in the case when |R + T - 1| > 1.d-4 for some band, it should print out > the value of R+T but continues to run anyway. > > All the best, > Alexander > > > > > > > Regards, > > Manoj Srivastava > > University of Florida, Gainesville, FL > > > > > > > > > > > > > > On Thu, 22 > > Oct 2009, Alexander Smogunov wrote: > > > > > On Thu, 2009-10-22 at 11:34 -0400, Manoj Srivastava wrote: > > > > Dear Alexander, > > > > Thanks for your answer. I just want to make sure. Imagine we have total > > > > number of channels in the left lead 2, so total number of Bloch's state > > > > are 4. 2 of them left going say a and b, and 2 right going say c and d. > > > > So, are you saying that for left going state a, the corresponding right > > > > going state is c? Are they ordered this way? > > > > > > what do you mean by corresponding? Left and right moving Bloch > > > states are in general not related one to another, you can even have > > > different number of them ... Only if you have some symmetry S which > > > brings kz to -kz conserving k_parallel, then the state with > > > \psi_{-kz} will be S \psi{kz}. This is true for example at 2D G point > > > when you have time reversal operation. > > > > > > Now the code simply arranges the propagating states in the order of > > > increasing |k_z|... > > > > > > Regards, Alexander > > > > > > > > > > > > > > Regards, > > > > Manoj > > > > > > > > > > > > On Thu, 22 Oct 2009, Alexander > > > > Smogunov wrote: > > > > > > > > > Dear Manoj. > > > > > > > > > > The output of complex k vectors is performed in > > > > > summary_band.f90 routine. If you want to see all > > > > > the complex k vectors, not only propagating ones, > > > > > you can change at the end of this routine: > > > > > > > > > > ------------------- > > > > > do i = 1, nchanl > > > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > > > enddo > > > > > ------------------- > > > > > > > > > > to > > > > > ------------------- > > > > > > > > > > do i = 1, 2*nstl > > > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > > > enddo > > > > > ------------------- > > > > > > > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > > > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > > > > the right states, another half, [nstl+1,2*nstl], - propagating or > > > > > decaying to the left. In each group, first nchanl states are propagating > > > > > states. > > > > > > > > > > The propagating states are normalized by the current and are arranged in > > > > > the above order at the end of jbloch.f90 routine, after the following > > > > > lines: > > > > > > > > > > ! > > > > > ! Right ordering (+, >, -, <) > > > > > ! > > > > > > > > > > > > > > > > > > > > Notice, that in the last versions the code gives in output > > > > > both propagating to the right and to the left states. > > > > > > > > > > Hope this helps, > > > > > Alexander > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > > > > Dear All, > > > > > > I am trying to figure out the left and right going Bloch's states in the > > > > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > > > > > > prints out Bloch's state moving in one direction. eg. in one of the > > > > > > calculation- > > > > > > k//=(0.375,-0.375) > > > > > > Nchannels of the left tip = 1 > > > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > > > > > > > Now if I want Bloch's state moving in right as well as left direction, I > > > > > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > > > > > AX=exp(ikd)BX, and out of those the ones with real solution would be our > > > > > > Bloch's state, so I get for each channel two solutions- > > > > > > kval (-0.275409421993275,1.823688001395235E-010) > > > > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > > > > > > > To figure out the direction, I can calculate current associated with these > > > > > > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > > > > > > its left moving Bloch's state. I can print out current from jbloch.f90 > > > > > > subroutine which are - > > > > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > > > > > other one right moving. Upto here its clear to me how to identify left and > > > > > > right moving states. > > > > > > > > > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > > > > > state. In another calculation where i get multiple Bloch's state- > > > > > > Nchannels of the left tip = 5 > > > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > > > > > -0.0746301 0.0000000 0.0000000 > > > > > > 0.1205527 0.0000000 0.0000000 > > > > > > 0.3112908 0.0000000 0.0000000 > > > > > > 0.4200218 0.0000000 0.0000000 > > > > > > -0.4935150 0.0000000 0.0000000 > > > > > > > > > > > > so i did the same trick i did above to first print out kz and then > > > > > > current, which gives me - > > > > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > > > > 1.24280031534940 9.313897787790 > > > > > > > > > > > > So, the first 5 are left moving and rest are right moving. But I dont know > > > > > > the pairs. for example for left moving state a, what is the corresponding > > > > > > right moving state whether its f or g ... j ? > > > > > > > > > > > > Any help would be appreciated. > > > > > > > > > > > > Regards, > > > > > > Manoj Srivastava > > > > > > University of Florida, Gainesville. > > > > > > > > > > > > _______________________________________________ > > > > > > Pw_forum mailing list > > > > > > Pw_forum at pwscf.org > > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > > > > > e-mail: smogunov at sissa.it > > > > > home-page: http://people.sissa.it/~smogunov > > > > > > > > > > _______________________________________________ > > > > > Pw_forum mailing list > > > > > Pw_forum at pwscf.org > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > -- > > > e-mail: smogunov at sissa.it > > > home-page: http://people.sissa.it/~smogunov > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > -- > e-mail: smogunov at sissa.it > home-page: http://people.sissa.it/~smogunov > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From manoj at phys.ufl.edu Fri Oct 30 19:04:16 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Fri, 30 Oct 2009 14:04:16 -0400 (EDT) Subject: [Pw_forum] Left and right going Bloch's states from PWCOND In-Reply-To: Message-ID: Just to add one more question. what is 'lorb' switch. I get following- lorb, & ! if .t. calculate the scattering (or Bloch) states so, i guess for transmission calculation i will have to set it to 't'. now- IF (lorb.and.ikind.eq.2) call errore('do_cond','lorb not working with ikind = 2',1) so, it seems that calculation for r and t for right and left going state, when left and right leads are different doesnot work. Is that correct? A possible work around may be rotating the whole system by 180 degree, which would be like treating left lead as right lead and scattering region rotated. Can I do that? Regards, Manoj On Fri, 30 Oct 2009, Manoj Srivastava wrote: > Dear Alexander, > Thank you very much! I read the code, it seems that there are > quiet a few changes in 4.1.1 compared to 4.0.4 :) .So, just to make sure, > if I want to calculate r and t for left going Bloch's state as well as > right going Bloch's state, I need to run the code two times, one time with > 'left_to_right' switch true, and another time false. I dont need to change > scattering region or interchange left and right leads. > > Regards, > Manoj Srivastava > On Thu, 29 Oct 2009, Alexander > Smogunov wrote: > > > Dear Manoj > > > > On Wed, 2009-10-28 at 01:49 -0400, Manoj Srivastava wrote: > > > Dear Alexander, > > > You are correct, left and right moving states are in general not > > > related. I was wrong about trying to create a pair. Thanks for > > > clarification. > > > I have a few more questions. > > > 1. I want to find out the reflection coefficient in the code. In > > > transmit.f90, transmission coefficient is |tmat(ig,n)|^2, where tmat is > > > vec1(ntran-n2d-npol*nocrosr+ig,n). (line 206, of version 4.0.5). > > > Reflection is not explicitly calculated in the code, but it should be > > > modulus square of vec1(2*n2d+npol*norbs+ig,n) (line 234). Am I correct? > > > > in the 4.1 version there is output of both t and r. > > > > > > > > > > 2. In the code reflection and transmission coefficients are calculated > > > for one direction lets say t++ and t+-, which is transmission and > > > reflection coefficient for Bloch' state moving from - to + , or in other > > > words, left to right. To calculate transmission and reflection for Bloch's > > > state going from right to left, what should I do? Should i just rotate my > > > system 180 degree, that way leads get interchanged and also scattering > > > area has orientation 180 different from before. > > > > > > > again in the 4.1 version you have possibility to calculate t and r for > > both directions, in the do_cond.f90 the transmit routine is called with > > an extra parameter which is true (false) if you want to calculate > > scattering states propagating from the left (from the right). > > > > > 3. In the subroutine sunitary.f90, is raux measure of conservation of > > > flux? If I sum all the transmission and reflecion coefficient for > > > a particular Bloch's state, I should get 1, and computationally a number > > > close to 1. How large could it be for S matrix not to be unitary, i m > > > getting raux more than one e.g. 1.0011 but code still considers smatrix > > > to be unitary. > > > > in the case when |R + T - 1| > 1.d-4 for some band, it should print out > > the value of R+T but continues to run anyway. > > > > All the best, > > Alexander > > > > > > > > > > > > Regards, > > > Manoj Srivastava > > > University of Florida, Gainesville, FL > > > > > > > > > > > > > > > > > > > > > On Thu, 22 > > > Oct 2009, Alexander Smogunov wrote: > > > > > > > On Thu, 2009-10-22 at 11:34 -0400, Manoj Srivastava wrote: > > > > > Dear Alexander, > > > > > Thanks for your answer. I just want to make sure. Imagine we have total > > > > > number of channels in the left lead 2, so total number of Bloch's state > > > > > are 4. 2 of them left going say a and b, and 2 right going say c and d. > > > > > So, are you saying that for left going state a, the corresponding right > > > > > going state is c? Are they ordered this way? > > > > > > > > what do you mean by corresponding? Left and right moving Bloch > > > > states are in general not related one to another, you can even have > > > > different number of them ... Only if you have some symmetry S which > > > > brings kz to -kz conserving k_parallel, then the state with > > > > \psi_{-kz} will be S \psi{kz}. This is true for example at 2D G point > > > > when you have time reversal operation. > > > > > > > > Now the code simply arranges the propagating states in the order of > > > > increasing |k_z|... > > > > > > > > Regards, Alexander > > > > > > > > > > > > > > > > > > Regards, > > > > > Manoj > > > > > > > > > > > > > > > On Thu, 22 Oct 2009, Alexander > > > > > Smogunov wrote: > > > > > > > > > > > Dear Manoj. > > > > > > > > > > > > The output of complex k vectors is performed in > > > > > > summary_band.f90 routine. If you want to see all > > > > > > the complex k vectors, not only propagating ones, > > > > > > you can change at the end of this routine: > > > > > > > > > > > > ------------------- > > > > > > do i = 1, nchanl > > > > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > > > > enddo > > > > > > ------------------- > > > > > > > > > > > > to > > > > > > ------------------- > > > > > > > > > > > > do i = 1, 2*nstl > > > > > > WRITE( stdout,'(3f12.7)') DBLE(kvall(i)), AIMAG(kvall(i)), eev > > > > > > enddo > > > > > > ------------------- > > > > > > > > > > > > Altogether there are 2*nstl (or 2*nstr) Bloch states in the left > > > > > > (or right) lead. First half, [1,nstl], are propagating or decaying to > > > > > > the right states, another half, [nstl+1,2*nstl], - propagating or > > > > > > decaying to the left. In each group, first nchanl states are propagating > > > > > > states. > > > > > > > > > > > > The propagating states are normalized by the current and are arranged in > > > > > > the above order at the end of jbloch.f90 routine, after the following > > > > > > lines: > > > > > > > > > > > > ! > > > > > > ! Right ordering (+, >, -, <) > > > > > > ! > > > > > > > > > > > > > > > > > > > > > > > > Notice, that in the last versions the code gives in output > > > > > > both propagating to the right and to the left states. > > > > > > > > > > > > Hope this helps, > > > > > > Alexander > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On Wed, 2009-10-21 at 14:13 -0400, Manoj Srivastava wrote: > > > > > > > Dear All, > > > > > > > I am trying to figure out the left and right going Bloch's states in the > > > > > > > lead from PWCOND. For a given (kx,ky)and energy we get kz. The code only > > > > > > > prints out Bloch's state moving in one direction. eg. in one of the > > > > > > > calculation- > > > > > > > k//=(0.375,-0.375) > > > > > > > Nchannels of the left tip = 1 > > > > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > > > > > > > 0.3157801 0.0000000 0.0000000 > > > > > > > > > > > > > > Now if I want Bloch's state moving in right as well as left direction, I > > > > > > > can go to kbloch.f90 subroutine, and print out all the eigen values of > > > > > > > AX=exp(ikd)BX, and out of those the ones with real solution would be our > > > > > > > Bloch's state, so I get for each channel two solutions- > > > > > > > kval (-0.275409421993275,1.823688001395235E-010) > > > > > > > kval (0.315780119742506,-3.611201785292708E-012) > > > > > > > > > > > > > > To figure out the direction, I can calculate current associated with these > > > > > > > Bloch's sate and if the current is +ive it is right moving , and if '-'ive > > > > > > > its left moving Bloch's state. I can print out current from jbloch.f90 > > > > > > > subroutine which are - > > > > > > > current eigenvalue -1.86502143831863 1.59149029314457 > > > > > > > > > > > > > > So, clearly the first state with kval=-0.2754094 is left moving and the > > > > > > > other one right moving. Upto here its clear to me how to identify left and > > > > > > > right moving states. > > > > > > > > > > > > > > I get confused when for a given (kx,ky,E), I have more than one Bloch' > > > > > > > state. In another calculation where i get multiple Bloch's state- > > > > > > > Nchannels of the left tip = 5 > > > > > > > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > > > > > > > > > > > > > -0.0746301 0.0000000 0.0000000 > > > > > > > 0.1205527 0.0000000 0.0000000 > > > > > > > 0.3112908 0.0000000 0.0000000 > > > > > > > 0.4200218 0.0000000 0.0000000 > > > > > > > -0.4935150 0.0000000 0.0000000 > > > > > > > > > > > > > > so i did the same trick i did above to first print out kz and then > > > > > > > current, which gives me - > > > > > > > kval (-0.420023481074359,1.979595081419732E-010) (call it a) > > > > > > > kval (0.420023367986768,2.500979698670295E-011) (b) > > > > > > > kval (-0.306507431678779,-1.236804629184431E-011) (c) > > > > > > > kval (-0.125376071175573,-6.134512510438736E-011) (d) > > > > > > > kval (-7.945001124706894E-002,6.683546930037856E-011)(e) > > > > > > > kval (0.106554601758169,-6.427946951285107E-011) (f) > > > > > > > kval (8.866867725358024E-002,8.342250371574646E-011) (g) > > > > > > > kval (0.325333314672671,1.260810749228185E-011) (h) > > > > > > > kval (-0.488725859521576,1.769197678346003E-010) (i) > > > > > > > kval (0.479509832763231,1.765499400037283E-010) (j) > > > > > > > > > > > > > > current eigenvalue -9.31389492882581 -1.24296522993488 > > > > > > > -1.21324078359658 -1.11950286753963 -1.08166842367443 > > > > > > > 1.08187482164864 1.11973146584263 1.21295295042188 > > > > > > > 1.24280031534940 9.313897787790 > > > > > > > > > > > > > > So, the first 5 are left moving and rest are right moving. But I dont know > > > > > > > the pairs. for example for left moving state a, what is the corresponding > > > > > > > right moving state whether its f or g ... j ? > > > > > > > > > > > > > > Any help would be appreciated. > > > > > > > > > > > > > > Regards, > > > > > > > Manoj Srivastava > > > > > > > University of Florida, Gainesville. > > > > > > > > > > > > > > _______________________________________________ > > > > > > > Pw_forum mailing list > > > > > > > Pw_forum at pwscf.org > > > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > -- > > > > > > e-mail: smogunov at sissa.it > > > > > > home-page: http://people.sissa.it/~smogunov > > > > > > > > > > > > _______________________________________________ > > > > > > Pw_forum mailing list > > > > > > Pw_forum at pwscf.org > > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > > > > > > > _______________________________________________ > > > > > Pw_forum mailing list > > > > > Pw_forum at pwscf.org > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > > e-mail: smogunov at sissa.it > > > > home-page: http://people.sissa.it/~smogunov > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > e-mail: smogunov at sissa.it > > home-page: http://people.sissa.it/~smogunov > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > From jcwu.suda at hotmail.com Sat Oct 31 10:23:15 2009 From: jcwu.suda at hotmail.com (wujianchun) Date: Sat, 31 Oct 2009 17:23:15 +0800 Subject: [Pw_forum] which is the local potential Message-ID: Dear All, I am calculating the formation energy of defects by PWSCF. I want to correct the E_vbm(E_vbm is the energy of the valence band maximum) so as to cancel the background charge neutralizing the defective supercell at the charge state q, E_vbm=E_vbm(bulk)+deltaV. Here deltaV is the averaged difference between the local potentials far from the defect in the defective supercell and the corresponding ones in the perfect supercell. The question is which one, 1 total potential V_bare+V_H+V_xc, 11 the V_bare+V_H potential or 12 the electric field potential in plot_num, corresponds the above-mentioned "local potentials"? PS: which is the unit of the potentials in the outfiles, eV or Ry? Thanks!regardsJianchun Wu _________________________________________________________________ ?? Windows 7????????????????? PC? http://www.microsoft.com/china/windows/buy/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091031/1273039e/attachment.htm From anvari_meh at physics.iust.ac.ir Sat Oct 31 10:46:03 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Sat, 31 Oct 2009 13:16:03 +0330 Subject: [Pw_forum] umklapp part in elph calculations Message-ID: Dear PWscf users I want to know that in elph calculations, are the elph interactions restricted in the unit cell (because it seems that k+q is laid in unit cell) or the code evaluates the umklapp process. Best Regards Mehrnaz Anvari Iran University of science & technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20091031/8185226e/attachment.htm