From rulin_cs at foxmail.com Thu Nov 10 08:33:56 2016 From: rulin_cs at foxmail.com (rulin_cs at foxmail.com) Date: Thu, 10 Nov 2016 15:33:56 +0800 Subject: [Wannier] How to pass MLWF to pw for a new calculation? Message-ID: <201611101533532102618@foxmail.com> Dear all, I am now doing DFT+U with pwscf, and I use pw2wannier90.x to find MLWFs. But how should I do in detail if I want to use these MLWFs to start a new pw calculation? Regards, Rulin Liu rulin_cs at foxmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From ch-wang at outlook.com Sun Nov 13 04:14:53 2016 From: ch-wang at outlook.com (Chong Wang) Date: Sun, 13 Nov 2016 03:14:53 +0000 Subject: [Wannier] how to read position matrix Message-ID: Hi everyone, For some reason, I need access to position matrix elements of Wannier functions, i.e. <0n|r|Rm>. (see also eq.39 in "Ab initio calculation of the anomalous Hall conductivity by Wannier interpolation".) I have found a subroutine named get_AA_R in get_oper.F90 in the source code. I have two questions: 1. Is it correct to just print out the AA_R matrix in get_AA_R subroutine? Are they just <0n|r|Rm>? 2. There's a comment saying: " Since Eq.(44) WYSV06 does not preserve the Hermiticity of the Berry potential matrix, take Hermitean part (whether this makes a difference or not for e.g. the AHC, depends on which expression is used to evaluate the Berry curvature. See comments in berry_wanint.F90) " I don't find a file named berry_wanint.F90. Can anyone explain this to me? Best! Chong Wang Institute for Advanced Study, Tsinghua University, Beijing, 100084 -------------- next part -------------- An HTML attachment was scrubbed... URL: From santubaidya2009 at gmail.com Wed Nov 16 09:10:58 2016 From: santubaidya2009 at gmail.com (Santu Baidya) Date: Wed, 16 Nov 2016 08:10:58 +0000 Subject: [Wannier] Problem in increasing dimension of neighbor hopping shells in the Real space Hamiltonian in wannier90 Message-ID: Dear wannier90 developers, I have calculated ?wannier90_hr.dat for my system which includes neighbor shells from [-4 3 3] to [4, 3, 3]. I want to increase the number of neighbor shells to [7 5 5] (for example). Please tell me which tag I should use to increase neighboring shells in wannier90_hr.dat. Thanking you, Santu BaidyaSeoul National UniversitySouth Korea -------------- next part -------------- An HTML attachment was scrubbed... URL: From szaboa at iis.ee.ethz.ch Wed Nov 16 10:30:58 2016 From: szaboa at iis.ee.ethz.ch (Aron Szabo) Date: Wed, 16 Nov 2016 10:30:58 +0100 Subject: [Wannier] Problem in increasing dimension of neighbor hopping shells in the Real space Hamiltonian in wannier90 In-Reply-To: References: Message-ID: <582C2752.3040006@iis.ee.ethz.ch> The easiest (and as far as I know maybe the only) way to do this is to increase the number of K-points in the scf simulation. The number of the real-space unit cell vectors in the hr_dat file are roughly the same as the number of K points in the simulation. If your simulation is already accurate enough with the current number of K points, then keep the charge density fixed and just do a non self-consistent simulation on a finer K point grid before doing the Wannier transformation. (It is also possible to calculate the hr_dat elements by yourself using the U matrix and the Bloch eigenenergies, or there might be even a command file option to increase those shells that I'm not aware of.) Best, Aron On 11/16/2016 09:10 AM, Santu Baidya wrote: > Dear wannier90 developers, > > I have calculated wannier90_hr.dat for my system which > includes neighbor shells from [-4 3 3] to [4, 3, 3]. I want to > increase the number of neighbor shells to [7 5 5] (for example). > > Please tell me which tag I should use to increase neighboring shells > in wannier90_hr.dat. > > > Thanking you, > > Santu Baidya > Seoul National University > South Korea > > > > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier -------------- next part -------------- An HTML attachment was scrubbed... URL: From rupsaguharoy88 at gmail.com Fri Nov 18 13:25:26 2016 From: rupsaguharoy88 at gmail.com (Rupsa GuhaRoy) Date: Fri, 18 Nov 2016 17:55:26 +0530 Subject: [Wannier] Error in x Wannier90 Message-ID: Dear all, I am trying to do a wannier fitting for a system form a GGA+U wien2k bandstructure calculation. Each step as specified in userguide is fine but the last step (x wannier90 -up/dn) each time it is showing the some error. Error: *Found keyword mp_grid more than once in input file* Could anyone please help me in this case? regards, Rupsa -------------- next part -------------- An HTML attachment was scrubbed... URL: From rupsaguharoy88 at gmail.com Sat Nov 19 06:39:48 2016 From: rupsaguharoy88 at gmail.com (Rupsa GuhaRoy) Date: Sat, 19 Nov 2016 11:09:48 +0530 Subject: [Wannier] Error in x Wannier90 In-Reply-To: References: Message-ID: Dear all, I need further help for my calculation. I am new to wien2wannier. Could anyone please tell me how/steps to extract hopping parameters/tight-binding terms from case_hr.dat? Thanks in advance. regards, Rupsa On Fri, Nov 18, 2016 at 5:55 PM, Rupsa GuhaRoy wrote: > Dear all, > > I am trying to do a wannier fitting for a system form > a GGA+U wien2k bandstructure calculation. > Each step as specified in userguide is fine but the > last step (x wannier90 -up/dn) each time it is showing the some error. > > Error: *Found keyword mp_grid more than once in input file* > > Could anyone please help me in this case? > > regards, > Rupsa > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rupsaguharoy88 at gmail.com Sat Nov 19 13:33:25 2016 From: rupsaguharoy88 at gmail.com (Rupsa GuhaRoy) Date: Sat, 19 Nov 2016 18:03:25 +0530 Subject: [Wannier] hopping terms Message-ID: Dear all, I need further help for my calculation. I am new to wien2wannier. Could anyone please tell me how/steps to extract hopping parameters/tight-binding terms from case_hr.dat? Thanks in advance. regards, Rupsa -------------- next part -------------- An HTML attachment was scrubbed... URL: From andreichibisov at yandex.ru Mon Nov 21 17:59:08 2016 From: andreichibisov at yandex.ru (Andrey Chibisov) Date: Tue, 22 Nov 2016 02:59:08 +1000 Subject: [Wannier] Platinum Message-ID: <297411479747548@web4h.yandex.ru> Dear Colleagues, I want to calculate the Seebeck coefficient for platinum. My input file is Pt.sa.win. I attached it. However, when I was running the Wannier90 issued a message: param_get_projection: Problem reading l state sp2d Error: examine the output/error file for details I can not understand how specify the Projections option -- Best regards, Andrey Chibisov. Ph.D. Numerical method of mathematical physics Laboratory, Computational Center, Russian Academy of Sciences. Khabarovsk, Russia Web page:?https://www.researchgate.net/profile/A_Chibisov http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en -------------- next part -------------- A non-text attachment was scrubbed... Name: Pt.sa.win Type: application/octet-stream Size: 28256 bytes Desc: not available URL: From santubaidya2009 at gmail.com Tue Nov 22 02:13:05 2016 From: santubaidya2009 at gmail.com (Santu Baidya) Date: Tue, 22 Nov 2016 01:13:05 +0000 Subject: [Wannier] Fixing Wannier centers in wannier90 Message-ID: Dear Wannier90 users and developers, I am using wannier90 to get wannier functions for a compound. I need to get wannier functions at particular sites. However, after getting MLWF I see wannier functions go to some other sites as they choose ML at those sites. Is there anyway I can get MLWF at the sites where I want, like projected wannier functions. Any suggestion would be appreciated. Thanking you, Santu BaidyaIBS-CCESSeoul National University -------------- next part -------------- An HTML attachment was scrubbed... URL: From meisam.a63 at gmail.com Wed Nov 23 16:08:14 2016 From: meisam.a63 at gmail.com (Mortaza Aghtar) Date: Wed, 23 Nov 2016 16:08:14 +0100 Subject: [Wannier] spin-orbit couplings of MLWFs Message-ID: Dear all, I am new to wannier program. I have followed example 17 to calculate the MLWFs for a non-collinear structure. Now my question is: Is there a way to print the spin-orbit coupling values for all the MLWFs? Regards, Mortaza Mortaza Aghtar, Ph.D. Ulm Universit?t Institut f?r Theoretiche Physik Albert-Einstein-Allee 11 89081 Ulm Office: 405/O25 Telephone: (+49) 731-50-22905 Group Page: http://qubit-ulm.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From andreichibisov at yandex.ru Fri Nov 25 02:20:20 2016 From: andreichibisov at yandex.ru (Andrey Chibisov) Date: Fri, 25 Nov 2016 11:20:20 +1000 Subject: [Wannier] spin orbit coupling with USP/PAW pseudopotentials Message-ID: <4778791480036820@web40m.yandex.ru> Dear Colleagues, I recently found a link: http://mailman.qe-forge.org/pipermail/wannier/2013-November/000725.html Please tell me, can I use USP/PAW pseudopotentials for Wannier90? Or not? -- Best regards, Andrey Chibisov. Ph.D. Numerical method of mathematical physics Laboratory, Computational Center, Russian Academy of Sciences. Khabarovsk, Russia Web page:?https://www.researchgate.net/profile/A_Chibisov http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en From pipidog at gmail.com Fri Nov 25 02:45:49 2016 From: pipidog at gmail.com (Shu-Ting Pi) Date: Thu, 24 Nov 2016 17:45:49 -0800 Subject: [Wannier] spin orbit coupling with USP/PAW pseudopotentials In-Reply-To: <4778791480036820@web40m.yandex.ru> References: <4778791480036820@web40m.yandex.ru> Message-ID: No, you cannot. Some codes can support Wannier90 using PAW such as ABINIT (but it has problem with SOC) or VASP. As for QE, you have to use NCPP. An easy way is to generate NCPP with SOC using ONCVPSP which I believe you can find some useful instructions by Google. On Thu, Nov 24, 2016 at 5:20 PM, Andrey Chibisov wrote: > Dear Colleagues, > I recently found a link: > http://mailman.qe-forge.org/pipermail/wannier/2013-November/000725.html > > Please tell me, can I use USP/PAW pseudopotentials for Wannier90? > Or not? > > -- > Best regards, > Andrey Chibisov. Ph.D. > Numerical method of mathematical physics Laboratory, > Computational Center, Russian Academy of Sciences. > Khabarovsk, Russia > Web page: https://www.researchgate.net/profile/A_Chibisov > http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier > -------------- next part -------------- An HTML attachment was scrubbed... URL: From taichizi at mail.ustc.edu.cn Fri Nov 25 05:30:15 2016 From: taichizi at mail.ustc.edu.cn (Rui Pang) Date: Fri, 25 Nov 2016 12:30:15 +0800 (GMT+08:00) Subject: [Wannier] How to make the wannierzation converge? Message-ID: <3ffa29ae.3839.15899bf83f7.Coremail.taichizi@mail.ustc.edu.cn> Dear all I am a new user to wannier90. To learn how to use this code, I add some modifications on example10, the graphite. I found that if the number of wannier functions is increased into all valence electrons, the calculation is diffcult to get convergence. I adjust dis_froz_max, trial_step, and cancle the distenglement, all of them cannot work. My question is how should I understand this? As I see some replies always attribute the convergence problems to the inccorect chosing of dis_froz_max and projection states. But it is clear that here the projection states is nothing wrong. And it seems that nothing need to be changed according to the priciple of distenglement. So how to understand this behavior and how to fix the related parameters to make the calculation converge? As it is an example released with wannier90, I think it is not necessary for me to attach my input file and other supplementary files. Thanks for your reply. Pang Rui -- Pang Rui, Lectureer School of physics and engneering Zhengzhou University China -------------- next part -------------- An HTML attachment was scrubbed... URL: From andreichibisov at yandex.ru Fri Nov 25 06:35:13 2016 From: andreichibisov at yandex.ru (Andrey Chibisov) Date: Fri, 25 Nov 2016 15:35:13 +1000 Subject: [Wannier] spin orbit coupling with USP/PAW pseudopotentials In-Reply-To: References: <4778791480036820@web40m.yandex.ru> Message-ID: <4232221480052113@web28j.yandex.ru> Dear Shu-Ting Pi, Thanks a lot for your comments. 25.11.2016, 11:45, "Shu-Ting Pi" : > No, you cannot. Some codes can support Wannier90 using PAW such as > ABINIT (but it has problem with SOC) or VASP. As for QE, you have to > use NCPP.? An easy way is to generate NCPP with SOC using ONCVPSP > which I believe you can find some useful instructions by Google. > > On Thu, Nov 24, 2016 at 5:20 PM, Andrey Chibisov wrote: >> Dear Colleagues, >> I recently found a link: >> http://mailman.qe-forge.org/pipermail/wannier/2013-November/000725.html >> >> Please tell me, can I use USP/PAW pseudopotentials for Wannier90? >> Or not? >> >> -- >> Best regards, >> Andrey Chibisov. Ph.D. >> Numerical method of mathematical physics Laboratory, >> Computational Center, Russian Academy of Sciences. >> Khabarovsk, Russia >> Web page:?https://www.researchgate.net/profile/A_Chibisov >> http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier --? Best regards, Andrey Chibisov. Ph.D. Numerical method of mathematical physics Laboratory, Computational Center, Russian Academy of Sciences. Khabarovsk, Russia Web page:?https://www.researchgate.net/profile/A_Chibisov http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en From lodvert at gmail.com Fri Nov 25 11:43:15 2016 From: lodvert at gmail.com (Lodvert TCHIBOTA POATY) Date: Fri, 25 Nov 2016 11:43:15 +0100 Subject: [Wannier] from setup_nnkp : error # 1 wrong excluded band index Message-ID: Hello Community ! Performing wannier functions for an isolated molecule of water, I have this error in the last step (library mode) : %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 0 from setup_nnkp : error # 1 wrong excluded band index %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% How to solve this problem ? pw input file : &CONTROL calculation = 'scf', prefix = 'H2O', pseudo_dir = '/afs/ictp.it/home/l/ltchibot/logitech/qe-6.0/pseudo', outdir = './tmp', / &SYSTEM ibrav = 0, nat = 3 ntyp = 2, ecutwfc = 35, ecutrho = 280, occupations = 'fixed' / &ELECTRONS diagonalization = 'cg' conv_thr = 1.D-7, mixing_beta = 0.2D0, / CELL_PARAMETERS bohr 17.0 0.0 0.0 0.0 17.0 0.0 0.0 0.0 17.0 ATOMIC_SPECIES H 1.00 H.pbe-rrkjus.UPF O 16.00 O.pbe-rrkjus.UPF ATOMIC_POSITIONS (angstrom) O 5.01757 1.83258 2.41933 H 5.48965 1.4012 3.11205 H 4.1684 1.4082 2.36026 K_POINTS Gamma wannier input file ## Valence + conduction states ## #num_wann = 18 #num_bands = 90 #dis_froz_max = -6.0 #dis_win_max = 4.0 #dis_num_iter = 100 ################################# num_iter = 100 translate_home_cell=true guiding_centres = true iprint = 2 ## Plotting ################# #restart = plot #wannier_plot = true #wannier_plot_format = cube #wannier_plot_list = 1,7,13 ############################# mp_grid : 1 1 1 gamma_only = true begin kpoints 0.0000 0.0000 0.0000 end kpoints begin atoms_cart O 5.01757 1.83258 2.41933 H 5.48965 1.4012 3.11205 H 4.1684 1.4082 2.36026 end atoms_cart ## Valence states ## begin projections O:px O:py H:s end projections #################### ## Valence + conduction states ## #begin projections #Bohr #H:s #c= 13.0, 12.7, 15.0 : s #c= 13.0, 17.3, 15.0 : s #c= 11.0, 16.1, 15.0 : s #c= 11.0, 13.9, 15.0 : s #c= 15.0, 13.9, 15.0 : s #c= 15.0, 16.1, 15.0 : s #C:pz #end projections ################################# begin unit_cell_cart bohr 17.00 0.00 0.00 0.00 17.00 0.00 0.00 0.00 17.00 end_unit_cell_cart pw2wan (library mode) input file &inputpp outdir = './tmp' prefix = 'H2O' seedname = 'water' write_amn = .true. write_mmn = .true. write_unk = .true. wan_mode = 'library' / Thanks for your suggestions. -- Lodvert TCHIBOTA POATY PhD student Materials for Renewable Energy Applications PERMANENT INSTITUTE UR-2N/IRSEN/FD-SME/UMNG Brazzaville - Republic of Congo Mail : lodvert at gmail.com Contact : +242 05 576 11 98 +242 06 914 45 42 PRESENT INSTITUTE The Abdus Salam International Centre for Theoretical Physics (ICTP) Section of Condensed Matter and Statistical Physics (CMSP) Strada Costiera, 11 I - 34151 Trieste Italy Mail : ltchibot at ictp.it Phone : +39 329 465 62 68 LB Room 305 From andreichibisov at yandex.ru Fri Nov 25 14:11:54 2016 From: andreichibisov at yandex.ru (Andrey Chibisov) Date: Fri, 25 Nov 2016 23:11:54 +1000 Subject: [Wannier] (no subject) Message-ID: <6658821480079514@web19j.yandex.ru> Dear Colleagues, I want to calculate the Seebeck coefficient for platinum. What do you think, what approach I need use for this? LDA or GGA? -- Best regards, Andrey Chibisov. Ph.D. Numerical method of mathematical physics Laboratory, Computational Center, Russian Academy of Sciences. Khabarovsk, Russia Web page:?https://www.researchgate.net/profile/A_Chibisov http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en From santubaidya2009 at gmail.com Wed Nov 30 08:24:55 2016 From: santubaidya2009 at gmail.com (Santu Baidya) Date: Wed, 30 Nov 2016 07:24:55 +0000 Subject: [Wannier] Question regarding calculation of neighbour cell hopping in wannier90_hr.dat Message-ID: <427ee795-310a-ee89-fbc7-1c3938272238@mixmax.com> Dear Wannier90 users, I am doing a wannier90 calculation to obtain real space hopping matrix. In the real space hopping matrix hopping elements are printed for a set of (R1 R2 R3)=(-5 -4 -4) (for example), where (R1,R2,R3) are lattice vectors .For each R1 = -5 R2 should vary from -4 , -3, -2 ,-1, 0, 1, 2, 3, 4 in wannier90_hr.dat. However, in my wannier90_hr.dat file I can not find few lattice vectors, e.g., (-5 0 1). Please tell me how these lattice vectors are printed in wannier90_hr.dat file. Also, please enlighten about the missing lattice vectors in the wannier90_hr.dat file. Thanking you, Dr. Santu BaidyaIBS-CCES, Seoul National UniversitySouth Korea -------------- next part -------------- An HTML attachment was scrubbed... URL: From ayhdnas06 at gmail.com Wed Nov 30 08:46:14 2016 From: ayhdnas06 at gmail.com (Sandhya Chintalapati) Date: Wed, 30 Nov 2016 15:46:14 +0800 Subject: [Wannier] How to get the wannier_plot if I use PAW potentials in non-collinear calculations? Message-ID: Dear Wannier90 users, I tried to calculate wannier_plot using PAW potentials in Quantum espresso in the non_collinear case. I found an error "write_unk not implemented with ncls". On the other hand, I could not find Norm conserving pseudo potentials in QE for relativistic case. I am new to QE and I don't know how to generate Norm Conserving pseudo potentials for relativistic case.Could you please suggest me what can be done in this regard?. I need the wannier_plot to check the wannier centers and find the proper energy windows for the convergence of the wannier spread. Is there any other way that I can get the wannier_plot using QE.?. Thank you. Best Regards, Sandhya. -------------- next part -------------- An HTML attachment was scrubbed... URL: