From martin.gmitra at gmail.com Thu Apr 7 13:17:41 2011 From: martin.gmitra at gmail.com (Martin Gmitra) Date: Thu, 7 Apr 2011 13:17:41 +0200 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT Message-ID: Dear Wannier90 users, We are interested to compare Hamiltonian in Wannier basis with a simple tight-binding nearest-neighbors model. My question is -- how one can reduce number of Wigner-Seitz grid-points nrpts such that resulting Hamiltonian (seedname_hr.dat) includes nearest neighbors only? Second question I do have is -- what are the units of the Hamiltonian matrix elements in the seedname_hr.dat file? With Best regards, Martin Gmitra From da.puggioni at gmail.com Thu Apr 7 13:52:54 2011 From: da.puggioni at gmail.com (danilo Puggioni) Date: Thu, 7 Apr 2011 13:52:54 +0200 Subject: [Wannier] hopping parameters Message-ID: Dear Wannier90 users, I am beginner in using Wannier90 code. I have a question about the hopping parameters that I can extract from the Hamiltonian. These parameters in my hamiltonian are complex. Is it correct? In PRB 56, 12847 Marzari and Vanderbilt conjecture that optimally localized Wannier functions are alway real then I suppose that the hopping parameter must be real. Thanks Danilo ---------------------------------------------------------------------------- Danilo Puggioni - e-mail: puggioni at dsf.unica.it Dipartimento Scienze Fisiche - Universita' di Cagliari Phone: +39-0706754840 Complesso Universitario di Monserrato - S.P. Monserrato-Sestu Km 0.700 I-09042 Monserrato (CA) - Italy ----------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From martin.gmitra at gmail.com Thu Apr 7 18:25:17 2011 From: martin.gmitra at gmail.com (Martin Gmitra) Date: Thu, 7 Apr 2011 18:25:17 +0200 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT In-Reply-To: <4D9DB800.9040403@mit.edu> References: <4D9DB800.9040403@mit.edu> Message-ID: Dear Nicola, Thanks for your suggestion. I have set bands_plot_mode=cut but playing with dist_cutoff does not introduce any change to Hamiltonian = seedname_hr.dat. What is than wrong? Martin On Thu, Apr 7, 2011 at 3:11 PM, Nicola Marzari wrote: > > Dear Martin, > > from previous emails of Y-S Lee: > > 1) the hamiltonian is in a wannier basis, each characterized by a wannier > center - > so set bands_plot_mode=cut and choose dist_cutoff appropriately. > > 2) The indeces refer to the final MLWFs, as labeled in the output > > Not sure about the units... > > ? ? ? ? ? ? ? ?nicola > > > > On 4/7/11 12:17 PM, Martin Gmitra wrote: >> >> Dear Wannier90 users, >> >> We are interested to compare Hamiltonian in Wannier basis with a >> simple tight-binding >> nearest-neighbors model. My question is -- how one can reduce number >> of Wigner-Seitz >> grid-points nrpts such that resulting Hamiltonian (seedname_hr.dat) >> includes nearest >> neighbors only? >> >> Second question I do have is -- what are the units of the Hamiltonian >> matrix elements >> in the seedname_hr.dat file? >> >> With Best regards, >> Martin Gmitra >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://www.democritos.it/mailman/listinfo/wannier > > > -- > --------------------------------------------------------------------- > 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 nicolas.poilvert at gmail.com Thu Apr 7 13:35:31 2011 From: nicolas.poilvert at gmail.com (nicolas poilvert) Date: Thu, 7 Apr 2011 07:35:31 -0400 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT In-Reply-To: References: Message-ID: Dear Martin, You don't need to reduce the number of Wigner-Seitz points in order to identify the nearest neighbors because seedname_hr.dat contains all the necessary information in order for you to extract those neighbors. The format goes as follows R1 R2 R3 m n Re() Im() where R1, R2, R3 are the integer coordinates of unit cell R (in the bases of the direct lattice vectors a1, a2, a3) with respect to the "home" cell (R=0). So all you need to do is extract all the matrix elements corresponding to the proper nearest neighbor cell R. For example if you had a 1D system like a nanotube in the z-direction, then you would probably want to extract the matrix elements corresponding to R = 0 0 1. You could of course reduce the number of k points in your calculation such that only the nearest neighbor cells would show up but then your calculation would most probably not be very converged for the Wannier functions (not enough k point sampling). Nicolas On Thu, Apr 7, 2011 at 7:17 AM, Martin Gmitra wrote: > Dear Wannier90 users, > > We are interested to compare Hamiltonian in Wannier basis with a > simple tight-binding > nearest-neighbors model. My question is -- how one can reduce number > of Wigner-Seitz > grid-points nrpts such that resulting Hamiltonian (seedname_hr.dat) > includes nearest > neighbors only? > > Second question I do have is -- what are the units of the Hamiltonian > matrix elements > in the seedname_hr.dat file? > > With Best regards, > Martin Gmitra > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > -- POILVERT Nicolas PhD candidate, Dpt of Materials Science and Engineering Massachusetts Institute of Technology 77, Massachusetts avenue Cambridge, MA 02139 USA work: (617) 452-4212 nicolas.poilvert at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From nicolas.poilvert at gmail.com Thu Apr 7 18:34:08 2011 From: nicolas.poilvert at gmail.com (nicolas poilvert) Date: Thu, 7 Apr 2011 12:34:08 -0400 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT In-Reply-To: References: <4D9DB800.9040403@mit.edu> Message-ID: If I remember correctly, using the "cut" mode will simply tell the code to truncate the real-space hamiltonian matrix and then use it for interpolating the bands. But the truncation is internal to the code, it does not modify the full hamiltonian matrix when you print it. Nicolas On Thu, Apr 7, 2011 at 12:25 PM, Martin Gmitra wrote: > Dear Nicola, > > Thanks for your suggestion. I have set bands_plot_mode=cut but playing > with > dist_cutoff does not introduce any change to Hamiltonian = seedname_hr.dat. > What is than wrong? > Martin > > > > On Thu, Apr 7, 2011 at 3:11 PM, Nicola Marzari wrote: > > > > Dear Martin, > > > > from previous emails of Y-S Lee: > > > > 1) the hamiltonian is in a wannier basis, each characterized by a wannier > > center - > > so set bands_plot_mode=cut and choose dist_cutoff appropriately. > > > > 2) The indeces refer to the final MLWFs, as labeled in the output > > > > Not sure about the units... > > > > nicola > > > > > > > > On 4/7/11 12:17 PM, Martin Gmitra wrote: > >> > >> Dear Wannier90 users, > >> > >> We are interested to compare Hamiltonian in Wannier basis with a > >> simple tight-binding > >> nearest-neighbors model. My question is -- how one can reduce number > >> of Wigner-Seitz > >> grid-points nrpts such that resulting Hamiltonian (seedname_hr.dat) > >> includes nearest > >> neighbors only? > >> > >> Second question I do have is -- what are the units of the Hamiltonian > >> matrix elements > >> in the seedname_hr.dat file? > >> > >> With Best regards, > >> Martin Gmitra > >> _______________________________________________ > >> Wannier mailing list > >> Wannier at quantum-espresso.org > >> http://www.democritos.it/mailman/listinfo/wannier > > > > > > -- > > --------------------------------------------------------------------- > > 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 > > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > -- POILVERT Nicolas PhD candidate, Dpt of Materials Science and Engineering Massachusetts Institute of Technology 77, Massachusetts avenue Cambridge, MA 02139 USA work: (617) 452-4212 nicolas.poilvert at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From martin.gmitra at gmail.com Thu Apr 7 19:25:33 2011 From: martin.gmitra at gmail.com (Martin Gmitra) Date: Thu, 7 Apr 2011 19:25:33 +0200 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT In-Reply-To: References: <4D9DB800.9040403@mit.edu> Message-ID: If the code would provide output for the "truncated" Hamiltonian -- which is internal, are the Hamiltonian far-away-elements from the center simply reduced and the remaining elements near the center untouched ? Or the center-like elements are changed as well? Best, Martin From nicolas.poilvert at gmail.com Thu Apr 7 19:35:55 2011 From: nicolas.poilvert at gmail.com (nicolas poilvert) Date: Thu, 7 Apr 2011 13:35:55 -0400 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT In-Reply-To: References: <4D9DB800.9040403@mit.edu> Message-ID: Normally in "cut" mode, the code simply looks at the distance between two wannier functions |wm> and |wn> and if their centres are more than "dist_cutoff" apart, then the matrix element is set to zero. But there are no routines to update the full hamiltonian matrix. Worst case scenario you can modify the code to do just that. But it seems that the simplest option is to parse the seedname_hr.dat matrix and extract all the matrix elements corresponding to the proper R vector. N. On Thu, Apr 7, 2011 at 1:25 PM, Martin Gmitra wrote: > If the code would provide output for the "truncated" Hamiltonian -- > which is internal, > are the Hamiltonian far-away-elements from the center simply reduced and > the > remaining elements near the center untouched ? Or the center-like elements > are > changed as well? > > Best, > Martin > -- POILVERT Nicolas PhD candidate, Dpt of Materials Science and Engineering Massachusetts Institute of Technology 77, Massachusetts avenue Cambridge, MA 02139 USA work: (617) 452-4212 nicolas.poilvert at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From martin.gmitra at gmail.com Sun Apr 10 12:37:58 2011 From: martin.gmitra at gmail.com (Martin Gmitra) Date: Sun, 10 Apr 2011 12:37:58 +0200 Subject: [Wannier] How to reduce neighbors for Hamiltonian in Wannier basis ? :: HR_PLOT In-Reply-To: References: <4D9DB800.9040403@mit.edu> Message-ID: Dear Wannier90 users, Is there possibility to extract hopping for nearest-neighbors atoms within a unit cell? For instance, graphene has the two atoms A, B in unit cell, can be hopping between A and B atoms obtained? Based on Wannier function definition with vector R [coordinates of unit cell with respect to the "home" cell (R=0)] it seems that it is not possible. Could you please prove/disprove or propose a way how one can obtain the hopping? Many thanks, Martin Gmitra From whw1985 at mail.ustc.edu.cn Mon Apr 25 14:53:38 2011 From: whw1985 at mail.ustc.edu.cn (Hongwei Wang) Date: Mon, 25 Apr 2011 20:53:38 +0800 (CST) Subject: [Wannier] seek help Message-ID: <13591507.1097761303736018587.JavaMail.coremail@mailweb> Dear Wannier90 users I am a Ph. D student in the Key Laboratory of Quantum Information, University of Science and Technology of China. I am tring to compute perovskite type structure compound CaMnO3,I have computed it with several spin configurations. I use the results of AFM spin configuration to clarify my question. In my system , there are three electons occupy three degenerate dxy dyz dxz t2g orbits of Mn, and the two degenerate dx2-y2 dz2 eg orbits are unoccupied. The following results are d orbits' on site energy of Mn atoms for spin up wannier functions part. (unit ev) dxy dyz dxz dx2-y2 dz2 Spin-up Mn atoms -2.134373 -2.101328 -2.101333 -0.544185 -0.496429 Spin-dn Mn atoms 0.901559 0.919423 0.919416 1.106947 1.154219 The results indicate that spin-up energy level and spin-dn energy level of 3d orbits has been splited by spin interactions, I want to know if the Energy difference between spin up and spin down energy level for the same d orbit is Hund coupling energy JH or Coulomb energy. ( see the formula 1 and 2 in PRL 100, 186402 (2008) ) I look forward to your reply, thanks for your enthusiastic help. Sincerely, Hongwei Wang