From yunpengwang85 at gmail.com Fri Mar 4 09:12:49 2011 From: yunpengwang85 at gmail.com (Yunpeng Wang) Date: Fri, 4 Mar 2011 16:12:49 +0800 Subject: [Wannier] shape of wannier function Message-ID: Hello all: In my calculation, the shape of wannier function is needed. As it known, with wannier_plot=.true., the wannier function is plotted as a scalar field in regular mesh. My question is: what is the value at each point? best wishes, Yun-Peng -- *Yunpeng Peter Wang* -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.mostofi at imperial.ac.uk Fri Mar 4 11:33:27 2011 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Fri, 04 Mar 2011 10:33:27 +0000 Subject: [Wannier] shape of wannier function In-Reply-To: References: Message-ID: <4D70BFF7.4060905@imperial.ac.uk> An HTML attachment was scrubbed... URL: From gianluca.giovannetti at gmail.com Mon Mar 7 09:15:16 2011 From: gianluca.giovannetti at gmail.com (Gianluca Giovannetti) Date: Mon, 7 Mar 2011 09:15:16 +0100 Subject: [Wannier] understanding the output of Wannier90 Message-ID: Dear All, i`m having a problem in understanding the output of wannier90 (file hr.dat). i have made a calculation for a molecular crystal. the unit cell consists of molecular sites (A and B) along the stacking direction a (with a lattice vector) as illustrated in the following: -----> a |o o | o o | ...... A B A B (here i`m plotting schematically two unit cells) Let label the molecules in this way: -----> a |o o | o o | ...... A1 B1 A2 B2 now the bonds B1-A2 are shorter while bonds A2-B2 are longer (the system is dimerized). the HOMOs (highest occupied molecular orbitals) of the two molecules A,B contained in the unit cell hybridize, giving rise to bonding and antinbonding bands around the Fermi level. then at Fermi level i have two states. the calculations performed with PW and W90 looks fine and the agreement of PW and W90 bands is good. once i plot using xcrystalden the Wannier functions of the such states at Fermi one is located at A2 and the other one is located to B1 (basically one is in the unit cell and the other one is located at the NN unit cell along a). now i go to look at the hr.dat file and i find: 0 0 0 1 1 3.334228 0.000000 0 0 0 2 1 0.222024 0.000000 0 0 0 1 2 0.222024 0.000000 0 0 0 2 2 3.334221 0.000000 1 0 0 1 1 0.000243 0.000000 1 0 0 2 1 0.203498 0.000000 1 0 0 1 2 0.000122 0.000000 1 0 0 2 2 0.000242 0.000000 -1 0 0 1 1 0.000243 0.000000 -1 0 0 2 1 0.000122 0.000000 -1 0 0 1 2 0.203498 0.000000 -1 0 0 2 2 0.000242 0.000000 i have some doubts in understanding the hopping parameters. i see the kinetic terms are different 0.222024/0.203498 as one would expect in a dimerized structure. this is fine. the values 3.334228 are the on-site energies of the tight-binding problem. my problem is the following. is 0.222024 referring to the the hopping between A1 and B1 (also between A2 and B2) or between B1 and A2? because the Wannier functions plotted in real space are sitting between adjacent different unit cells i would expect 0.222024 referring to hopping between B1 and A2 and 0.203498 referring to hopping between A1 and B1 (also between A2 and B2). then at smaller distances would correspond larger hoppings and at larger distances would correspond smaller hoppings. could you please confirm this way to interpret the file hr.dat? :-) i thank you in advance for our time. cheers, Gianluca Giovannetti -------------- next part -------------- An HTML attachment was scrubbed... URL: From gianluca.giovannetti at gmail.com Mon Mar 7 13:06:21 2011 From: gianluca.giovannetti at gmail.com (Gianluca Giovannetti) Date: Mon, 7 Mar 2011 13:06:21 +0100 Subject: [Wannier] understanding the output of Wannier90 In-Reply-To: <12957434.1299496623881.JavaMail.root@kistmail.kist.re.kr> References: <12957434.1299496623881.JavaMail.root@kistmail.kist.re.kr> Message-ID: Dear Young-Su, thank you for your replay. if i take the previous unit cell: ----->a | o o | o o| A1 B1 A2 B2 2 1 2 1 the first wannier function is located to B1 and the second one is located to A2. this information i was missing to report/say in my previous email. Following your suggestion then: 0 0 0 2 1 0.222025 0.000000 0 0 0 1 2 0.222025 0.000000 is meaning that hopping between B1 and A2 is 0.222025; 1 0 0 2 1 0.203502 0.000000 is meaning that hopping between A2 and B2 is 0.203502; -1 0 0 1 2 0.203502 0.00000 is meaning that hopping between B1 and A1 is 0.203502 This makes sense to me. cheers, Gianluca Giovannetti 2011/3/7 Young-Su Lee > Dear Gianluca Giovannetti, > > > > You can interpret hr.dat using the five indices preceding the H matrix > element: > > i j k m n indicates H element between the m_th WF at the final center > and the n_th WF at (the final center + (i,j,k) lattice vector) ? please > someone correct me, if this is wrong. > > > > The WF indices follow the order in the *wout file. > > The final center does not have to reside in the home cell, as you?ve > already noticed from the WF at A2. > > If your first WF is at A2 and the second one is at B1 > > > > 1 0 0 2 1 : btw B1 and A3 > > 1 0 0 1 2 : btw A2 and B2 > > -1 0 0 2 1 : btw B1 and A1 > > > > If you want to move the WF at A2 to A1, use translation_centre_frac = 0.5, > 0.5, 0.5, then all the WFs will be inside your home cell > (translate_home_cell only translates centers without modifying hr.dat). I > guess this is less confusing. > > > > Hope this helps. > > > > Young-Su > > > > > > * ----- Original Message -----* > From: Gianluca Giovannetti > To: wannier at quantum-espresso.org > Date: 2011-03-07 17:15:40 > Subject: [Wannier] understanding the output of Wannier90 > Dear All, > > i`m having a problem in understanding the output of wannier90 (file hr > .dat). > > i have made a calculation for a molecular crystal. > > the unit cell consists of molecular sites (A and B) along the stacking > direction a (with a lattice vector) as illustrated in the following: > > -----> a > > |o ? o | o ? o | ...... > ?A ? B ?A ?B > > (here i`m plotting schematically two unit cells) > > Let label the molecules in this way: > > -----> a > > |o ? ?o | o ? ?o | ...... > ?A1 B1 A2 B2 > > now the bonds B1-A2 are shorter while bonds A2-B2 are longer (the system is > dimerized). > > the HOMOs (highest occupied molecular orbitals) of the two molecules A,B > contained in the unit cell hybridize, giving rise to bonding and > antinbonding bands around the Fermi level. then at Fermi level i have two > states. > > the calculations performed with PW and W90 looks fine and the agreement of > PW and W90 bands is good. > > once i plot using xcrystalden the Wannier functions of the such states at > Fermi one is located at A2 and the other one is located to B1 (basically one > is in the unit cell and the other one is located at the NN unit cell along > a). > > now i go to look at the hr.dat file and i find: > > ??? 0 ? ?0 ? ?0 ? ?1 ? ?1 ? ?3.334228 ? ?0.000000 > ?? ?0 ? ?0 ? ?0 ? ?2 ? ?1 ? ?0.222024 ? ?0.000000 > ?? ?0 ? ?0 ? ?0 ? ?1 ? ?2 ? ?0.222024 ? ?0.000000 > ?? ?0 ? ?0 ? ?0 ? ?2 ? ?2 ? ?3.334221 ? ?0.000000 > > ?? ?1 ? ?0 ? ?0 ? ?1 ? ?1 ? ?0.000243 ? ?0.000000 > ?? ?1 ? ?0 ? ?0 ? ?2 ? ?1 ? ?0.203498 ? ?0.000000 > ?? ?1 ? ?0 ? ?0 ? ?1 ? ?2 ? ?0.000122 ? ?0.000000 > ?? ?1 ? ?0 ? ?0 ? ?2 ? ?2 ? ?0.000242 ? ?0.000000 > > ?? -1 ? ?0 ? ?0 ? ?1 ? ?1 ? ?0.000243 ? ?0.000000 > ?? -1 ? ?0 ? ?0 ? ?2 ? ?1 ? ?0.000122 ? ?0.000000 > ?? -1 ? ?0 ? ?0 ? ?1 ? ?2 ? ?0.203498 ? ?0.000000 > ?? -1 ? ?0 ? ?0 ? ?2 ? ?2 ? ?0.000242 ? ?0..000000 > > i have some doubts in understanding the hopping parameters. > > i see the kinetic terms are different?0.222024/0.203498 as one would expect > in a dimerized structure. this is fine. > > the values 3.334228 are the on-site energies of the tight-binding problem. > > my problem is the following. > is 0.222024 referring to the the hopping between A1 and B1 (also between A2 > and B2) ?or between B1 and A2? > > because the Wannier functions plotted in real space are sitting between > adjacent different unit cells i would expect 0.222024 referring to?hopping > between?B1 and A2 and 0.203498??referring to?hopping between?A1 and B1 (also > between A2 and B2). > then at smaller distances would correspond larger hoppings and at larger > distances?would correspond smaller?hoppings. > > could you please confirm this way to interpret the file hr.dat? :-) > > i thank you in advance for our time. > > cheers, > > Gianluca Giovannetti > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rcollyer at phys.lsu.edu Fri Mar 11 20:16:33 2011 From: rcollyer at phys.lsu.edu (Robert Collyer) Date: Fri, 11 Mar 2011 13:16:33 -0600 Subject: [Wannier] dist_cutoff and one_dim_axis Message-ID: <4D7A7511.9050500@phys.lsu.edu> All, I'm trying to use the dist_cutoff option with a dist_cutoff_mode of three_dim, but wannier90 (v 1.2) gives me the error "Error: one_dim_axis not recognised" because I have not specified one. According to the user_guide, one_dim_axis is not used when the dis_cutoff_mode is three_dim. But, with bands_plot and bands_plot_mode=cut, the code in parameters.F90 (line 652) requires it. With a cut off mode of three_dim, is the axis required by the rest of the code? A quick search does not seem to indicate that it is so, but I have not looked very closely. Also, if I supply a value for one_dim_axis, how will it affect my results? Thanks, Rob Collyer From a.mostofi at imperial.ac.uk Fri Mar 11 22:32:59 2011 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Fri, 11 Mar 2011 21:32:59 +0000 Subject: [Wannier] dist_cutoff and one_dim_axis In-Reply-To: <4D7A7511.9050500@phys.lsu.edu> References: <4D7A7511.9050500@phys.lsu.edu> Message-ID: <4D7A950B.9020507@imperial.ac.uk> An HTML attachment was scrubbed... URL: From whw1985 at mail.ustc.edu.cn Mon Mar 14 17:12:11 2011 From: whw1985 at mail.ustc.edu.cn (Hongwei Wang) Date: Tue, 15 Mar 2011 00:12:11 +0800 (CST) Subject: [Wannier] seek help Message-ID: <23155763.4554921300119131700.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 beginner in using the Wannier90 code, I can't well understand the Key word "guiding_centres". I wonder if you can tell me its physical picture and the situation in which we should use guiding_centres? In addition, I want to know how to set the Key word num_guide_cycles?num_no_guide_iter? conv_noise_amp? conv_noise_num reasonablely. can you tell me some experience to set them. Thanks for any of your help? Sincerely, Hongwei Wang From marzari at MIT.EDU Mon Mar 14 17:33:31 2011 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 14 Mar 2011 16:33:31 +0000 Subject: [Wannier] seek help In-Reply-To: <23155763.4554921300119131700.JavaMail.coremail@mailweb> References: <23155763.4554921300119131700.JavaMail.coremail@mailweb> Message-ID: <4D7E435B.5060004@mit.edu> Dear Hongwei, guiding centers were used to suggest how to properly choose the branch cut in the evaluation of complex logarithms - the logic is explained inside wann_phases (basically making sure that roughly every contribution that you add up in evaluating the position of the wannier center has a branch cut that is closest to giving that position), but these days you'd be better off by choosing carefully your projections, and forgetting about the guiding cetners (keywords are nevertheless explained in the user guide). With regard to conv_noise... , it must be something Arash or Jonathan coded? Looks like a way to get unstuck from very shallow minimizations (e.g. as in the rotation of the d-orbitals in a system where localization is almost invariant with respect to rotation). nicola On 3/14/11 4:12 PM, Hongwei Wang wrote: > 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 beginner in using the Wannier90 code, I can't well understand the Key word "guiding_centres". I wonder if you can tell me its physical picture and the situation in which we should use guiding_centres? In addition, I want to know how to set the Key word num_guide_cycles??num_no_guide_iter?? conv_noise_amp?? conv_noise_num reasonablely. > can you tell me some experience to set them. > Thanks for any of your help?? > > Sincerely, > Hongwei Wang > > > > _______________________________________________ > 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 a.mostofi at imperial.ac.uk Mon Mar 14 18:17:41 2011 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Mon, 14 Mar 2011 17:17:41 +0000 Subject: [Wannier] seek help In-Reply-To: <4D7E435B.5060004@mit.edu> References: <23155763.4554921300119131700.JavaMail.coremail@mailweb> <4D7E435B.5060004@mit.edu> Message-ID: <4D7E4DB5.4080702@imperial.ac.uk> An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: moz-screenshot.png Type: image/png Size: 18952 bytes Desc: not available URL: From marzari at MIT.EDU Mon Mar 14 18:30:02 2011 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 14 Mar 2011 17:30:02 +0000 Subject: [Wannier] seek help In-Reply-To: <4D7E4DB5.4080702@imperial.ac.uk> References: <23155763.4554921300119131700.JavaMail.coremail@mailweb> <4D7E435B.5060004@mit.edu> <4D7E4DB5.4080702@imperial.ac.uk> Message-ID: <4D7E509A.8030502@mit.edu> Thanks Arash! I had actually searched for conv_noise in the user guide, but for some mysterious reason Preview (on the mac) cannot find any occurrence of "conv_noise" (even if it's there; otoh "conv_tol" is found, or others...). Life is a mistery... nicola On 3/14/11 5:17 PM, Arash Mostofi wrote: > Dear Hongwei, > > As Nicola mentions, input parameters are explained in the user guide. > Regarding conv_noise_*, from p22: -- --------------------------------------------------------------------- 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 HTML attachment was scrubbed... URL: From rcollyer at phys.lsu.edu Mon Mar 14 20:35:46 2011 From: rcollyer at phys.lsu.edu (Robert Collyer) Date: Mon, 14 Mar 2011 14:35:46 -0500 Subject: [Wannier] how do you determine which Wannier orbital corresponds to which projection? Message-ID: <4D7E6E12.5060004@phys.lsu.edu> All, I know from a previous question posted to this list that you determine where each Wannier orbital is located in the Hamiltonian via the case_centres.xyz file. But, how do you determine which orbital is which, e.g. d_z^2, etc.? Is this even possible with the amount of mixing going on in the unitary transform? Thanks, Rob Collyer From a.mostofi at imperial.ac.uk Tue Mar 15 01:49:35 2011 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Tue, 15 Mar 2011 00:49:35 +0000 Subject: [Wannier] how do you determine which Wannier orbital corresponds to which projection? In-Reply-To: <4D7E6E12.5060004@phys.lsu.edu> References: <4D7E6E12.5060004@phys.lsu.edu> Message-ID: <4D7EB79F.4040900@imperial.ac.uk> An HTML attachment was scrubbed... URL: From da.puggioni at gmail.com Wed Mar 16 15:53:05 2011 From: da.puggioni at gmail.com (Danilo Puggioni) Date: Wed, 16 Mar 2011 15:53:05 +0100 Subject: [Wannier] hopping parameter 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. I don't understand. Thanks Danilo From hubuliangwei at yahoo.com.cn Fri Mar 25 15:01:18 2011 From: hubuliangwei at yahoo.com.cn (=?utf-8?B?5beNIOaigQ==?=) Date: Fri, 25 Mar 2011 22:01:18 +0800 (CST) Subject: [Wannier] how to get the van der waals energy Message-ID: <364579.30508.qm@web15202.mail.cnb.yahoo.com> Dear all: I am a beginner in using the Wannier90 code, I have done the following steps: 1 Run scf/nscf calculations with pw.x 2 Run wannier90.x generates seedname.nnkp 3 Run pw2wannier90.x generates seedname.mmn, seedname.amn, seedname.eig How could I get the van der waals energy using the formula EvdW (equation 4 of PRL 100, 053002, 2008)? Thanks for your help. ?Wei Liang PH.D Candidate in Electrochemistry College of Chemistry and Molecular Science Wuhan University, 430072, Hubei Province, China Email:hubuliangwei at yahoo.com.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: