From f_mirjani at ph.iut.ac.ir Mon Jun 2 10:48:18 2008 From: f_mirjani at ph.iut.ac.ir (fatemeh.mirjani) Date: Mon, 2 Jun 2008 12:18:18 +0330 (IRST) Subject: [Wannier] importing the data from WIEN2k to wannier90 Message-ID: <4797585.63231212396498128.JavaMail.root@mta.iut.ac.ir> Dear users; I'm writing an interface code for importing the data from WIEN2k(a full-potential code) to wannier90. Please let me know that for importing the data from WIEN2k to wannier90, calculating the Mmn & Amn matrix is enough? Or do I need any another file in the appropriate format for importing to wannier90? I'm anxiously looking forward your reply and guidelines. -- Best Regards Fatemeh Mirjani Computational Condensed Matter Research Lab. ICTP Affiliated Center in Isfahan, Physics Department, Isfahan University of Technology, Iran Email: f_mirjani at ph.iut.ac.ir Tel/Fax Office: +98311-3913746 Tel Lab.: +98311-3913731 From jry20 at cam.ac.uk Mon Jun 2 11:54:06 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Mon, 2 Jun 2008 10:54:06 +0100 (BST) Subject: [Wannier] importing the data from WIEN2k to wannier90 In-Reply-To: <4797585.63231212396498128.JavaMail.root@mta.iut.ac.ir> References: <4797585.63231212396498128.JavaMail.root@mta.iut.ac.ir> Message-ID: On Mon, 2 Jun 2008, fatemeh.mirjani wrote: > I'm writing an interface code for importing the data from WIEN2k(a > full-potential code) to wannier90. Please let me know that for importing > the data from WIEN2k to wannier90, calculating the Mmn & Amn matrix is > enough? Or do I need any another file in the appropriate format for > importing to wannier90? Fatemeh, You will also need a file containing the Kohn-Sham eigenvalues (*.eig), but this is a trivial file to generate. The mmn, amn and eig files are all that you will need to generate with wien2k. Please see page 43 of the wannier90v1.1 user guide for the precise specification of these files. Yours Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From marzari at MIT.EDU Mon Jun 2 13:14:00 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 02 Jun 2008 07:14:00 -0400 Subject: [Wannier] importing the data from WIEN2k to wannier90 In-Reply-To: <4797585.63231212396498128.JavaMail.root@mta.iut.ac.ir> References: <4797585.63231212396498128.JavaMail.root@mta.iut.ac.ir> Message-ID: <4843D5F8.2070404@mit.edu> fatemeh.mirjani wrote: > Dear users; > > I'm writing an interface code for importing the data from WIEN2k(a full-potential code) to wannier90. > Please let me know that for importing the data from WIEN2k to wannier90, calculating the Mmn & Amn matrix is enough? Or do I need any another file in the appropriate format for importing to wannier90? > > > I'm anxiously looking forward your reply and guidelines. > Dear Fatemeh, Jonathan is (of course) correct. Let me also strss, though, that this is a fairly complex undertaking - calculating the Mmn and Amn matrix elements in the FLAPW basis is not an easy tak, unless you have extensive experience with the details of the code. Prof Posternak (CC'd here - one of the pioneers of F/LAPW techniques) has some notes on the subject that could be useful - you might want to get in touch with him. Last - you might also want to output the periodic part of the Bloch orbitals on a real space mesh - this is done for visualization purposes along, but the M and A matrices shouldn't be calculated in such approximate representation. Last (again) - I vaguely remember that someone in the Wien2k community was planning to interface the code with Wannier - that might have indeed been you, but you might want also to make sure that there are no concurrent efforts going on. All the best, and best luck ! Esfahan Nesf-e Jahan. 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 vranjan at gmail.com Fri Jun 6 01:08:36 2008 From: vranjan at gmail.com (Vivek Ranjan) Date: Thu, 5 Jun 2008 19:08:36 -0400 Subject: [Wannier] spread In-Reply-To: <53402.152.14.74.117.1212179894.squirrel@webmail.ncsu.edu> References: <53402.152.14.74.117.1212179894.squirrel@webmail.ncsu.edu> Message-ID: <1abcddff0806051608t7161fa87vb3058776a32bd137@mail.gmail.com> I had asked the following question a few days back. I am trying to calculate wannier function of BN sheet. (i) I have a hard time reaching convergence. In about ten thousand steps the spread seem reasonable. However, if I run longer, spread increases once again on one of the sites (around 15 A). (ii) There is no problem in convergence if the BN-plane is staggered. (iii) I have attached one diagram with the mail. I asked for "B:sp2;pz" in projections. But it seems to be centered on N (green) and spread over three B (yellow). Seems like a large spread ? (iv) The diagram depicts three wannier functions. I want only one.... How do I get rid of these images of the central one ? I would appreciate some suggestion. Thanks, Vivek Physics, NCSU On Fri, May 30, 2008 at 4:38 PM, Vivek Ranjan wrote: > Hello, > > I am trying to calculate wannier function of BN sheet. The way the unit > cell is constructed, I have chosen (12,1,1) kpoint mesh for scf > calculation. Then I do an nscf calculation with 4 kpoints (4,1,1). The > spread of one of wannier function is 26 A. I run several steps (more than > 20000) and still it does not converge. The question is how to reduce this > spread ? > > I tried putting more kpoints by doing (12,2,2) scf calculations and 16 > kpoints (4,2,2) for nscf and wannier parts. The problem is instead of 16 > kpoints in the nscf that I hand put in the file, pwscf always generates 24 > kpoints. The additional kpoints are shifted kpoints that I put. It never > happened to me before and wondering why it is happening ? > > (12,1,1) scf, and (4,1,1) kpoints has no such problems. > > Thanks, > > Vivek, > Physics, NCSU. > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > -------------- next part -------------- A non-text attachment was scrubbed... Name: bn_00028.gif Type: image/gif Size: 149101 bytes Desc: not available URL: From jry20 at cam.ac.uk Fri Jun 6 13:13:15 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Fri, 6 Jun 2008 12:13:15 +0100 (BST) Subject: [Wannier] spread In-Reply-To: <1abcddff0806051608t7161fa87vb3058776a32bd137@mail.gmail.com> References: <53402.152.14.74.117.1212179894.squirrel@webmail.ncsu.edu> <1abcddff0806051608t7161fa87vb3058776a32bd137@mail.gmail.com> Message-ID: Vivek, Could you post your input and output files (pwscf and w90) somewhere we can read (eg on a webserver). At the moment I don't have enough information about what you are trying to do to give anything more than vague suggestions. > (iv) The diagram depicts three wannier functions. I want only one.... > How do I get rid of these images of the central one ? >> calculation. Then I do an nscf calculation with 4 kpoints (4,1,1). If you create WF from a 4x1x1 sampling of the BZ then the WF will live in a periodic cell 4x1x1 of the real-space unit cell. If you then plot the WF in a supercell (wannier_plot_supercell, which defaults to 2) you will see both the "home" WF and some of its periodic image. Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From nafiserb at gmail.com Sun Jun 8 09:43:20 2008 From: nafiserb at gmail.com (nafise rezaei) Date: Sun, 8 Jun 2008 12:13:20 +0430 Subject: [Wannier] center of MLWFs Message-ID: <785b26a90806080043h6f9ccf90i875e6728ea40230@mail.gmail.com> Dear all, I have a question about center of MLWFs. Do center of MLWFs change for valence states , when MLWFs be calculated in these two cases: once be calculated only for valence states and once be calculated for valence + conduction states? Best regards, From f_mirjani at ph.iut.ac.ir Sun Jun 8 10:50:39 2008 From: f_mirjani at ph.iut.ac.ir (fatemeh.mirjani) Date: Sun, 8 Jun 2008 12:20:39 +0330 (IRST) Subject: [Wannier] number of bands for a special atom and special kpoint Message-ID: <17775185.17071212915039033.JavaMail.root@mta.iut.ac.ir> Dear users; For exporting the required data from Wien2k to Wannier90 I have a problem. For example for GaAs: I can calculate the Mmn matrix for Ga and As separately not for GaAs! The number of bands for Ga is different from the number of bands for As. for each atom and each kpoint I have a number of bands. Totally, In a sample the number of occupied bands in Wien2k for each atom is different from another atoms. When I want to import the Mmn matrix that I have calculated in Wien2k, I don't know what should I do? Because in seedname.mmn Wannier90 code writes the Mmn matrix with a special number of bands for the total atoms of sample not for a special atom. Your help is greatly appreciated like your previous favor. -- Best Regards Fatemeh Mirjani | Email: f_mirjani at ph.iut.ac.ir Computational Condensed Matter Research Lab. | Tel/Fax Office: +98311-391374 ICTP Affiliated Center in Isfahan, | Tel Lab.: +98311-3913731 ----------Physics Department, Isfahan University of Technology, Iran---------------------- From jry20 at cam.ac.uk Mon Jun 9 11:51:17 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Mon, 9 Jun 2008 10:51:17 +0100 (BST) Subject: [Wannier] center of MLWFs In-Reply-To: <785b26a90806080043h6f9ccf90i875e6728ea40230@mail.gmail.com> References: <785b26a90806080043h6f9ccf90i875e6728ea40230@mail.gmail.com> Message-ID: On Sun, 8 Jun 2008, nafise rezaei wrote: > I have a question about center of MLWFs. > Do center of MLWFs change for valence states , when MLWFs be > calculated in these two cases: once be calculated only for valence > states and once be calculated for valence + conduction states? Nafise, In general, yes they will change. Just to address one possible confusion: When you create MLWF to describe valence and conduction states it isn't possible to talk about MLWF for the valence states - each MLWF has partial occupancy. You can look at the contribution of a given WF to a particular band at a particular k-point. (eg see bands_plot_project) The example from the original paper (Maximally localized Wannier functions for entangled energy bands I. Souza, N. Marzari and D. Vanderbilt, Phys. Rev. B 65, 035109 (2002)) is that of silicon. For the valence states the MLWF are centred on the bond centres. For MLWF to span the valence and low lying conduction states we can choose sp3 like MLWF - which are not bond centred. Hope this helps. Yours Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From jry20 at cam.ac.uk Mon Jun 9 12:20:42 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Mon, 9 Jun 2008 11:20:42 +0100 (BST) Subject: [Wannier] number of bands for a special atom and special kpoint In-Reply-To: <17775185.17071212915039033.JavaMail.root@mta.iut.ac.ir> References: <17775185.17071212915039033.JavaMail.root@mta.iut.ac.ir> Message-ID: On Sun, 8 Jun 2008, fatemeh.mirjani wrote: > For exporting the required data from Wien2k to Wannier90 I have a problem. > > The number of bands for Ga is different from the number of bands for As. > for each atom and each kpoint I have a number of bands. Fatemeh, I know very little about LAPW - maybe someone here does; although I suspect that this question is about technical details of Wien2k and might be better answered elsewhere (eg wien2k mailing list?) Firstly, you must agree that GaAs has 4 valence bands (a lower isolated 's' band, and three entangled 'p' bands). This is physics and doesn't care about which code you use. So you need to identify these bands within Wien2k. The Mmn matrix is computed between these bands. You say that Ga and As have different numbers of "bands". My guess is that the code is including semi-core states (eg the Ga 3d) but only within the relavant atomic sphere. MLWF for semi-core states would not be very interesting (after-all they are already localised), so you could simply exclude them from the Mmn and the wannier procedure. Yours Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From f_mirjani at ph.iut.ac.ir Mon Jun 9 14:44:02 2008 From: f_mirjani at ph.iut.ac.ir (fatemeh.mirjani) Date: Mon, 9 Jun 2008 16:14:02 +0330 (IRST) Subject: [Wannier] check Maximum Im/Re Ratio and plot the WFs (from Wien2k to Wannier90) Message-ID: <21006602.37591213015442377.JavaMail.root@mta.iut.ac.ir> Dear users; By asking question in mailing-list, I found that I need eig,Mmn,Amn files for importing the data from Wien2k to Wannier90. >From PWSCF to Wannier90, for checking the Im/Re ration in seedname.wout and plotting the WFs I should put "write_unk=.true." in seedname.pw2wan . If I want to check Maximum Im/Re Ratio and plot the WFs (from Wien2k to Wannier90) what should I do? In this case I don't have any seedname.pw2wan file. Do you have any suggestion about this? Your guideline is greatly appreciated like your previous favor. -- Best Regards Fatemeh Mirjani | Email: f_mirjani at ph.iut.ac.ir Computational Condensed Matter Research Lab. | Tel/Fax Office: +98311-3913746 ICTP Affiliated Center in Isfahan, | Tel Lab.: +98311-3913731 ----------Physics Department, Isfahan University of Technology, Iran---------------------- From nafiserb at gmail.com Mon Jun 9 15:51:00 2008 From: nafiserb at gmail.com (nafise rezaei) Date: Mon, 9 Jun 2008 18:21:00 +0430 Subject: [Wannier] center of MLWFs Message-ID: <785b26a90806090651p4f921223wf51b2f8feba97ed7@mail.gmail.com> Dear Jonathan; I do not understand exactly your reply. My question was about final center of MLWFs for valence states.in the paper of Maximally localized Wannier functions for entangled energy bands , any things about final center of MLWFs are not mentioned . You have spoken about initial center of MLWFs and the fact that initial center of MLWFs for valence states and for valence + low lying conduction states can be different. when I consider both of valence and conduction states ,will final center of MLWFs for valence states change respect with when only MLWFs for valence states be calculated? I appreciate your help. Best regards. From jry20 at cam.ac.uk Mon Jun 9 16:26:57 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Mon, 9 Jun 2008 15:26:57 +0100 (BST) Subject: [Wannier] center of MLWFs In-Reply-To: <785b26a90806090651p4f921223wf51b2f8feba97ed7@mail.gmail.com> References: <785b26a90806090651p4f921223wf51b2f8feba97ed7@mail.gmail.com> Message-ID: On Mon, 9 Jun 2008, nafise rezaei wrote: > I do not understand exactly your reply. My question was about final > center of MLWFs for valence states.in the paper of Maximally > localized Wannier functions > for entangled energy bands , any things about final center of MLWFs > are not mentioned . > > You have spoken about initial center of MLWFs and the fact that > initial center of MLWFs for valence states and for valence + low lying > conduction states can be different. > when I consider both of valence and conduction states ,will final > center of MLWFs for valence states change respect with when only > MLWFs for valence states be calculated? Nafise, My previous reply did concern the final (ie Max-loc) WF. [it also happened to apply to the functions used as an initial guess - but this is not so significant as these are rather arbitrary] If you are still confused by this point, I suggest that a useful next step would be to consider the specific system you are interested in: look at the MLWF for the valence and valence+conduction and see what happens (share with the list if don't understand what you see) Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From f_mirjani at ph.iut.ac.ir Tue Jun 10 08:56:50 2008 From: f_mirjani at ph.iut.ac.ir (fatemeh.mirjani) Date: Tue, 10 Jun 2008 10:26:50 +0330 (IRST) Subject: [Wannier] check Maximum Im/Re Ratio and plot the WFs (from Wien2k to Wannier90) In-Reply-To: <21006602.37591213015442377.JavaMail.root@mta.iut.ac.ir> Message-ID: <32039361.47731213081010071.JavaMail.root@mta.iut.ac.ir> Dear users; By asking question in mailing-list, I found that I need eig,Mmn,Amn files for importing the data from Wien2k to Wannier90. >From PWSCF to Wannier90, for checking the Im/Re ration in seedname.wout and plotting the WFs I should put "write_unk=.true." in seedname.pw2wan . If I want to check Maximum Im/Re Ratio and plot the WFs (from Wien2k to Wannier90) what should I do? In this case I don't have any seedname.pw2wan file. Do you have any suggestion about this? Your guideline is greatly appreciated like your previous favor. -- Best Regards Fatemeh Mirjani | Email: f_mirjani at ph.iut.ac.ir Computational Condensed Matter Research Lab. | Tel/Fax Office: +98311-3913746 ICTP Affiliated Center in Isfahan, | Tel Lab.: +98311-3913731 ----------Physics Department, Isfahan University of Technology, Iran---------------------- From jry20 at cam.ac.uk Tue Jun 10 12:45:31 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Tue, 10 Jun 2008 11:45:31 +0100 (BST) Subject: [Wannier] check Maximum Im/Re Ratio and plot the WFs (from Wien2k to Wannier90) In-Reply-To: <32039361.47731213081010071.JavaMail.root@mta.iut.ac.ir> References: <32039361.47731213081010071.JavaMail.root@mta.iut.ac.ir> Message-ID: On Tue, 10 Jun 2008, fatemeh.mirjani wrote: > By asking question in mailing-list, I found that I need eig,Mmn,Amn > files for importing the data from Wien2k to Wannier90. From PWSCF to > Wannier90, for checking the Im/Re ration in seedname.wout and plotting > the WFs I should put "write_unk=.true." in seedname.pw2wan . If I want > to check Maximum Im/Re Ratio and plot the WFs (from Wien2k to Wannier90) > what should I do? In this case I don't have any seedname.pw2wan file. Do > you have any suggestion about this? Fatemeh, Let me trace the steps: 1- You compute the eig, Mmn and Amn files using your favourite electronic structure code 2- You use wannier90 to obtain the MLWF. You'll get information on their spread and centres; you can use them as a tight-binding basis eg to compute fermi-surfaces or transport properties. To be more precise, we have obtained the MLWF in the basis of Bloch states from the original electronic structure code. This is good because we have thrown away the basis (eg planewaves or LAPW) and are working with small matrices (ie it's the minimal basis) 3- If you want to plot the MLWF (or discover if they are real) you must represent the MLWF in real space. For this you require the cell periodic parts of the original Bloch states on a grid. A relatively coarse regular grid is simplest. These are the files UNK* as detailed in the W90 user guide. Yours Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From qianxf at MIT.EDU Tue Jun 10 16:38:11 2008 From: qianxf at MIT.EDU (Xiaofeng Qian) Date: Tue, 10 Jun 2008 10:38:11 -0400 (EDT) Subject: [Wannier] center of MLWFs In-Reply-To: References: Message-ID: Dear Nafise, The initial centers of MLWFs are just the initial guess to start with for the construction of MLWFs. It also happends to the initial guess of atomic, bonding or hybrid orbitals. In general, they are not necessary to be the same as the final centers and final orbitals you obtain from the Wannier90 package. However what Jonathan mentioned has already answered your question and he is talking about the FINAL center of MLWFs. One key point is: if you include both valence and conduction states to calculate MLWFs, the final MLWFs are a linear combination of valence and conduction states. That means you can not tell which Wannier function is particularly for the valence states, or which Wannier function is particularly for the conduction states. Therefore, it is not very meaningful to compare the center of two sets of MLWFs constructed from valence states only and both valence+conduction states. For example, usually the valence states correspond the bonding-like Wannier functions while the low-lying conduction states correponds to the anti-bonding-like Wannier functions. Then if you construct the MLWFs using valence states only, finally you will obtain bonding-like Wannier functions whose centers are the center of those bonds. If you include both valence and conduction states to construct MLWFs, finally you will obtain atomic-like Wannier functions whose centers are atomic positions. Obviously bond centers and atomic positions are very different. But the latter ones have contributions from both valence and conduction states, therefore we can not find any particular Wannier functions which is purely from the transformation of valence states only. So the answer to your question is: the final centers of MLWFs do change when you use valence states only and valence+conduction states. But there is no specific MLWF which comes purely from valence states when you use valence+conduction states for the construction. So it is not valid to compare "the centers of MLWFs for the valence states". Best, Xiaofeng Qian > ------------------------------ > > Message: 4 > Date: Mon, 9 Jun 2008 18:21:00 +0430 > From: "nafise rezaei" > Subject: [Wannier] center of MLWFs > To: wannier at quantum-espresso.org > Message-ID: > <785b26a90806090651p4f921223wf51b2f8feba97ed7 at mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > Dear Jonathan; > > I do not understand exactly your reply. My question was about final > center of MLWFs for valence states.in the paper of Maximally > localized Wannier functions > for entangled energy bands , any things about final center of MLWFs > are not mentioned . > > You have spoken about initial center of MLWFs and the fact that > initial center of MLWFs for valence states and for valence + low lying > conduction states can be different. > when I consider both of valence and conduction states ,will final > center of MLWFs for valence states change respect with when only > MLWFs for valence states be calculated? > > I appreciate your help. > > Best regards. From hsuhan at tyr.cems.umn.edu Thu Jun 12 22:17:18 2008 From: hsuhan at tyr.cems.umn.edu (Han Hsu) Date: Thu, 12 Jun 2008 15:17:18 CDT Subject: [Wannier] Setting wf_collect=.true. before running pw2wannier90.x? Message-ID: <200806122017.m5CKHIGQ019935@vendetta.software.umn.edu> Hello wannier90 users and developers, Recently I started learning wannier90 by running examples. I am using PWscf ver 3.2.3, and I have already downloaded and compiled the latest wannier90 and pw2wannier90. When running example 5 (MLWF for diamond), I used 4 CPUs and 2 pools to do scf and nscf calculation with the input file provided in wannier90-1.1/Doc. Then, when I ran pw2wan, an error message "from davcio : error # 10" jumped out, whether I specified -npool for pw2wannier90.x or not. I can only run pw2wannier.x succesfully by adding a line wf_collect=.true. in the input file of nscf calculation. I am wondering if I always have to do so in any nscf calculations when multiple CPUs and pools are used. If this is the case, then I'll need to redo many of my finished calculations so that their MLWfs can be calculated. I appreciate any response and hint. Thank you! Han ********************************************************** Han Hsu, Ph.D. Postdoctoral Associate Department of Chemical Engineering & Materials Science University of Minnesota http://www.cems.umn.edu/~hsuhan ********************************************************** From a.mostofi at imperial.ac.uk Thu Jun 12 23:36:54 2008 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Thu, 12 Jun 2008 22:36:54 +0100 Subject: [Wannier] Setting wf_collect=.true. before running pw2wannier90.x? In-Reply-To: <200806122017.m5CKHIGQ019935@vendetta.software.umn.edu> References: <200806122017.m5CKHIGQ019935@vendetta.software.umn.edu> Message-ID: <485196F6.90605@imperial.ac.uk> Dear Han If you use multiple processors and only a single pool, then you don't necessarily need to use wf_collect: it should work as long as you use the same number of processors for the nscf and pw2wannier90 parts of the calculation. When multiple pools are used, however, you do need to set wf_collect=true to be able to run pw2wannier90 successfully. I should add that this isn't a fundamental limitation, it's just that no one has got around to coding it yet... Best regards Arash -- :------------------------------------------------------------: : Dr. Arash A. Mostofi :: a.mostofi at imperial.ac.uk : : Lecturer and RCUK Fellow :: : : Depts. of Materials & Physics :: : : Imperial College London :: T +44 (0)207 594 6753 : : London SW7 2AZ, United Kingdom :: F +44 (0)207 594 6757 : :------- http://www.cmth.ph.ic.ac.uk/people/a.mostofi -------: Han Hsu wrote: > Hello wannier90 users and developers, > > Recently I started learning wannier90 by running examples. I am using PWscf > ver 3.2.3, and I have already downloaded and compiled the latest wannier90 > and pw2wannier90. > > When running example 5 (MLWF for diamond), I used 4 CPUs and 2 pools to do > scf and nscf calculation with the input file provided in wannier90-1.1/Doc. > Then, when I ran pw2wan, an error message "from davcio : error # 10" jumped > out, whether I specified -npool for pw2wannier90.x or not. > > I can only run pw2wannier.x succesfully by adding a line wf_collect=.true. > in the input file of nscf calculation. I am wondering if I always have to > do so in any nscf calculations when multiple CPUs and pools are used. > > If this is the case, then I'll need to redo many of my finished > calculations so that their MLWfs can be calculated. > > I appreciate any response and hint. Thank you! > > Han > > ********************************************************** > Han Hsu, Ph.D. > Postdoctoral Associate > Department of Chemical Engineering & Materials Science > University of Minnesota > http://www.cems.umn.edu/~hsuhan > ********************************************************** > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier From hsuhan at cems.umn.edu Fri Jun 13 20:54:42 2008 From: hsuhan at cems.umn.edu (Han Hsu) Date: Fri, 13 Jun 2008 13:54:42 -0500 Subject: [Wannier] Setting wf_collect=.true. before runningpw2wannier90.x? In-Reply-To: <485196F6.90605@imperial.ac.uk> References: <200806122017.m5CKHIGQ019935@vendetta.software.umn.edu> <485196F6.90605@imperial.ac.uk> Message-ID: <000401c8cd86$f1271040$0fa55486@cems.umn.edu> Thank you very much, Arash. Perhaps the setting of wf_collect can be added to the example input files in the next release of Wannier90 if the code remains the same for this... Han -----Original Message----- From: wannier-bounces at quantum-espresso.org [mailto:wannier-bounces at quantum-espresso.org] On Behalf Of Arash Mostofi Sent: Thursday, June 12, 2008 4:37 PM To: wannier at quantum-espresso.org Subject: Re: [Wannier] Setting wf_collect=.true. before runningpw2wannier90.x? Dear Han If you use multiple processors and only a single pool, then you don't necessarily need to use wf_collect: it should work as long as you use the same number of processors for the nscf and pw2wannier90 parts of the calculation. When multiple pools are used, however, you do need to set wf_collect=true to be able to run pw2wannier90 successfully. I should add that this isn't a fundamental limitation, it's just that no one has got around to coding it yet... Best regards Arash -- Han Hsu wrote: > When running example 5 (MLWF for diamond), I used 4 CPUs and 2 pools to do > scf and nscf calculation with the input file provided in wannier90-1.1/Doc. > Then, when I ran pw2wan, an error message "from davcio : error # 10" jumped Wannier mailing list Wannier at quantum-espresso.org http://www.democritos.it/mailman/listinfo/wannier From hsuhan at cems.umn.edu Sat Jun 14 00:14:03 2008 From: hsuhan at cems.umn.edu (Han Hsu) Date: Fri, 13 Jun 2008 17:14:03 -0500 Subject: [Wannier] Question about searcg_shell: cell too long? In-Reply-To: <485196F6.90605@imperial.ac.uk> References: <200806122017.m5CKHIGQ019935@vendetta.software.umn.edu> <485196F6.90605@imperial.ac.uk> Message-ID: <000c01c8cda2$c999e040$0fa55486@cems.umn.edu> Hello Wannier90 users, I am trying to calculate MLWFs of LaCoO3. When using wannier90.x -pp to get nnkp file, an error occur: Unable to satisfy B1 with any of the first 12 shells Your cell might be very long, or you may have an irregular MP grid Try increasing the parameter search_shells in the win file (default=12) So I increase the number of shells. Same thing happened for search_shells = 13, ... 196. Then, when I set search_shells = 197, suddenly thousands nearest neighbors are found, and I got following error message kmesh_get: something wrong, found too many nearest neighbours LaCoO3 has R-3c symmetry, rhombohedral unit cell. In PWscf, its celldm(1)= 10.086259 a.u., and celldm(4)= 0.482720. For k_point mesh, 8 X 8 X 8 is used. The win file is attached at the end of this post. It is hard to imagine such cell could be too long. I really appreciate any response and hint. Thank you! Han !------------- ! Job Control !------------- !restart = plot ! To specify the restarting mode iprint = 3 exclude_bands = 1-16, 47-48 ! Exlude the undesired bands, search_shells = 197 !length_unit = bohr !------------- ! Wannierise !------------- num_iter = 100 !conv_window = 2 ! > 1 to activate !Conv_tol = 1.0E-10 ! (in Ang^2) !guiding_centres = T !------------ ! System !------------ num_wann = 30 !num_bands = 46 begin unit_cell_cart bohr 5.12953848 -2.96153745 8.16392898 0.00000000 5.92307490 8.16392898 -5.12953848 -2.96153745 8.16392898 end unit_cell_cart begin atoms_frac La 0.250000000 0.250000000 0.250000000 La 0.750000000 0.750000000 0.750000000 Co 0.000000000 0.000000000 0.000000000 Co 0.500000000 0.500000000 0.500000000 O 0.807523753 -0.307523753 0.250000000 O 0.250000000 0.807523753 -0.307523753 O -0.307523753 0.250000000 0.807523753 O 0.192476247 1.307523753 0.750000000 O 0.750000000 0.192476247 1.307523753 O 1.307523753 0.750000000 0.192476247 end atoms_frac begin projections Co:l=2, mr=2 Co:l=2, mr=3 Co:l=2, mr=5 O:l=0 O:l=1, mr=1 O:l=1, mr=2 O:l=1, mr=3 end projections mp_grid = 8 8 8 begin kpoints -0.375 -0.375 -0.375 -0.375 -0.375 -0.25 -0.375 -0.375 -0.125 -0.375 -0.375 0 -0.375 -0.375 0.125 -0.375 -0.375 0.25 -0.375 -0.375 0.375 -0.375 -0.375 0.5 -0.375 -0.25 -0.375 -0.375 -0.25 -0.25 -0.375 -0.25 -0.125 -0.375 -0.25 0 -0.375 -0.25 0.125 -0.375 -0.25 0.25 -0.375 -0.25 0.375 -0.375 -0.25 0.5 -0.375 -0.125 -0.375 -0.375 -0.125 -0.25 -0.375 -0.125 -0.125 -0.375 -0.125 0 -0.375 -0.125 0.125 -0.375 -0.125 0.25 -0.375 -0.125 0.375 -0.375 -0.125 0.5 -0.375 0 -0.375 -0.375 0 -0.25 -0.375 0 -0.125 -0.375 0 0 -0.375 0 0.125 -0.375 0 0.25 -0.375 0 0.375 -0.375 0 0.5 -0.375 0.125 -0.375 -0.375 0.125 -0.25 -0.375 0.125 -0.125 -0.375 0.125 0 -0.375 0.125 0.125 -0.375 0.125 0.25 -0.375 0.125 0.375 -0.375 0.125 0.5 -0.375 0.25 -0.375 -0.375 0.25 -0.25 -0.375 0.25 -0.125 -0.375 0.25 0 -0.375 0.25 0.125 -0.375 0.25 0.25 -0.375 0.25 0.375 -0.375 0.25 0.5 -0.375 0.375 -0.375 -0.375 0.375 -0.25 -0.375 0.375 -0.125 -0.375 0.375 0 -0.375 0.375 0.125 -0.375 0.375 0.25 -0.375 0.375 0.375 -0.375 0.375 0.5 -0.375 0.5 -0.375 -0.375 0.5 -0.25 -0.375 0.5 -0.125 -0.375 0.5 0 -0.375 0.5 0.125 -0.375 0.5 0.25 -0.375 0.5 0.375 -0.375 0.5 0.5 -0.25 -0.375 -0.375 -0.25 -0.375 -0.25 -0.25 -0.375 -0.125 -0.25 -0.375 0 -0.25 -0.375 0.125 -0.25 -0.375 0.25 -0.25 -0.375 0.375 -0.25 -0.375 0.5 -0.25 -0.25 -0.375 -0.25 -0.25 -0.25 -0.25 -0.25 -0.125 -0.25 -0.25 0 -0.25 -0.25 0.125 -0.25 -0.25 0.25 -0.25 -0.25 0.375 -0.25 -0.25 0.5 -0.25 -0.125 -0.375 -0.25 -0.125 -0.25 -0.25 -0.125 -0.125 -0.25 -0.125 0 -0.25 -0.125 0.125 -0.25 -0.125 0.25 -0.25 -0.125 0.375 -0.25 -0.125 0.5 -0.25 0 -0.375 -0.25 0 -0.25 -0.25 0 -0.125 -0.25 0 0 -0.25 0 0.125 -0.25 0 0.25 -0.25 0 0.375 -0.25 0 0.5 -0.25 0.125 -0.375 -0.25 0.125 -0.25 -0.25 0.125 -0.125 -0.25 0.125 0 -0.25 0.125 0.125 -0.25 0.125 0.25 -0.25 0.125 0.375 -0.25 0.125 0.5 -0.25 0.25 -0.375 -0.25 0.25 -0.25 -0.25 0.25 -0.125 -0.25 0.25 0 -0.25 0.25 0.125 -0.25 0.25 0.25 -0.25 0.25 0.375 -0.25 0.25 0.5 -0.25 0.375 -0.375 -0.25 0.375 -0.25 -0.25 0.375 -0.125 -0.25 0.375 0 -0.25 0.375 0.125 -0.25 0.375 0.25 -0.25 0.375 0.375 -0.25 0.375 0.5 -0.25 0.5 -0.375 -0.25 0.5 -0.25 -0.25 0.5 -0.125 -0.25 0.5 0 -0.25 0.5 0.125 -0.25 0.5 0.25 -0.25 0.5 0.375 -0.25 0.5 0.5 -0.125 -0.375 -0.375 -0.125 -0.375 -0.25 -0.125 -0.375 -0.125 -0.125 -0.375 0 -0.125 -0.375 0.125 -0.125 -0.375 0.25 -0.125 -0.375 0.375 -0.125 -0.375 0.5 -0.125 -0.25 -0.375 -0.125 -0.25 -0.25 -0.125 -0.25 -0.125 -0.125 -0.25 0 -0.125 -0.25 0.125 -0.125 -0.25 0.25 -0.125 -0.25 0.375 -0.125 -0.25 0.5 -0.125 -0.125 -0.375 -0.125 -0.125 -0.25 -0.125 -0.125 -0.125 -0.125 -0.125 0 -0.125 -0.125 0.125 -0.125 -0.125 0.25 -0.125 -0.125 0.375 -0.125 -0.125 0.5 -0.125 0 -0.375 -0.125 0 -0.25 -0.125 0 -0.125 -0.125 0 0 -0.125 0 0.125 -0.125 0 0.25 -0.125 0 0.375 -0.125 0 0.5 -0.125 0.125 -0.375 -0.125 0.125 -0.25 -0.125 0.125 -0.125 -0.125 0.125 0 -0.125 0.125 0.125 -0.125 0.125 0.25 -0.125 0.125 0.375 -0.125 0.125 0.5 -0.125 0.25 -0.375 -0.125 0.25 -0.25 -0.125 0.25 -0.125 -0.125 0.25 0 -0.125 0.25 0.125 -0.125 0.25 0.25 -0.125 0.25 0.375 -0.125 0.25 0.5 -0.125 0.375 -0.375 -0.125 0.375 -0.25 -0.125 0.375 -0.125 -0.125 0.375 0 -0.125 0.375 0.125 -0.125 0.375 0.25 -0.125 0.375 0.375 -0.125 0.375 0.5 -0.125 0.5 -0.375 -0.125 0.5 -0.25 -0.125 0.5 -0.125 -0.125 0.5 0 -0.125 0.5 0.125 -0.125 0.5 0.25 -0.125 0.5 0.375 -0.125 0.5 0.5 0 -0.375 -0.375 0 -0.375 -0.25 0 -0.375 -0.125 0 -0.375 0 0 -0.375 0.125 0 -0.375 0.25 0 -0.375 0.375 0 -0.375 0.5 0 -0.25 -0.375 0 -0.25 -0.25 0 -0.25 -0.125 0 -0.25 0 0 -0.25 0.125 0 -0.25 0.25 0 -0.25 0.375 0 -0.25 0.5 0 -0.125 -0.375 0 -0.125 -0.25 0 -0.125 -0.125 0 -0.125 0 0 -0.125 0.125 0 -0.125 0.25 0 -0.125 0.375 0 -0.125 0.5 0 0 -0.375 0 0 -0.25 0 0 -0.125 0 0 0 0 0 0.125 0 0 0.25 0 0 0.375 0 0 0.5 0 0.125 -0.375 0 0.125 -0.25 0 0.125 -0.125 0 0.125 0 0 0.125 0.125 0 0.125 0.25 0 0.125 0.375 0 0.125 0.5 0 0.25 -0.375 0 0.25 -0.25 0 0.25 -0.125 0 0.25 0 0 0.25 0.125 0 0.25 0.25 0 0.25 0.375 0 0.25 0.5 0 0.375 -0.375 0 0.375 -0.25 0 0.375 -0.125 0 0.375 0 0 0.375 0.125 0 0.375 0.25 0 0.375 0.375 0 0.375 0.5 0 0.5 -0.375 0 0.5 -0.25 0 0.5 -0.125 0 0.5 0 0 0.5 0.125 0 0.5 0.25 0 0.5 0.375 0 0.5 0.5 0.125 -0.375 -0.375 0.125 -0.375 -0.25 0.125 -0.375 -0.125 0.125 -0.375 0 0.125 -0.375 0.125 0.125 -0.375 0.25 0.125 -0.375 0.375 0.125 -0.375 0.5 0.125 -0.25 -0.375 0.125 -0.25 -0.25 0.125 -0.25 -0.125 0.125 -0.25 0 0.125 -0.25 0.125 0.125 -0.25 0.25 0.125 -0.25 0.375 0.125 -0.25 0.5 0.125 -0.125 -0.375 0.125 -0.125 -0.25 0.125 -0.125 -0.125 0.125 -0.125 0 0.125 -0.125 0.125 0.125 -0.125 0.25 0.125 -0.125 0.375 0.125 -0.125 0.5 0.125 0 -0.375 0.125 0 -0.25 0.125 0 -0.125 0.125 0 0 0.125 0 0.125 0.125 0 0.25 0.125 0 0.375 0.125 0 0.5 0.125 0.125 -0.375 0.125 0.125 -0.25 0.125 0.125 -0.125 0.125 0.125 0 0.125 0.125 0.125 0.125 0.125 0.25 0.125 0.125 0.375 0.125 0.125 0.5 0.125 0.25 -0.375 0.125 0.25 -0.25 0.125 0.25 -0.125 0.125 0.25 0 0.125 0.25 0.125 0.125 0.25 0.25 0.125 0.25 0.375 0.125 0.25 0.5 0.125 0.375 -0.375 0.125 0.375 -0.25 0.125 0.375 -0.125 0.125 0.375 0 0.125 0.375 0.125 0.125 0.375 0.25 0.125 0.375 0.375 0.125 0.375 0.5 0.125 0.5 -0.375 0.125 0.5 -0.25 0.125 0.5 -0.125 0.125 0.5 0 0.125 0.5 0.125 0.125 0.5 0.25 0.125 0.5 0.375 0.125 0.5 0.5 0.25 -0.375 -0.375 0.25 -0.375 -0.25 0.25 -0.375 -0.125 0.25 -0.375 0 0.25 -0.375 0.125 0.25 -0.375 0.25 0.25 -0.375 0.375 0.25 -0.375 0.5 0.25 -0.25 -0.375 0.25 -0.25 -0.25 0.25 -0.25 -0.125 0.25 -0.25 0 0.25 -0.25 0.125 0.25 -0.25 0.25 0.25 -0.25 0.375 0.25 -0.25 0.5 0.25 -0.125 -0.375 0.25 -0.125 -0.25 0.25 -0.125 -0.125 0.25 -0.125 0 0.25 -0.125 0.125 0.25 -0.125 0.25 0.25 -0.125 0.375 0.25 -0.125 0.5 0.25 0 -0.375 0.25 0 -0.25 0.25 0 -0.125 0.25 0 0 0.25 0 0.125 0.25 0 0.25 0.25 0 0.375 0.25 0 0.5 0.25 0.125 -0.375 0.25 0.125 -0.25 0.25 0.125 -0.125 0.25 0.125 0 0.25 0.125 0.125 0.25 0.125 0.25 0.25 0.125 0.375 0.25 0.125 0.5 0.25 0.25 -0.375 0.25 0.25 -0.25 0.25 0.25 -0.125 0.25 0.25 0 0.25 0.25 0.125 0.25 0.25 0.25 0.25 0.25 0.375 0.25 0.25 0.5 0.25 0.375 -0.375 0.25 0.375 -0.25 0.25 0.375 -0.125 0.25 0.375 0 0.25 0.375 0.125 0.25 0.375 0.25 0.25 0.375 0.375 0.25 0.375 0.5 0.25 0.5 -0.375 0.25 0.5 -0.25 0.25 0.5 -0.125 0.25 0.5 0 0.25 0.5 0.125 0.25 0.5 0.25 0.25 0.5 0.375 0.25 0.5 0.5 0.375 -0.375 -0.375 0.375 -0.375 -0.25 0.375 -0.375 -0.125 0.375 -0.375 0 0.375 -0.375 0.125 0.375 -0.375 0.25 0.375 -0.375 0.375 0.375 -0.375 0.5 0.375 -0.25 -0.375 0.375 -0.25 -0.25 0.375 -0.25 -0.125 0.375 -0.25 0 0.375 -0.25 0.125 0.375 -0.25 0.25 0.375 -0.25 0.375 0.375 -0.25 0.5 0.375 -0.125 -0.375 0.375 -0.125 -0.25 0.375 -0.125 -0.125 0.375 -0.125 0 0.375 -0.125 0.125 0.375 -0.125 0.25 0.375 -0.125 0.375 0.375 -0.125 0.5 0.375 0 -0.375 0.375 0 -0.25 0.375 0 -0.125 0.375 0 0 0.375 0 0.125 0.375 0 0.25 0.375 0 0.375 0.375 0 0.5 0.375 0.125 -0.375 0.375 0.125 -0.25 0.375 0.125 -0.125 0.375 0.125 0 0.375 0.125 0.125 0.375 0.125 0.25 0.375 0.125 0.375 0.375 0.125 0.5 0.375 0.25 -0.375 0.375 0.25 -0.25 0.375 0.25 -0.125 0.375 0.25 0 0.375 0.25 0.125 0.375 0.25 0.25 0.375 0.25 0.375 0.375 0.25 0.5 0.375 0.375 -0.375 0.375 0.375 -0.25 0.375 0.375 -0.125 0.375 0.375 0 0.375 0.375 0.125 0.375 0.375 0.25 0.375 0.375 0.375 0.375 0.375 0.5 0.375 0.5 -0.375 0.375 0.5 -0.25 0.375 0.5 -0.125 0.375 0.5 0 0.375 0.5 0.125 0.375 0.5 0.25 0.375 0.5 0.375 0.375 0.5 0.5 0.5 -0.375 -0.375 0.5 -0.375 -0.25 0.5 -0.375 -0.125 0.5 -0.375 0 0.5 -0.375 0.125 0.5 -0.375 0.25 0.5 -0.375 0.375 0.5 -0.375 0.5 0.5 -0.25 -0.375 0.5 -0.25 -0.25 0.5 -0.25 -0.125 0.5 -0.25 0 0.5 -0.25 0.125 0.5 -0.25 0.25 0.5 -0.25 0.375 0.5 -0.25 0.5 0.5 -0.125 -0.375 0.5 -0.125 -0.25 0.5 -0.125 -0.125 0.5 -0.125 0 0.5 -0.125 0.125 0.5 -0.125 0.25 0.5 -0.125 0.375 0.5 -0.125 0.5 0.5 0 -0.375 0.5 0 -0.25 0.5 0 -0.125 0.5 0 0 0.5 0 0.125 0.5 0 0.25 0.5 0 0.375 0.5 0 0.5 0.5 0.125 -0.375 0.5 0.125 -0.25 0.5 0.125 -0.125 0.5 0.125 0 0.5 0.125 0.125 0.5 0.125 0.25 0.5 0.125 0.375 0.5 0.125 0.5 0.5 0.25 -0.375 0.5 0.25 -0.25 0.5 0.25 -0.125 0.5 0.25 0 0.5 0.25 0.125 0.5 0.25 0.25 0.5 0.25 0.375 0.5 0.25 0.5 0.5 0.375 -0.375 0.5 0.375 -0.25 0.5 0.375 -0.125 0.5 0.375 0 0.5 0.375 0.125 0.5 0.375 0.25 0.5 0.375 0.375 0.5 0.375 0.5 0.5 0.5 -0.375 0.5 0.5 -0.25 0.5 0.5 -0.125 0.5 0.5 0 0.5 0.5 0.125 0.5 0.5 0.25 0.5 0.5 0.375 0.5 0.5 0.5 end kpoints !---------------- ! Disentanglement !---------------- !Edit later for metallic system !---------------- ! Post processing !--------------- ! To plot the WF !wannier_plot = T !wannier_plot_supercell = 2 !wannier_plot_list = 2 !restart = plot ! To plot Band Structure !bands_plot =T !begin kpoint_path !L 0.50000 0.50000 0.5000 G 0.00000 0.00000 0.0000 !G 0.00000 0.00000 0.0000 X 0.50000 0.00000 0.5000 !X 0.50000 -0.50000 0.0000 K 0.37500 -0.37500 0.0000 !K 0.37500 -0.37500 0.0000 G 0.00000 0.00000 0.0000 !end kpoint_path __________________________________________________ Han Hsu, Ph.D. Postdoctoral Associate Department of Chemical Engineering & Materials Science University of Minnesota http://www.cems.umn.edu/~hsuhan __________________________________________________ From jry20 at cam.ac.uk Sat Jun 14 12:21:23 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Sat, 14 Jun 2008 11:21:23 +0100 Subject: [Wannier] Question about searcg_shell: cell too long? In-Reply-To: <000c01c8cda2$c999e040$0fa55486@cems.umn.edu> References: <200806122017.m5CKHIGQ019935@vendetta.software.umn.edu> <485196F6.90605@imperial.ac.uk> <000c01c8cda2$c999e040$0fa55486@cems.umn.edu> Message-ID: <1D15C17D-6043-493C-A6CC-40A0004F8AA2@cam.ac.uk> On 13 Jun 2008, at 23:14, Han Hsu wrote: > > I am trying to calculate MLWFs of LaCoO3. When using wannier90.x - > pp to get > nnkp file, an error occur: > > Unable to satisfy B1 with any of the first 12 shells > Your cell might be very long, or you may have an irregular MP grid > Try increasing the parameter search_shells in the win file > (default=12) > > So I increase the number of shells. Same thing happened for > search_shells = > 13, ... 196. Then, when I set search_shells = 197, suddenly thousands > nearest neighbors are found, and I got following error message > > kmesh_get: something wrong, found too many nearest neighbours Han, This is subtle problem, and unfortunately the error message is misleading. The first two kpoint shells are actually sufficient to satisfy the so- called B1 condition. However, there is not enough precision in the specified lattice vectors, and so we don't quite hit the target accuracy for solving B1. I realise that specifying more sig. fig. in the lattice vectors is tough - but a perfectly reasonable solution is to lower the B1 tolerance: set kmesh_tol=0.00001 (this is 10x > than the default) and it'll work (no need to alter search_shells from the default 12) Note that lowering this tolerance will has no effect on the accuracy of the wannier functions. Thanks for pointing this out. In the future we should do two things: 1- Make a script to automatically create w90 input from pwscf inputs - and hence always include enough sig. figs. 2- The kmesh routine could be smarter; detect when we have only just failed to satisfy B1 and offer a more helpful error message. Yours Jonathan > -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From hsuhan at tyr.cems.umn.edu Sat Jun 14 19:06:06 2008 From: hsuhan at tyr.cems.umn.edu (Han Hsu) Date: Sat, 14 Jun 2008 12:06:06 CDT Subject: [Wannier] Question about searcg_shell: cell too long? Message-ID: <200806141706.m5EH66KW007332@vanguard.software.umn.edu> Hello Jonathan, Thank you for your reponse. Setting larger kmesh_tol works well. In fact, setting kmesh_tol = 0.000001005 is enough in my case. I guess this is beacause the structure of my LaCoO3 does not deviate too much from cubic perovskite...though I am not sure about this. Before knowing the setting for kmesh_tol, I followed the instruction written in the *.wout file, that is, setting devel_flag=kmesh_degen and writing a file for k-point shell. The file includes two shells, which should be sufficient. However, It did not work wither. Regards, Han > Han, > > This is subtle problem, and unfortunately the error message is > misleading. > > The first two kpoint shells are actually sufficient to satisfy the so- > called B1 condition. However, there is not enough precision in the > specified lattice vectors, and so we don't quite hit the target > accuracy for solving B1. I realise that specifying more sig. fig. in > the lattice vectors is tough - but a perfectly reasonable solution is > to lower the B1 tolerance: set > kmesh_tol=0.00001 (this is 10x > than the default) > and it'll work (no need to alter search_shells from the default 12) > Note that lowering this tolerance will has no effect on the accuracy > of the wannier functions. > > Thanks for pointing this out. In the future we should do two things: > > 1- Make a script to automatically create w90 input from pwscf inputs > - and hence always include enough sig. figs. > > 2- The kmesh routine could be smarter; detect when we have only just > failed to satisfy B1 and offer a more helpful error message. > > Yours > Jonathan ********************************************************** Han Hsu, Ph.D. Postdoctoral Associate Department of Chemical Engineering & Materials Science University of Minnesota http://www.cems.umn.edu/~hsuhan ********************************************************** From hsuhan at tyr.cems.umn.edu Tue Jun 17 01:33:21 2008 From: hsuhan at tyr.cems.umn.edu (Han Hsu) Date: Mon, 16 Jun 2008 18:33:21 CDT Subject: [Wannier] Different num_wann setting for pw2wannier90.x and wannier90.x.... Message-ID: <200806162333.m5GNXLlr000816@vendetta.software.umn.edu> ? Hello Wannier90 users and developers, I am dealing with an insulator. Its valence band contains 46 bands. Among these bands, 22 of them can be excluded, only the reminaing 24 bands are needed for MLWFs. When I ran pw2wannier90.x to calculate Amn, Mmn, and Unk, I did not realize only 24 bands are necessary, so I set num_wann=46 in my *.win file, and got 46X46 Amn and Mmn. Now I realize I only need 24 bands. Do I have to rerun pw2wannier90.x to get 24X24 Amn and Mmn? Or can I directly run wannier90.x by setting num_wann=24 with the previously calculated 46X46 Amn and Mmn? Thank you very much for your reponse. Han ********************************************************** Han Hsu, Ph.D. Postdoctoral Associate Department of Chemical Engineering & Materials Science University of Minnesota http://www.cems.umn.edu/~hsuhan ********************************************************** From f_mirjani at ph.iut.ac.ir Tue Jun 17 11:46:58 2008 From: f_mirjani at ph.iut.ac.ir (fatemeh.mirjani) Date: Tue, 17 Jun 2008 13:16:58 +0330 (IRST) Subject: [Wannier] the initial guess in the Amn matrix In-Reply-To: <3189687.32481213695528211.JavaMail.root@mta.iut.ac.ir> Message-ID: <15149141.32761213696018184.JavaMail.root@mta.iut.ac.ir> Dear users ; I have a question about the initial guess ( g_n(r) ) for Wanniers in the Amn matrix. g_n(r)=g(r)* sigma( c(l,m)* Y(l,m)) When we don't consider hybrids orbitals as a trial Wannier, the orbitals are simple (like as s,p,d,f) and the c(l,m) coefficients are equal to 1 . yes? But for considering the hybrids orbitals as a trial Wannier, I want to use the Table3.2 in page35 of Wannier_Userguide. For extracting the coefficients from this table we see l quantities are negative. I don't know how can I use these data whereas I should put the corresponding psi(l_c,m_c) (in the integral for Amn) that is not available in Wien2k. (because l is negative and we have psi(l,m) for positive l .) Your help is greatly appreciated like your previous favor. Kind Regards, Fatemeh -- +------------------------------------------------------------------------------+ |Fatemeh Mirjani | Email: f_mirjani at ph.iut.ac.ir | |Computational Condensed Matter Research Lab. | Tel/Fax Office: +98311-3913746 | |ICTP Affiliated Center in Isfahan, | Tel Lab.: +98311-3913731 | +----------Physics Department, Isfahan University of Technology, Iran----------+ From jry20 at cam.ac.uk Tue Jun 17 11:56:30 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: Tue, 17 Jun 2008 10:56:30 +0100 (BST) Subject: [Wannier] Different num_wann setting for pw2wannier90.x and wannier90.x.... In-Reply-To: <200806162333.m5GNXLlr000816@vendetta.software.umn.edu> References: <200806162333.m5GNXLlr000816@vendetta.software.umn.edu> Message-ID: On Mon, 16 Jun 2008, Han Hsu wrote: > I am dealing with an insulator. Its valence band contains 46 bands. Among > these bands, 22 of them can be excluded, only the reminaing 24 bands are > needed for MLWFs. > > When I ran pw2wannier90.x to calculate Amn, Mmn, and Unk, I did not realize > only 24 bands are necessary, so I set num_wann=46 in my *.win file, and got > 46X46 Amn and Mmn. > > Now I realize I only need 24 bands. Do I have to rerun pw2wannier90.x to > get 24X24 Amn and Mmn? Or can I directly run wannier90.x by setting > num_wann=24 with the previously calculated 46X46 Amn and Mmn? Han, If the 22 bands you want to exclude form an isolated set _below_ (or above) the remaining valence bands then you can just do the following: Set num_bands=46 and num_wann=24 Set the outer window for disentanglement to include only the upper 24 bands (ie put dis_min_win in the gap between the lower 22 and upper 24). This should cause w90 to try disentanglement - which will converge instantly as it has no work to do. This is identical to using exclude_bands. Any problems, just report back. If the 22 bands happen to form an isolated set within the original 24 - eg as for the 3d bands in ZnSe - then the only option is to use exclude_bands. No fundamental reason for this, it could be coded into w90. Jonathan -- Dr Jonathan Yates | Theory of Condensed Matter Group Corpus Christi College | Cavendish Laboratory Cambridge, CB2 1RH, UK | Cambridge, CB3 OHE, UK email jry20 at cam.ac.uk | Tel +44 (0)1223 337461 From hsuhan at cems.umn.edu Wed Jun 18 00:45:58 2008 From: hsuhan at cems.umn.edu (Han Hsu) Date: Tue, 17 Jun 2008 17:45:58 -0500 Subject: [Wannier] Different num_wann setting for pw2wannier90.x and wannier90.x.... In-Reply-To: References: <200806162333.m5GNXLlr000816@vendetta.software.umn.edu> Message-ID: <000901c8d0cb$e99f9f70$0fa55486@cems.umn.edu> Thank you, Jonathan. In my case, the 22 bands that I should have excluded when I ran pw2wan indeed form an isolated group below the 24 bands I am interested. I believe your suggestion, setting dis_win_min, works for this case. I did not try it, however, because I already calculated 24 X 24 Amn and Mmn way before I got any reponse.... Han -----Original Message----- From: wannier-bounces at quantum-espresso.org [mailto:wannier-bounces at quantum-espresso.org] On Behalf Of Jonathan Yates Sent: Tuesday, June 17, 2008 4:57 AM To: Wannier Mailinglist Subject: Re: [Wannier] Different num_wann setting for pw2wannier90.x and Han, If the 22 bands you want to exclude form an isolated set _below_ (or above) the remaining valence bands then you can just do the following: Set num_bands=46 and num_wann=24 Set the outer window for disentanglement to include only the upper 24 bands (ie put dis_min_win in the gap between the lower 22 and upper 24). This should cause w90 to try disentanglement - which will converge instantly as it has no work to do. This is identical to using exclude_bands. Any problems, just report back. If the 22 bands happen to form an isolated set within the original 24 - eg as for the 3d bands in ZnSe - then the only option is to use exclude_bands. No fundamental reason for this, it could be coded into w90. Jonathan _______________________________________________ Wannier mailing list Wannier at quantum-espresso.org http://www.democritos.it/mailman/listinfo/wannier From hsuhan at cems.umn.edu Wed Jun 18 01:59:40 2008 From: hsuhan at cems.umn.edu (Han Hsu) Date: Tue, 17 Jun 2008 18:59:40 -0500 Subject: [Wannier] A question regarding to restart mode.... Message-ID: <000001c8d0d6$34beaff0$0fa55486@cems.umn.edu> Hello Wannier90 developers and users, Now I have a question about how the restart mode works in Wannier90. Suppose I run wannier90.x by setting num_iter = 1000 in the *.win file. Unluckily, after 1000 iterations have completed, a satisfactory convergent result is not achieved. So I go back to the original *win file, reset num_iter = 2000, and add a line restart=default. When I run wannier90.x with the modified *win file, depending on the machine, either an error message "Error in input file: value of restart not recognized" displays, or no more iteration is computed, even no error message displays. Setting restart=default only works in the following way, at least on my machines. Suppose I set num_iter = 1000000, and within wall time limit, only 1000 iterations are completed. Then I add a line restart=default to the *win file, and the calculation can continue as expected. I am wondering if this happens to anyone else, or just to me? I already started to suspect if I compiled everything correctly now... Thank for any hints and responses. Han __________________________________________________ Han Hsu, Ph.D. Postdoctoral Associate Department of Chemical Engineering & Materials Science University of Minnesota http://www.cems.umn.edu/~hsuhan __________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From f_mirjani at ph.iut.ac.ir Wed Jun 18 09:11:53 2008 From: f_mirjani at ph.iut.ac.ir (fatemeh.mirjani) Date: Wed, 18 Jun 2008 10:41:53 +0330 (IRST) Subject: [Wannier] the UNK files (from Wien2k to Wannier90) Message-ID: <873157.43981213773113679.JavaMail.root@mta.iut.ac.ir> Dear users; If I want to extract the UNK files (from Wien2k to Wannier90), what should I do? The UNK files in Wannier90 are binary. Thanks in advance. Kind Regards, Fatemeh +------------------------------------------------------------------------------+ |Fatemeh Mirjani | Email: f_mirjani at ph.iut.ac.ir | |Computational Condensed Matter Research Lab. | Tel/Fax Office: +98311-3913746 | |ICTP Affiliated Center in Isfahan, | Tel Lab.: +98311-3913731 | +----------Physics Department, Isfahan University of Technology, Iran----------+ From a.mostofi at imperial.ac.uk Wed Jun 18 09:57:20 2008 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Wed, 18 Jun 2008 08:57:20 +0100 Subject: [Wannier] A question regarding to restart mode.... In-Reply-To: <000001c8d0d6$34beaff0$0fa55486@cems.umn.edu> References: <000001c8d0d6$34beaff0$0fa55486@cems.umn.edu> Message-ID: <4858BFE0.80801@imperial.ac.uk> Dear Han If you want the calculation to continue doing more wannier localisation steps you should set restart=wannierise rather than default. The fact that you get an unhelpful error message (or nothing at all) is not ideal and we'll look into it. Arash -- :------------------------------------------------------------: : Dr. Arash A. Mostofi :: a.mostofi at imperial.ac.uk : : Lecturer and RCUK Fellow :: : : Depts. of Materials & Physics :: : : Imperial College London :: T +44 (0)207 594 6753 : : London SW7 2AZ, United Kingdom :: F +44 (0)207 594 6757 : :------- http://www.cmth.ph.ic.ac.uk/people/a.mostofi -------: Han Hsu wrote: > Hello Wannier90 developers and users, > > > > Now I have a question about how the restart mode works in Wannier90. > > > > Suppose I run wannier90.x by setting num_iter = 1000 in the *.win file. > Unluckily, after 1000 iterations have completed, a satisfactory > convergent result is not achieved. So I go back to the original *win > file, reset num_iter = 2000, and add a line restart=default. When I run > wannier90.x with the modified *win file, depending on the machine, > either an error message ?Error in input file: value of restart not > recognized? displays, or no more iteration is computed, even no error > message displays. > > > > Setting restart=default only works in the following way, at least on my > machines. Suppose I set num_iter = 1000000, and within wall time limit, > only 1000 iterations are completed. Then I add a line restart=default to > the *win file, and the calculation can continue as expected. > > > > I am wondering if this happens to anyone else, or just to me? I already > started to suspect if I compiled everything correctly now?.. > > > > Thank for any hints and responses. > > > > Han > > __________________________________________________ > > Han Hsu, Ph.D. > > Postdoctoral Associate > > Department of Chemical Engineering & Materials Science > > University of Minnesota > > http://www.cems.umn.edu/~hsuhan > > __________________________________________________ > > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier From a.mostofi at imperial.ac.uk Wed Jun 18 09:57:37 2008 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Wed, 18 Jun 2008 08:57:37 +0100 Subject: [Wannier] the UNK files (from Wien2k to Wannier90) In-Reply-To: <873157.43981213773113679.JavaMail.root@mta.iut.ac.ir> References: <873157.43981213773113679.JavaMail.root@mta.iut.ac.ir> Message-ID: <4858BFF1.3040607@imperial.ac.uk> Dear Fatemeh There is a logical keyword "wvfn_formatted" in Wannier90 which you'll find useful. Arash -- :------------------------------------------------------------: : Dr. Arash A. Mostofi :: a.mostofi at imperial.ac.uk : : Lecturer and RCUK Fellow :: : : Depts. of Materials & Physics :: : : Imperial College London :: T +44 (0)207 594 6753 : : London SW7 2AZ, United Kingdom :: F +44 (0)207 594 6757 : :------- http://www.cmth.ph.ic.ac.uk/people/a.mostofi -------: fatemeh.mirjani wrote: > Dear users; > > If I want to extract the UNK files (from Wien2k to Wannier90), what should I do? > The UNK files in Wannier90 are binary. > > Thanks in advance. > > Kind Regards, > Fatemeh > > +------------------------------------------------------------------------------+ > |Fatemeh Mirjani | Email: f_mirjani at ph.iut.ac.ir | > |Computational Condensed Matter Research Lab. | Tel/Fax Office: +98311-3913746 | > |ICTP Affiliated Center in Isfahan, | Tel Lab.: +98311-3913731 | > +----------Physics Department, Isfahan University of Technology, Iran----------+ > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier From hsuhan at cems.umn.edu Wed Jun 18 21:16:07 2008 From: hsuhan at cems.umn.edu (Han Hsu) Date: Wed, 18 Jun 2008 14:16:07 -0500 Subject: [Wannier] A question regarding to restart mode.... In-Reply-To: <4858BFE0.80801@imperial.ac.uk> References: <000001c8d0d6$34beaff0$0fa55486@cems.umn.edu> <4858BFE0.80801@imperial.ac.uk> Message-ID: <001201c8d177$c2f37e30$0fa55486@cems.umn.edu> Dear Arash, Thank you very much for the response. I'll try setting restart=wannierise. I would like to point out that the manual might be a little bit misleading about this for a non-native English speaker like me. It says (on page 17) setting restart=wannierise would "Restart from the beginning of wannierise routine". For me, "the beginning of wannierise routine" sounds like starting the wannierization from scratch again..... This is why I chose restart=default over restart=wannierise. Again, thank you for the clarification. Han -----Original Message----- From: wannier-bounces at quantum-espresso.org [mailto:wannier-bounces at quantum-espresso.org] On Behalf Of Arash Mostofi Sent: Wednesday, June 18, 2008 2:57 AM To: 'Wannier Mailinglist' Subject: Re: [Wannier] A question regarding to restart mode.... Dear Han If you want the calculation to continue doing more wannier localisation steps you should set restart=wannierise rather than default. The fact that you get an unhelpful error message (or nothing at all) is not ideal and we'll look into it. Arash From vranjan at ncsu.edu Wed Jun 25 18:23:07 2008 From: vranjan at ncsu.edu (Vivek Ranjan) Date: Wed, 25 Jun 2008 12:23:07 -0400 (EDT) Subject: [Wannier] kpoint Message-ID: <42359.152.14.74.117.1214410987.squirrel@webmail.ncsu.edu> Hello, I have successfully run scf and nscf calculations using pwscf. In the next stage - wannier code to get generate overlaps - I input the same kpoints as automatically generated in scf steps (in units of 2*pi/a). But wannier stops making a complaint : Exiting....... Non-symmetric k-point neighbours! I cannot understand the reason. Can someone help in what could be wrong ? Thank you, Vivek Ranjan Department of Physics, NCSU, Raleigh, NC From jry20 at cam.ac.uk Wed Jun 25 23:45:53 2008 From: jry20 at cam.ac.uk (Jonathan Yates) Date: 25 Jun 2008 22:45:53 +0100 Subject: [Wannier] kpoint In-Reply-To: <42359.152.14.74.117.1214410987.squirrel@webmail.ncsu.edu> References: <42359.152.14.74.117.1214410987.squirrel@webmail.ncsu.edu> Message-ID: On Jun 25 2008, Vivek Ranjan wrote: >I have successfully run scf and nscf calculations using pwscf. In the next >stage - wannier code to get generate overlaps - I input the same kpoints >as automatically generated in scf steps (in units of 2*pi/a). But wannier >stops making a complaint : > >Exiting....... >Non-symmetric k-point neighbours! > >I cannot understand the reason. Can someone help in what could be wrong ? Vivek, Wannier90 expects the kpoints to form a regular grid. Conventionally this would be a regular grid in the reciprocal unit cell (as generated by the script utility/kmesh.pl) Note that W90 does not (at present) take account of any symmetry - ie you must give the kpoints in the full zone (not the symmetry reduced kpoints) The error message simply means that W90 thinks your kpoints do not form a regular grid. A few possibilities: you are using the symmetry reduced kpoints from the scf, you have not specified the kpoints to enough decimal places (eg for a 6 6 6 mesh). I suggest using the kmesh.pl script to generate the kpoints for W90 and nscf calculations. If none of these suggestions help, please post your input files (without them we can only guess at the problem) Jonathan -- Cavendish Lab, Cambridge, UK