From dwz4587 at 163.com Tue Mar 11 01:06:38 2014 From: dwz4587 at 163.com (dwz) Date: Tue, 11 Mar 2014 08:06:38 +0800 (CST) Subject: [Wannier] kmesh_get_bvector: Not enough bvectors found Message-ID: <3db276e.9fa.144ae7664b9.Coremail.dwz4587@163.com> Dear wannier experts: When trying to run wannier90 lib mode together with VASP, the wannier.wout reported such erorr -? kmesh_get_bvector: Not enough bvectors found.? It seems that wannier90 fail to find enough b-vectors for the shells defined (I chosed a default = 12). Then I improved the search_shells to 100, but same error appearred.The k-grid is 7*7*1 Monckhorst_Pack. Should I change the search_shell to higher values untill no error appears ? Or should I handle other things ? Any response will be appreciated. Yours. Wenzheng Dong Shandong University -------------- next part -------------- An HTML attachment was scrubbed... URL: From mail.fanwei at gmail.com Wed Mar 12 03:29:12 2014 From: mail.fanwei at gmail.com (Wei Fan) Date: Wed, 12 Mar 2014 11:29:12 +0900 Subject: [Wannier] Wannier center change Message-ID: Dear all, I am a new user of Wannier90. Recently, I am using VASP + Wannier90 for 5d systems. I have two questions. 1) why will the center of wanner functions shift during wannierising process? Initially, I set up wannier functions at 5d element's site. Even from the first step, the WF center shift. for example: wannier90.win: "Begin Projections c=0,0,0 :dyz;dxz;dxy: z=0,0,1: x=0.97,-0.2428,0 c=2.760835,2.760835,0 :dyz;dxz;dxy: z=0,0,1: x=0.97,0.2428,0 End Projections " I am expecting something like: WF 1 0,0,0 WF 2 0,0,0 WF 3 0,0,0 WF 4 2.760835,2.760835,0 WF 5 2.760835,2.760835,0 WF 6 2.760835,2.760835,0 but in wannier90.wout: " *------------------------------- WANNIERISE ---------------------------------* +--------------------------------------------------------------------+<-- CONV | Iter Delta Spread RMS Gradient Spread (Ang^2) Time |<-- CONV +--------------------------------------------------------------------+<-- CONV ------------------------------------------------------------------------------ Initial State WF centre and spread 1 ( 0.000000, 0.000000, 0.000000 ) 4.40959838 WF centre and spread 2 ( 0.000000, 0.000000, 0.000000 ) 4.40959827 WF centre and spread 3 ( -0.690210, -0.000002, 0.000000 ) 118.47980281 WF centre and spread 4 ( 2.760835, 2.760835, 0.000000 ) 4.13599014 WF centre and spread 5 ( 2.760835, 2.760835, 0.000000 ) 4.13599017 WF centre and spread 6 ( 2.760832, 2.070626, 0.000000 ) 210.53405406 Sum of centres and spreads ( 7.592293, 7.592295, 0.000000 ) 346.10503382 " 2) For metallic system, are there more physical meaning to include more states above Fermi level into the Frozen window? Or basically I can focus the inner window at anywhere I am interested. Thank you very much in advance! Wei Fan Postdoctoral researcher, Computational Condensed Matter Physics Group RIKEN, Japan -------------- next part -------------- An HTML attachment was scrubbed... URL: From member at linkedin.com Wed Mar 12 21:15:44 2014 From: member at linkedin.com (=?UTF-8?Q?=D8=A7=D9=85=DB=8C=D8=B1_=D8=A8=DB=8C=D8=A7=D9=86=DB=8C_?= =?UTF-8?Q?via_LinkedIn?=) Date: Wed, 12 Mar 2014 20:15:44 +0000 (UTC) Subject: [Wannier] Hong-Wei, please add me to your LinkedIn network Message-ID: <674235185.34903916.1394655344560.JavaMail.app@ela4-app2733.prod> LinkedIn ------------ ???? ????? requested to add you as a connection on LinkedIn: ------------------------------------------ Hong-Wei, I'd like to add you to my professional network on LinkedIn. - ???? Accept invitation from ???? ????? http://www.linkedin.com/e/41efdi-hsp1s71m-1s/p42GAvl-sDhJZ934E2dLPjc_8yn4e01gJFkVP-4btyP/blk/I537617566_130/3wOtCVFbmdxnSVFbm8JrnpKqlZJrmZzbmNJpjRQnOpBtn9QfmhBt71BoSd1p65Lr6lOfP0PclYSdzkTcjoTcPl9bQV7sD18m38JbP0OejkMdjAVej8Td3wMdzAQe3kLrCBxbOYWrSlI/eml-comm_invm-b-in_ac-inv28/?hs=false&tok=3bW_PeQcXvHm81 View profile of ???? ????? http://www.linkedin.com/e/41efdi-hsp1s71m-1s/rso/279387954/kdwd/name/130724507_I537617566_130/?hs=false&tok=3dLZPT8RPvHm81 ------------------------------------------ You are receiving Invitation emails. This email was intended for Hong-Wei Wang. Learn why this is included: http://www.linkedin.com/e/41efdi-hsp1s71m-1s/plh/http%3A%2F%2Fhelp%2Elinkedin%2Ecom%2Fapp%2Fanswers%2Fdetail%2Fa_id%2F4788/-GXI/?hs=false&tok=2OfwRFIdbvHm81 (c) 2012, LinkedIn Corporation. 2029 Stierlin Ct, Mountain View, CA 94043, USA. -------------- next part -------------- An HTML attachment was scrubbed... URL: From salazar at physics.utoronto.ca Wed Mar 26 03:04:07 2014 From: salazar at physics.utoronto.ca (Cuauhtemoc Salazar) Date: Tue, 25 Mar 2014 22:04:07 -0400 Subject: [Wannier] Problems with the updated pw2wanier90 interface Message-ID: <69693BDA-6A89-4117-9B1B-BEC85D453CBE@physics.utoronto.ca> Dear Forum Members, I am having problems to work out the examples provided in the wannier90-2.0.0 distribution that require the updated pw2wannier90.f90 code, which is provided by wannier90 at wannier90-2.0.0/pwscf/v5.0 Firstly, I successfully reproduced examples 01 to 05 using the original pw2wanier90.f90 code provided within QEspresso_5.0.2 (such example does *not* require the updated version of pw2wannier90, but I started with this simpler example before moving to the actual example of my interest, example18 --Berry curvature). Then, following wannier90-2.0.0/pwscf/README I did the replacement cp wannier90-2.0.0/pwscf/v5.0/pw2wannier90.f90 espresso-5.0.2/PP/src/. and re-built the PP part of QE: cd ~/espresso-5.0.2/PP make pp Then I tried to reproduce (again) the wannier90-2.0.0/examples/example05 (running pw2wannier90 with the same number of processors as pw.x, as indicated in the w90 tutorial guide, page 10) but got a number of errors, appended at the end of this message. I wonder if someone could have an advice to solve this, With regards, Temok [panther2:46075] *** Process received signal *** [panther2:46075] Signal: Segmentation fault (11) [panther2:46075] Signal code: Address not mapped (1) [panther2:46075] Failing at address: (nil) [panther2:46075] [ 0] /lib64/libpthread.so.0() [0x30de60f710] [panther2:46075] [ 1] /lib64/libc.so.6(memcpy+0x35) [0x31a0289985] [panther2:46075] [ 2] /home/temok/opt/openmpi-1.6.5/build-gcc/lib/libmpi.so.1(opal_convertor_pack+0x196) [0x7fbca07d61a6] [panther2:46075] [ 3] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_btl_sm.so(mca_btl_sm_sendi+0x23d) [0x7fbc9c7d065d] [panther2:46075] [ 4] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_pml_ob1.so(mca_pml_ob1_send_request_start_copy+0xb4) [0x7fbc9cfeeff4] [panther2:46075] [ 5] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_pml_ob1.so(mca_pml_ob1_isend+0x4b0) [0x7fbc9cfe7350] [panther2:46075] [ 6] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_tuned.so(ompi_coll_tuned_bcast_intra_generic+0x323) [0x7fbc9b91d2a3] [panther2:46075] [ 7] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_tuned.so(ompi_coll_tuned_bcast_intra_binomial+0xd8) [0x7fbc9b91d638] [panther2:46075] [ 8] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_tuned.so(ompi_coll_tuned_bcast_intra_dec_fixed+0xcc) [0x7fbc9b91505c] [panther2:46075] [ 9] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_sync.so(mca_coll_sync_bcast+0x78) [0x7fbc9bb2b6b8] [panther2:46075] [10] /home/temok/opt/openmpi-1.6.5/build-gcc/lib/libmpi.so.1(MPI_Bcast+0x130) [0x7fbca073c3b0] [panther2:46075] [11] /home/temok/opt/openmpi-1.6.5/build-gcc/lib/libmpi_f77.so.1(pmpi_bcast+0x7e) [0x7fbca0a935be] [panther2:46075] [12] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(bcast_integer_+0x192) [0x60d742] [panther2:46075] [13] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(__mp_MOD_mp_bcast_iv+0x97) [0x5feff7] [panther2:46075] [14] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(read_nnkp_+0x1bbf) [0x44220f] [panther2:46075] [15] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(MAIN__+0x9c9) [0x447b49] [panther2:46075] [16] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(main+0x2a) [0x72267a] [panther2:46075] [17] /lib64/libc.so.6(__libc_start_main+0xfd) [0x31a021ed1d] [panther2:46075] [18] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x() [0x42b819] [panther2:46075] *** End of error message *** -------------------------------------------------------------------------- mpirun noticed that process rank 0 with PID 46075 on node XXXXXXXX exited on signal 11 (Segmentation fault). -------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From giovanni.pizzi at epfl.ch Thu Mar 27 12:33:43 2014 From: giovanni.pizzi at epfl.ch (Giovanni Pizzi) Date: Thu, 27 Mar 2014 11:33:43 +0000 Subject: [Wannier] Problems with the updated pw2wanier90 interface In-Reply-To: <69693BDA-6A89-4117-9B1B-BEC85D453CBE@physics.utoronto.ca> References: <69693BDA-6A89-4117-9B1B-BEC85D453CBE@physics.utoronto.ca> Message-ID: <789C9A13-C9C1-48FB-854E-8E355292B0AF@epfl.ch> Dear Temok, There are different possible reasons for such an error, depending on which version of Quantum Espresso you are using. In particular, after the release of Wannier90 2.0, the QE code has changed some internal details, and therefore the pw2wannier90.f90 file that we provide is not working anymore with the most recent SVN version of QE. However, it should work with QE 5.0.2 patched with the 5.0.3 patch that you can find on the QE website. Could you try to use the 5.0.2 + 5.0.3 patch of Quantum Espresso (if you are not using it already) together with the pw2wannier90.f90 file provided with Wannier90 2.0, and report us if the problem is solved? Thanks, Giovanni -- Giovanni Pizzi Post-doctoral Research Scientist EPFL STI IMX THEOS MXC 340 (B?timent MXC) Station 12 CH-1015 Lausanne (Switzerland) Phone: +41 21 69 31124 On 26 Mar 2014, at 03:04, Cuauhtemoc Salazar wrote: Dear Forum Members, I am having problems to work out the examples provided in the wannier90-2.0.0 distribution that require the updated pw2wannier90.f90 code, which is provided by wannier90 at wannier90-2.0.0/pwscf/v5.0 Firstly, I successfully reproduced examples 01 to 05 using the original pw2wanier90.f90 code provided within QEspresso_5.0.2 (such example does *not* require the updated version of pw2wannier90, but I started with this simpler example before moving to the actual example of my interest, example18 --Berry curvature). Then, following wannier90-2.0.0/pwscf/README I did the replacement cp wannier90-2.0.0/pwscf/v5.0/pw2wannier90.f90 espresso-5.0.2/PP/src/. and re-built the PP part of QE: cd ~/espresso-5.0.2/PP make pp Then I tried to reproduce (again) the wannier90-2.0.0/examples/example05 (running pw2wannier90 with the same number of processors as pw.x, as indicated in the w90 tutorial guide, page 10) but got a number of errors, appended at the end of this message. I wonder if someone could have an advice to solve this, With regards, Temok [panther2:46075] *** Process received signal *** [panther2:46075] Signal: Segmentation fault (11) [panther2:46075] Signal code: Address not mapped (1) [panther2:46075] Failing at address: (nil) [panther2:46075] [ 0] /lib64/libpthread.so.0() [0x30de60f710] [panther2:46075] [ 1] /lib64/libc.so.6(memcpy+0x35) [0x31a0289985] [panther2:46075] [ 2] /home/temok/opt/openmpi-1.6.5/build-gcc/lib/libmpi.so.1(opal_convertor_pack+0x196) [0x7fbca07d61a6] [panther2:46075] [ 3] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_btl_sm.so(mca_btl_sm_sendi+0x23d) [0x7fbc9c7d065d] [panther2:46075] [ 4] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_pml_ob1.so(mca_pml_ob1_send_request_start_copy+0xb4) [0x7fbc9cfeeff4] [panther2:46075] [ 5] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_pml_ob1.so(mca_pml_ob1_isend+0x4b0) [0x7fbc9cfe7350] [panther2:46075] [ 6] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_tuned.so(ompi_coll_tuned_bcast_intra_generic+0x323) [0x7fbc9b91d2a3] [panther2:46075] [ 7] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_tuned.so(ompi_coll_tuned_bcast_intra_binomial+0xd8) [0x7fbc9b91d638] [panther2:46075] [ 8] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_tuned.so(ompi_coll_tuned_bcast_intra_dec_fixed+0xcc) [0x7fbc9b91505c] [panther2:46075] [ 9] /home/temok/opt/openmpi-1.6.5/build-gcc/./lib/openmpi/mca_coll_sync.so(mca_coll_sync_bcast+0x78) [0x7fbc9bb2b6b8] [panther2:46075] [10] /home/temok/opt/openmpi-1.6.5/build-gcc/lib/libmpi.so.1(MPI_Bcast+0x130) [0x7fbca073c3b0] [panther2:46075] [11] /home/temok/opt/openmpi-1.6.5/build-gcc/lib/libmpi_f77.so.1(pmpi_bcast+0x7e) [0x7fbca0a935be] [panther2:46075] [12] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(bcast_integer_+0x192) [0x60d742] [panther2:46075] [13] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(__mp_MOD_mp_bcast_iv+0x97) [0x5feff7] [panther2:46075] [14] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(read_nnkp_+0x1bbf) [0x44220f] [panther2:46075] [15] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(MAIN__+0x9c9) [0x447b49] [panther2:46075] [16] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x(main+0x2a) [0x72267a] [panther2:46075] [17] /lib64/libc.so.6(__libc_start_main+0xfd) [0x31a021ed1d] [panther2:46075] [18] ../../../QEspresso/espresso-5.0.2/bin/pw2wannier90.x() [0x42b819] [panther2:46075] *** End of error message *** -------------------------------------------------------------------------- mpirun noticed that process rank 0 with PID 46075 on node XXXXXXXX exited on signal 11 (Segmentation fault). -------------------------------------------------------------------------- _______________________________________________ Wannier mailing list Wannier at quantum-espresso.org http://www.democritos.it/mailman/listinfo/wannier -------------- next part -------------- An HTML attachment was scrubbed... URL: From ustc0200 at gmail.com Thu Mar 27 13:10:25 2014 From: ustc0200 at gmail.com (Xi Zhu) Date: Thu, 27 Mar 2014 20:10:25 +0800 Subject: [Wannier] projection choosen for MoS2 Message-ID: Dear Member; Could anyone give some suggestion on how to choose the "begin projection" part in MoS2 single layer? Thanks ! XI -------------- next part -------------- An HTML attachment was scrubbed... URL: From jinluocheng.phys at gmail.com Thu Mar 27 13:38:55 2014 From: jinluocheng.phys at gmail.com (JinLuo Cheng) Date: Thu, 27 Mar 2014 13:38:55 +0100 Subject: [Wannier] projection choosen for MoS2 In-Reply-To: References: Message-ID: <53341BDF.5050009@gmail.com> Dear Xi, I think the begin projection for highest 6 valence bands and lowest 5 conduction bands maybe taken as Mo:l=2 for d orbits S : l=1 for p orbits. There are totally 11 orbits. I did this in abinit, the wannier functions obtained look good. Best, JinLuo ----------------------------------------------------------- JinLuo Cheng B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium Email: jinluocheng.phys at gmail.com, jcheng at b-phot.org ----------------------------------------------------------- On 2014?03?27? 13:10, Xi Zhu wrote: > Dear Member; > > Could anyone give some suggestion on how to choose the "begin > projection" part in MoS2 single layer? > > Thanks ! > > > XI > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier -------------- next part -------------- An HTML attachment was scrubbed... URL: From ustc0200 at gmail.com Fri Mar 28 12:31:45 2014 From: ustc0200 at gmail.com (Xi Zhu) Date: Fri, 28 Mar 2014 19:31:45 +0800 Subject: [Wannier] projection choosen for MoS2 Message-ID: Dear JinLuo Thanks for the reply ! I have tested, the DISENTANGLE part converged very well, however, the WANNIERISE part cannot converge.... 2000 -0.225E-01 6.7455453610 1839.0573668511 444.86 <-- CONV O_D= 1587.08767851587.0876785 O_OD= 176.1668017 O_TOT= 1839.0573669 <-- SPRD Delta: O_D= -0.28779050.2877905E-01 O_OD= 0.6316350E-02 O_TOT= -0.2246270E-01 <-- DLTA can you provide any suggestion? Thanks very much ! Call Send SMS Add to Skype You'll need Skype CreditFree via Skype 2014-03-27 20:38 GMT+08:00 JinLuo Cheng : > Dear Xi, > I think the begin projection for highest 6 valence bands and > lowest 5 conduction bands maybe taken as > Mo:l=2 for d orbits > S : l=1 for p orbits. > There are totally 11 orbits. I did this in abinit, the wannier functions > obtained look good. > > Best, > JinLuo > > ----------------------------------------------------------- > JinLuo Cheng > > B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium > > Email: jinluocheng.phys at gmail.com, jcheng at b-phot.org > ----------------------------------------------------------- > > On 2014?03?27? 13:10, Xi Zhu wrote: > > Dear Member; > > Could anyone give some suggestion on how to choose the "begin projection" > part in MoS2 single layer? > > Thanks ! > > > XI > > > _______________________________________________ > Wannier mailing listWannier at quantum-espresso.orghttp://www.democritos.it/mailman/listinfo/wannier > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From szaboa at iis.ee.ethz.ch Fri Mar 28 13:28:06 2014 From: szaboa at iis.ee.ethz.ch (Aron Szabo) Date: Fri, 28 Mar 2014 13:28:06 +0100 Subject: [Wannier] projection choosen for MoS2 In-Reply-To: References: Message-ID: <53356AD6.1030407@iis.ee.ethz.ch> Dear Xi, If you use only the 7 highest valence bands, and 4 lowest conduction bands (by setting the exclude_bands option), then you don't need to do the disentangle part at all, because they are separated from the higher and lower bands. Check the band structure you got from your DFT calculation, to see if it looks correct, and set either the bands you want to use, or the energy window accordingly. You are probably using different bands now. Also, make sure that the wavefunction you are using was fully converged when creating the projections and overlaps. If everything goes fine, with the Mo:d and S:p initial projections all the 11 spreads should be around 1-2 Ang^2 even after the first step. Best, Aron Szabo (PhD student at IIS, ETH Zurich) On 03/28/2014 12:31 PM, Xi Zhu wrote: > Dear JinLuo > > Thanks for the reply ! > > I have tested, the DISENTANGLE part converged very well, however, the > WANNIERISE part cannot converge.... > > > 2000 -0.225E-01 6.7455453610 1839.0573668511 444.86 <-- > CONV > O_D= 1587.08767851587.0876785O_OD= 176.1668017 O_TOT= > 1839.0573669 <-- SPRD > Delta: O_D= -0.28779050.2877905E-01 O_OD= 0.6316350E-02 O_TOT= > -0.2246270E-01 <-- DLTA > > > can you provide any suggestion? > > Thanks very much ! > > > > Call > Send SMS > Add to Skype > You'll need Skype CreditFree via Skype > > > 2014-03-27 20:38 GMT+08:00 JinLuo Cheng >: > > Dear Xi, > I think the begin projection for highest 6 valence bands > and lowest 5 conduction bands maybe taken as > Mo:l=2 for d orbits > S : l=1 for p orbits. > There are totally 11 orbits. I did this in abinit, the wannier > functions obtained look good. > > Best, > JinLuo > > ----------------------------------------------------------- > JinLuo Cheng > > B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium > > Email:jinluocheng.phys at gmail.com ,jcheng at b-phot.org > ----------------------------------------------------------- > > On 2014?03?27? 13:10, Xi Zhu wrote: >> Dear Member; >> >> Could anyone give some suggestion on how to choose the "begin >> projection" part in MoS2 single layer? >> >> Thanks ! >> >> >> XI >> >> >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://www.democritos.it/mailman/listinfo/wannier > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier -------------- next part -------------- An HTML attachment was scrubbed... URL: From jinluocheng.phys at gmail.com Fri Mar 28 14:22:25 2014 From: jinluocheng.phys at gmail.com (JinLuo Cheng) Date: Fri, 28 Mar 2014 14:22:25 +0100 Subject: [Wannier] projection choosen for MoS2 In-Reply-To: <53356AD6.1030407@iis.ee.ethz.ch> References: <53356AD6.1030407@iis.ee.ethz.ch> Message-ID: <53357791.6060405@gmail.com> Dear Xi, Yes, Aron is right. It should be 7 highest valence bands and 4 lowest conduction bands. His suggestions are very likely the reason why you got those big spreads after 2000 steps (usually my max iter step is set as 200). Beside, if you are also using ABINIT, you need to turn on the option w90iniprj 2 and also turn on the guild centre option in Wannier90 : guiding_centres = TRUE to use the beginning projection. Best, JinLuo ----------------------------------------------------------- JinLuo Cheng B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium Email: jinluocheng.phys at gmail.com, jcheng at b-phot.org ----------------------------------------------------------- On 2014?03?28? 13:28, Aron Szabo wrote: > Dear Xi, > > If you use only the 7 highest valence bands, and 4 lowest conduction > bands (by setting the exclude_bands option), then you don't need to do > the disentangle part at all, because they are separated from the > higher and lower bands. Check the band structure you got from your DFT > calculation, to see if it looks correct, and set either the bands you > want to use, or the energy window accordingly. You are probably using > different bands now. Also, make sure that the wavefunction you are > using was fully converged when creating the projections and overlaps. > If everything goes fine, with the Mo:d and S:p initial projections all > the 11 spreads should be around 1-2 Ang^2 even after the first step. > > Best, > Aron Szabo > > (PhD student at IIS, ETH Zurich) > > On 03/28/2014 12:31 PM, Xi Zhu wrote: >> Dear JinLuo >> >> Thanks for the reply ! >> >> I have tested, the DISENTANGLE part converged very well, however, >> the WANNIERISE part cannot converge.... >> >> >> 2000 -0.225E-01 6.7455453610 1839.0573668511 444.86 >> <-- CONV >> O_D= 1587.08767851587.0876785O_OD= 176.1668017 O_TOT= >> 1839.0573669 <-- SPRD >> Delta: O_D= -0.28779050.2877905E-01 O_OD= 0.6316350E-02 O_TOT= >> -0.2246270E-01 <-- DLTA >> >> >> can you provide any suggestion? >> >> Thanks very much ! >> >> >> >> Call >> Send SMS >> Add to Skype >> You'll need Skype CreditFree via Skype >> >> >> 2014-03-27 20:38 GMT+08:00 JinLuo Cheng > >: >> >> Dear Xi, >> I think the begin projection for highest 6 valence bands >> and lowest 5 conduction bands maybe taken as >> Mo:l=2 for d orbits >> S : l=1 for p orbits. >> There are totally 11 orbits. I did this in abinit, the wannier >> functions obtained look good. >> >> Best, >> JinLuo >> >> ----------------------------------------------------------- >> JinLuo Cheng >> >> B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium >> >> Email:jinluocheng.phys at gmail.com ,jcheng at b-phot.org >> ----------------------------------------------------------- >> >> On 2014?03?27? 13:10, Xi Zhu wrote: >>> Dear Member; >>> >>> Could anyone give some suggestion on how to choose the "begin >>> projection" part in MoS2 single layer? >>> >>> Thanks ! >>> >>> >>> XI >>> >>> >>> _______________________________________________ >>> Wannier mailing list >>> Wannier at quantum-espresso.org >>> http://www.democritos.it/mailman/listinfo/wannier >> >> >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://www.democritos.it/mailman/listinfo/wannier >> >> >> >> >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://www.democritos.it/mailman/listinfo/wannier > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier -------------- next part -------------- An HTML attachment was scrubbed... URL: From ustc0200 at gmail.com Sat Mar 29 13:58:09 2014 From: ustc0200 at gmail.com (Xi Zhu) Date: Sat, 29 Mar 2014 20:58:09 +0800 Subject: [Wannier] projection choosen for MoS2 In-Reply-To: <53357791.6060405@gmail.com> References: <53356AD6.1030407@iis.ee.ethz.ch> <53357791.6060405@gmail.com> Message-ID: Thanks for your help, JingLuo and Aron, the problem is solved. Have a nice weekend. Xi 2014-03-28 21:22 GMT+08:00 JinLuo Cheng : > Dear Xi, > > Yes, Aron is right. It should be 7 highest valence bands and 4 lowest > conduction bands. His suggestions are very likely the reason why you got > those big spreads after 2000 steps (usually my max iter step is set as > 200). > > Beside, if you are also using ABINIT, you need to turn on the option > w90iniprj 2 > and also turn on the guild centre option in Wannier90 : > guiding_centres = TRUE > to use the beginning projection. > > > Best, > JinLuo > > ----------------------------------------------------------- > JinLuo Cheng > > B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium > > Email: jinluocheng.phys at gmail.com, jcheng at b-phot.org > ----------------------------------------------------------- > > On 2014?03?28? 13:28, Aron Szabo wrote: > > Dear Xi, > > If you use only the 7 highest valence bands, and 4 lowest conduction bands > (by setting the exclude_bands option), then you don't need to do the > disentangle part at all, because they are separated from the higher and > lower bands. Check the band structure you got from your DFT calculation, to > see if it looks correct, and set either the bands you want to use, or the > energy window accordingly. You are probably using different bands now. > Also, make sure that the wavefunction you are using was fully converged > when creating the projections and overlaps. If everything goes fine, with > the Mo:d and S:p initial projections all the 11 spreads should be around > 1-2 Ang^2 even after the first step. > > Best, > Aron Szabo > > (PhD student at IIS, ETH Zurich) > > On 03/28/2014 12:31 PM, Xi Zhu wrote: > > Dear JinLuo > > Thanks for the reply ! > > I have tested, the DISENTANGLE part converged very well, however, the > WANNIERISE part cannot converge.... > > > 2000 -0.225E-01 6.7455453610 1839.0573668511 444.86 <-- > CONV > O_D= 1587.08767851587.0876785 O_OD= 176.1668017 O_TOT= > 1839.0573669 <-- SPRD > Delta: O_D= -0.28779050.2877905E-01 O_OD= 0.6316350E-02 O_TOT= > -0.2246270E-01 <-- DLTA > > > can you provide any suggestion? > > Thanks very much ! > > > > Call > Send SMS > Add to Skype > You'll need Skype CreditFree via Skype > > > 2014-03-27 20:38 GMT+08:00 JinLuo Cheng : > >> Dear Xi, >> I think the begin projection for highest 6 valence bands and >> lowest 5 conduction bands maybe taken as >> Mo:l=2 for d orbits >> S : l=1 for p orbits. >> There are totally 11 orbits. I did this in abinit, the wannier functions >> obtained look good. >> >> Best, >> JinLuo >> >> ----------------------------------------------------------- >> JinLuo Cheng >> >> B-Phot, Floor 9, Building F Pleinlaan 2, 1050 Elsene, Belgium >> >> Email: jinluocheng.phys at gmail.com, jcheng at b-phot.org >> ----------------------------------------------------------- >> >> On 2014?03?27? 13:10, Xi Zhu wrote: >> >> Dear Member; >> >> Could anyone give some suggestion on how to choose the "begin projection" >> part in MoS2 single layer? >> >> Thanks ! >> >> >> XI >> >> >> _______________________________________________ >> Wannier mailing listWannier at quantum-espresso.orghttp://www.democritos.it/mailman/listinfo/wannier >> >> >> >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://www.democritos.it/mailman/listinfo/wannier >> >> > > > _______________________________________________ > Wannier mailing listWannier at quantum-espresso.orghttp://www.democritos.it/mailman/listinfo/wannier > > > > > _______________________________________________ > Wannier mailing listWannier at quantum-espresso.orghttp://www.democritos.it/mailman/listinfo/wannier > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > > -------------- next part -------------- An HTML attachment was scrubbed... URL: