[Wien] MAE of Co hcp
Dear Prof. Peter Blaha, Thank you for your suggestion about calculating MAE. I adopt it and successfully create the case.struct file using P1 symmetry. But when I used the w2web to initialize the calculation of a cubic system, the kgen program produces only 2052 k-point since I specified the divisions of the reciprocal unit-cell vectors, 16x16x16 without shift the k-mesh and not adding inversion symmetry (for clicking the complex calculation in the session information). So why there is only almost half of the k-point? Or I must manually create the klist file with command x kgen -so? Thank you in advance! Best regards, On Fri, Oct 15, 2010 at 8:01 PM, Peter Blaha pblaha at theochem.tuwien.ac.atwrote: No, you are NOT using P1 symmetry (but you have 24 symops.) As I said, you have to find a common symmetry for all your cases. Your struct file definitely does not do this. If you are unable to manage this by yourself, eventually you may have to use a trick like adding a 3rd atom at some arbitrary position, so that the WIEN2k initialization programs really find P1 symmetry with only ONE symop. (or the common one). Stop the initialization after kgen. After you have this, remove the 3rd atom from struct and in* files and run lstart and dstart to complete the initialization. The runsp save_lapw The do the spinorbit+lapw2 for the two directions. PS: Please note: As far as I remember, the MAE of Co is only a few mycroRy, i.e. one had to use enormous k-meshes to get anything meaningful (and LDA/GGA is wrong anyway ?). Check literature. Am 13.10.2010 10:27, schrieb Bin Shao: Dear Prof. Peter Blaha, With your suggestion, I recalculate the hcp Co in P1 symmetry with experiment parameters, and the k-mesh is 40x40x24. Then I get the energy in (001), (100) and (010) direction. But I find that the energy in (100) is large than (001) and that in (010) is smaller than (001). In my opinion, they should be the same in (100) and in (010) direction. Please give me some comments, and the attachment is my struct file. Thank you in advance! Best, On Tue, Oct 12, 2010 at 2:46 PM, Peter Blaha pblaha at theochem.tuwien.ac.at mailto:pblaha at theochem.tuwien.ac.at wrote: If you change the k-mesh, you need to recalculate also lapw1. However, the proper approach is to find a symmetry (usually the lower one), which accommodates both directions of the magnetization (usually you can run also the higher symmetry case with the low symmetry). Run already the non-SO calculation in this symmetry and then use only lapwso/lapw2 with the two directions in case.inso. Am 12.10.2010 05:24, schrieb Bin Shao: Dear all, I intend to calculate the MAE of hcp Co with force theorem. After the nosoc scf calculation, I add the soc non-selfconsistently with the directions of M || c and M || a, respectively. I use the initso_lapw to prepare the input files which creats new structs and new klists for spin-polarized case. Then I run the program by commands x lapwso -up -p x lapw2 -c -up -p x lapw2 -c -dn -p But here comes some errors in lapw2, dnlapw2.error 'FERMI' - number of k-points inconsistent when reading kgen 'FERMI' - check IN1 and KGEN files! ** testerror: Error in Parallel LAPW2 Before Co hcp, I calculated the case of Fe monolayer with the same approach, but no errors. How to deal with the problem or do I need to provide some more input files? Any suggestion will be appriciated, thank you in advance! Best regards, -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com mailto:binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at mailto: Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.at mailto:blaha at theochem.tuwien.ac.at WWW: http://info.tuwien.ac.at/theochem/ -- ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at mailto: Wien at
[Wien] MAE of Co hcp
Thank you very much for your suggestions! On Fri, Oct 15, 2010 at 8:01 PM, Peter Blaha pblaha at theochem.tuwien.ac.atwrote: No, you are NOT using P1 symmetry (but you have 24 symops.) As I said, you have to find a common symmetry for all your cases. Your struct file definitely does not do this. If you are unable to manage this by yourself, eventually you may have to use a trick like adding a 3rd atom at some arbitrary position, so that the WIEN2k initialization programs really find P1 symmetry with only ONE symop. (or the common one). Stop the initialization after kgen. After you have this, remove the 3rd atom from struct and in* files and run lstart and dstart to complete the initialization. The runsp save_lapw The do the spinorbit+lapw2 for the two directions. PS: Please note: As far as I remember, the MAE of Co is only a few mycroRy, i.e. one had to use enormous k-meshes to get anything meaningful (and LDA/GGA is wrong anyway ?). Check literature. Am 13.10.2010 10:27, schrieb Bin Shao: Dear Prof. Peter Blaha, With your suggestion, I recalculate the hcp Co in P1 symmetry with experiment parameters, and the k-mesh is 40x40x24. Then I get the energy in (001), (100) and (010) direction. But I find that the energy in (100) is large than (001) and that in (010) is smaller than (001). In my opinion, they should be the same in (100) and in (010) direction. Please give me some comments, and the attachment is my struct file. Thank you in advance! Best, On Tue, Oct 12, 2010 at 2:46 PM, Peter Blaha pblaha at theochem.tuwien.ac.at mailto:pblaha at theochem.tuwien.ac.at wrote: If you change the k-mesh, you need to recalculate also lapw1. However, the proper approach is to find a symmetry (usually the lower one), which accommodates both directions of the magnetization (usually you can run also the higher symmetry case with the low symmetry). Run already the non-SO calculation in this symmetry and then use only lapwso/lapw2 with the two directions in case.inso. Am 12.10.2010 05:24, schrieb Bin Shao: Dear all, I intend to calculate the MAE of hcp Co with force theorem. After the nosoc scf calculation, I add the soc non-selfconsistently with the directions of M || c and M || a, respectively. I use the initso_lapw to prepare the input files which creats new structs and new klists for spin-polarized case. Then I run the program by commands x lapwso -up -p x lapw2 -c -up -p x lapw2 -c -dn -p But here comes some errors in lapw2, dnlapw2.error 'FERMI' - number of k-points inconsistent when reading kgen 'FERMI' - check IN1 and KGEN files! ** testerror: Error in Parallel LAPW2 Before Co hcp, I calculated the case of Fe monolayer with the same approach, but no errors. How to deal with the problem or do I need to provide some more input files? Any suggestion will be appriciated, thank you in advance! Best regards, -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com mailto:binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at mailto: Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.at mailto:blaha at theochem.tuwien.ac.at WWW: http://info.tuwien.ac.at/theochem/ -- ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at mailto: Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA,
[Wien] MAE of Co hcp
No, you are NOT using P1 symmetry (but you have 24 symops.) As I said, you have to find a common symmetry for all your cases. Your struct file definitely does not do this. If you are unable to manage this by yourself, eventually you may have to use a trick like adding a 3rd atom at some arbitrary position, so that the WIEN2k initialization programs really find P1 symmetry with only ONE symop. (or the common one). Stop the initialization after kgen. After you have this, remove the 3rd atom from struct and in* files and run lstart and dstart to complete the initialization. The runsp save_lapw The do the spinorbit+lapw2 for the two directions. PS: Please note: As far as I remember, the MAE of Co is only a few mycroRy, i.e. one had to use enormous k-meshes to get anything meaningful (and LDA/GGA is wrong anyway ?). Check literature. Am 13.10.2010 10:27, schrieb Bin Shao: Dear Prof. Peter Blaha, With your suggestion, I recalculate the hcp Co in P1 symmetry with experiment parameters, and the k-mesh is 40x40x24. Then I get the energy in (001), (100) and (010) direction. But I find that the energy in (100) is large than (001) and that in (010) is smaller than (001). In my opinion, they should be the same in (100) and in (010) direction. Please give me some comments, and the attachment is my struct file. Thank you in advance! Best, On Tue, Oct 12, 2010 at 2:46 PM, Peter Blaha pblaha at theochem.tuwien.ac.at mailto:pblaha at theochem.tuwien.ac.at wrote: If you change the k-mesh, you need to recalculate also lapw1. However, the proper approach is to find a symmetry (usually the lower one), which accommodates both directions of the magnetization (usually you can run also the higher symmetry case with the low symmetry). Run already the non-SO calculation in this symmetry and then use only lapwso/lapw2 with the two directions in case.inso. Am 12.10.2010 05:24, schrieb Bin Shao: Dear all, I intend to calculate the MAE of hcp Co with force theorem. After the nosoc scf calculation, I add the soc non-selfconsistently with the directions of M || c and M || a, respectively. I use the initso_lapw to prepare the input files which creats new structs and new klists for spin-polarized case. Then I run the program by commands x lapwso -up -p x lapw2 -c -up -p x lapw2 -c -dn -p But here comes some errors in lapw2, dnlapw2.error 'FERMI' - number of k-points inconsistent when reading kgen 'FERMI' - check IN1 and KGEN files! ** testerror: Error in Parallel LAPW2 Before Co hcp, I calculated the case of Fe monolayer with the same approach, but no errors. How to deal with the problem or do I need to provide some more input files? Any suggestion will be appriciated, thank you in advance! Best regards, -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com mailto:binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at mailto:Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.at mailto:blaha at theochem.tuwien.ac.atWWW: http://info.tuwien.ac.at/theochem/ -- ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at mailto:Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671
[Wien] MAE of Co hcp
Dear Prof. Peter Blaha, With your suggestion, I recalculate the hcp Co in P1 symmetry with experiment parameters, and the k-mesh is 40x40x24. Then I get the energy in (001), (100) and (010) direction. But I find that the energy in (100) is large than (001) and that in (010) is smaller than (001). In my opinion, they should be the same in (100) and in (010) direction. Please give me some comments, and the attachment is my struct file. Thank you in advance! Best, On Tue, Oct 12, 2010 at 2:46 PM, Peter Blaha pblaha at theochem.tuwien.ac.atwrote: If you change the k-mesh, you need to recalculate also lapw1. However, the proper approach is to find a symmetry (usually the lower one), which accommodates both directions of the magnetization (usually you can run also the higher symmetry case with the low symmetry). Run already the non-SO calculation in this symmetry and then use only lapwso/lapw2 with the two directions in case.inso. Am 12.10.2010 05:24, schrieb Bin Shao: Dear all, I intend to calculate the MAE of hcp Co with force theorem. After the nosoc scf calculation, I add the soc non-selfconsistently with the directions of M || c and M || a, respectively. I use the initso_lapw to prepare the input files which creats new structs and new klists for spin-polarized case. Then I run the program by commands x lapwso -up -p x lapw2 -c -up -p x lapw2 -c -dn -p But here comes some errors in lapw2, dnlapw2.error 'FERMI' - number of k-points inconsistent when reading kgen 'FERMI' - check IN1 and KGEN files! ** testerror: Error in Parallel LAPW2 Before Co hcp, I calculated the case of Fe monolayer with the same approach, but no errors. How to deal with the problem or do I need to provide some more input files? Any suggestion will be appriciated, thank you in advance! Best regards, -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.atWWW: http://info.tuwien.ac.at/theochem/ -- ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com -- next part -- An HTML attachment was scrubbed... URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20101013/d288d967/attachment.htm -- next part -- A non-text attachment was scrubbed... Name: nosoc.struct Type: application/octet-stream Size: 1753 bytes Desc: not available URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20101013/d288d967/attachment.dll
[Wien] MAE of Co hcp
If you change the k-mesh, you need to recalculate also lapw1. However, the proper approach is to find a symmetry (usually the lower one), which accommodates both directions of the magnetization (usually you can run also the higher symmetry case with the low symmetry). Run already the non-SO calculation in this symmetry and then use only lapwso/lapw2 with the two directions in case.inso. Am 12.10.2010 05:24, schrieb Bin Shao: Dear all, I intend to calculate the MAE of hcp Co with force theorem. After the nosoc scf calculation, I add the soc non-selfconsistently with the directions of M || c and M || a, respectively. I use the initso_lapw to prepare the input files which creats new structs and new klists for spin-polarized case. Then I run the program by commands x lapwso -up -p x lapw2 -c -up -p x lapw2 -c -dn -p But here comes some errors in lapw2, dnlapw2.error 'FERMI' - number of k-points inconsistent when reading kgen 'FERMI' - check IN1 and KGEN files! ** testerror: Error in Parallel LAPW2 Before Co hcp, I calculated the case of Fe monolayer with the same approach, but no errors. How to deal with the problem or do I need to provide some more input files? Any suggestion will be appriciated, thank you in advance! Best regards, -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.atWWW: http://info.tuwien.ac.at/theochem/ --
[Wien] MAE of Co hcp
Dear Prof. Peter Blaha, Thank you for your suggestion. So if I use the space group P1 for all the cases, it will be ok just for more cpu time. On Tue, Oct 12, 2010 at 2:46 PM, Peter Blaha pblaha at theochem.tuwien.ac.atwrote: If you change the k-mesh, you need to recalculate also lapw1. However, the proper approach is to find a symmetry (usually the lower one), which accommodates both directions of the magnetization (usually you can run also the higher symmetry case with the low symmetry). Run already the non-SO calculation in this symmetry and then use only lapwso/lapw2 with the two directions in case.inso. Am 12.10.2010 05:24, schrieb Bin Shao: Dear all, I intend to calculate the MAE of hcp Co with force theorem. After the nosoc scf calculation, I add the soc non-selfconsistently with the directions of M || c and M || a, respectively. I use the initso_lapw to prepare the input files which creats new structs and new klists for spin-polarized case. Then I run the program by commands x lapwso -up -p x lapw2 -c -up -p x lapw2 -c -dn -p But here comes some errors in lapw2, dnlapw2.error 'FERMI' - number of k-points inconsistent when reading kgen 'FERMI' - check IN1 and KGEN files! ** testerror: Error in Parallel LAPW2 Before Co hcp, I calculated the case of Fe monolayer with the same approach, but no errors. How to deal with the problem or do I need to provide some more input files? Any suggestion will be appriciated, thank you in advance! Best regards, -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com mailto:binshao1118 at gmail.com ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.atWWW: http://info.tuwien.ac.at/theochem/ -- ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Bin Shao, Ph.D. Candidate College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: binshao1118 at gmail.com -- next part -- An HTML attachment was scrubbed... URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20101012/ee92a30b/attachment.htm
