Re: [Wien] lapwdm not working (anymore)?
Regarding "grep THETA case.outsymso". One gets two angles THETA and PHI. Thats what I wrote: >> grep THETA case.outsymso >> >> and check if theta (the angle between M and z) and phi agree with >> expectations according to the selected M-direction in case.inso. Are the real (a1,a2,a3) and reciprocal (b1,b2,b3) vectors printed out explicitly somewhere in Cartesian coordinates (just for sanity checks)? in many case.output* files. Check out yourself. I understand that for hexagonal lattice a1 is oriented 30deg away from the x axis. It seems that PHI angle is then given with respect the x axis. Yes, the hexagonal x axis and the cartesian one differ by 30 degree. The hexagonal and cartesian y-axis coincide. -- --- Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-158801165300 Email: peter.bl...@tuwien.ac.at WWW: http://www.imc.tuwien.ac.at WIEN2k: http://www.wien2k.at - ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] lapwdm not working (anymore)?
Dear Prof. Blaha, dear All, Regarding "grep THETA case.outsymso". One gets two angles THETA and PHI. Are the real (a1,a2,a3) and reciprocal (b1,b2,b3) vectors printed out explicitly somewhere in Cartesian coordinates (just for sanity checks)? I understand that for hexagonal lattice a1 is oriented 30deg away from the x axis. It seems that PHI angle is then given with respect the x axis. Best, Lukasz On 2024-11-26 17:06, Peter Blaha wrote: Dear WIEN2k users, Unfortunately since WIEN2k_23.1 a severe bug was introduced in SRC_symmetso/angle.f It may lead to a wrong symmetry reduction due to spin-orbit interactions. The bug is active in magnetic (spinpolarized) spin-orbit calculations at least for B and F cubic and orthorhombic CXY, CXZ and CYZ lattices, where the magnetization direction was multiplied by the primitive Bravais matrix and thus e.g the (0 0 1) direction was moved into (1 1 1) or (1 1 0), resulting into wrong Euler angles theta and phi and thus to wrong symmetry. It should NOT be active for P, H and R lattices. If you have done SO calculations with WIEN2k_23 or 24, please check grep THETA case.outsymso and check if theta (the angle between M and z) and phi agree with expectations according to the selected M-direction in case.inso. Please replace the attached angle.f subroutine in SRC_symmetso and recompile. Best regards Peter Blaha PS: Special thanks to Stefaan, who reported the problem of SO calculations. Am 26.11.2024 um 16:22 schrieb Peter Blaha: initso_lapw should reduce the symmetry operations to 16, not 12 (if M= 0 0 1). It makes cubic --> tetragonal Unfortunately, a quick test shows that in *outputsymso theta and phi is wrong and thus it reduces to a wrong symmetry. I'll debug it as quick as possible. Peter Am 26.11.2024 um 16:03 schrieb Stefaan Cottenier via Wien: Dear WIEN2k community, I am struggling with lapwdm, for calculating the orbital magnetic moment. This feature worked fine many years ago, but I am not able to get it working right now. Tests are done with WIEN2k 23.2, not with the current version (not available on the resources to which I currently have access). But I guess this does not matter for this test. bcc Fe is the test case: blebleble s-o calc. M|| 0.00 0.00 1.00 B 1 229 RELA 5.410352 5.410352 5.410352 90.00 90.00 90.00 ATOM -1: X=0. Y=0. Z=0. MULT= 1 ISPLIT=-2 Fe1 NPT= 781 R0=.5 RMT= 2.19000 Z: 26.0 LOCAL ROT MATRIX: 1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 0 NUMBER OF SYMMETRY OPERATIONS Initialization: init_lapw -prec 1 -rkmax 7.5 -numk 8000 init_so_lapw In the latter step, all defaults were accepted and symmetso was allowed to run. It reduces the number of symmetry operations from 48 to 12. There are several ‘warnings’ in case.outsymso, but I guess these are normal and indicate the 36 symmetry operations that are eliminated. The case is run by: runsp -so -cc 0.1 and converges in 12 iterations, without problems. I then prepare the following case.indmc file: -12. Emin cutoff energy 1 number of atoms for which density matrix is calculated 1 4 0 1 2 3 index of 1st atom, number of L's, L1 1 3 r-index, (l,s)index This should give the orbital contribution to the orbital moment for the s, p, d and f orbitals of the iron atom (not all of them relevant for this element and this property, I know). When running lapwdm, there is an error message in stdout : x lapwdm -c -so -up Error: check case.outputdmup, symmetry might be wrong The output file case.scfdmup has only a single line: Spin-polarized + s-o calculation, M|| 0.000 0.000 1.000 And the output file case.outputdmup terminates with an extra line (which is presumably an error message) after heaving dealt with the second symmetry operation: 2 Euler angles: a,b,c: 270.0 90.0 0.0 # of operation, phase, det: 2 4.71238896548353 0.000E+000 symm. operation 2 so-det= 0.000E+000 Did I overlook something, has something changed to the procedure (I can’t find any hint for this in the usersguide) or is this feature broken? Thanks, Stefaan ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/ wien@zeus.theochem.tuwien.ac.at/index.html ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/in
Re: [Wien] lapwdm not working (anymore)?
Thanks, Peter - that was very fast...! Stefaan -Oorspronkelijk bericht- Van: Wien Namens Peter Blaha Verzonden: dinsdag 26 november 2024 17:07 Aan: wien@zeus.theochem.tuwien.ac.at Onderwerp: Re: [Wien] lapwdm not working (anymore)? Dear WIEN2k users, Unfortunately since WIEN2k_23.1 a severe bug was introduced in SRC_symmetso/angle.f It may lead to a wrong symmetry reduction due to spin-orbit interactions. The bug is active in magnetic (spinpolarized) spin-orbit calculations at least for B and F cubic and orthorhombic CXY, CXZ and CYZ lattices, where the magnetization direction was multiplied by the primitive Bravais matrix and thus e.g the (0 0 1) direction was moved into (1 1 1) or (1 1 0), resulting into wrong Euler angles theta and phi and thus to wrong symmetry. It should NOT be active for P, H and R lattices. If you have done SO calculations with WIEN2k_23 or 24, please check grep THETA case.outsymso and check if theta (the angle between M and z) and phi agree with expectations according to the selected M-direction in case.inso. Please replace the attached angle.f subroutine in SRC_symmetso and recompile. Best regards Peter Blaha PS: Special thanks to Stefaan, who reported the problem of SO calculations. Am 26.11.2024 um 16:22 schrieb Peter Blaha: > initso_lapw should reduce the symmetry operations to 16, not 12 (if > M= > 0 0 1). It makes cubic --> tetragonal > > Unfortunately, a quick test shows that in *outputsymso theta and > phi is wrong and thus it reduces to a wrong symmetry. > > I'll debug it as quick as possible. > > Peter > > > Am 26.11.2024 um 16:03 schrieb Stefaan Cottenier via Wien: >> Dear WIEN2k community, >> >> I am struggling with lapwdm, for calculating the orbital magnetic >> moment. This feature worked fine many years ago, but I am not able to >> get it working right now. >> >> Tests are done with WIEN2k 23.2, not with the current version (not >> available on the resources to which I currently have access). But I >> guess this does not matter for this test. >> >> bcc Fe is the test case: >> >> bleblebles-o calc. M|| 0.00 0.00 >> 1.00 >> >> B1 229 >> >> RELA >> >>5.410352 5.410352 5.410352 90.00 90.00 90.00 >> >> ATOM -1: X=0. Y=0. Z=0. >> >> MULT= 1 ISPLIT=-2 >> >> Fe1NPT= 781 R0=.5 RMT= 2.19000 Z: 26.0 >> >> LOCAL ROT MATRIX:1.000 0.000 0.000 >> >> 0.000 1.000 0.000 >> >> 0.000 0.000 1.000 >> >> 0 NUMBER OF SYMMETRY OPERATIONS >> >> Initialization: >> >> init_lapw -prec 1 -rkmax 7.5 -numk 8000 >> >> init_so_lapw >> >> In the latter step, all defaults were accepted and symmetso was >> allowed to run. It reduces the number of symmetry operations from 48 >> to 12. There are several 'warnings' in case.outsymso, but I guess >> these are normal and indicate the 36 symmetry operations that are >> eliminated. >> >> The case is run by: >> >> runsp -so -cc 0.1 >> >> and converges in 12 iterations, without problems. >> >> I then prepare the following case.indmc file: >> >> -12. Emin cutoff energy >> >> 1 number of atoms for which density matrix is >> calculated >> >> 1 4 0 1 2 3 index of 1st atom, number of L's, L1 >> >> 1 3 r-index, (l,s)index >> >> This should give the orbital contribution to the orbital moment for >> the s, p, d and f orbitals of the iron atom (not all of them relevant >> for this element and this property, I know). >> >> When running lapwdm, there is an error message in stdout : >> >> x lapwdm -c -so -up >> >> Error: check case.outputdmup, symmetry might be wrong >> >> The output file case.scfdmup has only a single line: >> >> Spin-polarized + s-o calculation, M|| 0.000 0.000 1.000 >> >> And the output file case.outputdmup terminates with an extra line >> (which is presumably an error message) after heaving dealt with the >> second symmetry operation: >> >> 2 Euler angles: a,b,c:270.0 90.00.0 >> >> # of operation, phase, det: >> >> 2 4.71238896548353 0.000E+000 >> >> symm. operation2 so-det= 0.000E+000 >> >> Did I overlook something, has something change
Re: [Wien] lapwdm not working (anymore)?
Dear WIEN2k users, Unfortunately since WIEN2k_23.1 a severe bug was introduced in SRC_symmetso/angle.f It may lead to a wrong symmetry reduction due to spin-orbit interactions. The bug is active in magnetic (spinpolarized) spin-orbit calculations at least for B and F cubic and orthorhombic CXY, CXZ and CYZ lattices, where the magnetization direction was multiplied by the primitive Bravais matrix and thus e.g the (0 0 1) direction was moved into (1 1 1) or (1 1 0), resulting into wrong Euler angles theta and phi and thus to wrong symmetry. It should NOT be active for P, H and R lattices. If you have done SO calculations with WIEN2k_23 or 24, please check grep THETA case.outsymso and check if theta (the angle between M and z) and phi agree with expectations according to the selected M-direction in case.inso. Please replace the attached angle.f subroutine in SRC_symmetso and recompile. Best regards Peter Blaha PS: Special thanks to Stefaan, who reported the problem of SO calculations. Am 26.11.2024 um 16:22 schrieb Peter Blaha: initso_lapw should reduce the symmetry operations to 16, not 12 (if M= 0 0 1). It makes cubic --> tetragonal Unfortunately, a quick test shows that in *outputsymso theta and phi is wrong and thus it reduces to a wrong symmetry. I'll debug it as quick as possible. Peter Am 26.11.2024 um 16:03 schrieb Stefaan Cottenier via Wien: Dear WIEN2k community, I am struggling with lapwdm, for calculating the orbital magnetic moment. This feature worked fine many years ago, but I am not able to get it working right now. Tests are done with WIEN2k 23.2, not with the current version (not available on the resources to which I currently have access). But I guess this does not matter for this test. bcc Fe is the test case: blebleble s-o calc. M|| 0.00 0.00 1.00 B 1 229 RELA 5.410352 5.410352 5.410352 90.00 90.00 90.00 ATOM -1: X=0. Y=0. Z=0. MULT= 1 ISPLIT=-2 Fe1 NPT= 781 R0=.5 RMT= 2.19000 Z: 26.0 LOCAL ROT MATRIX: 1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 0 NUMBER OF SYMMETRY OPERATIONS Initialization: init_lapw -prec 1 -rkmax 7.5 -numk 8000 init_so_lapw In the latter step, all defaults were accepted and symmetso was allowed to run. It reduces the number of symmetry operations from 48 to 12. There are several ‘warnings’ in case.outsymso, but I guess these are normal and indicate the 36 symmetry operations that are eliminated. The case is run by: runsp -so -cc 0.1 and converges in 12 iterations, without problems. I then prepare the following case.indmc file: -12. Emin cutoff energy 1 number of atoms for which density matrix is calculated 1 4 0 1 2 3 index of 1st atom, number of L's, L1 1 3 r-index, (l,s)index This should give the orbital contribution to the orbital moment for the s, p, d and f orbitals of the iron atom (not all of them relevant for this element and this property, I know). When running lapwdm, there is an error message in stdout : x lapwdm -c -so -up Error: check case.outputdmup, symmetry might be wrong The output file case.scfdmup has only a single line: Spin-polarized + s-o calculation, M|| 0.000 0.000 1.000 And the output file case.outputdmup terminates with an extra line (which is presumably an error message) after heaving dealt with the second symmetry operation: 2 Euler angles: a,b,c: 270.0 90.0 0.0 # of operation, phase, det: 2 4.71238896548353 0.000E+000 symm. operation 2 so-det= 0.000E+000 Did I overlook something, has something changed to the procedure (I can’t find any hint for this in the usersguide) or is this feature broken? Thanks, Stefaan ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/ wien@zeus.theochem.tuwien.ac.at/index.html -- --- Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-158801165300 Email: peter.bl...@tuwien.ac.at WWW: http://www.imc.tuwien.ac.at WIEN2k: http://www.wien2k.at - SUBROUTINE ANGLE(XMS,alat,alpha,lattic,theta,phi) IMPLICIT REAL*8 (A-H,O-Z) character*4 lattic logical ortho dimension alat(3),alpha(3),br2(3,3) !*** ! DIMENSION XMS(3),dif(3),help(3) do i=1,3 dif(i)=xms(i) enddo pi=4.d0*atan(1
Re: [Wien] lapwdm not working (anymore)?
initso_lapw should reduce the symmetry operations to 16, not 12 (if M= 0 0 1). It makes cubic --> tetragonal Unfortunately, a quick test shows that in *outputsymso theta and phi is wrong and thus it reduces to a wrong symmetry. I'll debug it as quick as possible. Peter Am 26.11.2024 um 16:03 schrieb Stefaan Cottenier via Wien: Dear WIEN2k community, I am struggling with lapwdm, for calculating the orbital magnetic moment. This feature worked fine many years ago, but I am not able to get it working right now. Tests are done with WIEN2k 23.2, not with the current version (not available on the resources to which I currently have access). But I guess this does not matter for this test. bcc Fe is the test case: blebleble s-o calc. M|| 0.00 0.00 1.00 B 1 229 RELA 5.410352 5.410352 5.410352 90.00 90.00 90.00 ATOM -1: X=0. Y=0. Z=0. MULT= 1 ISPLIT=-2 Fe1 NPT= 781 R0=.5 RMT= 2.19000 Z: 26.0 LOCAL ROT MATRIX: 1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 0 NUMBER OF SYMMETRY OPERATIONS Initialization: init_lapw -prec 1 -rkmax 7.5 -numk 8000 init_so_lapw In the latter step, all defaults were accepted and symmetso was allowed to run. It reduces the number of symmetry operations from 48 to 12. There are several ‘warnings’ in case.outsymso, but I guess these are normal and indicate the 36 symmetry operations that are eliminated. The case is run by: runsp -so -cc 0.1 and converges in 12 iterations, without problems. I then prepare the following case.indmc file: -12. Emin cutoff energy 1 number of atoms for which density matrix is calculated 1 4 0 1 2 3 index of 1st atom, number of L's, L1 1 3 r-index, (l,s)index This should give the orbital contribution to the orbital moment for the s, p, d and f orbitals of the iron atom (not all of them relevant for this element and this property, I know). When running lapwdm, there is an error message in stdout : x lapwdm -c -so -up Error: check case.outputdmup, symmetry might be wrong The output file case.scfdmup has only a single line: Spin-polarized + s-o calculation, M|| 0.000 0.000 1.000 And the output file case.outputdmup terminates with an extra line (which is presumably an error message) after heaving dealt with the second symmetry operation: 2 Euler angles: a,b,c: 270.0 90.0 0.0 # of operation, phase, det: 2 4.71238896548353 0.000E+000 symm. operation 2 so-det= 0.000E+000 Did I overlook something, has something changed to the procedure (I can’t find any hint for this in the usersguide) or is this feature broken? Thanks, Stefaan ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html -- --- Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-158801165300 Email: peter.bl...@tuwien.ac.at WWW: http://www.imc.tuwien.ac.at WIEN2k: http://www.wien2k.at - ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
