Dear Prof. Blaha,I changed the version from wien0803 into wien0902, and obtained the same strange value. I believethere is a little bug in qtl. It can be justified from the total d electrons variation and the change ofMMTOT before and after rotation.By detailed investigation, i found that the DOS plot shows a good consistency with the rotation features,where the orbital D-Z2 keeps the same as old one, orbitals D-X2Y2 and D-XY is exchanged. And i alsofound that the qtl problem comes from the inappropriate fermi energy in spin-polarized calculation. Wouldyou please suggest me a method to get the right QTL value by the right fermi energy?Thanks for your attention.Best regards,Xueli Wang
From: wxlh...@msn.com To: wien at zeus.theochem.tuwien.ac.at Date: Thu, 22 Oct 2009 22:47:22 +0800 Subject: Re: [Wien] Local Rotation Problems Dear Prof. Blaha, Thanks for your reply. I did the spin-polarized and non-spin-polarized calculations with totally independent process from initialization in different directories, and ran spin-up and dn separately. I am also sure that i ran x lapw2 -up -qtl but not x lapw2 -qtl. I did try these for several examples but shows the similar results. Best regards, Xueli Wang > Date: Thu, 22 Oct 2009 15:04:21 +0200 > From: pblaha at theochem.tuwien.ac.at > To: wien at zeus.theochem.tuwien.ac.at > Subject: Re: [Wien] Local Rotation Problems > > I don't know what you are doing wrong. > > There is "no difference" between a spin-polarized and non-spin-polarized ca! lculation, except > that the effort doubles (you have to do it for spin-up and dn separately and > independently!!!) > and the normalization changes from 2 to 1. > > Did you mix spin-polarized and non-spinpolarized calculations in ONE > directory ??? > > Are you sure you ran x lapw2 -up -qtl ??? It looks as if you ran x > lapw2 -qtl > which gived the different normalization. > > Anyway, I'd try the qtl program for this purpose. > > > Xueli Wang schrieb: > > Dear Dear Wien2k users and dear Prof. Blaha, > > > > I have tried to rotation x,y by 45 degree around the z axis for a > > transition metal ion in an octahedron according > > to your direction in previous discussion: > > 1. NSYM=1 with identity in the first operation; > > 2. x kgen with full k-mesh; > > 3. x lapw1 -up/dn; > > 4. change Local Rotation Martrix; > > 5. x ! lapw2 -qtl -up/dn. > > > > It is okey for every non e spin polarized calculation or MMTOT=0, but problem happened as we execute the > > sp calculation and even more problem for the none-zero MMTOT: > > 1. It can only be succeed without parallel calculation, such as we could > > use! > > x lapw1 -up/dn, x lapw2 -qtl -up/dn, > > but not x lapw1 -up/dn -p, x lapw2 -qtl -up/dn -p; > > 2. As we check the QTL results after rotation in sp calculation, we found > > that the quantities is about 2 times of > > the "right" value; > > 3. We think that the local rotation can not change the charge density of > > the compound, so the x,y rotated by > > 45 degree will show an EXCHANGE in QTL between orbital dXY and dX2Y2, dZ2 > > should keep the SAME value > > as before, and the charge density will projected to the new orbital > > direction of dXZ, dYZ. The results of MMTOT=0 > > coincide with what we analysis, but not for none-zero MMTOT. > > I tried the same r! otation in TiO2 for different MMTOT, > > it shows the same problem. How could we explain this calculation? Is there > > any error in our rotation? > > > > Below is the QTL results for Ti atom in TiO2 before and after rotation in > > the two condition of MMTOT=0 and MMTOT=2, > > the old local rotation matrix is > > LOCAL ROT MATRIX: 0.7071068 0.7071068 0.0000000 > > -0.7071068 0.7071068 0.0000000 > > 0.0000000 0.0000000 1.0000000 > > and the new one is > > LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 > > 0.0000000 1.0000000 0.0000000 > > 0.0000000 0.0000000 1.0000000 > > > > S,P,D,F,PX,PY,PZ,D-Z2,D-X2Y2,D-XY,D-XZ,D-YZ > > MMTOT=0 > > Before > > QTL001: 1.0406 2.9865 0.6504 0.0232 0.9942 0.9941 0.9980 0.1061 0.1543 > > 0.0958 0.2! 028 0.0915 > > QTL001: 1.0406 2.9865 0.6504 0.0232 0.9942 0.9 941 0.9980 0.1061 0.1543 0.0958 0.2028 0.0915 > > After > > QTL001: 2.0813 5.9727 1.3008 0.0465 1.9883 1.9883 1.9961 0.2122 0.1916 > > 0.3085 0.2943 0.2943 > > QTL001: 1.0407 2.9864 0.6504 0.0233 0.9942 0.9942 0.9981 0.1061 0.0958 > > 0.1543 0.1472 0.1472 > > > > MMTOT=2 > > Before > > QTL001: 1.0394 2.9814 0.9011 0.0232 0.9924 0.9924 0.9966 0.1547 0.1968 > > 0.2326 0.1751 0.1418 > > QTL001: 1.0380 2.9787 0.5225 0.0209 0.9909 0.9915 0.9964 0.0889 0.1270 > > 0.0650 0.1733 0.0685 > > After > > QTL001: 2.0768 5.9613 1.0590 0.0459 1.9842 1.9842 1.9928 0.1834 0.1348 > > 0.2586 0.2411 0.2411 > > QTL001: 1.0383 2.9812 0.5307 0.0234 0.9923 0.9923 0.9967 0.0900 0.0683 > > 0.1274 0.1225 0.1225 > > > > Thank you in advance, > > Xueli Wang > > > > > > ------------------------------------------------------------------------ > > > > ______! _________________________________________ > > 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.at WWW: > 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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20091028/050cd3fc/attachment-0001.htm>