Re: [Wien] AFM type II
Dear Gerhard, nice ... thanks a lot for the references! Best regards, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-29 13:16, schrieb Fecher, Gerhard: Dear Martin, this concerns your remark: "With two magnetic species, say, Mn and Cu, you would wind up with different size of the moment on Mn and Cu. I know of no case where exact compensation into an AFM structure occures by accident in such a situation." You may have the situation of a completely compensated ferrimagnet exampels are: CrMnSb (or VFeSb) in the cubic C1b structure H. van Leuken and R. A. de Groot, Phys. Rev. Lett. 74, 1171 (1995) or more complicated Mn1.5FeV0.5Al Rolf Stinshoff et al; Phys. Rev. B 95, 060410(R) (2017) However, it is by purpose rather than by accident. This was already found by Neel in his work on antiferromagnets (probably it is mentioned in the Nobel lecture) Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von pieper [pie...@ifp.tuwien.ac.at] Gesendet: Mittwoch, 29. Januar 2020 12:49 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] AFM type II No one can give you an honest answer without knowing the structure you put these elements in. Zr, S, Se are almost certainely non-magnetic, but there are quite a few structures with magnetic moments on Cr, Cu, and of course on Mn. To make terminology more complicated, remember that AFM means fully compensated magnetic moments. The net magnetization of a unit cell of some AFM structure is zero. This does not happen by coincidence, it is because by symmetry all moments have the same size, and there are as many of them pointing in one direction as there are pointing in exactly the opposite direction. So, IF your compound REALLY is AFM by experiment (NO net magnetization), you almost certainely have only one magnetic species in there (probably Mn). With two magnetic species, say, Mn and Cu, you would wind up with different size of the moment on Mn and Cu. I know of no case where exact compensation into an AFM structure occures by accident in such a situation. You always get something with net moment - and these are called ferrimagnetic structures. And since we are at it: there are canted and helical strucutures where the moments are not collinear (not within the scope of Wien2k), there are spin density waves, ... Scanning this thread my advice would be to study a (good) book on solid state physics, with special attention payed to its chapter discussing magnetic order. If it doesn't have such a chapter its not a good book - at least not for you. Do NOT use wikipedia or this mailing list and its archive as a substitute for such a reading. It will not work. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-27 17:45, schrieb djamel slamnia: WHEN I STUDY A COMPOUND CONTAINS THIS ELEMENETS CU MN CR ZR S SE BETWEEN THEM WITCH ONE TO PUT IT SPIN UP OR DOWN AND NON-MAGNETIC ??? Le lundi 27 janvier 2020 à 14:22:39 UTC+1, Gavin Abo a écrit : As previously mentioned [1], a short literature survey showed that AFM type II and III are terms used for _fcc_ and _bcc_ lattices. Since spacegroup 156 is not one of those, it might be inappropriate to use those terms for spacegroup 156 having a _primitive_ lattice [2] of the hexagonal crystal family. If you do a more extensive literature survey yourself and find a paper (article, book, etc.) that defines the AFM magnetic orders for spacegroup 156, then reference and use it for what the AFM order is. If there is not any notations and terms for AFM magnetic orders for spacegroup 156, you might have to make your own figure or write in your own words what the definition is should any AFM magnetic orders exist for it. Keep in mind that as mentioned before in the mailing list archive, the initial configuration can be set in case.inst with "instgen_lapw -ask" [3]. Though, you need to check the final magnetic order that comes out of the scf [4], because the configuration set in case.inst with instgen_lapw is just the initial one that could change [5,6]. It is also possible to try to force a magnetic order using dmatup/dn matrices but the final magnetic order is still what comes out of the scf and could be different [6-11]. Therefore, it likely not beneficial to name the AFM order before starting a calculation such that you would li
Re: [Wien] AFM type II
No one can give you an honest answer without knowing the structure you put these elements in. Zr, S, Se are almost certainely non-magnetic, but there are quite a few structures with magnetic moments on Cr, Cu, and of course on Mn. To make terminology more complicated, remember that AFM means fully compensated magnetic moments. The net magnetization of a unit cell of some AFM structure is zero. This does not happen by coincidence, it is because by symmetry all moments have the same size, and there are as many of them pointing in one direction as there are pointing in exactly the opposite direction. So, IF your compound REALLY is AFM by experiment (NO net magnetization), you almost certainely have only one magnetic species in there (probably Mn). With two magnetic species, say, Mn and Cu, you would wind up with different size of the moment on Mn and Cu. I know of no case where exact compensation into an AFM structure occures by accident in such a situation. You always get something with net moment - and these are called ferrimagnetic structures. And since we are at it: there are canted and helical strucutures where the moments are not collinear (not within the scope of Wien2k), there are spin density waves, ... Scanning this thread my advice would be to study a (good) book on solid state physics, with special attention payed to its chapter discussing magnetic order. If it doesn't have such a chapter its not a good book - at least not for you. Do NOT use wikipedia or this mailing list and its archive as a substitute for such a reading. It will not work. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-27 17:45, schrieb djamel slamnia: WHEN I STUDY A COMPOUND CONTAINS THIS ELEMENETS CU MN CR ZR S SE BETWEEN THEM WITCH ONE TO PUT IT SPIN UP OR DOWN AND NON-MAGNETIC ??? Le lundi 27 janvier 2020 à 14:22:39 UTC+1, Gavin Abo a écrit : As previously mentioned [1], a short literature survey showed that AFM type II and III are terms used for _fcc_ and _bcc_ lattices. Since spacegroup 156 is not one of those, it might be inappropriate to use those terms for spacegroup 156 having a _primitive_ lattice [2] of the hexagonal crystal family. If you do a more extensive literature survey yourself and find a paper (article, book, etc.) that defines the AFM magnetic orders for spacegroup 156, then reference and use it for what the AFM order is. If there is not any notations and terms for AFM magnetic orders for spacegroup 156, you might have to make your own figure or write in your own words what the definition is should any AFM magnetic orders exist for it. Keep in mind that as mentioned before in the mailing list archive, the initial configuration can be set in case.inst with "instgen_lapw -ask" [3]. Though, you need to check the final magnetic order that comes out of the scf [4], because the configuration set in case.inst with instgen_lapw is just the initial one that could change [5,6]. It is also possible to try to force a magnetic order using dmatup/dn matrices but the final magnetic order is still what comes out of the scf and could be different [6-11]. Therefore, it likely not beneficial to name the AFM order before starting a calculation such that you would likely want to identify the name of the magnetic order after having finished the converged calculation. [1] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg19515.html [2] https://en.wikipedia.org/wiki/Crystal_structure#Lattice_systems [3] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10044.html [4] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg17516.html [5] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg03243.html [6] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg06739.html [7] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14259.html [8] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05054.html [9] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg13124.html [10] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16281.html [11] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16286.html On 1/26/2020 1:01 PM, djamel slamnia wrote: THANKS AGAIN SIR I NEED TO KNOW WHAT IS THE AFM ORDERS FOR P3M1 (156) ??? TYPE II OR III THANKS IN ADVANCE ___ 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/
Re: [Wien] AFM type II
I guess you are talking about the conventional classification of AFM ordering used in neutron diffraction. Consider AFM as a standing wave of spin orientations. Different Types of AFM are then distinguished by the wave vector of the standing wave in the crystal lattice. AFM-I is the AFM order with the shortest possible wavelength, that is moments on nearest neighbor planes are antiparallel. In an fcc structure the nearest neighbors of an atom at (0,0,0) are on the two planes 'above' and 'below' along the space diagonal of the cube at positions (1/2,1/2,0) and so on. The wave vector of the standing wave describing spin orientation is perpendicular to these planes of parallel spins. The length of the wave vector q in reciprocal space is such that going the distance d to the next plane with parallel moments (half the space diagonal) in that direction result in an identical situation in the wave function cos(qx), that is q*d=2*pi. For AFM-II moments on planes with next nearest neighbors are antiparallel. And so on. The longer the wavelength of the standing wave (or the shorter q in reciprocal space) the more unit cells in the crystal lattice you will need to represent the AFM structure (depending on the distance between lattice planes in your structure without AFM order). Good luck with figuring out the directions and lengths of wave vectors in your structure yourself, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-26 21:01, schrieb djamel slamnia: THANKS AGAIN SIR I NEED TO KNOW WHAT IS THE AFM ORDERS FOR P3M1 (156) ??? TYPE II OR III THANKS IN ADVANCE Le dimanche 26 janvier 2020 à 20:51:41 UTC+1, Gavin Abo a écrit : THE DEFINITION FOR THE COMPOUND A=B = 3.74 A ALPHA = BETA = 90 GAMMA = 120 for AFM type I : i creat superstructure x super cell target lattice H : x =1, y = 1 , Z =2 then x sgroup, program define automatically the space group the same of my original space group 156 without warrning As you have described above (for Z=2), your attempt at creating a supercell has failed as "x sgroup" collapsed the supercell structure back to the non-supercell structure. As mentioned on the FAQ page for supercell construction, you need to displace an atom, change an atom, or use a special label: http://susi.theochem.tuwien.ac.at/reg_user/faq/supercells.html In order to keep the supercell without "x sgroup" reducing it back to the original structure, refer to previous posts in the mailing list archive about breaking the symmetry. A few of the many posts about that as examples are at the three links below: https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg18380.html https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg01866.html https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15517.html but when try to do it for type II : x super cell P : x =1, y = 1 , Z =1 x sgroup could not define the space group In section "3.12 Setting up a new case" on page 29 in the WIEN2k 19.1 usersguide [ http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf ], there is the statement: "Alternatively with the new StructGen you can specify the spacegroup and only the inequivalent positions. The equivalent ones will be generated automatically." This means spacegroups in WIEN2k are defined according to the inequivalent positions and not by the equivalent positions. For the case above (Z=1), it is likely that "x supercell" found some equivalent positions in the original structure and automatically added special labels to them changing them into inequivalent positions. Thus, a supercell structure was successfully created. If you want the supercell structure to reduce back to the original structure, you would likely just need to remove all or some of the special labels in StructGen before running "x sgroup". For understanding the inequivalent and equivalent positions with WIEN2k spacegroups, the example in the post at the following link might helpful: http://zeus.theochem.tuwien.ac.at/pipermail/wien/2013-January/018171.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/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/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/index.html
Re: [Wien] density matrix is not positive
If I understand correctly, then the SCF cycle does not wade through some river (to stay in the pedagogic example). It walks along the shore, sometimes with one foot in the water (the density generated by mixer). The other foot (the density generated from the wave functions) always comes down on dry land. This way the SCF can step across cracks in reality that are narrower than the step width. This also explains why one apparently can be rather careless about semi-positiveness of dmats when manipulating them to stabilize certain solutions with the -orb option, correct? I can appreciate that this has advantages if there are a lot of rivers. I don't understand why this is always better than going for unitary transformations only, because it does add a lot of definitely wrong directions to explore. But I am happy to trust the experts. So thanks a lot for the explanations, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-08 16:10, schrieb Laurence Marks: Yet more pedagogy... Over the last few years I have been trying to add to the mixer algorithms so it can sense when their might be piranha around. One of these is large fluctuations of the potential; if these occur it becomes more cautious. Another currently being tested for a future release is taking a step backwards if something was not right. Slowly the mixer is becoming smarter...I hope. _ Professor Laurence Marks "Research is to see what everybody else has seen, and to think what nobody else has thought", Albert Szent-Gyorgi www.numis.northwestern.edu [1] On Wed, Jan 8, 2020, 08:40 Laurence Marks wrote: For pedagogical reasons, let me explain why forcing dmat to positive definite may not be good. The mixer is solving a fixed point problem, which with +U is for both the dmat and density. Currently it can explore all values, including ones which are not physically reasonable. If we limit it to positive definite then there are some forbidden regions. The problem may be slower if the mixer can't explore them. As an illustration, suppose your home is on the other side of a shallow stream. If you demand that your feet don't get wet, you may have to walk a long way to a bridge to get home. If you are OK with getting your feet wet, you just ford the river. Fine, so long as there are no pirhana fish (ghost bands) in the stream. _ Professor Laurence Marks "Research is to see what everybody else has seen, and to think what nobody else has thought", Albert Szent-Gyorgi www.numis.northwestern.edu [1] On Wed, Jan 8, 2020, 07:01 Laurence Marks wrote: Being more specific, the unmixed density matrix (dmat) produced by one scf iteration is necessarily positive definite. The mixer produces a new dmat that is a linear combination of prior dmats used, i.e. dmat_(n+1) = sum_i c_i * dmat_i If some of the coefficients c_i are negative the new density matrix may not be positive definite. A trivial example used two 1x1 matrices M=1 and M=2. As Peter said, at convergence (:MV) the dmat must be positive definite since the input is equal to the output. N.B., one could constrain the dmat to be positive definite. However, mathematically this may converge slower, not faster. _ Professor Laurence Marks "Research is to see what everybody else has seen, and to think what nobody else has thought", Albert Szent-Gyorgi www.numis.northwestern.edu [1] On Wed, Jan 8, 2020, 06:40 Peter Blaha wrote: The "new" (unmixed) dmat is always ok, only through mixing of the individual elements this may happen. When reaching scf, the "mixed" dmat = "unmixed" dmat and thus this should no longer occur. On 1/8/20 1:19 PM, pieper wrote: Sorry for interrupting, but I am intrigued. I was not aware that the density matrix can leave the physically meaningfull realm of semi-positive operators. May I spoil this thread with the question why the mixer produces such non-physical trial states? And if it does, what forces drive rho back to proper density operators during convergence? --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-04 11:50, schrieb Peter Blaha: I repeated your calculations. Yes, intermediately I also got these density matrix warnings, but NOT with final convergence. Most likely, this warning should be removed from mixer (Laurence !), as it does not give any meaningful hint or it should be printed only if one is nearly converged (:DIS is lower than ...) I have a few other comments: You seem to start a Ferro-magnetic calculation. As was mentioned before, maybe an AFM structure is more stable. So check other possible magnetic configurations. You seem to start with init_lapw, and thus with the default (ferromagnetic) case.inst For magnetic oxides it is highly recommended to s
Re: [Wien] density matrix is not positive
Sorry for interrupting, but I am intrigued. I was not aware that the density matrix can leave the physically meaningfull realm of semi-positive operators. May I spoil this thread with the question why the mixer produces such non-physical trial states? And if it does, what forces drive rho back to proper density operators during convergence? --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2020-01-04 11:50, schrieb Peter Blaha: I repeated your calculations. Yes, intermediately I also got these density matrix warnings, but NOT with final convergence. Most likely, this warning should be removed from mixer (Laurence !), as it does not give any meaningful hint or it should be printed only if one is nearly converged (:DIS is lower than ...) I have a few other comments: You seem to start a Ferro-magnetic calculation. As was mentioned before, maybe an AFM structure is more stable. So check other possible magnetic configurations. You seem to start with init_lapw, and thus with the default (ferromagnetic) case.inst For magnetic oxides it is highly recommended to start with: instgen -ask This lets you define a possible AFM arrangement and in particular you should set the O-atoms as non-magnetic. This save a lot of scf cycles. -rkmax 7.5 is quite some "overkill". This rkmax value is for the small O-atoms and RKmax=6 (beginning) and 7 (final convergence) is sufficient. On the other hand, a HDLO for Nd-4f states is highly recommended. I never start the scf-cycle with "-orb" right after intitialization, but always do (at least a crude) GGA calculation before switching on -orb. Otherwise you may even more likely end up in some local minimum but not the ground state. I got a half-metallic solution with a spin-up 4f peak down at -3.5 eV and another one at EF. You expected that the 4f states are all removed from EF, but your compound seems to be "special" (is this compound experimentally known ??). The stoichiometry of Nd4 Ni3 O8 suggests Ni2+ ions, leaving a (Nd4)10+ charge state, i.e. a Nd2.5+ state, which is very unusual. But I got 34f electrons, which hints to a Nd3+ and in fact one of your Oxygens has p-states above EF. With this half-filled 4f-DOS peak at EF also spin-orbit could be very important. And as always in such LDA+U calculations, play with the initial case.dmat file. In my case I have 2 4f orbitals "fully" occupied (and thus shifted down), but 2 others are partly occupied (close to 1/2) and thus not much shifted and pinned to EF. As mentioned before, additional symmetry splitting with AFM for the 2 equivalent Nd atoms may lead to further splitting. Am 03.01.2020 um 10:30 schrieb Fan: Thank you for your reply, Prof. Blaha. In fact, when I change parameters in inorb, I always do 'rm case.dmat*' and 'rm case.vorb*' , so I think it should not be the problem. But I do value your suggestions, so I started a new session, only used the struct file of Nd4Ni3O8 which is attached here, and then init_lapw -b -rkmax 7.5 -numk 2000 -sp -vxc 5 -lvns 6 create inorb and indm by copying them from templates and modify them accordingly runsp_lapw -p -orb -i 200 -ec 0.0001 -cc 0.001 -NI during the scf cycle, QTL-B warning appeared, but the value was very small (2~3, still could be a trouble?), the density matrix warnings appeared at about the 20th cycle and persisted. With regards. Fan ___ 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/index.html
Re: [Wien] x symmetso error
Hi, in your struct file there is no indication that you did initso. It has been said again and again and it is in the UG: SOC may change the symmetry operations applicable for a given crystal structure, depending on the direction of the quantization axis (magnetization). initso will detect this and present you with a new struct file. Accept this file! Without actually trying I suspect that it will do so (lower symmetry) in your structure if you point it along 100, while all symmetry operations will be preserved if you point it along 001. And be carefull to use the same structure file (with lowest symmetry) for ALL calculations where you want compare total energies (e.g. magnetic anisotropy) Best regards --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2019-12-16 10:48, schrieb Riyajul Islam: Dear Wien2k users, I'm using wien2k 19.1 version. I'm trying to include SOC in my calculation after GGA+U was done. I did GGA+U+SOC calculations along 001 magnetization direction without any errors. But while I chose 100 direction, I get an error during x symmetso forrtl: severe (174): SIGSEGV, segmentation fault occurred Image PCRoutineLine Source symmetso 004514E3 Unknown Unknown Unknown libpthread-2.23.s 2AD9F2D14390 Unknown Unknown Unknown libiomp5.so2AD9F355C871 Unknown Unknown Unknown libiomp5.so2AD9F355C5E4 Unknown Unknown Unknown libiomp5.so2AD9F355E4C5 Unknown Unknown Unknown libiomp5.so2AD9F355F89F Unknown Unknown Unknown libiomp5.so2AD9F355411B Unknown Unknown Unknown libiomp5.so2AD9F3554007 Unknown Unknown Unknown libiomp5.so2AD9F3553D52 Unknown Unknown Unknown libiomp5.so2AD9F355454D Unknown Unknown Unknown libiomp5.so2AD9F34DDD61 Unknown Unknown Unknown ld-2.23.so [1] 2AD9F2AEBE15 Unknown Unknown Unknown libc-2.23.so [2] 2AD9F3850FF8 Unknown Unknown Unknown libc-2.23.so [2] 2AD9F3851045 Unknown Unknown Unknown symmetso 0046D1E4 Unknown Unknown Unknown symmetso 0040C191 MAIN__359 symmetso.f symmetso 004044F2 Unknown Unknown Unknown libc-2.23.so [2] 2AD9F3837830 __libc_start_main Unknown Unknown symmetso 004043E9 Unknown Unknown Unknown 0.6u 0.0s 0:00.71 100.0% 0+0k 0+42368io 0pf+0w error: command /home/edison/Wien2k19.1/symmetso upsymmetso.def failed Here I'm also attaching the structure file. Kindly help me solve this problem. Kind regards -- Riyajul Islam Ph.D Scholar National Institute of Technology Nagaland Chumukedima, Dimapur Nagaland 797103, India Links: -- [1] http://ld-2.23.so [2] http://libc-2.23.so ___ 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/index.html
Re: [Wien] In which approach the magnetism is treated in the Wien2k Code ?
Hi, please read the user guide (and, probably, a good book on solid state physics). Wien2k is an all electron code. There is a switch to take into account the spin degree of freedom for the electrons (spin polarized calculations). In such a calculation the spin of each electron is taken into account, regardless of wether its itinerant or localized. Best regards --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2019-12-10 21:46, schrieb Abderrahmane Reggad: Helli wien users I want to know in which approach the magnetism is treated in wien2k code? We know that the Fe metal in bcc structure is ferromagnetic and is considered as having a dual nature itinerant and localized. How is teated this magnetic state in wien2k code? Best regards -- Dr. Abderrahmane Reggad Engineering Physics LaboratoryFaculty of Material Sciences, Ibn Khaldoun University, Tiaret, 14000, AlgeriaTel: +213(0)561861963 - Algeria ___ 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/index.html
Re: [Wien] Spin-polarization and spin-orbit coupling
Reading your questions I get the impression that you miss a crucial point of what and how Density Functional Theory (DFT) does. Therefore I would like to extend on what Peter Blaha said. Maybe it helps to clarify some of his answers. Pauli's Hamiltonian represents the energy of an electron in some electromagnetic field (your external magnetic field). DFT aims to find the ground state of a system of many interacting Fermions (electrons) represented by any Hamiltonian for the interaction you care to throw at it. The mean interaction energy of a given electron spin with all others in the system is the same as its energy in some fictitious internal field. DFT works via a subtle reinterpretation of the role of the charge density, which has far reaching consequences. In 1964 Hohenberg and Kohn proved a fascinating property of the charge density: This single, real valued function of (without spin) one spacial variable uniquely determines the many particle wave function of interacting electrons in an external Coulomb potential (due to, in this case, nuclear charges). Calculation of the spacial charge density from the wave function by an expectation value is not a one way process. If two solutions of Schroedingers equation for the interacting many particle system in a given external Coulomb potential give the same charge density, then the two wave functions with all their N space variables for N electrons are the same. One can, in principle, go back: the charge density determines the wave function - and with it everything, including the energy. Furthermore, the ground state charge density can be determined by a variational principle and its Euler-Lagrange equations (the Kohn-Sham equations). They can be solved very effectively in a self consistent, iterative process: start from some charge densitiy (the closer to the solution the better to avoid local minima). Then calculate the potentials - solve the Kohn-Sham equations - calculate the charge density - compare it to previous ones. If the difference is larger than some threshold, mix up a variation of it and calculate the potentials ... The most simple case is if your favorite interaction Hamiltonian has Coulomb potentials from local charges and no spin contribution. Spin-up and -down charge densities are identical (Local Densitsity Approximation, LDA, for a certain type of electric potential). If there are exchange interaction energies present (products of the spin operators of pairs of interacting electrons) the densities of the two spin directions become different (Local Spin Density Approximation, LSDA, the electric potential stays the same). Such a Heisenberg exchange interaction does, however, not depend on the oriention of interacting spins in the crystal lattice. Only the strength of the equivalent internal field has a meaning, not its direction. In contrast, a spin-orbin coupling in the Hamiltonian does depend on the orientation of the spin moment in the lattice. With the interaction energy depending on spin orientation, so does the equivalent internal field. The system is magnetically anisotropic. You have to specifiy the orientation of the moment to determine the energy of the graound state. There is, of course, a huge number of books an reviews on DFT. The UG cites enough stuff to keep you busy for a long time. Personally, I like a review of A. Becke: THE JOURNAL OF CHEMICAL PHYSICS 140, 18A301 (2014) --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2019-09-24 08:23, schrieb Luigi Maduro - TNW: Dear WIEN2k users, I have three questions concerning the inclusion of spin in a material in WIEN2k. The three questions concern the two terms where a spin-dependent term appears in the Pauli Hamiltonian for magnetic systems, which are: Question 1) In the Pauli Hamiltonian a term appears which is a dot product of the spin-matrices of the system and an effective magnetic field. The effective magnetic field is a summation of an external magnetic field and an exchange-correlation term. The exchange-correlation term B_xc, is expressed as a derivative of the density w.r.t. the magnetization (in the LDA framework) and that B_xc is parallel to the magnetization density vector. If I understand correctly then the material of interest is magnetic when B_xc is nonzero. When doing a spin-polarized calculation, what happens then to the external magnetic field term? Is the external magnetic field term set to zero? Question 2) The other term in the Pauli Hamiltonian is the spin-orbit coupling (SOC) term, which is proportional to (1/r x dV/dr ) (dV/dr = the derivative of the potential w.r.t. the radial coordinate). When doing a calculation including SOC the script init_so asks for the magnetization direction (in hkl). In a non-spin polarized calculation with SOC the magnetization direction has no meaning, is this correct? Question 3) If the system of interest
Re: [Wien] Magnetic moments converging in a different direction to the one they are defined
Fe3O4 being an old but unsatisfied love of mine a few additional comments: Determining exchange constants by spin reversal only makes sense if the changes in electronic structure are small (see e.g. P. Novak et. al PHYSICAL REVIEW B 71, 184433, 2005). This (usually) works best in insulators, it is a delicate problem in metals, and much more so in Fe3O4 with its Vervey transition. Here a very intricate coupling between electronic and structural degrees of freedom is at work. I seem to recall that the low temperature phase is a comlicated mess (see e.g. Novak et al, PRB 61, 1256, 2000 and references therein). So complicated that, as far as I remember, in the early 2000nds S.Cottenier, R. Laskowski, J. Rusz, M. Rots and P. Novak gave a talk on a Wien2k Workshop calculating exchange interactions in magnetite using the non-collinear magnetism version NCM-Wien2k. Unfortunately I don't have time to search for literature on that one, but you probably don't want to get into NCM anyway. However, I don't think you can avoid DFT+U or +EECE - at least not for Fe3O4. I am sure you can find a lot of literature on DFT+U and +EECE of Fe3O4, among others by Novak, Madsen, ... This may introduce an additional parameter in your comparisons of your structures. If you are using an older version of Wien2k, upgrade! Wien26_16 had a bug with DFT+U (see https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15590.html). Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2019-04-16 10:29, schrieb Penny, Charles: Dear all, I am running spin-polarised calculations on a range of iron-spinel structures (namely, magnetite (Fe3O4), maghemite (gamma-Fe2O3) and greigite (Fe3S4)) with the objective of calculating magnetic exchange energies in these minerals. This requires calculating total energies of lot of different spin configurations. This process has worked well for magnetite and maghemite, but I have encountered a problem with greigite. When I run a calculation on a spin configuration of greigite that isn’t the ferrimagnetic ground state (e.g. a ferromagnetic configuration) the calculation converges to the ferrimagnetic solution, with the sublattice moments pointing in opposing directions. In the examples below, I have used a low-symmetry unit cell with eight unique iron atoms which allows me to calculate the required number of spin configurations for estimating J_ij. Atoms 1-4 correspond to A site iron atoms in the spinel structure, atoms 5-8 correspond to B site iron atoms and atoms 9-16 are sulphur atoms. In a ferrimagnetic system the A and B sites have opposing moments and sulphur atoms are non-magneitc. When I define a ferrimagnetic spin configuration, the calculation proceeds as expected, with the final moments looking like; rkmax_8_k_500.scf::MMINT: MAGNETIC MOMENT IN INTERSTITIAL = -0.05116 rkmax_8_k_500.scf::MMI001: MAGNETIC MOMENT IN SPHERE 1= 2.47349 rkmax_8_k_500.scf::MMI002: MAGNETIC MOMENT IN SPHERE 2= 2.47348 rkmax_8_k_500.scf::MMI003: MAGNETIC MOMENT IN SPHERE 3= 2.47348 rkmax_8_k_500.scf::MMI004: MAGNETIC MOMENT IN SPHERE 4= 2.47348 rkmax_8_k_500.scf::MMI005: MAGNETIC MOMENT IN SPHERE 5= -3.01699 rkmax_8_k_500.scf::MMI006: MAGNETIC MOMENT IN SPHERE 6= -3.01699 rkmax_8_k_500.scf::MMI007: MAGNETIC MOMENT IN SPHERE 7= -3.01699 rkmax_8_k_500.scf::MMI008: MAGNETIC MOMENT IN SPHERE 8= -3.01699 rkmax_8_k_500.scf::MMI009: MAGNETIC MOMENT IN SPHERE 9= -0.03675 rkmax_8_k_500.scf::MMI010: MAGNETIC MOMENT IN SPHERE 10= -0.03675 rkmax_8_k_500.scf::MMI011: MAGNETIC MOMENT IN SPHERE 11= -0.03675 rkmax_8_k_500.scf::MMI012: MAGNETIC MOMENT IN SPHERE 12= -0.03675 rkmax_8_k_500.scf::MMI013: MAGNETIC MOMENT IN SPHERE 13= -0.03675 rkmax_8_k_500.scf::MMI014: MAGNETIC MOMENT IN SPHERE 14= -0.03675 rkmax_8_k_500.scf::MMI015: MAGNETIC MOMENT IN SPHERE 15= -0.03675 rkmax_8_k_500.scf::MMI016: MAGNETIC MOMENT IN SPHERE 16= -0.03675 rkmax_8_k_500.scf::MMTOT: SPIN MAGNETIC MOMENT IN CELL = -14.88108 Final energy; rkmax_8_k_500.scf::ENE : ** TOTAL ENERGY IN Ry = -43322.30312592 However, when I define a ferromagnetic spin configuration the system converges to a ferrimagnetic solution with final moments; k_500_rkmax_8.scf::MMINT: MAGNETIC MOMENT IN INTERSTITIAL = 0.05118 k_500_rkmax_8.scf::MMI001: MAGNETIC MOMENT IN SPHERE 1= -2.47348 k_500_rkmax_8.scf::MMI002: MAGNETIC MOMENT IN SPHERE 2= -2.47347 k_500_rkmax_8.scf::MMI003: MAGNETIC MOMENT IN SPHERE 3= -2.47346 k_500_rkmax_8.scf::MMI004: MAGNETIC MOMENT IN SPHERE 4= -2.47346 k_500_rkmax_8.scf::MMI005: MAGNETIC MOMENT IN SPHERE 5= 3.01697 k_500_rkmax_8.scf::MMI006: MAGNETIC MOMENT IN SPHERE 6= 3.01697 k_500_rkmax_8.scf::MMI007: MAGNETIC
Re: [Wien] Augmented Plane Wave
Dear Pablo, I suspect your problem occurs because you left out the word which "Augmented" refers too: It is an "Augmented Plane Wave Method", that is, the Method is augmented (including additional basis functions), not the plane waves (in amplitude or intensity). Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 2019-03-22 02:49, schrieb delamora: Dear Wien users, I have a question about the name of "Augmented Plane Wave" I had the idea that when the wave enters the Muffin Tin sphere the amplitude of the wave increased. Trying to see this I found that when a wave crosses a step function, https://quantummechanics.ucsd.edu/ph130a/130_notes/node149.html When the incoming wave exp(ikx) reaches an upwards step function there is a reflected wave R exp(-ikx) and a transmitted wave T exp(ik'x) what this article shows is; 1 + R = T That is, the amplitudes of the incoming wave and the reflected wave add to the amplitude of the transmitted wave If I take this into a square well then I would understand that the waves inside the well have the total amplitude equal to the incoming and transmitted wave. That is, when the wave enters the Muffin Tin the amplitude of wave is not AUGMENTED. So why is this method called "Augmented Plane Wave"? Saludos Pablo ___ 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/index.html
Re: [Wien] Problem with DOS
Assuming that by 'very fine band' you mean a very narrow band my first guess would be that it is missing in your DOS because it sits between two energies on the energy axis of the plot. Focus the energy range where your DOS is calculated on the interval where the band actually is. Good luck, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 14.01.2019 23:49, schrieb Wien2k User: Dear wien2k users: The band structure has given a very fine band but when I plot the corresponding density of state I can not find the peak corresponding to this band (or the value of this peak is less than the format of output file of dos (F?.8)) how can I get the peak value even if it is less than E-8 ___ 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/index.html
Re: [Wien] somethint about the symmetry in the struct file
In addition to Gavin's comments: I believe you want to do a permutation of the list of symmetry operations placed by symmetry at the end of case.struct. You want to do this in file case.struct_st, NOT in case.struct? To my best knowledge you can change whatever you want in file case.struct_st. It is a book-keeping output of program symmetry. Its single use is at the end of symmetry to (optionally) replace an existing file case.struct. After that is done you almost certainely may tidy it up as you whish, or place comments in there, or even delete it. As Gavin said, try it. Just DO NOT replace case.struct with the edited file - or do that only if you know exactly what you are doing (which you apparently do not know). What do you mean by 'the first matrix is not unitary'? The one operation that is always present in the list is the identity, or unity. You most probably mean that one, and you want to have it at the top of the list? May I ask WHY you want change the sequence of symmetry operations? Good luck, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 16.01.2019 04:11, schrieb 姜若诗: Hello,I want to ask something about the symmetry matrix in the case.struct. If I change the order of the symmetry matrix in the case.struct. For example, the first matrix is not unitary,I change the eighth matrix and the first matrix in the case.struct_st in ordet to make sure the first one is unitary,then I run it.Will it have some influence to my result? Wish your reply. Sincerely Jasmine ___ 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/index.html
Re: [Wien] EFG: theory Vs experiment for a case
As an addendum to previous comments by the experts I strongly recommend to always ask (and answer) yourself the question: what symmetries does the experimental result imply, and what symmetries are present in my model? You state that samples prepared in a precedure 'where synthetic parameters can be controlled' (whatever that means) 'shows STO with probe showing very pure HFI'. I take it that TDPACS finds Hf in a unique environment. All of the spectra is accounted for by a single value abs(V_zz)=1.69e21 v/m^2, eta=0.52? No (sizeable) broadening, no other lines from different EFG's? In that case: 1) Whatever model you set up in case.struct, your Hf-site MUST NOT have a threefold (or higher) rotation symmetry in its point symmetry group (check outputs of structure initialzation programs). 2) Your structure (with your defect of the day) should have a significantly lower total energy than other possibilities - otherwise a superposition of the corresponding spectra is expected. Up to now your calculations without neighboring oxygen defects seem to be useless because you did not take into account the first point. With respect to the second point I am very dubious about your calculations assuming oxygen defects in some neighborhood. Some Hf-vacancy pair may have a calculated EFG resembling the experimental one. But in an unrelaxed structure this is meaningless. And even if you insist to take something like that serious, you will have to come up with a really good reason why that special defect structure is the only one present in your sample! Looking at your result that the artificial V_zz from your unrelaxed 3x2x2 supercell already has the same size as the experimental one I strongly suspect that you will get a pretty good simulation of the experimental result by some very small rearrangement of atoms in the nearest, at most the next neighbor shell - WITHOUT oxygen vacany. Put Hf on a Ti-site and really DO a relaxation. Start it from a structure with appropriate (low) point symmetry at Hf. (shift some neighboring atom(s) a tiny bit out of symmetry in some plausible direction). And remember, EFG is short range - from my experience even a 2x2x2 supercell may give you a good idea what happens. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 19.12.2018 09:48, schrieb Ashwani Kumar: System under study is SrTiO3 doped with Hf-181 tdpac probe for TDPAC spectroscopy by thermal equillibriation method (1250 C for >12 hours). the doping is substitutional as Hf(IV) and Ti(IV) has nearby ionic radius and same oxidation state. Bulk STO shows no efg, no assymetry parameter as STO is cubic lattice and Hf/Ti present at octahedral symmetry but in another synthetic procedure where synthesis parameters can be controlled shows STO with probe showing very pure HFI. From literature, STO is known to have oxygen vacancies depending on synthesis route. Calculation: STO structure is optimized and minimized. Created supercell 3x2x2 as it allows me to use 5 processor with 8 gb ram (with 3x3x3, calc. restricted to only 2 processors due to RAM limitation. So long scf run time. i will try now) . Problem of breaking symmetry did not strike me at the time of calculation. Replaced one Ti by Hf. Then supercell calculation was done for oxygen vacancies in first cordination and fourth nearest neighbour cordination. The efg matched for the later case (oxygen vacancy at fourth nearest neighbour) but not assymetry parameter. Minimization of forces on supercell was not done ( i thought basic lattice unit was already optimized). What i understood from previous replies of this thread is 1.Must do 3x3x3 supercell calculation, 2. Electron electron correlation can also be included for improvement of calculation, 3. Need to check calculation with reduced RMT of Sr and Ti. To what % agreemnet wrt experimental efg data, efg values from theoretical calculation can be accepted. Thanks Dr. Cottenier for the suggestion, i had already subscribed to HF course A (ID: iak). I will subscribe to Course B after finishing course A. Thanks, A. Kumar On Wed, Dec 19, 2018 at 12:08 AM Ashwani Kumar wrote: hi, thanks for reply. the assymetry parameter, (Vxx-Vyy)/Vzz, is zero (wien2k calculation) whereas i got 0.52 from TDPAC (Time dependent perturbed angular correlation) spectroscopy for a SrTiO3 (STO) defect structure. EFG component is -1.63 x10^21 V/m2 (wien2k, lapw2 -efg) and i obtained 1.69 x 10^21 V/m2 (calculated from TDPAC results). STO has cubic lattice so no efg and no assymetry parameter (for both wien2k and TDPAC) but defect STO structure showed very pure hyperfine interactions with assymetry parameter : 0.52. So i am not having confidence over my wien2k calculation because : 1. i am getting assymetry parameter =0 2. Negative EFG which i understood from previous answers that negative s
Re: [Wien] Qestion about DOS results
... Good luck, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 11.12.2018 23:06, schrieb shaymlal dayananda: Dear all Sorry, my reply to the original mail chain is waiting for the moderator approval! So I am sending this as a new email. I have actually considered hubard-U (4.5 eV is included) and spin orbital coupling also added. My structure is U3O8, case.indmc -12. Emin cutoff energy 2 number of atoms for which density matrix is calculated 1 1 3 index of 1st atom, number of L, L1 2 1 3 index of 1st atom, number of L, L1 0 0krad, kls case.inorb 1 2 0 nmod, natorb, ipr PRATT 1.0BROYD/PRATT, mixing 1 1 3 iatom nlorb, lorb 2 1 3 iatom nlorb, lorb 1 nsic 0..AMF, 1..SIC, 2..HFM 0.3307 0.00U J (Ry) Note: we recommend to use U_eff = U-J and J=0 0.3307 0.00U J I am having a followup question for your comments. 1. Can I conclude FM for my system? because: atom-1(uranium1) is non-magnetic, atom-2 uranium (multiplicity is 2 for this atom) has a magnetic moment of 0.71935. These two uranium has parallel magnetism. 2. This system is U3O8, (not U2O5). It has only 11 atoms in the unit cell as 1 (U), 2(U), 2(U), 3(O), 4(O), 4(O), 5 (O), 6(O), 6(O), 7(O), 7(O). With this, is this possible to accept the obtained DOS? (And do we necessarily should get different DOS for FM/AFM cases for their spin UP/DN cases? Thank you Shayam ___ 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/index.html
Re: [Wien] Regarding Local environments of individual atom in a structure
High Sandeep, I think this has not been answered, so I will give it a try: Yes, case.outputnn would be one place to look for information about local environments. By way of an example I copied a few lines from case.outputnn of a Co hcp structure. The lines starting with % are comments I inserted ... ATOM: 1 EQUIV. 1 Co AT 0.3 0.7 0.25000 % Specifies the environment of which atom is listed in the following lines. % Position in crystal coordinates. RMT( 1)=2.34000 AND RMT( 1)=2.34000 % RMT of that atom (there are 2 Co per unit cell) and of a nearest neighbor SUMS TO 4.68000 LT. NN-DIST= 4.71810 % nearest neighbor distance must be greater than sum of RMT's ATOM: 1 CoAT -0. 0. -0.2500 IS 4.71810 A.U. 2.49671 ANG % Position in crystal coordinates of the first nearest neighbor, & and distance in a.u. and Angstrom ATOM: 1 CoAT -0. 0. 0.7500 IS 4.71810 A.U. 2.49671 ANG % The same for the second nearest neighbor ATOM: 1 CoAT 0.6667 1. -0.2500 IS 4.71810 A.U. 2.49671 ANG ATOM: 1 CoAT 0.6667 1. 0.7500 IS 4.71810 A.U. 2.49671 ANG ATOM: 1 CoAT 0.6667 0. -0.2500 IS 4.71810 A.U. 2.49671 ANG ATOM: 1 CoAT 0.6667 0. 0.7500 IS 4.71810 A.U. 2.49671 ANG % There are 6 nearest neighbors in the hcp structure, all at the same distance. ATOM: 1 CoAT 1. 0.6667 0.2500 IS 4.73773 A.U. 2.50710 ANG % Position in crystal coordinates and distance % of the first next nearest neighbor atom ... There are also 6 next nearest neighbors in this case, which are listed in the following lines. If you need further shells set the associated call parameter for nn to higher values (see user guide). If you need (binding) angles you can calculate them from the position data (remember to transform to cartesian coordinates), or use your favorite crystal structure viewer (xcrysden works fine). Take information about local symmetry from case.outputs (or case.outputsgroup). In case.outputs you will find for each atom in case.struct a list of symmetry elements for that position. For atom 1 in th above case of hcp-Co this reads: ... ATOM: 1 Co operation # 1 1 Co operation # 4 2 || 120 Co operation # 5 2 || 210 Co operation # 6 2 || 1-10 Co operation # 14 m (s-h) n z Co operation # 18 m (s-v) n 100 Co operation # 19 m (s-v) n 010 Co operation # 20 m (s-v) n 110 Co operation # 21 3 || z Co operation # 22 3 || -z Co operation # 23 S3 (-6) || z Co operation # 24 S3 (-6) || -z % There is identity, three 2-fold axes, four mirror planes, %two 3-fold axes (along +/-z), and two 6-fold screw axes pointgroup is -6m2 (neg. iatnr!!) % The point group with these symmetry elements axes should be: -6 || z, m n y z-rotation vector: 0. 0. 1. y-rotation vector: 0. 1. 0.2 % Orientation of the local axes, highest symmetry along z %(here 6-fold rotation), next along y (mirror plane normal y) WARNING: LOCAL ROTATION MATRIX CHANGED LOCAL ROT MATRIX: NEWOLD 1.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 % Local rotation matrices belonging to this orientation (see user guide) ... I hope this helps. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 18.09.2018 12:34, schrieb sandeep Kumar: Dear Prof. Peter Blaha and WIEN2k Users, I am trying to find the local environments of an individual atom in a structure. In VASP code, it is written clearly in the OUTCAR file that how many atoms and which atom are nearest neighbour around an individual atom in a structure. We can also analysis local environments from the structure itself using VESTA. Can we do in the WIEN2k code using case.outputnn file ? Could you please explain it to me? Thanks Sandeep -- Dr. Sandeep Kumar (Postdoctoral Research Fellow) Institute for Nanotechnology & Advanced Materials, Department of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel ___ 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/index.html
Re: [Wien] charged and neutral cell
As I said before: when you remove an atom from case.struct, you remove the nuclear charge AND all Z electrons. This is neutral. If your perfect structure was neutral, then the new structure is neutral as well. Don't try to tell Wien2k that Ba should be missing 2 electrons and O should have them. It makes no sense: with Wien2k you have a very sophisticated tool which has been designed by extremely clever people to do with enormous precision exactly what you apparently want to do by hand: distribute electrons in a solid. Throw the program a handfull of neutral atoms, tell it their positions - thats what the struct file is for - and let it do it's magic. As a guidance, I really recommend the UG, and use the w2web interface. In short (fill in the gaps from the UG): - create a case directory for the perfect structure (perhaps using the web interface w2web). Generate a valid case.struct file (via w2web's structeditor, from a cif file ... whatever). Walk through initialization. Run the scf. Inspect the file case.scf. Calculate DOS, do a band structure spagetti plot ... I gues you have done this already. - create a new case directory and copy the struct file of the perfect structure to this directory. Use supercell to generate your 2x2x2 unit cell (perhaps P, not centered). Rename it to match the directory name. Use w2webs the structeditor to kill one Ba atom, if that is what you want to do first. (A 2x2x2 unit cell will be a pedagogical exercise, it is too small to accomodate the structural relaxation such a massive defect will need). Make sure that the RMT's in the struct file are at least 5% from touching since you definitely want to give Wien2k a chance to relax the defect structure (see UG on structural relaxation). Walk step by step through initialization: nn first. It will be unhappy with the multiplicities and creat a new struct. Accept the this new struct file. Next sgroup, which will most probably again make its own changes. Accept them. With a little bit of luck this struct file is then ok for the symmetry program. If the symmetry program crashes it may be due to different rounding of exact positions in sgroup and symmetry - there was a recent thread about in the mailing list. If symmetry is happy you are probably in the green with the defect structure - inspect the structure with your favorite crystal structure viewing tool, perhaps xcrysden. Proceed with initialization. If you want run a simple scf without structural relaxation as a reference. For anything beyond pedagogical exercises I think scf runs with structural relaxation are mandatory. - rinse and repeat with whatever defect structure you are interested in, each in fresh case directory. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 30.11.2017 20:27, schrieb chin Sabsu: Dear Pieper Sir, Thank you for detailed reply. Okay, I understood few things now. But if we remove one atom from the super cell then how we can say that the defective cell is charge neutral? I assume that I will be convinced from your next reply that "how a defective supercell is considered charge neutral (A^+2 from 2x2x2 cell of ABO3)". Now Please guide me how I can create fully charged A or B or oxygen vacancies defects in ABO3 system? Thank you in advance. Sincerely Chin On Thursday, 30 November 2017 6:26 PM, pieper <pie...@ifp.tuwien.ac.at> wrote: No, if you remove atoms in the case.struct file, you remove atoms, that is neutral entities. Just don't play around with the entry Z for the number of electrons (which you should not do if you are not sure what you are doing). It defaults to the nuclear charge - use the w2w2b interface and leave the entry empty in structgen, it will be filled in automatically. You need not do anything special during initialization to maintain charge neutrality. As for the second question, readjust your concept of valence to align with how it is used within Wien2k. In a nutshell, distinction between core and valence electrons in Wien2k is made by their binding energy to a specific nuclear charge. If the binding energy is large enough, the total charge density of the electron is to high accuracy within the sphere of radius RMT around that nuclear charge - this is a core electron. If this is not the case, its a valence electron. It is much more difficult to be a core electron in Wien2k than it is in your classical thinking of Ba2+ and O2-. Leave that distinction to the program lstart during initialization. (There is a lot more, look into the introductory section of the UG and continue reading from there). If you want to calculate defect structures, start from the perfect structure. Perhaps begin with examples that come with the Wien2k-package. After a successful scf run inspect the file case.scf. Search (grep) for :NEC You will find two numbers, the total number of charges, nucle
Re: [Wien] charged and neutral cell
No, if you remove atoms in the case.struct file, you remove atoms, that is neutral entities. Just don't play around with the entry Z for the number of electrons (which you should not do if you are not sure what you are doing). It defaults to the nuclear charge - use the w2w2b interface and leave the entry empty in structgen, it will be filled in automatically. You need not do anything special during initialization to maintain charge neutrality. As for the second question, readjust your concept of valence to align with how it is used within Wien2k. In a nutshell, distinction between core and valence electrons in Wien2k is made by their binding energy to a specific nuclear charge. If the binding energy is large enough, the total charge density of the electron is to high accuracy within the sphere of radius RMT around that nuclear charge - this is a core electron. If this is not the case, its a valence electron. It is much more difficult to be a core electron in Wien2k than it is in your classical thinking of Ba2+ and O2-. Leave that distinction to the program lstart during initialization. (There is a lot more, look into the introductory section of the UG and continue reading from there). If you want to calculate defect structures, start from the perfect structure. Perhaps begin with examples that come with the Wien2k-package. After a successful scf run inspect the file case.scf. Search (grep) for :NEC You will find two numbers, the total number of charges, nuclear and electrons. They should be equal within the accuracy lstart uses to decide wether or not charge density leaks out of the RMT-sphere. You will also find :NOE Which is the total number of not-core electrons, or :CTOnnn which is the total charge (in one spin channnel) in sphere nnn, and many other projections of the charge density on specific volumes, energy windows, spin or orbital states. Use the UG section on the scf file as an introduction. In adition you can use analysis programs like AIM (see UG, section on AIM) to calculate details. Then create a new case directory and restart there by modifying your file case.struct - perhaps using programs like structgen and supercell. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 30.11.2017 06:12, schrieb chin Sabsu: Dear Peter Sir and other experts, Could you please explain how to deal charged vacancy and neutral vacancy? If I remove an atom (say x^+2) from the unit cell then the Unit cell will be charged unit cell with charge -2e. If I remove an atom (say x^-2) from the unit cell then the Unit cell will be charged unit cell with charge +2e. Are the above statements are correct? If above statements are correct then please suggest me how to make the system neutral after creating vacancy. Below are the lines from case.in2 and case.inm where, as far as I know, I need to do some changes. Assuming that I have xx NE in case.in2 and I remove an atom of vacancy +2 then how I should modify these two files so that the final defective cell remains charge neutral? case.in2 TOT (TOT,FOR,QTL,EFG,FERMI) -13.2xx.0 0.50 0.05 1 EMIN, NE, ESEPERMIN, ESEPER0, iqtlsave case.inm MSR1 0.0 YES (BROYD/PRATT, BG charge (-1 for core hole), norm) I have one another question: How the number NE (here I kept it xx) in case.in2 changes case by case? In case of BaZrO3 it, NE, is 40 but I could not find any way to calculate this number manually because Ba is having +2, Zr also +2 and O is -2. Total number of atoms are 5(Ba*1, Zr*1 and O*3) in the unit cell and below is electronic configuration of atoms. Ba: _1s22s2_2p6_3s2_3p63d10_4s2_4p64d10_5S2_5P6_6S2_ Zr: _1s22s2_2p6_3s2_3p63d10_4S2_4P6_4D25S2_ _O:__1s2 2S2 2P4_ _Bold are treated as F (valence) states during initialization._ Looking forward to hearing from any of the expert. Sincerely Chin ___ 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/index.html
Re: [Wien] Ueff
Until one of the real experts (I am not among them) finds time to answer: My guess is that both is correct, Ueff=U-J as well as Ueff=U-J/2. That is, the meaningful thing to do is set J=0 (see UG section on ORB), and not estimate it from DFT. The common ground between ORB or DFT and the magnetism of atoms as in Hunds rules is a multiband Hubbard model. For the latter, the multiple correlation energy parameters of such a model are summarized into one intra-orbital U, one inter-orbital V, and a Hund's rule parameter J reducing the inter-orbital correlation energy for same spin direction. [1]. For the former, the parameter Ueff represents on-site contributions to the correlation energy which are underestimated by L(S)DA. Ueff is, therefore, an approximation that has to estimate the contribution already accounted for by L(S)DA - which leads to options 1,2,3 for different self-interaction corrections. I doubt that there is a general and safe way to assign specific parts of the correlation energy to U, V, and J. At least, the authors of the ORB code recommend to avoid the ambiguities [2], and to set J=0. [1] see for example, the talk 'Localized Electrons with Wien2k' by Elias Assmann Wien, workshop 2013, ldau.pdf among the documentations on the Wien2k site [2] G. K. H. Madsen and P. Novák, Europhys. Lett. 69, 777 (2005) --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 28.11.2017 08:42, schrieb karima Physique: Dear prof. P. Blaha and wien2k users : I am still waiting for your answers and I thank you in advance. my question was : Does Ueff=U-J or Ueff=U- 1/2 J? how to estimate J using constrained DFT.? Thank you in advance 2017-11-27 12:05 GMT+01:00 karima Physique <physique.kar...@gmail.com>: Dear Wien2k users: Does Ueff=U-J or Ueff=U- 1/2 J? how to estimate J using constrained DFT.? Thank you in advance ___ 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/index.html
Re: [Wien] Question Regarding NMR Calculations and Nuclear quadrupole coupling Constant
To clarify this: Yes, there is a line shift by nuclear quadrupole interaction, and, depending on the size of the EFG, it can be significant for the determinatation of chemical shift or even Knight shift. It appears in perturbation theory beyond first order, which describes the familiar quadrupole splitting. For this you might want to look at M. H. Cohen and F. Reif, Quadrupole Effects in Nuclear Magnetic Resonance Studies of Solids, Solid State Physics, Vol 5 Academic Press, 1957 editors F. Seitz and D. Turnbull Of course, nowadays you can also do a numerical diagonalization of the Hamiltonian for magnetic field and quadrupole interaction. Best regards, --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 22.11.2017 13:33, schrieb Peter Blaha: This is a question to NMR-experimentalists. They usually know how they obtain the CS and quadrupol splitting from their experimental data. I don't think the quadrupole moment influences the value of the CS. On 11/22/2017 11:24 AM, sandeep Kumar wrote: Dear Professor Peter Blaha and Dr. Robert Laskowski, It is known thatquadrupolar nuclei such as 17O the resonance frequency is a combination of the chemical shift and the isotropic quadrupole coupling (goes like Cq^2/w0, Cq is the coupling and w0 is the Larmor frequency) and for a perfectly symmetric environment it should be zero. I have calculated NMR chemical shift and quadrupole coupling constantof an insulator systems. My question is how we can calculate the contributions of Cq in chemical shifts and how much the quadrupole moment influences the chemical shift? Please correct me if I am wrong. Thanks Sandeep Kumar -- Dr. Sandeep Kumar, Post-doc Department of Chemistry, The Lise Meitner-Minerva Center for Computational Quantum Chemistry & The Institute for Nanotechnology and Advanced Materials, Bar-Ilan University, Ramat-Gan 52900, Israel ___ 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/index.html
Re: [Wien] HDLO and LVNS
High Pablo, I think there is a reasonable amount of information in the UG, including a citation for further reading (which I never followed): UG Version 17.1, ch. 7.5, first par.: When adding such LOs, make sure their E-parameters are far away from each other (typically at least 1 Ry, often 2.4 Ry), otherwise ghostbands may occur. In addition (or sometimes, alternatively) a second-derivative (HDLO) LO for improved E-linearization of valence bands (in particular for d and f-states and large RMTs)[Karsai et al., 2017] can be defined. So, HD indeed stands for 'higher derivative' in general, and in Wien2k 2nd derivatives are implemented. Furthermore, see the 'Interpretive comments' for case.in1 in the UG, there especially the comments on NAPWL. Maybe you also want to note that there is a warning in the OPTICS section that HDLO's are not supported in that package. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 12.11.2017 23:46, schrieb delamora: Thank you, but I think that the usersguide should have a minimal explanation, for example, the two lines that you wrote. - DE: Wien <wien-boun...@zeus.theochem.tuwien.ac.at> en nombre de Gavin Abo <gs...@crimson.ua.edu> ENVIADO: domingo, 12 de noviembre de 2017 12:57 p. m. PARA: wien@zeus.theochem.tuwien.ac.at ASUNTO: Re: [Wien] HDLO and LVNS I believe HD stands for high derivative. See where it says high derivative LO (HDLO) in the Computer Physics Communications Vol. 220 p. 230 (2017) article: https://doi.org/10.1016/j.cpc.2017.07.008 On 11/12/2017 8:22 AM, delamora wrote: Dear Peter Blaha, In the 17.1 version the HDLO and LVNS appear; atom 1 has a large sphere , consider setting HDLOs and/or larger LVNS I searched in the usersguide and I did not find much information; For LVNS there is no explanation, all I found is in 5.1.3; -lvns L -> in batch mode: LVNS_max (default: 4) and for HDLO it is not clear what 'HD' stand for Cheers Pablo de la Mora ___ 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/index.html
Re: [Wien] Number of charge concentration in the system
No, that is the total number of valence electrons (see User Guide). For results you should first look at case.scf (though volume and total electron number is hardly a result of Wien2k ...). For example use the grep command in a terminal set in your case directory: grep VOLUME *.scf :VOL : UNIT CELL VOLUME = 149.45692 grep NUCLEAR *.scf :NEC01: NUCLEAR AND ELECTRONIC CHARGE 54.053.99861 (if I do these grep in a dir with a Wien2k case for hcp-Co, a hexagonal structure with 2 Co per unit cell) grep ELECTRONS *.scf :NOE : NUMBER OF ELECTRONS = 30.000 (the number of valence electrons, like file case.in2) Or view case.scf in your favorite editor and search for 'UNIT CELL', 'NUCLEAR', 'VALENCE', or labels likes ':VOL', ':NOE', ':NEC', ... See the UG for a list and an explanation what (most of) the labels mean. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 02.11.2017 02:28, schrieb halim said: Dear Professor Pieper, Thank you very much for your answer and clarification, and detailed answer. Do you mean the total numbers of electrons is NE printed in file.in2? Thank you for your answer. Halim Said Le Mercredi 1 novembre 2017 22h26, pieper <pie...@ifp.tuwien.ac.at> a écrit : You want to know the overall electron concentration? Why use Wien2k for that one? If your unit cell is spanned by crystallografic basis vectors a,b,c then the volume is V=a*(b x c). If your unit cell contains n_i atoms of elements nuclear charge Z_i the total number of electrons is N=sum_i(n_i*Z_i). The concentration is c=N/V. And yes, volume V as well as total number of electrons N are printed in Wien2k's result file ... --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 01.11.2017 05:19, schrieb halim said: Dear Wien2k users community, I would like to ask you if we can get the results of total number of concentration of the system from wien2k code Looking forward your answer. Halim Said ___ 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/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/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/index.html
Re: [Wien] Number of charge concentration in the system
You want to know the overall electron concentration? Why use Wien2k for that one? If your unit cell is spanned by crystallografic basis vectors a,b,c then the volume is V=a*(b x c). If your unit cell contains n_i atoms of elements nuclear charge Z_i the total number of electrons is N=sum_i(n_i*Z_i). The concentration is c=N/V. And yes, volume V as well as total number of electrons N are printed in Wien2k's result file ... --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 01.11.2017 05:19, schrieb halim said: Dear Wien2k users community, I would like to ask you if we can get the results of total number of concentration of the system from wien2k code Looking forward your answer. Halim Said ___ 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/index.html
Re: [Wien] Querry in a resultant structure
In my humble opinion YOU have to decide wether your structure looks like what you want. I only can/will expand somewhat on my previous advice to look at case.outputnn and to compare structures in xcrysden: Since xcrysden is difficult about arrows, do this in colors: You have 4 purple marbles in your unit cell. There are 3 ways to replace two of them by identical green marbles, depicted as AFM I - III in your figure. So, to avoid the fuss of talking xcrysden into plotting arrows, take the supercell generated .struct file and produce 3 different .struct files by replacing two Ni with, say, Co. Look at these with xcrysden and decide which is what. Rename all Ni-colored atoms Ni1, all Co-colored atoms Ni2 (and perhaps make all S identical) and ask sgroup what it thinks of the idea. As for the mysterious downgrading from rhombohedral to monoclinic: Look at your structures along the (1,1,1) direction. Is there a 3-fold symmetry? I hope this helps, --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 25.10.2017 22:39, schrieb Abderrahmane Reggad: Hello again Here is the 3 different AFM configurations https://ibb.co/mqySFm Here are the 4 independent Ni atoms https://ibb.co/nt2nFm Now the big problem lies in if is it possible tp get a monoclinic structure (space group #11) with orthorhombic lattice parameters as we know that the rhombohedral structure in wien2k is represented bt a hexagonal lattice parameters. Best regards ___ 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/index.html
Re: [Wien] Querry in a resultant structure
It is a valid procedure to generate a unit cell that supports placing antiparallel spins on the two Ni sublattices containing Ni1 and Ni2 atoms. I cannot tell you wether the new structure is 'true' or not. Concerning the crystal structure you simply should check youerself with xcrysden (or some similar program): There should be no visible change in the positions when viewing the two structures from the same angle. You also should compare the distances to nearest and next nearest neighbors for the old and new structure - they are listed in the respective case.outputnn and should be identical. Concerning the magnetic structure I don't know what exactly you mean with 'type III' af. If you are uneasy with the concept of an antiferromagnetic wave vector you might try to visualize what the difference to type I and type II is. Then look into outputnn again and decide if you really want these neighbors being antiparallel, or if it's the neighbours at some other distance. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 24.10.2017 14:04, schrieb Abderrahmane Reggad: Thank you pieper for your answer I am studying the magnetic order of type III with Ni1,Ni2 up and Ni3, Ni4 down according the figure included. This is the procedure that I adopted: -supercell 1x1x1 P to make the 4 atoms Ni indepedent a - I make the Ni1 and Ni2 to be the atom Ni1 and the atoms Ni3 and Ni4 to be the atom Ni2. - sgroup to search a new space group - I got the new strycture with monoclinic space group but with orthorhombic lattice parameters. I want to know if this new structure is true or not. The non magnetic structure is as follows: NiS-MnP P LATTICE,NONEQUIV.ATOMS: 2 62_Pnma MODE OF CALC=RELA unit=ang 10.056514 6.692230 10.722761 90.00 90.00 90.00 ATOM -1: X=0.0050 Y=0.2500 Z=0.2000 MULT= 4 ISPLIT= 8 -1: X=0.9950 Y=0.7500 Z=0.8000 -1: X=0.5050 Y=0.2500 Z=0.3000 -1: X=0.4950 Y=0.7500 Z=0.7000 Ni NPT= 781 R0=0.5000 RMT= 2.08Z: 28.0 LOCAL ROT MATRIX:0.000 1.000 0.000 0.000 0.000 1.000 1.000 0.000 0.000 ATOM -2: X=0.2000 Y=0.2500 Z=0.5700 MULT= 4 ISPLIT= 8 -2: X=0.8000 Y=0.7500 Z=0.4300 -2: X=0.7000 Y=0.2500 Z=0.9300 -2: X=0.3000 Y=0.7500 Z=0.0700 S NPT= 781 R0=0.0001 RMT= 1.70Z: 16.0 LOCAL ROT MATRIX:0.000 1.000 0.000 0.000 0.000 1.000 1.000 0.000 0.000 8 NUMBER OF SYMMETRY OPERATIONS Best regards ___ 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/index.html
Re: [Wien] Querry in a resultant structure
I don't know what your question is, but hope that a comment might help. Apparently you did not change size and shape of the unit cell (using a 1*1*1 supercell), so the lattice constants and angles are the same. However, you told sgroup that there is not one type of Ni in the unit cell but two. This change in the decoration of the bricks your crystal is built of broke some symmetries in their arrangement - now you are down to monoclinic. I don't know exactly what af wave vector you want to establish in your structure (+-+-, or ++--++--, or ... counting along which direction?). Is your problem that your Ni1 and Ni2 are not the two you actually wanted to have antiparallel moments? If 'afIII' fits into this unit cell (Ni1 and Ni2 do have antiparallel moments), and if the atoms are where you want them, run symmetry, set spin directions in case.inst, run lstart ... Type III sounds comlicated enough to not fit into this unit cell - meaning that you cannot complete one period of the +/- count in your chosen direction within the unit cell. In that case you have to change shape and/or size of the unit cell. Maybe the options provided by supercell (double the cell along basis directions ...) are sufficient. There is no guarantee for this to be the case. You may have to work out the fitting new shape and the positions of the atoms in the new unit cell using additional tools. I hope this helps, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 22.10.2017 20:57, schrieb Abderrahmane Reggad: Dear wien users I am studing the magnetic structure of type III for a orthorhombic structure. After doing the supercell and labelling the independent atoms for spin-up and spin-dn atoms and using the sgroup command I got the following structure: The space group of a monoclinic structure and the lattice parameters are those of orthorhombic structure. NiS-MnP-afmIII P LATTICE,NONEQUIV.ATOMS: 4 11_P21/m MODE OF CALC=RELA unit=bohr 10.998349 10.005529 6.864228 90.00 90.00 90.00 ATOM -1: X=0.25021227 Y=0.1424 Z=0.2500 MULT= 2 ISPLIT= 8 -1: X=0.74978773 Y=0.8576 Z=0.7500 Ni1NPT= 781 R0=0.5000 RMT= 2.21Z: 28.0 LOCAL ROT MATRIX:1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 ATOM -2: X=0.75024661 Y=0.50012576 Z=0.7500 MULT= 2 ISPLIT= 8 -2: X=0.24975339 Y=0.49987424 Z=0.2500 Ni2NPT= 781 R0=0.5000 RMT= 2.21Z: 28.0 LOCAL ROT MATRIX:1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 ATOM -3: X=0.08986359 Y=0.75012482 Z=0.7500 MULT= 2 ISPLIT= 8 -3: X=0.91013641 Y=0.24987518 Z=0.2500 S 1NPT= 781 R0=0.0001 RMT= 1.81Z: 16.0 LOCAL ROT MATRIX:1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 ATOM -4: X=0.41013993 Y=0.25012471 Z=0.7500 MULT= 2 ISPLIT= 8 -4: X=0.58986007 Y=0.74987529 Z=0.2500 S 2NPT= 781 R0=0.0001 RMT= 1.81Z: 16.0 LOCAL ROT MATRIX:1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 4 NUMBER OF SYMMETRY OPERATIONS Best regards ___ 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/index.html
Re: [Wien] negative position in rstruc. Please report .
In my experience this can happen because of rounding issues with sgroup (my guess is in the output format). The structure file generated by sgroup (which you seem to have correctly accepted) may have atoms in slightly off positions, very close to the symmetric positions they should occupy. This may confuse the next program in initialization, namely symmetry. Last time I stumbled into this error message I either -let supercell generate a shifted unit cell with the origin at a different atom. However, there were cases where I was unable to find a shift producing a viable unit cell. - or located rounding errors in the positions generated by sgroup. After the sgroup step, open the struct file in w2web's structeditor and look for symmetry related coordinates like, for example, 0.499. Set them to 0.5 (at a glance, in your struct put 0. instead 0f 0.3334 etc.). I hope one of the two works in your case as well --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 20.10.2017 10:25, schrieb mohamed mahdi: i m trying to performe a calculation for ErCo4.5Si0.5 , after making a supercell and accepting the structer from wien2k : when running int again I got this error the .struct ERROR: negative position in rstruc. Please report . ___ 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/index.html
Re: [Wien] Spin Magnetic susceptibility
My two cent on this: Spins of itinerant electrons or, more specifically, electrons in the immediate vicinity of the Fermi surface usually will dominate the magnetic field response at T=0 (Pauli susceptibilty) - if there is a Fermi surface, i.e. if you have a metal, and if you do not consider magnetic order. However, I would not jump to the conclusion that this is the 'only' spin contribution. Basically chi=M/H. The spin part of the magnetization M is the field (H) induced occupation difference between spin-up and -down channel. It occures because of the Zeeman interaction between electron spin and field. Wether or not the electrons are localized doesn't matter. An example might be the large spin susceptibility of antiferromagnetic insulators. Best regards, --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 02.10.2017 18:44, schrieb karima Physique: Thank you very much and as you said, we are still waiting for a confirmation from Prof. P. Blaha and Wien2k users. 2017-09-30 21:53 GMT+02:00 Yundi Quan <yq...@ucdavis.edu>: Hi, As I understand it, the susceptibility you obtain by shifting DOS is the Pauli susceptibility, i.e. itinerant electrons near the Fermi level. But I might be wrong. On Sat, Sep 30, 2017 at 8:40 AM, karima Physique <physique.kar...@gmail.com> wrote: Dear Prof. P. Blaha and Wien2k user; Is the spin part of the magnetic susceptibility due to the contribution of delocalized electrons only? ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien [1] SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html [2] ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien [1] SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html [2] Links: -- [1] http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien [2] 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/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/index.html
Re: [Wien] The expression of spin susceptibility
Hi Karima, I am a strong believer in the idea that one learns best what one works out by oneself, so I prefer to just point you in a direction: Take chi=M/B, where B is in T (beware of a shortcut here between B and mu_0*H, mu_0 being the vacuum permeability), and M is some density of magnetic moments (per volume, weight, particle number, you-name-it). The units on the left hand side of that equation are stated as [m^3/(unit cell content)]. Set aside that Peter Blaha admitted there might be a typo there, the use of 10^6 cm^3 there indicates that this is meant to be a cgs molar susceptibility. On the right hand side the units are [mu_B/(bohr^3*T)]. With bohr^3 a volume this is clearly a volume susceptibility. Did you try to convert mu_B/bohr^3 to cgs or SI units? You will find conversion factors for mu_B (magnetic moment) and bohr (length) e.g. on the NIST website. From there you should be able to convert to cgs and be able to tell us what exactly the typo is ... Best regards, --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 23.09.2017 23:55, schrieb karima Physique: Dear Prof. P. Blaha and wien2k users; what the meaning of the number 6.258116 in the expression of spin susceptibility? ___ 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/index.html
[Wien] Fermi-contact hyperfine fields
Dear Wien2k users, I have a question about the Fermi-contact fields printed at the HFFnnn labels in the scf file. Back in 2010 Peter Blaha advised in this mailing list that the Fermi-contact interaction can be underestimated at least for 3d-metals like Fe by 10 - 20% [www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg02201.html] and Pavel Novak indicated that in his opinion it might be difficult to correct this misbehaviour by some suitable local or semi-local Vxc [www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg03011.html]. Since this was pre-NMR-package and at the beginning of hybrids I wonder wether the situation has improved in the meantime? The NMR-package as I understand it calculates shielding current distributions so I don't expect it improves on Fermi-contact fields - is that correct? And is there a way to get improved values from hybrids that I am unaware of? Best regards, Martin Pieper -- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] WIEN2k_17.1
Hi Pavel, if this is not the trivial missing space, this looks to me as if libxc's ./config has difficulties with f90 files for your ifort. What's your ifort version? Maybe try to run ./config without any parameters and look what it says about the compilers it finds. Thats what the INSTALL file of libxc seems to say (and what worked with me, but I am using gfortran). Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 21.09.2017 11:41, schrieb nov...@fzu.cz: Dear WIEN users, I successfully compiled WIEN2k_17.1 with the exception of lapw0. As required in User's Guide (11.1.2) I downloaded libxc-3.00. However, after ./configure FC=ifort--prefix=$LIBXCDIR error appeared: checking for Fortran flag to compile .f90 files... unknown configure: error: Fortran could not compile .f90 files I then changed in configure f90 to f, but a new error appeared: configure: error: in `/storage/praha1/home/novakpa/libxc-3.0.0': configure: error: f program to find the size of a Fortran integer failed Can you help please. Pavel ___ 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/index.html
Re: [Wien] 2 problems with w2web
Certainely not the recommended way to do things, but the following solved my problem with save_lapw in the w2web Utils menue: In file:///opt/WIEN2k/SRC_w2web/htdocs/util/savelapw.pl change line save the results of an eels calculation only (innes, elnes, broadspec, qtl files, but no figures) to save the results of an eels calculation only (innes, elnes, broadspec, qtl files, but no figures) that is, change the name of the checkbox from 'p' to 'u', and change accordingly line $cmdline .= " -eels" if ($p); to $cmdline .= " -eels" if ($u); No recompile necessary. As I said before, I am no good around perl, so don't ask me why this works, i.e. where $p becomes true. I don't think the browser is to blame, since I had the misbehaviour with Firefox, Opera, and Konqueror (KDE's own browser). Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 16.09.2017 13:28, schrieb Peter Blaha: No, in my test the default save_lapw from "Utils" saves everything (do not click on any option) Yes, :HFFxxx is not printed during MSR1a (minimization) mode, but only once the script has changed back to MSR1. This is a bit difficult to fix 1) The default call of the save_lapw script from the 'Utils' menue lists (and saves) only "qtl*, elnes, innes, inq, inb, broadspec, klist and kgen files saved under ./Co-hcp-test". This probably is the second last of the alternative options one can check in w2web's 'save_lapw' menue. I was unable to call the default behaviour since there is no box to check for this. I guess this is w2web working on a wrong default, the save_lapw command from terminal works as advertised. I noticed this only when I updated my Linux (now OpenSuSe Leap 42.2) and simultaneously Wien2k (now Wien2k 17.1). So the problem might be connected with my new perl version (perl 5, version 18, subversion 2 (v5.18.2)). 2) The plot feature of w2web's scf analysis script crashes when structural relaxation is active and HFFnnn's are selected for the plot. Probable reason (I ran into this when writing a python script for analysis)): It appears that in scf cycles with structural relaxation active the HFFnnn labels are printed (perhaps HFF's are only calculated?) to case.scf only during the finalizing iterations. This makes the number of iterations larger than the number of HFFnnn values. Maybe the module filling in the plot arrays is confused by this. Best regards, Martin Pieper ___ 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] 2 problems with w2web
I encountered two minor problems with w2web of Wien2k 17.1 (I am no good around perl so my analysis is probably poor or simply wrong): 1) The default call of the save_lapw script from the 'Utils' menue lists (and saves) only "qtl*, elnes, innes, inq, inb, broadspec, klist and kgen files saved under ./Co-hcp-test". This probably is the second last of the alternative options one can check in w2web's 'save_lapw' menue. I was unable to call the default behaviour since there is no box to check for this. I guess this is w2web working on a wrong default, the save_lapw command from terminal works as advertised. I noticed this only when I updated my Linux (now OpenSuSe Leap 42.2) and simultaneously Wien2k (now Wien2k 17.1). So the problem might be connected with my new perl version (perl 5, version 18, subversion 2 (v5.18.2)). 2) The plot feature of w2web's scf analysis script crashes when structural relaxation is active and HFFnnn's are selected for the plot. Probable reason (I ran into this when writing a python script for analysis)): It appears that in scf cycles with structural relaxation active the HFFnnn labels are printed (perhaps HFF's are only calculated?) to case.scf only during the finalizing iterations. This makes the number of iterations larger than the number of HFFnnn values. Maybe the module filling in the plot arrays is confused by this. Best regards, Martin Pieper -- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] About the magnetic moment of vanadium in vanadium sulphide
Hi, A. Reggad There are no questions in your last posts so the case seems to be closed. However, I am curious, and there seems to be something to learn in addition to the nice comments made by G. Fecher: First, is the band gap (of NiS) really a good observable to fix the HF mixing (for VS)? The user guide (ch. 4.5.7, first par.) says: The onsite-exact- exchange/hybrid methods apply HF only inside the atomic spheres and only to one particular orbital. Thus you can use it only for localized electrons (see [Tran et al., 2006] for details). Onsite- exact-exchange will NOT improve gaps in sp-semiconductors. These sulfides are not sp-semiconductors, but how reliable will this be here? Are those d-electrons localized? Second, do you, A. Reggad, consider your 0.0002 mu_B to be zero? What about the AF order the Japanese experimentalists reported in the publication you yourself forwarded us? Any comments on the magnetic moment Prof. Fecher told us about? Did you follow his advice? Did you even read it? --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 11.09.2017 15:55, schrieb Abderrahmane Reggad: Hi Laurence The amount of 0.05 of exact exchange is the value that gives a gap value equal to the experimental one for nickel sulphide and close values for magnetic moments for some 3d transition metal sulphides like CrS, FeS, CoS and NiS. You can check my article about the nickel sulphide http://www.sciencedirect.com/science/article/pii/S0921452617303915 Best regards ___ 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/index.html
Re: [Wien] (no subject)
That's why I told you to look critically at both, the experiments and the calculations: in experiments a high Neel temperature may hide an AF state. However, you do realize that changing the convergence criteria alone almost certainely will NOT give you the large moment moment Gerhard Fecher found instead of your practically zero 0.05 mu_B? You will have to follow his suggestions about the initial state! --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.09.2017 02:08, schrieb Abderrahmane Reggad: Hi Before doing the calculation again with the new energy and charge criterion I let you this old article http://journals.jps.jp/doi/abs/10.1143/JPSJ.14.196 Best regards ___ 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/index.html
Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide
I agree, there is some confusion. There probably always is. The confusion here is mainly about definition para-, dia-, ferro-, antiferro-, heli-, or non- magnetic. This comes up every now and then. As I said before, I like the following distinction. a) Define Non-Magnetic ONLY in context of calculations, and there as a model where in the converged ground state all local moments are smaller than the numerical noise. The rest were defined much earlier from an experimental point of view. They all are meant to distinguish response to external magnetic field - a field that DOES NOT EXIST in your off the shelf DFT! b) Various kinds of dia- and para-magentism specify wether the magnetization induced by an applied field is opposite (diamagnetic) or parallel (paramagnetic) to the field. c) Magnetically ordered states (FM, AFM of various types, ferri-, heli-, spin density wave-, ...) are PM in this sense. They are distinguished by the spacial arrangement of non-zero static magnetic moments (all parallel, exactly compensated antiparallel, not compensated antiparallel, not collinear, varying in size continuously, ...). So in response to your points: 1.: Yes, but a PM response just might be due to some unnoticed AF ordered state. Because of Ockham's razor one would have to convince me of that, but who know's? 2.: Yes, you can get a non-magnetic ground state from a spin-resolved calculation. 'Local moments' always refers to the moments within RMT's (or whatever the DFT code supports to assign local moments). The electron spins are of course always there, but spin up and down can compensate - and often will compensate due to spin degenerate ground states. 3.: No, I would be very surprised (and suspicious) if the calculations yield some exactly compensating moment on S. For exact compensation, that is an AF arrangement, you have to have the same element pointing moments in opposite directions. Even the crystalografic lattice sites occupied by the antiparallel moments should be the same. I know of no counter example but would be interested if there is one. In VS Vanadium just might split one of its sublattices into such an AF arrangement. 4.: There is a DOS plotted in the work A. Reggad linked in his earlier question. 5.: That DOS certainely looks metalic and the authors claim agreement with experiment. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 07.09.2017 11:02, schrieb E.A.Moore: I have been following this thread and I think there is some confusion. 1. On the thread it said that the experiment showed it was Pauli paramagnetic. This is the type of magnetism displayed by some metals e.g. sodium which is only apparent if you apply a magnetic field. 2. If you include spin in your calculation (GGA or GGA + U) you can only get ferromagnetic, antiferromagnetic or ferromagnetic states. (An earlier thread deals with how to get paramagnetic states). I think you can only get a nonmagnetic state if you do not include spin? A material with Pauli paramagnetism will be antiferromagnetic in straight forward spin-including calculations. 3. I assume the 0.05 muB refers to the magnetic moment on V. If vanadium sulphide is antiferromagnetic and the magnetic moment on Vanadium is 0.05 muB, then there must be a balancing magnetic moment on the sulphur. 4. I suspect this compound might be alloy-like. Is there considerable mixing of V and S in the valence bands? 5. Assuming your formula is VS, it might be worth noting that VO shows some metallic physical properties. Elaine A. Moore Reader in theoretical chemistry The Open University -Original Message- From: Wien [mailto:wien-boun...@zeus.theochem.tuwien.ac.at] On Behalf Of Fecher, Gerhard Sent: 07 September 2017 08:12 To: A Mailing list for WIEN2k users Subject: Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide 0.05 muB does not mean that it is antiferromagnetic ! what was your charge convergence criterion ? You did never answer my question whether you started the EECE calculation from a converged GGA calculation. Why do you like to have an afm state when the experiment tells it is not ? Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Abderrahmane Reggad [jazai...@gmail.com] Gesendet: Donnerstag, 7. September 2017 00:26 An: wien@zeus.theochem.tuwien.ac.at Betreff: Re: [Wien] About the magnetic moment o
Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide
Your problem is 'if it's possible to get' AF VS with 0.05 mu_B/Vanadium in DFT? IF (capitals are on purpose here) you did everything right to properly converge your calculations to the necessary numerical precision, then you give the answer yourself: Yes, it is possible if one uses GGA+U or +EECE. Wether or not this is an improved model of VS remains open. You still did very little to convince Prof. Fecher or me that you actually did things right to achieve that precision. And you did even less to convince me (us) that experiments might have missed the small moment. Even if you are convinced of both, that you did everything right and the experimentalists did something wrong, you still should discuss very carefully the physics behind your decision to improve the model of a metal by introducing a local orbital potential. You probably don't need to convince us, but keep it in mind for fights with the referees of a publication on this. Good luck with your calculations Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 07.09.2017 00:26, schrieb Abderrahmane Reggad: Hi Martin The problem is that I want to know if it's possible to get a such value of 0.05 MB for atomic magnetic moment for the AFM state of vanadium sulphide in NiAs structure. Hafner and Hobbs have found all the calculations converged to the non magnetic state because they have used the GGA method. To get the AFM state they have to use either the EECE or GGA+U methods. I hope you touch the problem Best regards ___ 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/index.html
Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide
I took an admittedly brief look at your link, and I still don't understand where your problem really is. I have made calculations for 5 3d transition metal sulphides with the method PBE+EECE and I have found that all the results are ok with the exception of the vanadium sulphide. I have made 3 calculations (nm, fm and afmI ) to determine the magnetic state for all the compounds and I found them to have afm state. What exactly is the discrepancy between your results and the state Hobbs and Hafner found? You did not answer Prof. Fechers and my question about the size of the moment you calculated. Nor my question of wether you find an insulator or a metal. Especially in case of the latter I might ask about your insistance on Hund's rule ... The problem is about the atomic magnetic moment of vanadium and for the other results which are not reliable because there is one study where they used the non magnetic and paramagnetic states with the same meaning which is contrary to what said Pr Blaha and they mentionned that they have found to be non magnetic without details. At a glance, those authors seem, in fact, to distinguish between 'non magnetic' and 'paramgentic': NM is used with respect to results of their calculations for cases where in the ground state all moments are zero (or below some numerical threshold). PM is used for states observed in experiments where it is defined by the response of the sample to a magnetic field (which they - and you - don't calculate). Not a bad idea, as far as I am concerned. And they do give details on VS in section 3.1 - in fact, a lot more details than you give us here about your calculations. Best regards --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 06.09.2017 03:59, schrieb Abderrahmane Reggad: I have made calculations for 5 3d transition metal sulphides with the method PBE+EECE and I have found that all the results are ok with the exception of the vanadium sulphide. I have made 3 calculations (nm, fm and afmI ) to determine the magnetic state for all the compounds and I found them to have afm state. The problem is about the atomic magnetic moment of vanadium and for the other results which are not reliable because there is one study where they used the non magnetic and paramagnetic states with the same meaning which is contrary to what said Pr Blaha and they mentionned that they have found to be non magnetic without details. You will find the study in the following link https://www.researchgate.net/profile/J_Hafner/publication/216328826_Magnetism_and_magneto-structural_effects_in_transition-metal_sulfides/links/00463525eaa9e1230f00.pdf Best regards ___ 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/index.html
Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide
High A. Reggad, substitute V for Cr in Prof. Fecher's questions, otherwise I am with him. In case you refer by 'almost zero' to the 0.05 mu_B I seem to recall from your original question, and now wonder about some fundamental discrepancy between the supposed experimental Pauli paramagnetism and your supposed antiferromagnetism: Take a VERY close look at both, experiment AND calculations. It is not trivial to experimentally distinguish a PM from an AF with such a small moment, especially if you have no idea that there is a Tc, let alone where it is. The, say T-dependant SQUID, measurement could start already in the AF state if Tc is above RT and never realize the state is already AF. Or maybe Tc is below the lowest T the experimental set-up could reach? It is equally not trivial to establish an AF ground state with that precision from DFT. Did you consider MMI of V as convergence criterium? Did you check convergence to that precision with respect to RKMAX, k-mesh, ... ? Did you keep the k-mesh (symmetries) between your calculations of the PM and the AF? Did you force the AF structure in your calculation? Did you try what happens if you don't impose it? Are you shure about your structural data? Did you do structural relaxation? Does structural relaxation influence your result of a magnetic ground state? What about the influence of the xc-potential? Did you do eece with LSDA or with PBE-GGA or with ...? And what about other physical properties? Maybe most important, is the stuff metallic/insulating in experiment/calculations? In case your question is about this horrible violation of Hund's rules, I repeat my former suggestion: Take a close look at the assumptions these rules rely on. And never forget that any law and rule of physics is valid only within some domain more or less clearly defined by such assumptions. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 05.09.2017 07:45, schrieb Fecher, Gerhard: About what moment are you talking, the total magnetic moment or the magnetic moment of the Cr atoms ? Did you start your EECE calculation from a regular GGA calculation that had no magnetic moments at the Cr ? Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Abderrahmane Reggad [jazai...@gmail.com] Gesendet: Dienstag, 5. September 2017 00:29 An: wien@zeus.theochem.tuwien.ac.at Betreff: Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide Thanks martin Experimentally they found that the vanadium sulphide is a pauli paramagnetic but I have found it to be antiferromagnetic like other transition metal sulphides but the magnetic moment value equals almost zero despite the fact that vanadium has 3 inpaired electrons. Best regards ___ 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/index.html
Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide
Be aware that Hund's rules do rest on certain assumptions about the relative strength of intra- and inter-atomic couplings. There are, after all, a lot of para- or diamagnetic materials around. Try to look that up. Is your Vanadium sulphide magnetically ordered from experiment? In your case the 0.05 mu_B is probably an instance of the problems frequently encountered in strongly correlated systems: There are several states with very similar total energy, but with very different configurations of charges and spins. Search the mailing list for magnetic ordered states, or for orbital order. The scf cycle of DFT will tend to stay near the configuration it started from. If you start from a standard non-magnetic configuration set by lstart it won't look at magnetically ordered states. Try to start the scf cycle from magnetic configurations - you can set them up in lstart - and compare the total energies. Be carefull to change only the magnetic starting configuration, NOT RMT's, RKMAX, K-mesh, crystal symmetry, whatever. Note that the starting spin configurations you can set in lstart may well be not good enough. You may have to modify the density matrix of the problematic orbitals by hand and let the scf cycle dig itself in in that state by using the option -orbc. This also has been explained in posts in this mailing list before - perhaps most detailed by prof. Tran. Good luck with your calculations --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 02.09.2017 12:24, schrieb Abderrahmane Reggad: Dear wien2k users I have calculated the magnetic moment of vanadium sulphide in NiAs structure using the EECE méthod with alpha parameter equals to 0.05. despite the fact that the number of inpariated electrons is equal to 3 I found that the magnetic moment of vanadium is equal to 0.05 MB which is inconsistent with the prediction of Hund. Best regards ___ 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/index.html
Re: [Wien] How to know if the M+2 has a high spin from calculation?
No, Wien2k probably won't 'determine alone which spin state is energetically more stable'. States with different spin configurations (PM, FM, AF, ...) frequently have small differences in total energy but large differences in their electronic configuration. Due to this the scf cycle tends to converge in local minima associated with the starting configuration. Thats why you should set different spin configurations in lstart and see where the scf cycle leads you. The solution with lowest total energy is what you seek. For your low/high spin configuration this might be enough. There are more complicated cases where you may have to guide the scf to the correct symmetry by populating the density matrices accordingly - this has been discussed in the mailing list a lot - see also option -orbc in the UG. Good luck Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 27.08.2017 13:06, schrieb Abderrahmane Reggad: Hello again I have found in the literature that the spin state configuration is like the magnetic configuration. So we have to make 2 calculations: one for the high spin configuration and another one for the low spin configuration and we look after for the configuration more energetically stable. I want to know if the things are so in the wien2k code or it behaves differently and can determine alone which spin state is the more energetically stable. Best regards ___ 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/index.html
Re: [Wien] How to simulate the ionic state of a compound?
High, Abderrahmane, If I introduce the Ni atom and I change the Z number to 27 or 26 it still nickel because the atom is ont only Z and I change Z the number of neutrons and protons still inchanged. NO! Z is the NUCLEAR charge. It sets the number of electrons automatically. If you change that you have another element! Try it by entering only the chemical symbol (e.g. Ni) in the corresponding field of the w2web struct editor. Leave the field Z empty. Press 'save and continue editing'. It will have filled in the correct Z automatically. One can circumvent this automatism, but it only makes sense if you know what you do. As I said before: find out yourself, start with a text book and the UG. You might learn enough in the process to know what you are doing. An other question: We suppose that the ionic state of NiO is a result from the electron density , can we force this ionic state to be Ni+3S-3 and it's yes how? Once again: You want to create some charge distribution of your own design where atom A lost electrons from and atom B caught them in what you decide is their own shell. This is NOT what DFT or Wien2k does! DFT gives you a (usually very good) approximation of where the electrons actually are in space. This may (in ionic crystals) or may not (with covalent or metallic bonds) agree more or less with your ideas. Note that the agreement depends not only on the charge distribution itself! It heavily depends on how you decide by your experimental and/or computational methods which electron belongs where. Thats where the different definitions of charges Prof. Marks pointed out come from. And try to get your hands on a good text book on solid state physics. It will help you far more than anyone in this mailing list can in his/her free time. Best regards Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 24.08.2017 02:23, schrieb Abderrahmane Reggad: Thank you for all for the rich discussion If I introduce the Ni atom and I change the Z number to 27 or 26 it still nickel because the atom is ont only Z and I change Z the number of neutrons and protons still inchanged.As I mentionned when I have changed Z I have observed that the total energy changes. An other question: We suppose that the ionic state of NiO is a result from the electron density , can we force this ionic state to be Ni+3S-3 and it's yes how? Best regards 2017-08-24 0:43 GMT+01:00 Abderrahmane Reggad <jazai...@gmail.com>: Hello The reason for my question is that I don't how does a DFT code work and I want to know if the ionic state is a result or an assumption. Thank you for contribution to clarify the picture. Best regards ___ 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/index.html
Re: [Wien] How to simulate the ionic state of a compound?
Not knowing what your average Fe knows I still would guess it probably doesn't. Maybe thats why nothing else changes when you relabel atoms in xcrysden. However, upon adding two protons (and some neutrons) in Wien2k, Fe seems to know it became Ni - we learned from the initial question that it says so through ETOT. ;-) Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 23.08.2017 11:16, schrieb Fecher, Gerhard: Funny discussion, I wonder wether iron (Z=26) knows that it is Nickel just if one gives it a new name ? Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Víctor Luaña Cabal [vic...@fluor.quimica.uniovi.es] Gesendet: Mittwoch, 23. August 2017 00:33 An: A Mailing list for WIEN2k users Cc: Victor Luaña Betreff: Re: [Wien] How to simulate the ionic state of a compound? On Tue, Aug 22, 2017 at 11:41:53PM +0200, pieper wrote: my response to Abderrahmane Reggad appears perhaps a little harsh. It was not meant that way. I wanted to emphasize that in my view the idea of telling Wien2k (or any other DFT program) its result (where the electrons are) and to simulate properties from there is completely backwards. Martin, It was not harsh at all in my perception, but I canot say what Abderrahmane thought about your words. I perceived as a normal exchange of ideas in a subject that I love. I agree that using a questionable populaton technique to obtain a number of electrons associated to an ion from a expensive DFT calculation is not neccesarily a good idea. Providing a number to say this is a Ni(+1.98) O(-2.02) crystal should not be the main result from a good article. However, a good physically based theory of chemical bonding in molecules, solids and clusters provides far more that that. Let me be very particular in describing QTAIM (quantum theory of atoms in molecules, aka Richard F. W. Bader et al work) as a beautiful and well founded physical teory on chemical bonding. Bader's QTAIM can be used to partition every property into contributions of a particular atom or functional group. Then you can find an explanation why many perovskite oxides, for instance, have a very similar bulk modulus, and the explanation comes from the fact that the oxide is the major contribution, and that contribution is common to the incumbent crystals. This is, however, what I wanted to point out: DFT (or Wien2k) tells you where the electrons are. Thats its central result. It does not make any sense (to me) to use a DFT program to - as A. Reggad put it - "simulate the NiO compound in its ionic state". If NiO would be a ionic compound then DFT would (hopefully, when set up properly) calculate an electron density with a lot of weight at O and a lot less at Ni as a RESULT. The simulation of any property one wishes to study can proceed from there. I agree, of course. That should be the spirit of an ab initio calculation. However, DFT sometimes separates from the ab initio family when the xc functional is chosen not because of its properties or for exploring the sensibility of the predictions to the functional but because a spurious predictive agreement to whatever 'experiment'. I believe finding explanations to peculiar facts is the real motive of computational research, at least until the point is achieved at which theoretical calculations become competitive with experimental measurements in terms of precision, exactitude and cost. And if the electron density of NiO does not really resemble the ionic picture, why use the ionic model to simulate things? Yes, but having a predictive measurement of the ionicity lets you examine how ionicity depends on thermodynamic conditions: is the effect of pressure more important than the effect of temperature or pH ... on a geological scale? *Chemical bonding* can be at the core of geophysics, research on materials, ... and many more fields. However, I'm a peculiar chemist and I love Physical Review since I was working on my phd thesis, long ago. Best regards from sunny north Spain, Víctor -- . ."Half of the US people use twitter to form its opinion and half / `' \ also elect the US president. I only hope they are not the same /(o)(o)\ half". --From a sentence by Gore Vidal /`. \/ .'\ / '`'` \ "[Technocrats, academics and journalists] are often motivated | \'`'`/ | by hope while the publi
Re: [Wien] How to simulate the ionic state of a compound?
High Victor, my response to Abderrahmane Reggad appears perhaps a little harsh. It was not meant that way. I wanted to emphasize that in my view the idea of telling Wien2k (or any other DFT program) its result (where the electrons are) and to simulate properties from there is completely backwards. Of course I agree with you that the electron density of a solid is a real observable - even in the XXI century. Quantum mechanics was invented to calculate such properties and the density is, after all, the D in DFT. I also agree that there are any number of solids where a ionic description makes perfect sense. I never tried but I am very confident that the ionic nature of alkali halides will be evident from the result of a DFT calculation. This is, however, what I wanted to point out: DFT (or Wien2k) tells you where the electrons are. Thats its central result. It does not make any sense (to me) to use a DFT program to - as A. Reggad put it - "simulate the NiO compound in its ionic state". If NiO would be a ionic compound then DFT would (hopefully, when set up properly) calculate an electron density with a lot of weight at O and a lot less at Ni as a RESULT. The simulation of any property one wishes to study can proceed from there. And if the electron density of NiO does not really resemble the ionic picture, why use the ionic model to simulate things? Best regards, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 22.08.2017 20:45, schrieb Víctor Luaña Cabal: On Tue, Aug 22, 2017 at 07:00:07PM +0200, pieper wrote: DFT in general and Wien2k especially are there to tell you with remarkably high precision what the charge distribution in a given structure actually looks like. Thats FAR better than any hand-waving Ni is 2+, O 2-. If you don't like what DFT tells you and want to draw some fictitious ionic representation of NiO, take xcrysden, plot the structure, and add labels saying that the (maybe) red circles are Ni2+ and the (maybe) blue circles are O2-. It should be obvious to you from your lectures and readings on solid state physics that the total energy depends on Z. If you change the nuclear charge Z you change the element at that position. One can adjust the number of electrons and their starting distribution in Wien2k, but I plainly won't tell you how. You will learn much more if you find out yourself - start with the User Guide and a solid state physics text book. martin, Nothing against your description and I agree basically with everything. However, the electron density (\rho) of a molecule or solid is a real and mensurable property that exists in 3D real position space, so the topology of \rho can be examined properly: that's Bader topological analysis. A mostly physical and matematical description of molecules and solids. Similarly, there are other scalar properties that are physically well defined. By Bader topological analysis we know that some systems are truly ionic in nature, like alkali halides or simple oxides, but that is not any surprise and crystallographers have used that since Laue and the Braggs. Other materials are more interesting, however, and pressure (thermodynamics, in general) modifies behavior of BP from being mostly boron phosphide to phosporus boride, including a pressure range in which electrons behave as independent chemical items and we have an electride. [Phys. Rev. B 63 (2001) 125103, Polarity inversion in the electron density of BP crystal, Paula Mori-Sánchez et al.] In molecules, for instance, we have the Electrostatic potential as a well defined scalar property. In some ways it brings us from XX century (the century of quantum mechanics and electron density, quite complex due to the effect of correlation and basis sets) to XIX century with Maxwell and Faraday (positive and negative charge distributions are what we need to know). Unfortunately, in solids there is a real indetermination with the zero of the EP, that it is basically arbitrary, as much as I know. Nice subject, if you let me tell, by the way. Best regards, Víctor Luaña -- . ."Half of the US people use twitter to form its opinion and half / `' \ also elect the US president. I only hope they are not the same /(o)(o)\ half". --From a sentence by Gore Vidal /`. \/ .'\ / '`'` \ "[Technocrats, academics and journalists} are often motivated | \'`'`/ | by hope while the public at large tends to be more focused | |'`'`| | on fear." -- Russell Mead (The Washington Post, 2017) \/`'`'`'\/ ==(((==)))===+=== ! Dr.Víctor Luaña, in silico chemist & prof. ! ! Departamento de Química Física y Analítica ! ! Universidad de Oviedo, 33006-Oviedo, Spain ! ! e-mail: <vic...@fluor.quimica.uniovi.es> ! ! ph
Re: [Wien] How to simulate the ionic state of a compound?
DFT in general and Wien2k especially are there to tell you with remarkably high precision what the charge distribution in a given structure actually looks like. Thats FAR better than any hand-waving Ni is 2+, O 2-. If you don't like what DFT tells you and want to draw some fictitious ionic representation of NiO, take xcrysden, plot the structure, and add labels saying that the (maybe) red circles are Ni2+ and the (maybe) blue circles are O2-. It should be obvious to you from your lectures and readings on solid state physics that the total energy depends on Z. If you change the nuclear charge Z you change the element at that position. One can adjust the number of electrons and their starting distribution in Wien2k, but I plainly won't tell you how. You will learn much more if you find out yourself - start with the User Guide and a solid state physics text book. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 22.08.2017 12:19, schrieb Abderrahmane Reggad: Dear wien users I want to simulate the NiO compound in its ionic state. We know that the Ni and O atoms have in their neutral state the atomic numbers 28 and 8 respectively. Now, if we want to study the ionic state of the compound Ni+2O-2, should we modify the atomic numbers to Z=26 for Ni and Z=10 for O, or the wien2k code do the operation instead of us? NB: When I change the Z number for Ni atom I found a change in the energy value. Best regards ___ 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/index.html
Re: [Wien] Questions about Na2IrO3 (PBE+SOC+U calculation with magnetic order)
Of course! The local moments in an AF insulator can be anything. I just saw that there is only one type of magnetic atom in the unit cell (Ir) and did not think properly. Sorry for that one. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 07.08.2017 09:18, schrieb Peter Blaha: The Ir moments are read from case.scf file, but why should it be near an integer, and what is the unit of the reported moments in case.scf file? If the unit is bohr magneton, then for a spin 1/2, should it be ~1.7? For a FERROmagnetic insulator, the TOTAL spin moment/cell must be integer. The moments of individual atoms can have any value. For an ANTIferromagnet, the total spin moment/cell must be zero, and of course the atomic moments can again have any value. What you get by :MMT or MMIxxx are spin moments only. The moments are given in Bohr magnetons (mu_B), but are the spin moments, not the effective moments. Thus if you have one unpaired electron the spin moment is one. PS: For your compound, the orbital moments are probably as important as the spin moments and you must add them to get the total moment. The gap can be roughly seen along the k-path I plotted and is checked by DOS plot, which shows it's an insulator. Yours sincerely, Hung-Yu On Sun, Aug 6, 2017 at 11:03 AM, pieper <pie...@ifp.tuwien.ac.at <mailto:pie...@ifp.tuwien.ac.at>> wrote: A few thoughts on this: ad 1) Presumably Ir is the magnetic ion, so what you describe seems to be ok with me. I did not check your .struct file, however. ad 2) See chapters 4.5.6 and 7.4 of the UG (Wien2k 17.1) on ther orb-program, and the references therein. Personally I found the .pdf file of his talk on LDA+U very usefull that E. Assmann posted on the Wien2k-site. Note the recommendation in the UG for the SIC-mode of orb for of strongly correlated systems: set J=0 and use only U_eff=U-J. The value of U_eff is something you will have to decide, perhaps based on the approach described by Madsen and Novak cited in the UG. You also might want to take a look at eece as an alternative (UG chapter 4.5.7) >From many comments here in the mailing list and from the UG (again e.g. chapter 5.5.7, 4.5.8) I take it that PBE is what you should do if you want to calculate spacial charge and spin distributions, but to calculate gaps you may have to switch to numerically much more costly hybrid methods. You probably can (mis)use U_eff as a free parameter to adjust the gap in your PBE calculation to your favorite value. However, the physical meaning of the value would be dubious (imho), and there is no guarantee that the Ir-moments simultaneously come near your favorit 'theoretical moments' (whatever the actual value and origin of those is). ad 3) Don't bother with the starting values of local moments for atomic configurations, and for an antiferromagnet the interstitial moment obviously should stay close to zero. But the moments you give in your table are very far from integer. Are this Ir-spin moments from case.scf or did you add orbital moemnts calculated by lapwdm? The spin moments should be somewhere near integer for an insulator. So, how did you determine the gap in the table? Did you plot a DOS? Is this really an insulator, or are there in fact bands crossing E_F? You might severly misjudge the (direct?) gap depending on where in k-space it is and the points in your k-list. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 <tel:%2B43-%280%29316-380-8564> Am 04.08.2017 19:22, schrieb Hung Yu Yang: Dear WIEN2k developers and users, I am trying to do a calculation on Na2IrO3, which has a band gap ~340meV and a zigzag antiferromagnetic order in its ground state, and I have some questions as follow: 1. To assign the zigzag antiferromagnetic order, what I did is that I first made a cif file that has two inequivalent Ir atoms, and let WIEN2k decide the symmetry for me. I adopted the generated structural file (attached in this mail), checked the cif file in some visulization software and made sure they were structurally equivalent except that there were 2 inequivalent Ir atoms instead of 1. Is this the proper way to generate structural files for magnetically ordered systems? 2. After generating the desired structural file, I put up on Ir1 atom and down on the other (Ir2), used several different combinations of U and J, and I got the following results: U(eV) J(eV)
Re: [Wien] Questions about Na2IrO3 (PBE+SOC+U calculation with magnetic order)
A few thoughts on this: ad 1) Presumably Ir is the magnetic ion, so what you describe seems to be ok with me. I did not check your .struct file, however. ad 2) See chapters 4.5.6 and 7.4 of the UG (Wien2k 17.1) on ther orb-program, and the references therein. Personally I found the .pdf file of his talk on LDA+U very usefull that E. Assmann posted on the Wien2k-site. Note the recommendation in the UG for the SIC-mode of orb for of strongly correlated systems: set J=0 and use only U_eff=U-J. The value of U_eff is something you will have to decide, perhaps based on the approach described by Madsen and Novak cited in the UG. You also might want to take a look at eece as an alternative (UG chapter 4.5.7) From many comments here in the mailing list and from the UG (again e.g. chapter 5.5.7, 4.5.8) I take it that PBE is what you should do if you want to calculate spacial charge and spin distributions, but to calculate gaps you may have to switch to numerically much more costly hybrid methods. You probably can (mis)use U_eff as a free parameter to adjust the gap in your PBE calculation to your favorite value. However, the physical meaning of the value would be dubious (imho), and there is no guarantee that the Ir-moments simultaneously come near your favorit 'theoretical moments' (whatever the actual value and origin of those is). ad 3) Don't bother with the starting values of local moments for atomic configurations, and for an antiferromagnet the interstitial moment obviously should stay close to zero. But the moments you give in your table are very far from integer. Are this Ir-spin moments from case.scf or did you add orbital moemnts calculated by lapwdm? The spin moments should be somewhere near integer for an insulator. So, how did you determine the gap in the table? Did you plot a DOS? Is this really an insulator, or are there in fact bands crossing E_F? You might severly misjudge the (direct?) gap depending on where in k-space it is and the points in your k-list. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 04.08.2017 19:22, schrieb Hung Yu Yang: Dear WIEN2k developers and users, I am trying to do a calculation on Na2IrO3, which has a band gap ~340meV and a zigzag antiferromagnetic order in its ground state, and I have some questions as follow: 1. To assign the zigzag antiferromagnetic order, what I did is that I first made a cif file that has two inequivalent Ir atoms, and let WIEN2k decide the symmetry for me. I adopted the generated structural file (attached in this mail), checked the cif file in some visulization software and made sure they were structurally equivalent except that there were 2 inequivalent Ir atoms instead of 1. Is this the proper way to generate structural files for magnetically ordered systems? 2. After generating the desired structural file, I put up on Ir1 atom and down on the other (Ir2), used several different combinations of U and J, and I got the following results: U(eV) J(eV) Ueff=U-J (eV) Moment(Ir1, Ir2) Gap(meV) 2.1 0.6 1.5 0.22532,-0.22439~700meV 2.4 02.4 0.30105,-0.30109~750meV 30.6 2.4 0.23225,-0.23235~900meV 31.5 1.5 0.17203,-0.17210~900meV First, in this test, it seems that the gap is affected by U only, not Ueff=U-J. Second, the moments of Ir seemed to be closely related to J. In this situation, what is the proper way to assign U and J? I am not sure how much it means if I just try to tune U and J until they match the experimental gap and/or theoretical moments. 3. I also have a question about the unit of moment given here; I understand that the moments depend on the RMT sizes and I assumed the unit is in bohr magneton, but the first value shown for Ir atom is 0.76697 and I am not sure in what unit is this value and how it assigns the initial moment for a certain element (say, Ir in my case). Also, the interstitial moments are nearly 0 among all the cycles. For this calculation, kpoints=450, RKmax=8 and I had to use the TEMP scheme for it to converge. I appreciate any reply from you. Yours sincerely, Hung-Yu ___ 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/index.html
Re: [Wien] Spin part of the magnetic susceptibility
There have been several comments on this. I can only guess at what exactly you want to calculate, but to me it seems that calculating a (zero-field) spin-resolved DOS would be a good start. Then you shift the spin-up and spin-down parts rigidly by mu_B*B by hand, and count how many mu_B you get per Tesla field. If you want/need anything more elaborate you will have to provide more information on what kind of measurement in what kind of ferromagnetic metal or semiconductor you actually are interested in (and/or read Ashcroft & Mermin). --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 21.07.2017 13:33, schrieb Wien2k User: Dear Wien2k Users;I am still waiting for an answer from Prof P.Blaha or Wien2k users 2017-07-20 2:03 GMT+02:00 Wien2k User <wien2k.u...@gmail.com>: dear wien2k users; How to calculate the spin part of the magnetic susceptibility for a ferromagnetic metal since the magnetic moment is different from zero. ___ 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/index.html
Re: [Wien] magnetic susceptibility for a ferromagnetic metal or for a ferromagnetic semiconductor
Note that the internal field in a metal (or an insulator) is not homogeneous. This is because of the inhomogeneous distribution of electrons. You don't have a homogeneous distribution of induced magnetic moments, and with that the susceptibility also is inhomogenous. chi is chi(r), a function of where in the material you do your measurement. The property of metals calculated with very impressive accuracy by Laskowski and Blaha and their NMR module of Wien2k is this LOCAL susceptibility at the position of the nuclei, chi(r=r_atom1), ... chi(r=r_atomn). This is what you measure in nuclear magnetic resonance, NMR, hence the name of the module. I guess the 'molar magnetic susceptibility' you want to calculate is what you get if you put the material in a (perhaps SQUID-) magnetometer: The average of chi(r) over the whole sample volume. From Peter Blahas comments here I take it that one could calculate this average on basis of the results from the NMR module: The spin and current distributions are there. My guess is that it will be not as trivial as it might look at first sight. For example I seem to recall that he mentioned that this is difficult to converge - at least in normal metals its a small effect, after all. This is not meant to discourage trying, just expect problems. The NMR module is, however, for 'normal' metals. My warning was about high expectations in calculating the macroscopic, homogeneous susceptibility of ferromagnets (or other magnetically oŕdered materials). Wien2k gives invaluable information about the size of local magnetic moments, their exchange energies, moments of itinerant electrons in a band structure, ... with the caveats mentioned earlier even on the influence of a magnetic field. However, the susceptibility is about how the applied field couples the ground state determined by DFT to any low lying excited states - and especially in magnetically ordered systems these excited states might well be beyond the scope of DFT. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 20.07.2017 18:12, schrieb karima Physique: Thank you very much for your detailed answer What I understood from what you wrote is that the DFT does not accurately estimate the magnetic susceptibility. honestly what is encouraged me to take an interest in this property is the paper of Prof. Robert Laskowski and Prof Peter Blaha (doi: 10.1021/acs.jpcc.5b05947 [3]) where they compared their results for many metals and found values very similar to the experimental ones. Now I do not know is what the estimate of molar magnetic susceptibility is possible with wien2k or its estimate is not always accurate. 2017-07-20 16:54 GMT+02:00 pieper <pie...@ifp.tuwien.ac.at>: From what I have understood from userguide and Prof P Blaha's replies; For semiconductor and insulator; there is the orbital part of the magnetic susceptibility only but for the metals there is also the spin part No! This is not correct. Let me expand a little on my view of this topic in the hope to clarify the problem at least somewhat: In a nutshell, you want Wien2k to find the state minimizing the total energy of some solid because you suspect that this is the state whose properties you observe in some experiment, and someone told you that DFT is the method of choice to do the search. Magnetic moments contribute M*B to that energy with M and B being vectors in a scalar product. In a first approximation two contributions from electrons to M can be considered separately, namely the spin moment, and the moment from an orbital momentum (or the current associated with a non-zero orbital momentum). Both parts contribute to the energy in the field. That is why Peter Blahas very first statement was: A magnetic field influences the spin and orbital degrees of freedom. It always does. Both contributions to M. No matter wether you have an insulator or a metal. The influence on specific properties you are interested in may be too small to bother, but somewhere else it will show up. There are different ways to tell Wien2k where to look for the state with minimum energy in an applied magnetic field (or electric field). This is where symmetry enters stage: Wien2k searches the state with lowest energy in the subspace compatible with the symmetries specified by the input files. The symmetry of the crystal structure is in the struct file. Concerning magnetic moments you can tell Wien2k to - consider only states that are symmetric with respect to spin, i.e. do a non-spin-polarized calculation. Doesn't make sense if you are interested in the effect of a magnetic field or want to model ferromagnets, but for the majority of materials and properties it's perfect to save time and calculate just spin +1/2, then carry the result over to spin -1/2. - take into account the two spin channels separately, i.e. do a spin-polarized calcu
Re: [Wien] magnetic susceptibility for a ferromagnetic metal or for a ferromagnetic semiconductor
From what I have understood from userguide and Prof P Blaha's replies; For semiconductor and insulator; there is the orbital part of the magnetic susceptibility only but for the metals there is also the spin part No! This is not correct. Let me expand a little on my view of this topic in the hope to clarify the problem at least somewhat: In a nutshell, you want Wien2k to find the state minimizing the total energy of some solid because you suspect that this is the state whose properties you observe in some experiment, and someone told you that DFT is the method of choice to do the search. Magnetic moments contribute M*B to that energy with M and B being vectors in a scalar product. In a first approximation two contributions from electrons to M can be considered separately, namely the spin moment, and the moment from an orbital momentum (or the current associated with a non-zero orbital momentum). Both parts contribute to the energy in the field. That is why Peter Blahas very first statement was: A magnetic field influences the spin and orbital degrees of freedom. It always does. Both contributions to M. No matter wether you have an insulator or a metal. The influence on specific properties you are interested in may be too small to bother, but somewhere else it will show up. There are different ways to tell Wien2k where to look for the state with minimum energy in an applied magnetic field (or electric field). This is where symmetry enters stage: Wien2k searches the state with lowest energy in the subspace compatible with the symmetries specified by the input files. The symmetry of the crystal structure is in the struct file. Concerning magnetic moments you can tell Wien2k to - consider only states that are symmetric with respect to spin, i.e. do a non-spin-polarized calculation. Doesn't make sense if you are interested in the effect of a magnetic field or want to model ferromagnets, but for the majority of materials and properties it's perfect to save time and calculate just spin +1/2, then carry the result over to spin -1/2. - take into account the two spin channels separately, i.e. do a spin-polarized calculation, and introduce B as 'orbital potential' in the program orb. Note that a field you put into .inorb acts on the electrons in the single atomic shell you specify, and only for the part within the muffin-tin radius. Everything else is indirectly adjusted according to interactions. I am (of course) with Peter Blaha: One can expect very reasonable estimates for field induced moments at the given shell, at least if the field induced polarization of electrons not affected by .inorb is small. The option also shows a Zeeman splitting of spin-up and -down bands due to the interactions, but I expect that depending on the case this can rapidly turn into a paedagogical example rather than an estimate of the spin susceptibility. - take into account the fact that the energy contribution M*B quite possibly reduced symmetry. M as well as B are vectors, after all, and transform as such under symmetry operations. You can switch on spin-orbit interaction to tell Wien2k about this. (Be aware that Wien2k always assumes that M and B are colinear). As Gerhard Fecher pointed out, a reduced symmetry in a magnetic field is NOT taken into account in the orb program. With only the orb option active Wien2k may search the needle (the energy minimum) in a completely wrong haystack (subspace). - do a calculation of the linear response of the all-electron wave function (representing spins AND orbital moments) to an applied field. As I understand it this is what the NMR module does. However, it calculates only the local response at the nuclei, not the integrated macroscopic susceptibility. Note that all this does not represent anything like the macroscopic susceptibility you will detect with a ferromagnetic (or any other magnetically ordered) material in your magnetometer. The susceptibility in these cases even at zero Kelvin (where DFT arguably applies) is dominated by things the spins can and will do, but Wien2k is completely unaware of (magnons, domain walls ...). Finally, don't take the UG or this mailing list as a substitute for a textbook. That is not their intention. Follow the advice of Prof. Fecher and read up on the subject of magnetism in solids. Perhaps start with something on solid state physics in general, not a specialized treatment of magnetism. Personally I like the introduction to Solid State Physics by Ashcroft and Mermin. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 19.07.2017 16:02, schrieb karima Physique: From what I have understood from userguide and Prof P Blaha's replies; For semiconductor and insulator; there is the orbital part of the magnetic susceptibility only but for the metals there is also the spin part and I
Re: [Wien] question
That is VERY little information you give! I suspect you are an inexperienced user trying to work from the command line without having worked at least through the examples of the user guide. You don't tell which program you started but the message tells you that its the wrong program: it has no -orb and -dm switches. I suggest you use the w2web interface and the examples in the UG to make yourself more familiar with the package ... Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 14.07.2017 16:10, schrieb Dj Fati: Could you please suggest an answer to my question When i do a calculation on the method GGA+U it gives me error option-orb does not exixt option- dm does not exist thanks ___ 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/index.html
Re: [Wien] Questions about imposing external magnetic field on no-magnetic system
In case no one has answered this up to now: ad 1) The procedure itself is ok. You might want switch on SO first and converge that without the orbital potential to establish a zero-field base line. Remember to put in LARGE fields - your off-the-shelf lab field of 10 T will not show up at any energy precision you can achieve. Estimate the energy of 1 mu_B in 10 T field in Ry units to see that. Note that your not-so-recent version of Wien2k is not the best for the task. The latest version is 17.1. With 16.1 came the NMR package which should be much better suited to calculate the effects of a magnetic field. ad 2) If you apply a magnetic field experimentally in the lab you do it at all atoms. I suppose you want to model that situation. imho it makes little sense to exempt one or two of your atoms from the field. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 10.07.2017 12:20, schrieb Peng Bingrui: Dear professor Blaha and WIEN2K users I'm running WIEN2K of 14 version on Linux system. I'm going to impose external magnetic field on LaPtBi, a no-magnetic material. The procedure that I'm going to use is : 1、Do a no-SO calculation : runsp_c_lapw. 2、Do a SO calculation : runsp_c_lapw -so -orb, while including external magnetic field as orbital potential in the same time. My questions are: 1、Whether this procedure is OK ? If it is not OK, what is the right one ? 2、Which atoms and which orbitals should I treat with orbital potential ? The electron configurations of these 3 atoms are: La (5d1 6s2) ; Pt (4f14 5d9 6s1); Bi (4f14 5d10 6s2 6p3). Thanks very much for your attention. Sincerely yours, Bingrui Peng from the Department of Physics, Nanjing University, China ___ 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/index.html
[Wien] minor problems in 17.1 with w2web and siteconfig_lapw
Installation of Wien2k 17.1 went very smoothely, thanks a lot for the new siteconfig_lapw! However, I encountered a small problem when I wanted to point the linker to the correct LIBXC dir: Make of my LIBXCD Version (dowmnloaded June.2017) produced a dir structure (LIBXCrootpath)/lib64 and /include, while siteconfig_lapw has a hard-wired (LIBXCrootpath)/lib and /include. When I finally figured this out it was easy to change the dir name from /lib64 to /lib, but ... And w2web gives me difficulties with save_lapw: there is no way to the default save_lapw (all input files, clms etc). If all option boxes are left untouched it uses the last possibility in the list (I forgot which one that is). save_lapw from the terminal works just fine. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] Questions in "initso_lapw" when using "runsp_c_lapw" to run my task
A Hubbard U does not make much sense in a non spin polarized calculation - and I seem to recall that this is mentioned in the UG. initso_lapw is, on the other hand, for spin-orbit interaction - there the question if you have a spin polarized case makes sense. This is also explained in the UG. So the general advise seems to be stiduy the UG - and if possible some of the references therein. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 08.07.2017 17:08, schrieb Peng Bingrui: Dear wien2k users I have a question that I'd like to discuss with you. My purpose is to use LDA+U to treat no-magnetic system. So I use "runsp_c_lapw -orb" to run my task, as told by the user_guide. And in "initso_lapw", it asks me "do you have a spinpolarized case ?" and "do you want to use the new structure for SO calculations ?". I answer "Yes" to both the two quetions. I'm not very sure whether what I do is right. I'd like to listen your experience and advices about this. Thank you very much for your attention. Bingrui Peng from the Department of Physics,Nanjing University,China ___ 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/index.html
Re: [Wien] Convergence problem in mbj potential calculation
You keep getting back to this match between the (presumably powder-)xrd you calculated from your .struct file and the experimental one. This does NOT imply that your structural model describes the physical situation correctly! With your - in Gerhards words - 'interesting approach' you should be able to produce any number of different .struct files, and calculating their powder-xrd find a similar 'match with the experimental one'. Did you calculate the xrd with any of the alternative distributions of Li, Ni and Ti on their sites in your unit cell? Did you try the same with the many, many other distributions you can generate in larger supercells? I am prettty sure you will find something that matches your xrd data. It will make as much sense as what you have now ... Be aware of the fact that structural models of Wien2k have symmetry properties different from your physical situation: in your model the metal atoms occupy lattice sites regularly, breaking certain symmetries of the underlying lattice. The actual distribution in your case is (apparently more or less) random, which overall breaks no symmetry, but locally breaks every symmetry somewhere. If the distribution of Li, Ni, and Ti really is random on one shared site I know of only one (expensive) way to proceed: try to reduce the influence of short range periodicity in your models by increasing the size of the supercell and putting Li, Ni, and Ti at random positions. Do structural relaxations, look what changes in your (GGA-) calculations, and keep fingers crossed that things you are interested in converge for supercells of a size you still can handle. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 15.06.2017 07:12, schrieb shamik chakrabarti: Dear Gerhard, You are right. The system can be written as (Li0.5Ti0.25Ni0.25)O. However, my mbj simulation is converged with MSR1 mixing in 221 cycles. And the DOS has come as metallic in this case too. Now, I may try -eece also. So far, in GGA, GGA+U and in mbj the DOS has come as metallic. However, if the structure is wrong it will not give the correct DOS in any case. Although, the xrd is matched with experimental one, there is an issue of longer Ti-O bond. I have to see how can I get the correct structure. Any suggestion in this regard will be helpful for us. with regards, with regards, On Wed, Jun 14, 2017 at 8:00 PM, Fecher, Gerhard <fec...@uni-mainz.de> wrote: Interesting approach as in Li2NiTiO4 the Li, Ni, and Ti atoms are randomly distributed on the 4a site with oxygen on 4b of the NaCl lattice therefore one better should write (Li0.5Ti0.25Ni0.25)O Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von shamik chakrabarti [shamik...@gmail.com] Gesendet: Mittwoch, 14. Juni 2017 15:39 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Convergence problem in mbj potential calculation Dear Gerhard & Laurence, Thank you very much for your reply & suggestions. The structure was initially of Li2NiTiO4. The space group is Fm-3m. We have created a 1x1x1 supercell (with target lattice primitive) and then distributed Li, Ni & Ti with 2:1:1 ration at 4b site. 4a site contains oxygen. Then we have optimize the volume and force per atom of the structure sequentially & results in the present structure. The xrd generated by the present structure matches exactly with the experimental xrd of Li2NiTiO4. However, we have shifted the mixing procedure from PORT to MSR1 and the simulation is converging well now. with regards, On Wed, Jun 14, 2017 at 3:17 PM, Laurence Marks <l-ma...@northwestern.edu<mailto:l-ma...@northwestern.edu>> wrote: I have said this before, but it is worth repeating: the most common reason a calculation does not converge is because the structure/model is unphysical/unchemical. Your Ti-O bonds are too long, the structure is "obviously" wrong. Scientific rigor is not optional. What is the R-factor? Are the temperature factors sane? Is there an independent measurement of the composition, e.g. ICP? Have you measured that it is FM? --- Professor Laurence Marks "Research is to see what everybody else has seen, and to think what nobody else has thought", Albert Szent-Gyorgi http://www.numis.northwestern.edu [1] Corrosion in 4D http://MURI4D.numis.northwestern.edu [2] Partner of the CFW 100% gender equity project, www.cfw.org/100-perce
Re: [Wien] Import error, no module named numpy
The Traceback error message is from python. Its trying to do what berrypi tells it to do, so module berrypi is found and probably installed ok. However, berrypi wants python to import numpy - and that is not found. Most probably something went wrong when you installed numpy (or mathplotlib?). Try in a terminal to run python and import the module manualy. Enter at your user prompt: user-prompt:~> python Python 2.7.12 (default, Jun 28 2016, 06:57:42) [GCC] on linux2 Type "help", "copyright", "credits" or "license" for more information. import numpy # math library This probably will not work as shown here for my terminal, meaning that numpy is either not installed, or in some unexpected place and you will have to point python to that place. You will have to look into what you did to install numpy (or a module including it, like mathplotlib). Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 20.05.2017 19:43, schrieb Jameela Fatheema: Dear all, I am running wien2k version 14 on ubuntu 16.04 The purpose of my calculations is to get born effective charges for AlN. The scf has been run successfully for 1000 kpoints. And I am trying to run the command “berrypi -k 6:6:6” but I am getting the following error. ubuntu@ubuntu-7G-Series:~/project/AlNitride/12may$ berrypi -k 6:6:6 Traceback (most recent call last): File "/home/ubuntu/WIEN2k_14/SRC_BerryPI/BerryPI/berrypi", line 20, in import numpy # math library ImportError: No module named numpy I have tried to make sure that python , numpy , berrypi and wannier90 have been installed correctly but the error remains. I would like to know how to solve this problem. Kindly if anyone can help. Thank you Regards Jameela Fatheema Department of Physics University of Peshawar ___ 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/index.html
Re: [Wien] paramagnetic or diamagnetic
In general I am as afraid as Gerhard Fecher that this question is at least very difficult to answer. By definition one needs to calculate the low field response of the electrons, and there is a whole bunch of contributions, from the single electron orbital and spin-Pauli contributions Peter mentions to quasiparticle contributions that are in my understanding simply not present in the ground state calculated by DFT. And there are questions of ground states with spin corelations, temperature and field dependencies leading to changes from overall paramagnetic to diamagnetic response ... Reading Peters response I wonder about even a single electron contribution and the NMR package in the case of metals: Does this package calculate only the (paramagnetic) Pauli spin susceptibility? Or is their Landau diamagnetism also included? For free electrons it amounts to 1/3 of the Pauli spin susceptibility (see your favorite textbook on the theory magnetism), so it is by no means safe to simply assume its small. Greetings, Martin Pieper Am 08.05.2017 20:19, schrieb Peter Blaha: In an insulator/semiconductor you have only the orbital part of the susceptibility. This can be calculated using our NMR package and such a material will be diamagnetic. In metals you have in addition a spin suszeptibility, which you can trivially calculate using spin-polarized calc. and an external field. Usually this part is paramagnetic. And then you have to see, which part dominates See also our NMR package. Am 08.05.2017 um 16:28 schrieb Fecher, Gerhard: I am afraid that this question can not be answered and I doubt if any answer on this can be generalised to all kinds of materials. As an experimentalist my answer will be: measure the susceptibility and it will tell you what your material is. As you do not apply any magnetic field in your (non-spinpolarized) calculation, the induced magnetic moment will be zero and a) tells you that this is true for both, diamagnetic or paramagnetic What about b) ? I tried it for Pt and indeed I find that the application of a magnetic field induces a magnetic moment (spin polarized calculation !) that is parallel to the applied field, and linearly dependent on its size, as expected for a paramagnet. However, I did not check whether the electrons in the closed shells behave diamagnetic as they should. I doubt that this will work for all materials as in most cases the induced moment will be just to low to decide even if you use brute force (very high field, very much k-points etc.) If a ferro- or other "magnetic" solution is close, then the application of the field may break the symmetry in such a way that you run into this state instead of staying in the paramagnetic state. Diamagnetism will probably not bee seen in Semiconductors. You may try semimetallic graphite which is a "strong" diamagnet to see whether it is possible to see any antiparallel allignment of induced magnetic moments. I did not further check, maybe there are some codes available to calculate the suscebtibility of para- or diamagnetic materials. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von karima Physique [physique.kar...@gmail.com] Gesendet: Montag, 8. Mai 2017 14:48 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] paramagnetic or diamagnetic Thank you very much for your answer I started a calculation in several magnetic phases (non-magnetic, ferromagnetic and antiferromagnetic) and I found that the non-magnetic phase is the most stable. so how can I know if the studied material is a paramagnetic or diamagnetic material? Thank you in advance 2017-05-08 8:06 GMT+02:00 Fecher, Gerhard <fec...@uni-mainz.de<mailto:fec...@uni-mainz.de>>: What distinguishes a paramagnetic from a diamagnetic material ? a) at zero magnetic field the induced magnetic moment is zero for both b) at external magnetic field the induced magnetiuc moment is parallel / antiparallel to the applied field. c) both is true d) none is true There was already a discussion about paramagnetism, see https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15029.html Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - Universi
Re: [Wien] change of lattice parameters while alloying
Accept the suggested structure. I always restart with initailization the new structure from nn. If you replace one of the atoms occupying a position with multiplicity .gt. 1 (with more than one position line in struct) you break the symmetries inducing that multiplicity. You may well wind up with a larger primitive unit cell volume. That will change the basis vectors a,b,c. Note that depending on what you want to compare in your series of calculations it probably is a good idea to do all calculations in the same space group - even if you are left with very low symmetry. Do this by giving all Cr in the pristine cell you want to substitute a differnet index (Cr1, Cr2 ...). And remember that you do not calculate an (random) alloy, you calculate a periodic structure. Watch out for superstructure effects, perhaps by checking larger supercells with the same Br-concentration but different configuration. Martin Pieper Am 09.05.2017 01:55, schrieb Dr. K. C. Bhamu: Dear Wien2k Users I am alloying a quaternery compound: replacing Cr with Br one by one. What I see is, the pristine compound is cubic with 225 SG. When I replace Cl by Br one by one then lattice parameters deviates from cubic to different structure. like a=b, c or a,b,c. SG also changes. During initialisation I accepted the new structure suggested by the w2web. Should I accept the suggested structure? As per my understanding from previous posts from mailing list we should accept the suggested structure which is with lower symmetry. Is the change of lattice parameters is normal behaviour of material or I am doing any mistake? Sincerely Bhamu ___ 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 --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] DOS of slab
Well, the colleague 'desires to obtain the DOS of the just top monolayer'. He (or she) might dodge your "philosophical" question of what space is occupied by the topmost layer since the DOS is, after all, not a volume density but an energy density. Even if he solves all your questions and specifies what space qualifies as topmost layer he still has to decide which states contribute how much. Counting electrons in that volume does not solve the problem as stated. For example, what about the unoccupied DOS above EFermi? The no less "philosophical" question then is to specify which states he wants to give the honor of being part of team 'topmost layer' - then to project on them. And my best wishes for the most difficult part: convince his audience that the choice is convincing for the point to be made. ;-) Martin Pieper Am 08.05.2017 08:29, schrieb Fecher, Gerhard: This might give rise to a "philosophical" question: What is the space ocupied by the topmost layer ? To use the DOS of the atoms in the "topmost" layer will neglect the interstitial between the atoms, that is, the DOS will be not the one of the complete "topmost" layer but only the one inside the MT spheres around the atoms that are defined to be in a selected layer. (Note, already for bulk materials, the total DOS is not the sum of the atomic DOSs) However, to what layer does the interstitial belong ? What if the surface is buckled ? And further, is the vacuum part of the "topmost" layer or not ? I wonder whether the Critic 2 code of the Oviedo Group in Spain can help by using a Baader analysis, and to calculate the DOS inside of the space filling basins instead of spheres. This might improve the situation but still leaves some questions open. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden ____ Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von pieper [pie...@ifp.tuwien.ac.at] Gesendet: Donnerstag, 4. Mai 2017 10:22 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] DOS of slab Assuming that you meant this to be a question, and that nobody answered up to now: If you want a plot of some partial DOS for a bulk crystal, you select an atom in the unit cell and possibly some orbital. Now you want to calculate the partial DOS of the topmost layer ... did you try (to understand) what happens if you select the atom(s) in your topmost layer? Am 29.04.2017 08:15, schrieb emami seyyed amir abbas: Dear users. I am trying to calculate the electronic and magnetic properties of a thin film. I created a slab of 5 monolayer and run the scf. I desire to obtain the DOS of the just top monolayer but there is no option in DOS program for this task. In DOS program i can plot DOS just for specific atom not layer. best regards. ___ 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 --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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/index.html --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] DOS of slab
Assuming that you meant this to be a question, and that nobody answered up to now: If you want a plot of some partial DOS for a bulk crystal, you select an atom in the unit cell and possibly some orbital. Now you want to calculate the partial DOS of the topmost layer ... did you try (to understand) what happens if you select the atom(s) in your topmost layer? Am 29.04.2017 08:15, schrieb emami seyyed amir abbas: Dear users. I am trying to calculate the electronic and magnetic properties of a thin film. I created a slab of 5 monolayer and run the scf. I desire to obtain the DOS of the just top monolayer but there is no option in DOS program for this task. In DOS program i can plot DOS just for specific atom not layer. best regards. ___ 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 --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] Band gap calculation of C diamond structure
The problem will be in the .struct you started from. You say yourself that the structure does not look like diamond in xcrysden. Don't change the .struct file in a case directory and restart the scf. The calculation relies on consistency of various input files. The standard advice seems to be: start a new case in fresh directory, generate (preferably from w2web) a new struct file, inspect it with xcrysden (there is no meaning in wasting CPU-time for scf-cycles on wrong structures), proceed with the initialization (check the nearest neighbor distances in outputnn to see if you got the units in struct right) and start the scf only when everything went ok. ... and keep fingers crossed --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 13.02.2017 09:18, schrieb Abhilash Patra: Dear Wien2k users, I am running WIEN2k_14.2 on my Lenovo workstation with fedora 23. I want to calculate the band gap of C, Si, and Ge with the diamond structure. To generate my struct file I used: No. of Atoms-1, Space group- 227_Fd-3m and a=b=c=3.567 and one carbon atom at (0,0,0). Then set RMT automatically (reduce RMT--0%). Then it set RMT to 1.18. I initiate calculation with all the default values and run SFC as there is no warning in the STDOUT. But when I see the structure in the Xcrysden the figure is not looking like diamond structure. And the band plot shows no gap : GAP: -9. Ry = -.eV ( metallic ) I have already tried the structure : Atom=1, FCC, a=3.567, one atom at (0,0,0) and other position at (0.25,0.25,0.25). This gives diamond-likee structure but in the STDOUT file it gives some warning like warning: !!! Struct file is not consistent with space group found. Number and name of space group: 227 (F d -3 m) [origin choice 2] and SCF gives error like head: cannot open ‘WIEN2k.inm’ for reading: No such file or directory head: cannot open ‘WIEN2k.inm’ for reading: No such file or directory no WIEN2k.clmsum(_old) file found, which is necessary for lapw0 ! stop error Please suggest me about the struct file which will give the results for all like LDA, PBE and mGGAfunctionals Thanks -- Abhilash Patra Research Scholar(Ph.D.) School of Physical science NISER,BBSR ___ 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/index.html
Re: [Wien] d-eg orbitals are not effected by applying Ueff
From the abstract of the paper you cited in your last mail I gather that this an alloy. So I suspect that there is some cite disorder with Co as well as Ga occupying the lattice sites randomly? In that case Co would have a number of different configurations, due to the random distribution, that are not represented in your .struct file. As for the gap: This was a question, not stating some fact. Unfortunately I did not follow the literature on GGA+U, so I am not aware of any studies with the same problem as yours. But I am also not aware of any study saying that a (reasonably sized) U MUST produce a gap. Here the real experts might provide more useful comments. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 28.01.2017 01:21, schrieb venkatesh chandragiri: Dear Prof. Pieper, the answers are as follows, I UNDERSTAND THAT THIS IS AN ALLOY? IF YES, WHAT KIND OF SUPERCELL DO YOU USE TO MODEL THE SYSTEM? I do not used any supercell configuration. The case.struct file contains the two atoms as given below. ARE THESE T2G AND EG SYMMETRIES ACTUALLY PRESENT IN ALL POSSIBLE CONFIGURATIONS WITHIN THE ALLOY (PROBABLY NOT)? IS IT POSSIBLE THAT THE SYMMETRY YOU ARTIFICIALLY IMPOSE IS AT THE BOTTOM OF THIS? Experimentally, at room temperature, we have refined the crystal structure of the Co50Ga50 alloy and the same used for the generating the case.struct. I do not understand the your meaning of " all possible configurations ". WHY DO YOU BELIEVE THAT U SHOULD ACT IN THE SAME WAY ON DIFFERENT SYMMETRIES? COULD IT BE THAT LOCALIZING T2G PAYS OFF MUCH BETTER THAN EG? I do not have experience before on the DFT +U studies, but have seen in literature that one could create artifical energy gap by applying U in order to explain the insulating behaviour of few oxides. It would be helpful for me, if you provide me some references where the localizing t2g pays off much better than eg by applying U. DID YOU TRY OTHER CONFIGURATIONS / SUPERCELLS / SYMMETRIES? I tried DFT + SO + U , but this also leads to the same feature in DOS like DFT + U case. No supercells are tried. Could you please suggest me next step or possible configurations so that I can give a try. thanks for your reply and looking forward to your help venkatesh === Co50Ga50.struct === Co50Ga50 P LATTICE,NONEQUIV.ATOMS: 2 221_Pm-3m MODE OF CALC=RELA unit=bohr 5.438632 5.438632 5.438632 90.00 90.00 90.00 ATOM 1: X=0. Y=0. Z=0. MULT= 1 ISPLIT= 2 Ga1 NPT= 781 R0=0.5000 RMT= 2.2400 Z: 31.0 LOCAL ROT MATRIX: 1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 ATOM 2: X=0.5000 Y=0.5000 Z=0.5000 MULT= 1 ISPLIT= 2 Co1 NPT= 781 R0=0.5000 RMT= 2.3500 Z: 27.0 LOCAL ROT MATRIX: 1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 48 NUMBER OF SYMMETRY OPERATIONS -1 0 0 0. 0-1 0 0. 0 0-1 0. 1 -1 0 0 0. 0-1 0 0. 0 0 1 0. 2 -1 0 0 0. 0 0-1 0. 0-1 0 0. 3 -1 0 0 0. 0 0 1 0. 0-1 0 0. 4 -1 0 0 0. 0 0-1 0. 0 1 0 0. 5 -1 0 0 0. 0 0 1 0. 0 1 0 0. 6 -1 0 0 0. 0 1 0 0. 0 0-1 0. 7 -1 0 0 0. 0 1 0 0. 0 0 1 0. 8 0-1 0 0. -1 0 0 0. 0 0-1 0. 9 0-1 0 0. -1 0 0 0. 0 0 1 0. 10 0 0-1 0. -1 0 0 0. 0-1 0 0. 11 0 0 1 0. -1 0 0 0. 0-1 0 0. 12 0 0-1 0. -1 0 0 0. 0 1 0 0. 13 0 0 1 0. -1 0 0 0. 0 1 0 0. 14 0 1 0 0. -1 0 0 0. 0 0-1 0. 15 0 1 0 0. -1 0 0 0. 0 0 1 0. 16 0-1 0 0. 0 0-1 0. -1 0 0 0. 17 0-1 0 0. 0 0 1 0. -1 0 0 0. 18 0 0-1 0. 0-1 0 0. -1 0 0 0. 19 0 0 1 0. 0-1 0 0. -1 0 0 0. 20 0 0-1 0. 0 1 0 0. -1 0 0 0. 21 0 0 1 0. 0 1 0 0. -1 0 0 0. 22 0 1 0 0. 0 0-1 0. -1 0 0 0. 23 0 1 0 0. 0 0 1 0. -1 0 0 0. 24 0-1 0 0. 0 0-1 0. 1 0 0
Re: [Wien] d-eg orbitals are not effected by applying Ueff
Not really a comment, just a few questions one might consider (I probably would, but I am half an expert at best): I understand that this is an alloy? If yes, what kind of supercell do you use to model the system? Are these t2g and eg symmetries actually present in all possible configurations within the alloy (probably not)? Is it possible that the symmetry you artificially impose is at the bottom of this? Why do you believe that U should act in the same way on different symmetries? Could it be that localizing t2g pays off much better than eg? Did you try other configurations / supercells / symmetries? Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 26.01.2017 19:17, schrieb venkatesh chandragiri: Dear Prof. Lyudmila Dobysheva Thanks for your reply. There was a old paper on this alloy, which show similar kind of results, ie. deep pseudo gap with finite states at Ef. please see the link below http://iopscience.iop.org/article/10.1088/0305-4608/12/2/009/meta [1] In this article they discussed the raise in resitivity using the Mott theory of electron localization when electrons are captured by these states around Ef, so called localised states. However, I want to apply U for the Co-d states in order to explain the raise in resistivity at low temperatures by creating artificial gap. But although, I applied U= 5 eV on Co-d states, d-t2g states are only moved well below the Fermi level while d-eg are not effected. This leads to presence of finite states from d-eg at Ef. please share any idea you have about this.. Can someone has any another comment on this..? thanks venkatesh Links: -- [1] http://iopscience.iop.org/article/10.1088/0305-4608/12/2/009/meta ___ 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/index.html
Re: [Wien] How to include the localized d orbitals in the atomic spheres?
You should recognize that the local atomic spheres are just a theoretical construct to arrive at a (very good) set of basis functions. It has nothing to do with wether or not the d-orbitals you describe with them are localized or not, or with the size of the gap, the magnetic moment ... It defines just the coordinate system one uses to describe the state. It has as much to do with what the state is like, as the use of cartesian or spherical coordinates has to do with where on earth you are. You also cannot somehow push the electrons out of the intermediate space - it would mean some additional potential with simply is not there in your material. And using spheres you certainely cannot avoid some intermediate space between them. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 30.11.2016 13:28, schrieb Abderrahmane Reggad: Thank you Dr Pieper for the clarification. How much does the including of the localized d orbitals inside the atomic spheres improve the band gap and the magnetic moment ? And how can we include all the localized d orbitals inside the atomic spheres? Best regards Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] Discrepancy in the simulation of the paramagnetic state
My (and probably Xavier's) concern with Regaard's question was something else. I have no problem whatsoever with you finding an approximation for Pt using wave functions. After all, your ground state model has zero static local moments, as has the Pt you want to model. ;-) However, the approximation seems at least dubious if the ground state model and the low temperature state of the material differ. If the material enters some magnetic state and the spin-polarized(!) DFT model does not one might look for a problem with the structure data, some structural phase transition, ... So I am with Xavier, and I would at least advise to be careful with the idea I understood Regaad did somehow get: Artificially compensate spins (e.g. via LDA instead of LSDA) to find an approximation for the paramagnetic phase at elevated temperature of a low temperature magnet. There is at least one difference between the material and the model: the model will NOT be paramagnetic (obtain a positive magnetization in an applied magnetic field). Wether or not this (or any other differences induced by the forced spin compensation) poses a problem will depend on what situation one wants to model. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 28.11.2016 08:33, schrieb Fecher, Gerhard: I hope you agree that Pt is paramagnetic I did two calculations for Pt, one was spin polarized the other not. The results are identical, no resulting magnetic moment (indeed, I started with one in the spin polarized case), did I play a trick or did Wien2k play a trick ? but may be Wien2k can not be used to calculate the electronic structure of Pt, because it is paramagnetic (Pt, not Wien2k !). I hope you agree that Pt is paramagnetic even at Zero temperature. why do I need to include temperature effects to calculate the ground state of Pt (at 0 K, where else) ? ... and what should MtC calculations tell me about it ? Remark 1: Calculations may be "spin polarized" (LSDA) or not (LDA) or they may be even more sophisticated "non-colinear spin polarized" or they may be for "disordred local moments" or for "spin spirals", or ???, just to name some. Remark 2: Materials may be diamagnetic, paramagnetic (Langevin, Pauli, van Vleck), ferromagnetic (localised moments, itinerant), ferrimagnetic (collinear, non-collinear), etc.. Therefore, I repeat my question: How do you distinguish diamagnetic, paramagnetic, ferromagnetic, and ... states ? The answer is for you, not for me. I tried to calculate for Pt using Hohenberg Kohn DFT, but I could not find the functional, all I found was some approximation using wave functions. Don't worry I will not ask a question about it ;-) Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Xavier Rocquefelte [xavier.rocquefe...@univ-rennes1.fr] Gesendet: Sonntag, 27. November 2016 12:46 An: wien@zeus.theochem.tuwien.ac.at Betreff: Re: [Wien] Discrepancy in the simulation of the paramagnetic state Just to add one more point to this funny discussion, the term "paramagnetic" is sometimes used in the DFT litterature in an improper way. It could clearly lead to misunderstanding for researchers who do not know so much on how magnetic properties could evolve with temperature and applied magnetic field. When you see in a paper "paramagnetic state" simulated using DFT ... it is NOT paramagnetic at all, it is simply a trick which must be considered with care as previously mentionned by Peter, Eliane and Martin. If you want to simulate a paramagnetic state you need to include the temperature effects, i.e. you should consider the spin dynamics and the competition between magnetic exchange interactions and thermal fluctuations. This could be done, at least, using Monte-Carlo calculations based on an effective hamiltonian constructed on top of DFT parameters (including magnetic exchange and anisotropy at least). Best Regards Xavier Le 27/11/2016 à 10:01, Fecher, Gerhard a écrit : How do you distinguish a diamagnetic, a paramagnetic, a ferromagnetic, and an antiferromagnetic state. Think ! This will answer your question, hopefully. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Feche
Re: [Wien] How to include the localized d orbitals in the atomic spheres?
Look into section 7.3 of the user guide: ORB (Calculate orbital potentials) The very first sentence reads: orb calculates the orbital dependent potentials, i.e. >>>potentials which are nonzero in the atomic spheres only <<So, in the region between the atomic spheres the potentials arising from U or hybrids are set to zero. Not strictly what happens in the material, but a way to improve the model significantly and keep the calculation viable. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 28.11.2016 23:21, schrieb Abderrahmane Reggad: Thank you Prof Cottenier for your answer My question is made according to the following statement: " The DFT+U and EECE are applied only inside atomic spheres " What does it mean that and how to realize it ? Best regards -- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] Discrepancy in the simulation of the paramagnetic state
My two cents concerning this problem: What moments do you consider to be zero, Reggad? With Ni being a METAL I would claim that the electron spin moment is the one to inspect. Certainely you (or the authors you read) do not propose to set the electron spin to zero? Remember that DFT calculates the situation at T=0. Simulating paramagnetism at high temperature by forcing the spin-polarization to zero may work for some purposes, but will be bad if you are interested in magnetic properties: With zero moments nothing will happen if you switch on a magnetic field! With 4f-electrons there is a good chance that one can consider the whole 4f-shell as a quantum system of its own, instead of the single electrons. Then each lattice site carrys some moment. The paramgnetic phase of such an ensemble is by definition the one where the expectation value of the total magnetization vanishes due to incoherent random fluctuations of the 4f-moments. imho, the view of this phase as all the moments pointing in random directions has its merits, but one should not over-interprete the picture. Such a state is most probably not even an eigenstate of the Hamiltonian, let alone the ground state that DFT is concerned with. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 26.11.2016 22:30, schrieb Abderrahmane Reggad: Thank you Prof Blaha for your quick answer. The Ni atom is 3d transition metal . But my question is about the simulation of the paramagnetic state. There are many people that considere that the paramagnetic state is the non-spin polarierd one and the magnetic moment is zero, but you say no and the magnetic moments exist in arbitrary directions and my quoting is about that. I have given 2 examples for that discrepancy with your statement. Best regards-- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] question about tetra's choice of Emax
Sure, Cu has a moment in any number of cases with more or less complex band structures. This was just recalling the basic textbook model of itinerant ferromagnetism of elemental metals from the 1960's. To me the simple band structure underlying that (not too bad) model looks much like the cases where the tetra's and lapw1's automatic setting of Emax might have difficulties. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 21.11.2016 09:48, schrieb Lyudmila Dobysheva: 17.11.2016 16:47, pieper wrote: One might (correctly) expect Cu with 11 electrons for these bands to be a happy paramagnet: the spins of 10 e in the d-band compensate, and the exchange for the one in the 4s is too small to redistribute anything from there, 4s is spin balanced. Furthermore it costs too much energy to increase the 4s population above half full, the exchange gain in an unbalanced 4d is not sufficient. I don't follow your discussion, but for the sake of accuracy, I want to comment that Cu atoms have magnetic moment in the Cu oxides. Best regards Lyudmila Dobysheva -- Phys.-Techn. Institute of Ural Br. of Russian Ac. of Sci. 426001 Izhevsk, ul.Kirova 132 RUSSIA -- Tel.:7(3412) 432045(office), 722529(Fax) E-mail: l...@ftiudm.ru, lyuk...@mail.ru (office) lyuk...@gmail.com (home) Skype: lyuka17 (home), lyuka18 (office) http://ftiudm.ru/content/view/25/103/lang,english/ -- ___ 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/index.html
Re: [Wien] orbital moment-mBJ
High Komal, please consult the LAPWDM sections in the user guide. There calculation of the orbital moment is explained. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 17.11.2016 13:35, schrieb Komal Bapna: Dear Wein users I am using mBJ potential to perform spin-polarised calculations on a perovskite structure. The scf file shows spin magnetic moment. Can I have the info about its orbital moment also using "runsp" command as I have used or I need to give some more command . Please reply Regards -- KOMAL ___ 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/index.html
Re: [Wien] question about tetra's choice of Emax
Thanks for the quick response! Sorry, I did not fully catch the problem of consistency between what is calculated in lapw1 and integrated in tetra. However, your PS and PPS leave me worried about where things go sideways with something as simple as fcc Co should be. As for the version, my VERSION file says WIEN2k_14.2 (Release 15/10/2014) which to my knowledge is not too bad? It does not include the dynamical Emax in lapw1? With that version, kgen 1 vectors, spin-resolved, and everything else default in w2web, I get from tetra something like the file below. Emax is apparently changed to a value below E_F. I understand that this might lead to problems already in the scf because the wrong states are occupied, but in that case again I cannot rely on the DOS, even when I increase Emax in .in1 and run lapw1 ... I have to re-run the scf, correct? - Co-fcc.outputtup -- Co-fcc # IAV : 0 NPRINT : 1 2 CASES FOR DOS: ATOM L cc-Co LATTICE CONST.= 6.79300 6.79300 6.79300 FERMI ENERGY= 0.56702 48 <; NMAT <; 63 SPIN=2 NATO= 2 JATOM 1 MULT= 1 ISPLIT= 2 tot,0,1,2,D-eg,D-t2g,3 CASE 1 : ATOM NUMBER 0 COLUMN READ 0 DOSTYPE=total-DOS CASE 2 : ATOM NUMBER 1 COLUMN READ 1 DOSTYPE= 1:total We will add0 DOS-cases together: BAND LIMITS OF BAND 1 ARE -3.81359 -3.79544 BAND LIMITS OF BAND 2 ARE -3.80710 -3.79544 BAND LIMITS OF BAND 3 ARE -3.80235 -3.79544 BAND LIMITS OF BAND 4 ARE -0.07833 0.28402 BAND LIMITS OF BAND 5 ARE 0.24033 0.37076 BAND LIMITS OF BAND 6 ARE 0.32045 0.48759 BAND LIMITS OF BAND 7 ARE 0.37076 0.51438 BAND LIMITS OF BAND 8 ARE 0.40511 0.51502 BAND LIMITS OF BAND 9 ARE 0.43442 1.04422 EMAX reduced due to lower HIGHEST BAND-minimum EMIN, DE, EMAX : -0.5 0.00200 0.43442 EMIN= -0.5 EMAX= 0.43442 EFACTR=499.6948 ESTEP = 0.00200 ENERGY BAND1 THROUGH9 ENERGY CHANNEL:1 TO 468 NUMBER OF K-POINTS: 165 NUMBER OF TETRAHEDRONS: 693 # BAND9 #EF= 0.56702 NDOS= 2 NENRG= 468Gaussian bradening: 0.00300 NUMBER OF ELECTRONS UP TO EF :0. DOS in states/Ry/spin smearing 1 0.3989422748506432.00 smearing 1 0.3989422748506432.00 # ENERGY0 total-DOS1 1:total -0.5 0.00 0. 0.00 0. -0.49800 0.00 0. 0.00 0. -0.49600 0.00 0. 0.00 0. . . . -- --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 16.11.2016 07:41, schrieb Peter Blaha: The DOS is calculated up to a value for which we can guarantee that the DOS is correct and complete. Of course, the DOS up to the highest band-maximum would be non-zero, but there is a chance that some (maybe a lot) of DOS is missing and a user would not notice this and "misintewrprete" this wrong DOS. Now he can trust that the calculated DOS is ok, and if he needs higher DOS, he has to increase the emax in case.in1 and/or in case.int. PS: You are probably using an older WIEN2k version, because now we use a "dynamical" Emax in case.in1, which takes the actual EF into account. But of course in cases of very steep bands above EF, the default in1 file may still be insufficient. PPS: In your situation it could even be, that the scf calculation is "wrong", since you occupied the wrong bands .... Am 15.11.2016 um 19:12 schrieb pieper: Hello, mailing list, yesterday I had for the first time some dispute with the way tetra automatically chooses its input energy range. I would like to understand why the particular automatic choice of Emax was introduced. Until then it worked so well that I didn't even notice that Emax is automatically adjusted, but then I wanted to illustrate Wien2k by calculating the example fcc Ni. I took Co instead of Ni, lattice constant adjusted to 6.637, RMT to touching spheres. I used w2web for initialiazation and initiating scf, kgen with 1 vectors, spin-polarized in an FM starting configuration, no spin-orbit coupling. Then I chose DOS from the Tasks menu, skipped the optional steps lapw1 and qtl, used lapw2 -qtl, configured .int to calculate just the total DOS (the default). When I proudly presented the result of dosplot to a visitor the plot ended below the Fermi energy ... Increasing Emax in .int did nothing, as the experts probably could have told me beforehand. The (in this case in my view annoying) automatic choice of Emax that kicks in is indicated in the header of .outputtup (as well as in the user guide - and yes, I know one should read it): . . . BAND LIMITS OF BAND 8 ARE 0.40511 0.51502 B
[Wien] question about tetra's choice of Emax
Hello, mailing list, yesterday I had for the first time some dispute with the way tetra automatically chooses its input energy range. I would like to understand why the particular automatic choice of Emax was introduced. Until then it worked so well that I didn't even notice that Emax is automatically adjusted, but then I wanted to illustrate Wien2k by calculating the example fcc Ni. I took Co instead of Ni, lattice constant adjusted to 6.637, RMT to touching spheres. I used w2web for initialiazation and initiating scf, kgen with 1 vectors, spin-polarized in an FM starting configuration, no spin-orbit coupling. Then I chose DOS from the Tasks menu, skipped the optional steps lapw1 and qtl, used lapw2 -qtl, configured .int to calculate just the total DOS (the default). When I proudly presented the result of dosplot to a visitor the plot ended below the Fermi energy ... Increasing Emax in .int did nothing, as the experts probably could have told me beforehand. The (in this case in my view annoying) automatic choice of Emax that kicks in is indicated in the header of .outputtup (as well as in the user guide - and yes, I know one should read it): . . . BAND LIMITS OF BAND 8 ARE 0.40511 0.51502 BAND LIMITS OF BAND 9 ARE 0.43442 1.04422 EMAX reduced due to lower HIGHEST BAND-minimum EMIN, DE, EMAX : -0.5 0.00200 0.43442 The problem in this case is that the 'lower highest Band-minimum' is BELOW the Fermi-energy: EF and DOS at fermi level *** 0.56702 0.00 0.00 So one MUST go back to the (in principle optional) lapw1 step with some larger Emax in .in1 (as indicated in the DOS menu and in the UG). Increasing Emax there appears to me a little clumsy since one has to guess an Emax that will generate a band with a band minimum above E_F. Why not have tetra choose Emax as the minimum of the Emax input in .int and the highest Band-MAXIMUM? --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 ___ 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] No convergence during Volume Optimization
The 'difference between energy and force approaches' is just that: a different approach to find the equilibrium positions in space for the atoms and their electrons within the structural model given by the .struct file. The equilibrium is the energy minimum in the parameter space under consideration. An energy minimum means that the derivatives with respect to (atom) positions - that is the forces - vanish. Knowing the derivatives (forces) helps, of course, a lot in finding the minimum - they point the way to the next (local) minimum. In the user guide you see: --User guide -- 5.3.1 Lattice parameters (Volume, c/a, lattice parameters) Package optimize The auxilliary program optimize (x optimize) generates from an existing case.struct (or case initial.struct, which is generated at the first call of optimize) a series of struct files with various volumes (or c/a ratios, or other modified parameters) (depending on your input): . . . After execution of this script you should have a series of scf-files with energies corresponding to the modified parameters, which should allow you to find the corresponding equillibrium parameters. For the volume optimization an analysis tool is available, other tools are under development). -- The minimum total energy defines the equilibrium. The derivatives with respect to lattice parameters are not easy to obtain during the scf cycle so for lattice parameters an 'energy approach' is used. For the internal parameters this is different. Persons as ingenious as Prof. Marks can calculate the derivatives with respect to internal parameters from the charge distribution at affordable computational cost during the scf. --User guide -- 5.3.2 Minimization of internal parameters (min lapw) Most of the more complicated structures have free internal structural parameters, which can either be taken from experiment or optimized using the calculated forces on the nuclei. Starting with WIEN2k 11.1 there are two possibilities to determine the equilibrium position of all individual atoms automatically (obeying the symmetry constraints of a certain space group). One can use either the shell script min lapw, together with the program mini, which will run a scf-cycle, update the positions using the calculated forces and restarts a new scf cycle. This continues until forces drop below a certain value; or use the normal scf-scripts run lapw -min where in case.inm the switch MSR1 will be modified to MSR1a such that the charge density and the positions are simultaneously opti- mized during the scf-cycle. -- The first option uses what you call the 'force approach': in equilibrium, no forces should push the atoms around. The second option indicates a mixed approach: The positions of the atoms (according to forces acting on them) and the ones of the electrons (to minimize the total energy) are BOTH adjusted in each step of the scf. As I said, there can be many reasons why your calculation did not reach convergence for some structural parameters. Did the scf stop without errors because the maximum number of iterations was reached? If yes, what was this number of iterations? Maybe your convergence criteria are too strong for the numerical precision you set by parameters like RKMAX or the k-mesh? Maybe the scf oscillates between several good solutions (is it a magnetic case?). Maybe your starting configuration and/or the model Hamiltonian is completely off or missing some ingredient and the poor scf wanders helpless around, lost in a multidimensional world ... Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 13.11.2016 19:14, schrieb Abderrahmane Reggad: Thank you Dr Pieper for your interesting to my questions. I have optimized the atomic positions before doing calculation. Tha thing that I didn't understand is that the convergence is reached for some points but not for others. For the "optimization notes " , there is no mention on the difference betwwen the energy and force approaches. Best regards -- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] No convergence during Volume Optimization
Look at the curve energy vs. volume and decide yourself wether its worth the while to increase the number of iterations (assuming that's the reason for the 'jump to the next point'). It also is a good idea to look for reasons why the convergence is slow (problem in the struct file? Bad position for some atom? Lattice constants very far from equilibrium? ... there are a lot of less trivial possibilities) Concerning your other question wether or not optimization with respect to energy or with respect to forces is the same I recommend reading the section on structure optimization in the user guide, and perhaps the 'Optimization Notes' of L. D. Marks that you will find on the Wien2k website. I am completely unable to improve on those. Best luck Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 12.11.2016 12:31, schrieb Abderrahmane Reggad: Dear Wien Users Sometimes , It happens that during volume optimization , the calculation doesn't converge to the chosen energy criteria and jumps to the next point . In this case, should we delete this point and choose another point or we could accept this calculation. For example , with energy criteria value of 0.0001 and with some points (we take 3 points as example)/ - 6 percent point: ETEST: .00083 - 0 percent point: ETEST: .00034 - 6 percent point ETEST: .00013 Best regards -- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] Minimization before Volume Optimization ?
sgroup (or other scripts in the initialization) does'nt change the positions of the atoms, it changes the basis used to describe the positions. 'Optimization' is done with respect to internal forces, or energy, both unknown at that step. Good luck Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 11.11.2016 14:43, schrieb Abderrahmane Reggad: Now I want to know if a low symmetry structure (orthorhombic) derived from its high symmetry structure (hexagonal NiAs )has its atomic positions optimized (because they are created with sgroup program) or need to be optimized . Best regards -- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] Is there an effect of SO on the ground state
Of course inclusion of SO interactions may lead to another magnetic ground state. For example, at least in principle it certainely will do so if the symmetry of your Hamiltonian including SO breaks some symmetry of the ground state you found without SO. So you might want to look wether or not initso induces changes of your symmetry. However, the effect in whatever you are interested in might be too small to make a real difference. Therefore, you also might compare typical SO interaction energies for elements in your structure with the energy differences between the ground states you worked out without SO. If you have light elements (small SO) and large Heisenberg exchange, then probably nothing will happen. In addition you should be aware of the fact that Wien2k (or any other DFT program) will find (if you are lucky and everything works out) the ground state within the symmetry you set up during initialization. If you set up a hexagonal structure with some AF magnetic structure on top of it (maybe in a larger unit cell) your result from the scf cycle will have that symmetry. The actual ground state of the material you are interested in might be completely different. A drastic example: SO interaction may lead to non-collinear helical spin structures - which are outside the scope of standard Wien2k. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 30.10.2016 14:55, schrieb Abderrahmane Reggad: Thanks Dr pieper for the rich information within your answer. Perhaps I didn't formulated my question well . I am not interesting exactly to the ground state energy , but to the magnetic ground state. As I mentionned before , I want to determine the magnetic ground state from 5 configurations: nm,fm, afm1,afm2 and afm3 for hexagonal crystal structure (NiAs structure exatly); and I want to know if the inclusion or not of the SO coupling affect this magnetic ground state.i.e. without SO , I got a magnetic ground state , and if I include the SO, may be this leads to another magnetic ground state. Best regards -- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] Is there an effect of SO on the ground state
My 2 cents to this: Not a particularly meaningful question. The only meaningful thing about energies, as in eigenvalues of the Hamiltonian, is their DIFFERENCE. The evolution with time of the state of an isokated quantum system (Schrödinger picture) is completely governed by the differences of the eigenvalues of the Hamiltonian. The choice of where you put zero is up to you. A common choice for zero is the ground state energy of the Hamiltonian. In this case clearly zero is zero, there is no effect of SO. Other choices are more practical in various circumstances: The mean of the spektrum of the Hamiltonian, or the Fermi energy (definition discussed recently here), or some fictitious reference limit like the state with all charges at infinite distance ... The ground state energy then probably will depend on whether or not you put SO into your Hamiltonian. After all you consider SO because it changes energy differences and might split degeneracies. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 29.10.2016 23:20, schrieb Abderrahmane Reggad: Hello again I am waiting for an answer to my question . My question is about the effect of the inclusion of the Spin-Orbit Coupling on the ground state energy. I want to know if the SO affect the ground state energy also or It only causes the splitting of the degenerate state energies. Any information will be fruitful for me. I hope find an answer to my question -- Mr: A.Reggad Laboratoire de Génie Physique Université Ibn Khaldoun - Tiaret Algerie ___ 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/index.html
Re: [Wien] 答复: HELP! cd $PWD; $t $exe ${def}_${loop}.def $loop; rm -f .lock_$lockfile[$p] ) >> .time2_$loop LAPW2 END
Since I am unable to think of a reason to HIDE not even the error
messages or warnings, but only the flag indicating them, I stick to my
first guess: echo messages from your job(s) informing you what is
currently done. To stop this you probably will have to look into your
shell commands, but there I am no epert and I dont know anything about
your server anyway.
---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564
Am 21.07.2016 17:40, schrieb Jing Qun:
Thanks.
These are neither ERROR messages nor WARNINGS.
During the calculations, everything is OK.
But these messages were not found during the calculation using other
HPC server, I wonder if these messages are hidden ERROR or WARNINGS.
-
发件人: Wien <wien-boun...@zeus.theochem.tuwien.ac.at> 代表
pieper <pie...@ifp.tuwien.ac.at>
发送时间: 2016年7月21日 8:44
收件人: A Mailing list for WIEN2k users
主题: Re: [Wien] HELP! cd $PWD; $t $exe ${def}_${loop}.def $loop;
rm -f .lock_$lockfile[$p] ) >> .time2_$loop LAPW2 END
Question is, is there anything wrong? Are these parts of ERROR
messages
or at least WARNINGS, or are they just echoes, your job(s) telling
you
what they are doing?
---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564
Am 21.07.2016 07:07, schrieb Jing Qun:
> Dear all users,
> When I submit a job, some messages like "
> cd $PWD; $t $exe ${def}_${loop}.def $loop; rm -f
.lock_$lockfile[$p] )
>>> .time2_$loop
> LAPW2 END " will be obtained. Can anyone tell what's wrong with it?
>
>
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien [1]
Wien -- A Mailing list for WIEN2k users [1]
zeus.theochem.tuwien.ac.at
A Mailing list for WIEN2k users. Please post questions, suggestions
or comments about WIEN2k ONLY in this list. Please follow the
following "Nettiquette" (depending ...
> SEARCH the MAILING-LIST at:
>
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
[2]
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien [1]
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
[2]
Links:
--
[1] http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[2]
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/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/index.html
Re: [Wien] HELP! cd $PWD; $t $exe ${def}_${loop}.def $loop; rm -f .lock_$lockfile[$p] ) >> .time2_$loop LAPW2 END
Question is, is there anything wrong? Are these parts of ERROR messages
or at least WARNINGS, or are they just echoes, your job(s) telling you
what they are doing?
---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564
Am 21.07.2016 07:07, schrieb Jing Qun:
Dear all users,
When I submit a job, some messages like "
cd $PWD; $t $exe ${def}_${loop}.def $loop; rm -f .lock_$lockfile[$p] )
.time2_$loop
LAPW2 END " will be obtained. Can anyone tell what's wrong with it?
___
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/index.html
Re: [Wien] x nn failed
There is far too little information to really help, but: - is your .struct file ok? Do you have one in the case directory? Does the structure look ok when viewed with xcrysden? - did you succeed with the TiC example as described extensively in the UG? - Were there any WARNINGS or even ERRORS during installation? - What does the error message file of nn say? A recommendation: As a beginner use the w2web interface! It goes a long way to help keeping the correct sequence of tasks, setting defaults and so on. Good luck --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 18.05.2016 16:52, schrieb ben abdallah houda: Dear users I have just finish installation of wien2k code, but when I start with x nn it writes x nn failed. Please can someone help me. Thanks for your help Envoyé depuis Yahoo Mail pour Android [1] Links: -- [1] https://overview.mail.yahoo.com/mobile/?.src=Android ___ 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/index.html
Re: [Wien] Fwd: Strain
Hello Muhammad! A DFT code will not simply say *crack* at some point of strain. Wien2k will always distribute the force you apply according to the perfect translational symmetry of your crystal structure - in contrast to nature where things like cracking or breaking occure at weak links - points where this translational symmetry is broken and the bonding is weaker. DFT will do its very best to distribute the electrons in any structure you throw at it and tell you (an upper bound of) the minimum energy of the configuration. I am no expert but I understand that from a calculated variation of total energy with lattice parameter you can estimate the intrinsic strength of a structure within some model for the process. In addition Wien2k gives information where the total energy is stored, for example the forces on atoms in your structure. Therefore, if you look at the total energy of your strained structure you will see that it is larger - at 5% distortion perhaps considerably so. Depending on site symmetry some calculated internal forces may indicate where in the structure the 'springs' you pulled are located. Furthermore the phonon code might inform you of instabilities in your structure. Good luck, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 04.12.2015 15:25, schrieb Muhammad Sajjad: Dear Tomas and Fecher Thank you very much. It is working perfect with lattice type F and symmetry operations equal to 16. But one thing is confusing me that I applied 5 % strain to bulk Si and still it is semiconductor with gap 0.2eV. Is it correct? as 5 % is too much strain and bulk materials tolerate very low strain. In my thinking it should break. On Fri, Dec 4, 2015 at 5:14 PM, Fecher, Gerhard <fec...@uni-mainz.de> wrote: but your structure becomes tetragonal you should reset the symmetry operations to (generate) to be recalculated during initialisation Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: wien-boun...@zeus.theochem.tuwien.ac.at [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Muhammad Sajjad [sajja...@gmail.com] Gesendet: Freitag, 4. Dezember 2015 15:08 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Fwd: Strain I am sorry having no idea about it. Simply I used optimized structure and increased a by a0*1.001 and decreased b & c by keeping lattice type F. Then run init_lapw -b. On Fri, Dec 4, 2015 at 4:51 PM, Fecher, Gerhard <fec...@uni-mainz.de<mailto:fec...@uni-mainz.de>> wrote: you applied a tetragonal distortion along x but how did you manage that you still have 48 symmetry operations ? Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: wien-boun...@zeus.theochem.tuwien.ac.at<mailto:wien-boun...@zeus.theochem.tuwien.ac.at> [wien-boun...@zeus.theochem.tuwien.ac.at<mailto:wien-boun...@zeus.theochem.tuwien.ac.at>] im Auftrag von Muhammad Sajjad [sajja...@gmail.com<mailto:sajja...@gmail.com>] Gesendet: Freitag, 4. Dezember 2015 14:04 An: wien Betreff: [Wien] Fwd: Strain Dear All I got Si structure from wien2k examples and after optimization it is SILICON F LATTICE,NONEQUIV.ATOMS: 1 MODE OF CALC=RELA unit=ang 10.305626 10.305626 10.305626 90.00 90.00 90.00 ATOM 1: X=0.1250 Y=0.1250 Z=0.1250 MULT= 2 ISPLIT= 2 1: X=0.8750 Y=0.8750 Z=0.8750 Si NPT= 781 R0=0.0001 RMT= 2.22 Z: 14.0 LOCAL ROT MATRIX: 1.000 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 48 NUMBER OF SYMMETRY OPERATIONS mBJLDA gives band gap of 1.19 eV (perfect). Now I apply strain (0.1 %) and structure is SILICON F LATTICE,NONEQUIV.ATOMS: 1 MODE OF CALC=RELA unit=ang 10.315932 10.300477 10.300477 90.00 90.00 90.00 ATOM -1: X=0.1250 Y=0.1250 Z=0.1250 MULT= 2 ISPLIT=-2 -1: X=0.8750 Y=0.8750 Z=0.8750 Si NPT= 781 R0=0.0001 RMT= 2.22 Z: 14.0 LOCAL ROT MATRIX: 0.000 0.000
Re: [Wien] Calculate spin orbit coupling with external magnetic field (ORB package)
Re-reading the thread my guess is that (as usual) Peter was right from the start: Nobody tried before ... if symmetso is compatible with nmod=3 in inorb. Your difficulties indicate that it's not. One additional thing I can think of to check there is wether the direction of M in your case.inso and the field direction in your case.inorb are compatible. However, such an incompatibility would not imply that SO and external field in ORB are incompatibel. At least I myself initialized SO (using w2web) always BEFORE turning on any orbital potential and did at least one SCF + save. Actually, the reason was some warning sentence in the UG sec. 7.3 about possible instabilities of the SCF for LDA+U without SO. Maybe I circumvented your initialization problems that way. Good luck Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 10.11.2015 22:06, schrieb Jing-Han Chen: Thank Gerhard and Martin for checking and trying this question with me. I am using wien2k 14.2 and case.indmc. Beside the issues pointed out by Gerhard and Martin, another thing I noticed is that symmetso has reading error of case.inorb only if the structure has more than one site. It might be the reason why the error does not appear in Pt. 2015-11-10 5:10 GMT-06:00 Fecher, Gerhard <fec...@uni-mainz.de>: indeed, the same with me as I just noted: the problem is that w2web creates .indm and likes to create a new one if its missing, even if you have a correct .indmc. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: wien-boun...@zeus.theochem.tuwien.ac.at [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von pieper [pie...@ifp.tuwien.ac.at] Gesendet: Dienstag, 10. November 2015 11:47 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Calculate spin orbit coupling with external magnetic field (ORB package) Being a stupid and lazy person I always use the w2web interface, and the *_so versions stay hidden with that. I would assume that they are temporary versions, probably built by the script to be renamed at the end to case.indm(c) and case.inorb. I would like to point out again what Gerhard said: this is a magnetic case so you have to use complex versions, especially case.indmc! Good luck Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.11.2015 21:59, schrieb Jing-Han Chen: Thanks for the comments from Martin, Gerhard and Peter. I assumes all inputs accurate since ORB and SOC can be run individually. It works successfully if I change case.inorb to some other name before x symmetso and create it manually. I have one further question. The script symmetso generates case.inorb_so and case.indm_so. Should they be the same as case.inorb case.indm respectively for SOC with external magnetic field, or should they be kept as empty after symmetso? 2015-11-09 3:40 GMT-06:00 pieper <pie...@ifp.tuwien.ac.at>: The same with me: I would have to dig through old archives to find out what I actually did, but I am fairly sure that I used SO + external field a few years ago (probably Wien2k 10 or 12) - and don't recall any incompatibilities between SO and external field at the time. Keep fingers crossed that it checks out with an input file error Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.11.2015 08:51, schrieb Fecher, Gerhard: I tried it once for Pt and it worked most probably there is an error in one of the input files inorb, indm, inso, or wherever else (maybe "c" versions of the input files are needed, or the m-directions are not consistent). I don't remember any conflict between initso and inorb. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: wien-boun...@zeus.theochem.tuwien.ac.at [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Peter Blaha [pbl...@theochem.tuwien.ac.at] Gesendet: Montag, 9. November 2015 07:17 An: A Mailing list for WIEN2k users Betreff
Re: [Wien] large deviation of atomic volume in BiNi compound
Dear Tomas, at the moment I do not see something being obviously suspicious. Maybe the culprit is some structural phase transition invalidating the experimental structure you compare with. You might get one or two ideas from Stefaan Cottenier's work? Error Estimates for Solid-State Density-Functional Theory Predictions: An Overview by Means of the Ground-State Elemental Crystals K. Lejaeghere , V. Van Speybroeck , G. Van Oost & S. Cottenier http://dx.doi.org/10.1080/10408436.2013.772503 Best regards, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 10.11.2015 13:05, schrieb Tomas Kana: Dear Martin and Gerhard, Thank you for your suggestions. Gerhard, thank you for mentioning this experimental work. Will you please send me the pdf of the article? I do not have access to it. Regarding Martin's questions: I tried to include magnetism of the constituents by performing spin polarized calculations, too, but the equilbrium volume was the same. The forces within the hexagonal unit cell were not given in case.scf (I think there was too much symmetry operations). However, I recently tried to express the hexagonal unit cell in a orthorhombic base and cancel the symmetry operations by using inequivalent atoms (I send the structure file in attachment). The volume was still wrong but I know the values of the forces. For the experimental atomic volume they were at most 0.84 mRy/a.u. With best regards Tomas Kana Since you ask for ideas and without really looking at the problem: Assuming that the experimental numbers are correct, is this a room temperature structure? The calculations are, of course, ground state zero Kelvin, so things might go south if there is a phase transition somewhere. Considering the elements you deal with maybe magnetic? What are the forces in your calculations? Good luck, Martin ___ 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/index.html
Re: [Wien] large deviation of atomic volume in BiNi compound
Since you ask for ideas and without really looking at the problem: Assuming that the experimental numbers are correct, is this a room temperature structure? The calculations are, of course, ground state zero Kelvin, so things might go south if there is a phase transition somewhere. Considering the elements you deal with maybe magnetic? What are the forces in your calculations? Good luck, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 10.11.2015 10:21, schrieb Tomas Kana: Dear Wien2k users, I came across a problem with equilibrium atomic volume of the BiNi compound. The experimental lattice is hexagonal with a = 4.079 Angstroem, c = 5.359 Angstroem (P. Villars, Pearson's Handbook: Crystallographic Data for Intermetallic Phases) However, the equilibrium volume turns out to be more than 16 % higher than the experimental one. I wonder since the equilibrium volume of pure Bi and Bi3Ni comes out with much better agreement with experiment (about 4 to 5 % deviation). I used GGA (no spin orbit coupling), Rmt*Kmax = 8.8, lmax = 10, Gmax = 16, 5000 k-points in the whole Brillouin zone. I enclosethe structure file in attachment. I tried LDA that gives better agreement with experiment (about 10 % deviation) but I think this is still too much. I have tried to use gaussian smearing instead of the tetrahedron method, increase Rmt*Kmax to 11, increase k-points to 20 000 in the whole Brillouin zone but nothing helped. In the mailing list I found someone had similar problem with a more complicated structure containing bismuth, but that was not solved: http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg10479.html Do you have any idea? Thank you in advance With best regards Tomas Kana Institute of Physics of Materials, Academy of Sciences of the Czech Republic Brno, Czech Republic ___ 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/index.html
Re: [Wien] Calculate spin orbit coupling with external magnetic field (ORB package)
Being a stupid and lazy person I always use the w2web interface, and the *_so versions stay hidden with that. I would assume that they are temporary versions, probably built by the script to be renamed at the end to case.indm(c) and case.inorb. I would like to point out again what Gerhard said: this is a magnetic case so you have to use complex versions, especially case.indmc! Good luck Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.11.2015 21:59, schrieb Jing-Han Chen: Thanks for the comments from Martin, Gerhard and Peter. I assumes all inputs accurate since ORB and SOC can be run individually. It works successfully if I change case.inorb to some other name before x symmetso and create it manually. I have one further question. The script symmetso generates case.inorb_so and case.indm_so. Should they be the same as case.inorb case.indm respectively for SOC with external magnetic field, or should they be kept as empty after symmetso? 2015-11-09 3:40 GMT-06:00 pieper <pie...@ifp.tuwien.ac.at>: The same with me: I would have to dig through old archives to find out what I actually did, but I am fairly sure that I used SO + external field a few years ago (probably Wien2k 10 or 12) - and don't recall any incompatibilities between SO and external field at the time. Keep fingers crossed that it checks out with an input file error Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.11.2015 08:51, schrieb Fecher, Gerhard: I tried it once for Pt and it worked most probably there is an error in one of the input files inorb, indm, inso, or wherever else (maybe "c" versions of the input files are needed, or the m-directions are not consistent). I don't remember any conflict between initso and inorb. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: wien-boun...@zeus.theochem.tuwien.ac.at [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Peter Blaha [pbl...@theochem.tuwien.ac.at] Gesendet: Montag, 9. November 2015 07:17 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Calculate spin orbit coupling with external magnetic field (ORB package) Probably nobody has ever tried it with a magnetic field in case.inorb. Move cse.inorb to some other name before x symmetso and create it manually (Just check, if the number of non-equivalent atoms has changed or not). Am 08.11.2015 um 20:38 schrieb Jing-Han Chen: Dear All I am trying to calculate the effect of spin-orbit coupling while the external magnetic field is specified by ORB package. However, it continues to give an error of reading the case.inorb during "x symmetso", one of initso_lapw step. I wonder whether SOC is actually not compatible with the nmod=3 ORB. Did anyone have the experience about this? -- -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-165300 FAX: +43-1-58801-165982 Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at WWW: http://www.imc.tuwien.ac.at/staff/tc_group_e.php -- ___ 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/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/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/index.html
Re: [Wien] Calculate spin orbit coupling with external magnetic field (ORB package)
The same with me: I would have to dig through old archives to find out what I actually did, but I am fairly sure that I used SO + external field a few years ago (probably Wien2k 10 or 12) - and don't recall any incompatibilities between SO and external field at the time. Keep fingers crossed that it checks out with an input file error Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.11.2015 08:51, schrieb Fecher, Gerhard: I tried it once for Pt and it worked most probably there is an error in one of the input files inorb, indm, inso, or wherever else (maybe "c" versions of the input files are needed, or the m-directions are not consistent). I don't remember any conflict between initso and inorb. Ciao Gerhard DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy: "I think the problem, to be quite honest with you, is that you have never actually known what the question is." Dr. Gerhard H. Fecher Institut of Inorganic and Analytical Chemistry Johannes Gutenberg - University 55099 Mainz and Max Planck Institute for Chemical Physics of Solids 01187 Dresden Von: wien-boun...@zeus.theochem.tuwien.ac.at [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Peter Blaha [pbl...@theochem.tuwien.ac.at] Gesendet: Montag, 9. November 2015 07:17 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Calculate spin orbit coupling with external magnetic field (ORB package) Probably nobody has ever tried it with a magnetic field in case.inorb. Move cse.inorb to some other name before x symmetso and create it manually (Just check, if the number of non-equivalent atoms has changed or not). Am 08.11.2015 um 20:38 schrieb Jing-Han Chen: Dear All I am trying to calculate the effect of spin-orbit coupling while the external magnetic field is specified by ORB package. However, it continues to give an error of reading the case.inorb during "x symmetso", one of initso_lapw step. I wonder whether SOC is actually not compatible with the nmod=3 ORB. Did anyone have the experience about this? -- -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-165300 FAX: +43-1-58801-165982 Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at WWW: http://www.imc.tuwien.ac.at/staff/tc_group_e.php -- ___ 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/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/index.html
Re: [Wien] Can we perform layered AFM calculations with Wien2k?
Hello Krishnaveni Parthasarathy Yes you can - though admittedly I did not look up your reference, and I might misinterprete your idea of 'layered AFM'. And I suppose you consider collinear magnetic moments. If that is not the case you might look into the (unsupported) WienNCM. In general, study the UG, especially the chapters on AFM calculations. You can think of any collinear AFM as ferromagnetic layers stacked along one particular crystalographic direction, the direction to the nearest antiparallel neighbor. For calculations with Wien2k you need a .strcut with a unit cell that has at least two magnetic atoms along that direction (perhaps from supercell if the crystalografic primitve cell is not large enough). Since you want them to be magnetically inequivalent the two have to be on inequievalent crystalografic sites. If they are on crystalografic equivalent positions you have to split the position and number the atoms to tell Wien2k that they are not same (perhaps Mn1 and Mn2). Note that this results in a different symmetry of your structure. If you want to compare total energies for different magnetic structures be careful to calculate all of them on the same crystalographic unit cell, changing only in the initialization which magnetic atom belongs to the group pointing 'up' and which one to the group pointing 'down'. If you have, for example, constructed a supercell with four inequivalent Mn-sites Mn1 .. Mn4, you can calculate AFM structures corresponding to Mn1, Mn2 up and Mn3, Mn4 down, or Mn1,3 up and Mn2,4 down, or ... This will correspond to ferromagnetic layers of Mn1, Mn2 stacked with antiparallel ferromagnetic layers Mn3,4, or ferromagnetic layers of Mn1,3 ... Not all of these arrangements are necessarily different. If the direction Mn1 - Mn2 is symmetry related to the one Mn1-Mn3 in the underlying crystalografic lattice, then there is no need to calculate with the stacking in these two equivalent directions. And obviously, for a complicated AF wave vector (the direction of the stacking), you will need large supercells so think, before you start. Good luck, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 28.09.2015 18:43, schrieb Krishnaveni. S: Ref: Electronic Structure, Chemical Bonding, and Finite-Temperature Magnetic Properties of Full Heusler Alloys Authors: YASEMIN KURTULUS, MICHAEL GILLEßEN, RICHARD DRONSKOWSKI* DOI 10.1002/jcc.20308 online in Wiley InterScience (www.interscience.wiley.com [1]). Ref to the article (Page 93) mentioned above, I understand that one can perform layered calculation in AFM calculations. In page 93, the authors have mentioned as below: “Here, the symmetry has been slightly reduced to obtain a model in which the Mn planes can be treated separately. This particular alignment is characterized by alternating planes along (001) of spin-up and spin-down manganese moments, and the identically oriented manganese moments within each plane are coupled by the p orbitals of the Z elements. An alternative anti ferromagnetic model with alternating magnetic planes along (111) is also thinkable, but earlier total-energycalculations11 have already clarified that the (001) model is characterized by lower energies in almost all cases.” The author has specified that these calculations have been done in VASP. Can some one help me understand if it is possible to perform similar type of calculations in Wien2k? -- Thanks and regards Krishnaveni Parthasarathy 8939675012 Links: -- [1] http://www.interscience.wiley.com ___ 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/index.html
Re: [Wien] Error in init_lapw
Dear Paresh, your initial .struct is almost certainely wrong. The WARNINGs from nn indicate that you put two different atoms at the same locations, probably by entering them into equivalent crystalografic positions. Take nn's WARNINGs seriously: your calculations will almost certainely crash if already nn has problems. At least your calculations will be meaningless if you don't understand its complaints since in that case you don't understand what the structure the program works with looks like. Start a new case in a fresh directory. Correct your structure, and inspect it with your favorite viewer (perhaps with xcrysden as recommended in the user guide). When it looks ok try nn again. Best luck, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 22.09.2015 09:12, schrieb Paresh Chandra Rout: Dear all, I am facing a warning message while setting nn-bondlength as follows nn (12:18:17) specify nn-bondlength factor: (usually=2) [and optionally dlimit, dstmax (about 1.d-5, 20)] 2 WARNING: Mult not equal. PLEASE CHECK outputnn-file WARNING: ityp not equal. PLEASE CHECK outputnn-file NN created a new Bi2FeReO6_G-type.struct_nn file . But when I use Old case.struct file I am able to enter into the next step . But when I accept to use the new case.struct case it is just get stuck there giving the error message as follows forrtl: severe (174): SIGSEGV, segmentation fault occurred Image PC Routine Line Source nn 0040779C MAIN__ 396 nn.f nn 0040312C Unknown Unknown Unknown libc.so.6 003F57A1ED5D Unknown Unknown Unknown nn 00403029 Unknown Unknown Unknown 0.022u 0.002s 0:06.38 0.3% 0+0k 0+480io 0pf+0w error: command /home/paresh/Downloads/WIEN2k_14.2/nn nn.def failed n stop error n I am proceeding with the old case.struct file . Is there any problem if I use old case.struct file for the calculation ? If yes kindly give me some direction to resolve the above problem . Any help would be highly appreciated . Kind Regards Paresh Chandra Rout Research Scholar Indian Institute of Science Education and Research Bhopal ___ 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/index.html
Re: [Wien] Diference in EFG values
Hello Stefaan and Muhammad From the Wien2k UG, chapter 7.1 on lapw0, I take it that Wien2k calculates the potential, and from that the EFG, from the TOTAL electron density. For lapw0 explicitely including interstitials, for the decomposition in lapw2 (chapter 7.7) explicitely only the electron density in the atomic sphere. So as a reminder to Muhammad there may appear (minor) differences within Wien2k between the two methods. But in my understanding both calculations in Wien2k always include core, semi-core, and valence electrons (within the atomic sphere)? The difference to VASP can then occure because that code only considers core states for the EFG? Best regards, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 15.09.2015 13:55, schrieb Stefaan Cottenier: According to my understanding, core-state with tag F are valence states. AM I right? If yes then why it is taking 3P states as valence states? The valence electrons for Co are 3d7, 4s2. You use the words 'valence electrons' in the traditional 'chemical' way as states outside the noble gas core. In the context of the DFT, the term 'valence electrons' refers to all states that are not forced to behave as in a free atom (i.e. they allow to feel the presence of the surrounding crystal). You cannot avoid to use Co-3p as valence states with wien2k, the numerics would break down if you didn't. Your output shows that wien2k has Co-3s and Co-3p as valence states, whereas vasp considers these as core states (i.e. they will not contribute to the EFG in vasp). Co-3s will never contribute to the EFG (spherical), whereas Co-3p could (albeit not very likely). First consider the suggestion by Peter Blaha: make sure you have identical XC-functionals in both codes, and inspect whether all magnetic moments and the DOS and band structure pictures agree for both codes. If that is all right, then use the procedure to which I refered in my previous post to isolate the EFG contribution of Co-3p. If that is nearly zero, I don't know what is going on. If it would be equal to the vasp-wien2k difference, then you have found the origin of your problem. 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/index.html
Re: [Wien] need your help please
Hello Sikander, you might want to read some introductionary chapter of your favorite book on statistical physics or on thermodynamics ... My first try at a short answer would be: Calculate microscopic properties from a proper Hamiltonian and density matrix for the system. Calculate macroscopic properties from ensemble averages of these. I hope this does not add to your confusion, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 15.09.2015 10:20, schrieb sikander Azam: Resp. all I am confused a bit that the Properties like, Electronic structure, optical and thermoelectric properties are microscopic or macroscopic properties. Regards SIkander ___ 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/index.html
Re: [Wien] Energy vs Volume is linear!
Hello Marzieh, I took the freedomn to use your data to test a little fitting procedure I just wrote. I tried to fit parameters a, b, c of the function y=a(x-b)**2+c to your data. The fit crashed - probably like the one you may have used - when presented the data and asked to fit the parabola from scratch. The reason for such a behaviour of a fit procedure frequently is bad user behaviour: The program is quite understandably unable to guess reasonable starting values for the fit parameters from such data. Presented with starting values a=1.00e-06, b=9.50e+03, c=-8.857000e+04 the fit finds without complaining a very good fit with a=5.687893e-07, b=9.612186e+03, c=-8.856770e+04 So you should do your DFT volume optimization around a volume of 9612 - and probably give a little thought to the problem of why you started so far off the optimum at larger volumes. Good luck Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.09.2015 03:43, schrieb Marzieh Gh: Dear Prof.Blaha & Tran I have calculated optimization of supercell (1*1*2). But Energy vs Volume is linear! case.outputeos: 10575.3364 -88567.151612 11162.8553 -88566.366642 12925.4112 -88561.471064 11750.3745 -88565.095045 12337.8929 -88563.437182 Is this normal? If no what is reason? What do I do? Please help me Best Regards -- Marzieh Ghoohestani PhD Student of Computational Nano Physics Nano Research Center, Department of Physics University of Technology, Isfahan, Iran ___ 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/index.html
Re: [Wien] f orbital under an external magnetic field
Dear Bin Shao, unfortunately I am travelling and won't be able to contribute during the next days. I am looking forward to comments from people with experience in calculations with rare earths. May I just ask why you go for the energy and not for the magnetization or the susceptibility? If there is some change of the crystal field ground state this should show. From your calculation you get the size of the magnetic moments for a given field, from that you get a susceptibility. From what you say something happens around 4 T. I cannot guess from the information I have what, but I would expect it to show in the susceptibility as well. Good luck with this interesting problem Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 06.08.2015 15:47, schrieb Bin Shao: Dear Martin Pieper， Thank you for your comments! Actually, I intend to demonstrate that the energy difference between the ground state of Er^3+ (S=3/2; L=6; J=15/2) and the excited state (S=3/2; L=0; J=3/2) can be tuned by the external magnetic field, With the magnetic filed and the crystal field, the excited state splits into four states, |+3/2, |+1/2, |-1/2, and |-3/2. For the 45 Tesla magnetic field, the delta energy between the |+3/2 and |-3/2 is over 10 meV. Since we can not directly get the excited state in wien2k, even by forcing the occupation number, the calculation will still be trick. However, because the spin quantum number of the two states is the same (S=3/2), there is no spin flip from the ground state to the excited state. In this case, we can estimate the energy difference between the ground state and the excited state by calculating the energy difference between the occupied states of f electron in minority spin of the ground state and the unoccupied counterparts in minority spin of the ground state. The energy difference should become smaller with increasing the magnetic field, which can be attributed to the lower in energy of the |-3/2 state relative to the |+/-3/2 state with no magnetic field. Since the energy shift is in the magnitude of meV, we can not seen this shift from the dos calculation due to the smear of the dos. Since the f band is usually very local and the band is very flat, so I checked the eigenvalues of the 7 f-electron at the Gamma point and try to show the energy shift from the variations of the eigenvalues. However, the results show that there is only an energy shift from the 0 T to 4 T. When the magnetic filed is increasing, the eigenvalues are almost the same as that of 4 T. This most probably is the old problem of the energy zero in disguise. This may be the problem. But I have calculated all the energy differences between the 3 unoccupied and 4 occupied states of f electron in minority spin, the 12 (3*4) values are keep the same trend while the magnetic filed is varied and they are all flat. For the different f states, they get different J and the energy shifts (g_J*mu_B*J*B) induced by the magnetic filed should be also different. So I am confused. It should be noted that the energy difference is independent to the energy zero. Best, Bin On Thu, Aug 6, 2015 at 7:23 PM, pieper pie...@ifp.tuwien.ac.at wrote: As an afterthought: This most probably is the old problem of the energy zero in disguise. The Zeeman interaction you estimated and as accounted for in Wien2k is basically g*mu_B*S*B. It gives you the energy difference between a moment pointing up and one pointing down. However, it has a vanishing trace, the zero is at B=0 and the center stays there. Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 [3] Am 06.08.2015 04:55, schrieb Bin Shao: Dear all, I made calculations of a compound with Er^3+(4f^11 5d^0 6s^0, ground state S=3/2, L=6, J=15/2) doping under an external magnetic field. I got the corresponding occupation of Er^3+ with 7 electrons in majority spin and 4 electrons in minority spin. With soc including, I got eigenvalues at Gamma point of the Er^3+ under the magnetic field from 4 Tesla to 45 Tesla. However, the picture indicates that the eigenvalues with the different magnetic fields almost keep the same as that of 4 T. Why? According to a simple estimation, the magnetic field of 45 T will introduce an energy shift about 10 meV, that would definitely be seen from the figure. Any comments will be appreciated. Thank you in advance! Best regards, Bin ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien [1] SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html [2] ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at
Re: [Wien] (no subject)
Dear Sikander Azam, there seems to be agreement that DFT calculates the ground state, and that this state is occupied at T=0 K. So there are two cases: 1) If comparison with your experiment works at some finite temperature it stands to reason that the probability of observing the ground state is large. Thermodynamics tells you that this should be the case when excitation energies (at least for operators related to your experiment) are large compared to thermal energy. 2) If comparison with your experiment does not work ... you might look in this mailing list if you did something wrong in the calculations. Or your computational model of the situation is wrong. Or your system is not in the ground state. Or ... Best luck with your coparisons Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 09.06.2015 12:02, schrieb sikandar azam: Dear All Please answer me this question explain why zero kelvin DFT based calculations are compared with experimentally calculated values at 0 K temp with regards sikander ___ 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/index.html
Re: [Wien] correct parameters for geometry optimization in magnetic states
Dear Prof. Kervan, the easiest thing to do is use the case.struct file you have for the AFM calculation and initialize the calculation in a fresh directory with the same parameters as for AFM calculation, only point all magnetic moments in the same direction (for FM) or set them zero (for PM). Usually the scf-cycle does not change the initial magnetic state from FM (or PM) to AFM but stays in the local minimum of that spin arrangement. Good luck, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 26.05.2015 09:39, schrieb Selcuk KERVAN: Dear wien2k users, During the optimization for PM, FM and AFM states, ground state energy diffrences are too small. For PM and FM calculations, space group and number of k points are the same. But for AFM states, space group and number of k points are different. So, how can I find correctly the most stable magnetic states. Thanks in advance, ** Prof. Dr. Selçuk KERVANProf. Dr. Selçuk KERVAN Gazi Üniversitesi Gazi University Polatlı Fen-Edebiyat FakültesiPolatlı Science and Arts Faculty Fizik Bölümü Department of Physics 06900 Polatlı06900 Polatlı ANKARA TÜRKİYE ANKARA TURKEY ___ 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/index.html
Re: [Wien] error on SCF calculation
Dear Rachida, Pavel asked you to check if lapw1c exists - the c at the end is important! Does ZnS need the complex version? Did you have any error messages during initialization? Are you able to do SCF's on example structures (the TiC example)? Best regards, Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 20.05.2015 15:54, schrieb rachida lamouri: Dear Pavel, thank you for your contribution. I checked if lapw1 exists and there is, and for version 14.2 I have tried and I had the same problem. I thought it may be due to an error in the installation, if there is a trick or something to change during installation. Best regards. Rachida. From: pavel.ondra...@email.cz To: wien@zeus.theochem.tuwien.ac.at Date: Wed, 20 May 2015 14:58:52 +0200 Subject: Re: [Wien] error on SCF calculation Dear Rachida, the hup: Command not found. line is harmless, however the lapw1c: Command not found is not. It looks like the lapw1c doesn't exist. It is possible it wasn't compiled properly (especially since there are multiple known problems when compiling Wien2k 13 with gfortran). Check if the lapw1c binary exists and if not then check your compile log for errors (file compile.msg in SRC_lapw1 subfolder). Also Wien2k 14.2 is the recommended version with many compilation fixes, so please upgrade to the latest version. Best regards Pavel On Wed, 2015-05-20 at 13:18 +0100, rachida lamouri wrote: hello, I am running wien2k_13 with gfortran compiler. when i try to run SCF calculation for ZnS, i get this error msg : hup: Command not found. STOP LAPW0 END /home/rachida/wien2k/lapw1c: Command not found. stop error If you could help me I would be very honorable. Pending a favorable response, please accept my respectful greetings. best regards. ___ 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/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/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/index.html
Re: [Wien] AFM calculations for YBCO6
Good evening,
Sorry, I commented first without remembering the structure correctly.
The antiferromagnetic and superconducting phase diagram of Y123O(6+x)
has been studied experimentally very extensively in the 90's with - in
my opinion - very clear results on the exact AF phases from neutron
diffraction and NMR. I don't want to drown anyone in literature, so
being an NMR guy I strongly suggest to look into the NMR work:
E. Brecht and W. W. Schmahl and H. Fuess and S. Schmenn and H.
L{\u}tgemeier and N. H. Andersen and B. Lebech and Th. Wolf,
Neutron diffraction and NQR study of the intermediate turn angle phase
formed during AFI to AFII reordering in
{YBa$_2$Cu$_{3-x}$Al$_x$O$_{6+\delta}$},
Physical Review B, 1997, 56, no.2, p 940
Allow me to suggest in addition my own work
K. Nehrke and M. W. Pieper and T. Wolf,
Local magnetic properties of PrBa2Cu3O6+x: {NMR} and {NQR} in crystals,
Phys. Rev. B, 1996, 53, 1
Perhaps you can work from there through the references for details and
additional studies.
There are two crystalografic different Cu sites in the perovskite, the
so-called chain- and twice as many plane-sites. The variable oxygen
content is between the Cu chain sites. In accord with what Prof.
Delamora noted, the Cu on chains never carrys a magnetic moment. Cu in
the planes forms a (well studied) spin-1/2 2D Heisenberg square lattice
AF (nearest neighbors antiparallel) at low oxygen concentration (the
Y123O6 end of the phase diagram), and the famous high-temperature
superconductor when the oxygen sites between chain-Cu are filled, doping
the electron holes into the planes.
If the system orders AF the planes ALWAYS form a square lattice spin-1/2
AF with nearest neighbors antiparallel. So you need to assign numbers
like Cu1 (for up) and Cu2 (for down) for Cu in the planes. The chain Cu
is on a different crystalografic site and doesn't need numbering.
What I did not remember correctly is that since you have two plane site
Cu even in the primitive cell you can, in fact, get away with a 1*1*1
cell. BUT look in XCRYSDEN or something like it to make sure that your
numbering makes nearest neighbors in the planes antiparallel! A
configuration with parallel spins along the a- or b-axis should have
significantly higher E_tot.
If you consider not only Y123O6 but introduce oxygen between Cu
chain-sits you loose the 4-fold symmetry of the c-axis even in the
non-magnetic case. This is why, as far as I recall, at least a 2*2*1
supercell is necessary. To describe the different stacking sequences
along the c-axis you will have to at least double the cell along the
c-axis as well. See the above mentioned Phys Revs and the literature
cited therein for the variety of AF phases that come about by different
stacking of these planes along the c-axis depending on the oxygen
content, or on the rare earth element.
As an additional comment: To my knowledge all evidence indicates that
there is NO coexistence of AF-order and superconductivity. At
intermediate oxygen content a phase separation occures into
superconducting and AF phases.
Good luck with your calculations,
Martin Pieper
---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564
Am 16.05.2015 21:15, schrieb delamora:
This is a problem I had studied before, but non magnetic, so I was
curious to try the magnetic case; w
With YBa2Cu3O7 you have magnetism and superconductivty! braking one of
Matthias rules
The result I got for, YBa2Cu3O6, was that Cu1 MM=0.0007 or non
magnetic as far as the calculations show.
Cu2=0.5873 with U=4eV
So, you have to study only Cu2 for magnetism.
From my point of view you the most and I would guess only ordering of
the inplane spins is like a chess board, alternating un and dn.
the interplanar Cu2-Cu2 ordering can be up-up or up-dn, but since
there are no oxygen atoms between the planes the difference should be
quite small.
I did the calculation with RxK=7 and 100 k-points with 4 parallel
cores, it took half an hour.
In the supercell output you have to delete all the numbers
Ba 1 = Ba, O 1 = O and only number 1 and 2 the Cu as you want to
order them
I did;
Cu 00x and 1/2,1/2,x as Cu 1 and later as 'up'
and for 0,1/2,x and 1/2,0,x as Cu 2 and later as 'dn'
Now you can have 00,-x as 'dn' and 0,1/2,-x as up and this gives
another ordering
De: wien-boun...@zeus.theochem.tuwien.ac.at
wien-boun...@zeus.theochem.tuwien.ac.at en nombre de Madesis
Ioannis(John) imade...@physics.uoc.gr
Enviado: sábado, 16 de mayo de 2015 01:09 p. m.
Para: Wien
Asunto: Re: [Wien] AFM calculations for YBCO6
Mr. Delamora, first of all thank you for your dedication, and thorough
examination of my problem.
I haven't fully tested your solution, however, there are plenty of AFM
orderings that I wish to test such as A, G and C type, all of which are
different combinations of in-plane
Re: [Wien] AFM calculations for YBCO6
Good eveneing, Ioannis Madesis As you observed yourself: The (primitve) unit cell of your structure does not have enough Cu1 to support AFM on that sublattice. For that you need at least two Cu1 atoms. You achieve that by doubling (at least) the unit cell (perhaps using supercell) and assigning two different numbers to the Cu-sites you want to assign opposite spins to. As far as I recall you should double the unit cell along the a- AND the b-axis to be able to assign the correct spin directions. If you want to describe AF stacking sequences along the c-axis you will have to double the unit cell along that axis. Once you assigned numbers to Cu according to their two spin directions in your supercell (say, leave Cu1 for up, rename the Cu1 you want to point down into Cu3) you can let nn and symmetry do the job of finding the corresponding space group - which will necessarily be different from 123 (consult the UG for this procedure). Good luck Martin Pieper --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 15.05.2015 14:32, schrieb Madesis Ioannis(John): Good evening everyone I am having trouble with the struct file for these calculations. To be more specific: This material has 2 nonequivalent Cu atoms, Cu1 and Cu2. Each one of these sits at the four corners of the unit cell, Cu1 at z=0 and Cu2 at approximately z=1/3, and the space group is 123. In order to achieve AFM ordering, I need to have 2 types of Cu1, and the spacegroup doesn't let me do that. Once I place Cu1(up), all 4 corners have Cu1(up). What can I do to solve this? ___ 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] A few (more) elementary -so questions (with onsite -eece)
Sorry, the comment was confusing. The spin moments specified in the initial configuration of course point along the axis specified by the direction. What I wanted to point out is that for symmetry reasons this is the only meaningfull component you can specify. In this sense symmetry comes first. The direction is more than just the spin direction. Its the only direction along which any vector operator that Wien2k can represent in the given case can point. Someone should correct me if I am wrong, but in my understanding this is (one of) the reason(s) that Wien2k can only cope with collinear magnetic moments: The axial symmetry is introduced as a global symmetry. The direction of the axis is the same in all muffin-tin spheres. I expect the Wienncm code handles this on basis of the local symmetry. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 05.05.2015 19:31, schrieb Laurence Marks: I would be interested in clarification from others, but from what I can see in the code it appears that this is the spin direction that is used, not just the direction of breaking the symmetry. I may be wrong. On Tue, May 5, 2015 at 11:47 AM, pieper pie...@ifp.tuwien.ac.at wrote: I definitely am not an expert for -so, therefore I will not shoot down anything, only a comment: From my point of view from magnetism I would ask for some caution with identifying the direction given in .inorb and .inso with 'the spin direction'. As Gerhard pointed out earlier in this thread, it's all about symmetry: The specified direction only sets up the symmetry of the case to be compatibel with whatever has a rotational invariance with that (quantization) axis - be that a spin, or orbital moment, magnetization, a magnetic field, ... The symmetry of the basis has to allow for a magnetization otherwise it won't appear when you calculate expectation values. Personally I find Pavel's 'lecture on spin-orbit.ps [1]' here in the Wien documentation files (I hope it's still there) very illuminating. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 05.05.2015 14:51, schrieb Laurence Marks: Me a culpa, I should have checked the mailing list first for the answers. That said, this issue has come up enough times in the past that I think the UG should be tweaked so it is clearer. Let me try my interpretation, so I can be shot down if needed. Within Wien2k magnetic effects can be approximately included in a number of ways. Some such as the spin-orbit coupling assume a direction for the spin vector (for all electrons actively considered), others such as Bext in orb specify a direction for an applied magnetic field (in Tesla) and use the same direction for the spin vector. (The two spin states are then either parallel or anti parallel to the specified direction.) When a direction is specified in case.inso or case.inorb this fixes the spin vector and (if used) the external magnetic field direction. Via the output files from lapwdm (case.scfdmXX) one can monitor how the angular momentum changes [1]. By using different directions for the spin vector (and field) one can probe how the energy changes and/orbital occupancies with assumed directions for the spin/external magnetic field. To escape the assumption that the spin vectors all have one direction the Wienncm code has to be used. [1] My addendum. Changes in the occupancies can be a soft electronic mode, i.e. very small changes in the energy for quite large changes in the density. The older mixing algorithms (MSEC1 or MSEC3) are not so good for soft modes and can stagnate. MSR1 is better and the next release (7.0) is much better. With onsite -eece /or -orb it may help to push the mixer by either forcing a larger step (echo .2 .msec or echo .1 .pratt) or stopping, doing a force on the orbital potential (x orb -up; x orb -dn) then restarting with -NI. It is probably wise to check how the orbital momentum is converging (grep :ORB0 *.scf, perhaps other) and make sure that the mixing is not starving (grep GREED: *.scf and check the values are not small, e.g. 0.035). ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu [2] [1] MURI4D.numis.northwestern.edu [3] [2] Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On May 4, 2015 6:22 PM, Laurence Marks l-ma...@northwestern.edu wrote: Typo: although I remember don't symmetry operations being split into these two classes everywhere in the code On Mon, May 4, 2015 at 6:04 PM, Laurence Marks l-ma...@northwestern.edu wrote: I am a newbie at -so, so a few simple questions. a) What is the meaning of the orbital moment in case.scfdm
Re: [Wien] A few (more) elementary -so questions (with onsite -eece)
I definitely am not an expert for -so, therefore I will not shoot down anything, only a comment: From my point of view from magnetism I would ask for some caution with identifying the direction given in .inorb and .inso with 'the spin direction'. As Gerhard pointed out earlier in this thread, it's all about symmetry: The specified direction only sets up the symmetry of the case to be compatibel with whatever has a rotational invariance with that (quantization) axis - be that a spin, or orbital moment, magnetization, a magnetic field, ... The symmetry of the basis has to allow for a magnetization otherwise it won't appear when you calculate expectation values. Personally I find Pavel's 'lecture on spin-orbit.ps' here in the Wien documentation files (I hope it's still there) very illuminating. --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 05.05.2015 14:51, schrieb Laurence Marks: Me a culpa, I should have checked the mailing list first for the answers. That said, this issue has come up enough times in the past that I think the UG should be tweaked so it is clearer. Let me try my interpretation, so I can be shot down if needed. Within Wien2k magnetic effects can be approximately included in a number of ways. Some such as the spin-orbit coupling assume a direction for the spin vector (for all electrons actively considered), others such as Bext in orb specify a direction for an applied magnetic field (in Tesla) and use the same direction for the spin vector. (The two spin states are then either parallel or anti parallel to the specified direction.) When a direction is specified in case.inso or case.inorb this fixes the spin vector and (if used) the external magnetic field direction. Via the output files from lapwdm (case.scfdmXX) one can monitor how the angular momentum changes [1]. By using different directions for the spin vector (and field) one can probe how the energy changes and/orbital occupancies with assumed directions for the spin/external magnetic field. To escape the assumption that the spin vectors all have one direction the Wienncm code has to be used. [1] My addendum. Changes in the occupancies can be a soft electronic mode, i.e. very small changes in the energy for quite large changes in the density. The older mixing algorithms (MSEC1 or MSEC3) are not so good for soft modes and can stagnate. MSR1 is better and the next release (7.0) is much better. With onsite -eece /or -orb it may help to push the mixer by either forcing a larger step (echo .2 .msec or echo .1 .pratt) or stopping, doing a force on the orbital potential (x orb -up; x orb -dn) then restarting with -NI. It is probably wise to check how the orbital momentum is converging (grep :ORB0 *.scf, perhaps other) and make sure that the mixing is not starving (grep GREED: *.scf and check the values are not small, e.g. 0.035). ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu [1] MURI4D.numis.northwestern.edu [2] Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On May 4, 2015 6:22 PM, Laurence Marks l-ma...@northwestern.edu wrote: Typo: although I remember don't symmetry operations being split into these two classes everywhere in the code On Mon, May 4, 2015 at 6:04 PM, Laurence Marks l-ma...@northwestern.edu wrote: I am a newbie at -so, so a few simple questions. a) What is the meaning of the orbital moment in case.scfdm* ? Is that the average direction projected to the global axis system? b) What is the physical significance of the orbital moment being parallel (or not quite parallel) to the direction used in case.inso? c) I understand that the results for different directions of B in case.inso reflect the magnetic anisotropy, but what are the units of field (if any)? d) What else is worth looking at? The partial orbital moment (:POM) seems relevant, but what exactly is it? e) I am blindly trusting that initso knows what it is doing, and have left the B symmetry operations in case.struct (although I remember symmetry operations being split into these two classes everywhere in the code). This seems to conflict with Pavel's notes, although those may be too old. Thanks. -- Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu [1] Corrosion in 4D: MURI4D.numis.northwestern.edu [2] Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi -- Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu [1] Corrosion in 4D: MURI4D.numis.northwestern.edu [2] Co-Editor, Acta Cryst A Research is to see what everybody
Re: [Wien] Physical significance of magnetization direction with -so
I hope my five cent might be usefull: If you do have magnetic moments, be they ferro-, ferri-, or antiferromagnetic, or induced by an external field, the results can depend on the orientation of the moments. In addition, keep in mind that the various magnetic moments one likes to think of may not be constants of motion, or good quantum numbers, so they cannot be used to specify the eigenstates. Inasmuch as S and L are good quantum numbers Hund's first and second rules for single ions state that the ground state of an electron shell in the Coulomb potential of the nuclear charge will have maximum total spin moment S=\sum s_i and maximum total angular momentum L=\sum l_i compatibel with the Pauli principle. L and S are constants of motion, they commute with the Hamiltonian of the nuclear Coulomb potential. L and S can, therefore, be used to enumerate the eigenstates. The energy is given by the size of L and S. The direction of L and S is unimportant (for a single ion without magnetic field applied!) the nuclear Coulomb potential is radial symmetric. Each state of the shell is (2L+1)(2S+1)-fold degenerate. This degeneracy is partially split by spin-orbit coupling and by the electrostatic crystal field. You find the Hamiltionian for spin-orbit coupling H_so=\lambda (S*L) with S and L the vectors of the total spin and angular momentum of the electrons in one shell (say, d-, or f-shell) from considering the energy of an electron spin s in the magnetic field due to the relative motion of the nuclear charge with respect to the electron on its path at angular momentum l around the nucleus. With this H_so only J=L+S stays a constant of motion, neither S nor L. You can find the magnitude of J by Hund's third rule if H_so stays the most important player down to that energy. However, this is still just the radial symmetric potential of a single nuclear charge, the energy of the eigenstates is given by specifying J and you can point J in any direction. In a crystal the charge distribution of the surrounding ions also acts on the electrons in the shell. This crystal field Hamiltonian obviously is not spherical symmertric, it will reduce the rotational symmetry to the point symmetry of the site. If there is a magnetic moment on an ion the energy will depend on its direction in the lattice. This is the source of the magnetic single-ion anisotropy. What the eigenstates of the combined Hamiltonian look like depends on the relative size and symmetry of the contributions. For outer d-shells crystal fields usually dominate over H_so leading e.g. to what is called the quenching of the orbital angular momentum (L=0) which is sensitive to magnetic fields. For 4f-shells which are shielded by outer d- and s-shells H_so is frequently dominant and Hund's third rule often survives, allowing to calculate J and consider the dependance of the energy on its direction in the lattice. Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 03.05.2015 19:37, schrieb Laurence Marks: An elementary question: do the results of -so depend upon the magnetization direction used in initso, or should they in principle be independent of it? -- Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu [1] Corrosion in 4D: MURI4D.numis.northwestern.edu [2] Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi Links: -- [1] http://www.numis.northwestern.edu [2] http://MURI4D.numis.northwestern.edu ___ 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/index.html
Re: [Wien] Crystal field splitting
Dear Salman, unfortunately I have neither means (no Wien2k on my notebook) nor time right now to quickly provide an answer to your problems. Anyhow, directing you to literature you can cite is probably better than an elaborate email lecture on crystal field splittings (not being really an expert I am not the ideal person to give such a lecture anyway). To me the best way to proceed seems to be be: - get your hands on some textbook on magnetism with a chapter on crystal field effects. Kei Yosida: Theory of Magnetism, Springer Series in Solid State Science vol 122 might be one choice. Personally, when I ran into crystal field effects a rather long time ago, I liked the article of M. T. Hutchings: Point charge ... in F. Seitz, D. Turnbull, Solid State Physics, vol. 16, p. 227, Academic Press 1964. - read (understand) that chapter. You should know at this stage which parameters appear in the crystal field Hamiltonian if you know L and the point symmetry. I did not use Pavel Novaks new package for Wien2k up to now but I expect that you can understand its I/O on that basis. And you should be aware of the limitations of the concept, especially in a 3d-metal! - Find the configuration of B nearest neighbors for the two non-equivalent Ni sites in your structure. The idea is that charges farther out are pretty much shielded. Simply look at the structure using Xcrysden or something similar to identify this configuration. In Wien2k, outputnn gives you the number and position of these B atoms. My guess (a guess, not more!) is, that one Ni site has a (distorted) octahedral configuration, the other a (also distorted) tetrahedral one. CF splitting usually is considered in steps of decreasing importance of the contribution in the Hamiltonian: first approximate the situation by a cubic crystal field, then take into account, say, tetragonal distortions of the ideal octahedron, and so on, lowering symmetry in each step. I hope this helps, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 03.11.2014 13:41, schrieb Salman Zarrini: +++ Dear Martin and Delamora, Many thanks for your answer, actually, controversial state for me here is 3d orbitals of Ni elements crystalline together with boron in a orthorhombic structure, Pnma space group. I can see different kind of point group in case.outputsgroup(1,1,C1 for sort one, m,m,Cs for atom sort 2 and m,m,Cs for atom sort 3 and a mmm,2/m,2/m 2/m,D2h in the end for the whole structure)but I can not make a link between them and potential crystal filed splitting for this structure, for convenience the struct file has been enclosed, I would be thankful if you guided me to find the proper crystal filed. Cheers, Salman +++ ___ 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/index.html
Re: [Wien] Crystal field splitting
Dear Salman, to be a little bit more precise, crystal field splitting is a local concept: The Coulomb interaction of all electrons in a given shell characterized by its angular momentum with the surrounding charge distribution is represented by a minimal set of angular momentum operators. To make sense this should to be a shell of reasonably well localized electrons (usually a 4f-shell). If you have a metal the bandwidth of that particular band should be small, the conduction electrons should be contributed by other shells. It also means that it is NOT the overall crystal symmetry you have to consider, but the point symmetry group of the crystalografic site occupied by the (Rare Earth?) element you are interested in. When your structure has sites with, say, octahedral and tetrahedral symmetry the crystal field Hamiltonian (and its eigenvalues, the splitting) will reflect these symmetries. The point group symmetries are tabulated, e.g. for the Wyckoff positions at the Bilbao server (http://www.cryst.ehu.es/) Best regards, Martin --- Dr. Martin Pieper Karl-Franzens University Institute of Physics Universitätsplatz 5 A-8010 Graz Austria Tel.: +43-(0)316-380-8564 Am 03.11.2014 03:56, schrieb delamora: The crystal field splitting depends on the crystal symmetry!!, but for a compound with metallic characteristics the bandwidth will be larger that the field splitting. De: wien-boun...@zeus.theochem.tuwien.ac.at wien-boun...@zeus.theochem.tuwien.ac.at en nombre de Salman Zarrini salman.zarr...@tu-darmstadt.de Enviado: domingo, 02 de noviembre de 2014 04:57 p.m. Para: wien@zeus.theochem.tuwien.ac.at Asunto: [Wien] Crystal field splitting Dear Wien2k users, I was wondering that how can I find out which kinds of crystal field splitting (Octahedral, Tetrahedral, Pentagonal bipyramidal,...) have been applied on my bulk metallic structure? Best regards, Salman ___ 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/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/index.html
