[Wien] How to calculate U for 3d-fully filled

2015-06-01 Thread Tuan Vu
Thank Pro. Gavin Abo and Pro. Mehmet Topsakal i just read paper Calculating the effective U in APW methods. NiO // Georg K. H. Madsen and Pavel Novak For NiO , element Ni (3d^6 4s^2) - has 8 d electrons. we can make two calculations. One with 4.5 ↑, 4 ↓ and one with 4.5↑, 3 ↓ constrained

[Wien] LDA-U error for SrVO3 during orb

2015-06-01 Thread Soumen Bag
Dear Wien2k experts, I want to do LDA-U non magnetic calculation for SrVO3 using *LSDA* correlation function. I am getting following error during *ord* execution. = *end-of-file during read*, unit 10, file ~/SRVO/SRVO3.dmat = i used following

[Wien] Partially Occupied Wyckoff site

2015-06-01 Thread Farshad Nejadsattari
Dear Wien2k community, I am working on an Iron Selenide superconductor with a formula (Li0.8Fe0.2)OHFeSe. The 5 Wyckoff positions are given below: [image: Inline image 1] My question is how can one include the Li/Fe1 contribution in the master case.struct file as the 3rd position is partially

Re: [Wien] Partially Occupied Wyckoff site

2015-06-01 Thread Pascal Boulet
Dear Farshad, There is no way but using a supercell with 100 of these sites populated with 81 Li atoms and 19 Fe atoms. Otherwise, you have to use an approximate model of the real structure. HTH Pascal Le 1 juin 2015 à 15:48, Farshad Nejadsattari fneja...@uottawa.ca a écrit : Dear Wien2k

Re: [Wien] Partially Occupied Wyckoff site

2015-06-01 Thread Farshad Nejadsattari
Thank you Pascal for your kind response, I will try both ways to see what I get. regards, Farshad On Mon, Jun 1, 2015 at 10:03 AM, Pascal Boulet pascal.bou...@univ-amu.fr wrote: Dear Farshad, There is no way but using a supercell with 100 of these sites populated with 81 Li atoms and 19

Re: [Wien] LDA-U error for SrVO3 during orb

2015-06-01 Thread Gavin Abo
As described in section 4.5.6 Orbital potentials on page 47 in the WIEN2k 14.2 usersguide [ http://www.wien2k.at/reg_user/textbooks/usersguide.pdf ], non-magnetic LDA+U calculations must be done spin-polarized (runsp_c_lapw -orb). It looks like you ran x orb instead of x orb -up (or x orb