Re: [Wien] how to define new z and x axis in case.qtl file by visualizing structure

2020-04-09 Thread Wasim Raja Mondal
I am not able to figure out how to give h,k,l values to define x and z axis. On Thu, Apr 9, 2020 at 8:39 PM Wasim Raja Mondal wrote: > Dear Prof. Blaha, > Thank you for pointing it out. Do you have any > guess value for putting h,k,l value in case.qtl file to define new

Re: [Wien] how to define new z and x axis in case.qtl file by visualizing structure

2020-04-09 Thread Wasim Raja Mondal
Dear Prof. Blaha, Thank you for pointing it out. Do you have any guess value for putting h,k,l value in case.qtl file to define new x and z axis for V2O3 corundum structure? Regards Wasim On Thu, Apr 9, 2020 at 2:13 AM Peter Blaha wrote: > But in that case you will NOT

Re: [Wien] Treating 4f states as core

2020-04-09 Thread Ali Baghizhadeh
Thank you very much for your comments Prof. Laurence. Your paper is one of my references to do my set of calculations, and interprate my EELS data later on. Basically h-REFeO3 (RE: Lu, Yb, Sc ..) is AFM basal plane with non-collinear spins of Fe3+. Magnetism of Yb should contribute on c-axis

Re: [Wien] Treating 4f states as core

2020-04-09 Thread Laurence Marks
Treating the Ln atoms as core ignores the crystal field splitting. In my opinion it is an approximation used in to get the calculation to converge, except for La & Gd I cannot justify it. The Hubbard +U puts the 4f very low in energy, in effect "shoves them out of the way". Based upon our

Re: [Wien] Treating 4f states as core

2020-04-09 Thread Ali Baghizhadeh
Dear Prof. Laurence Thank you. In the mentioned paper, they treated Yb 4f electrons as valance while some for HoMnO3 or DyFeO3, treated 4f electrons as core. I do not know what is the best approach? For most h-LuFeO3 family there is experimentally a weak FM, and there are some ideas on its

Re: [Wien] Treating 4f states as core

2020-04-09 Thread Laurence Marks
Let me turn this thread around a bit, as it seems to be endless. There are several possible reasons you are not getting an AFM solution, the two main ones being: a) You are doing the calculation wrong. b) It is not AFM. What is the evidence that it should be AFM? Your opinion, or uncontroversial

Re: [Wien] Treating 4f states as core

2020-04-09 Thread Ali Baghizhadeh
Dear Prof. Blaha I did run simulation accordingly. The MMTOT increases as SCF cycles increases. After 100 cycles I could not get convergence criteria (runsp_lapw -p -ec 0.0001 -cc 0.001). I will try to run it again, after 40 cycles I may switch to LDA+U on Fe and Yb with following lines in

Re: [Wien] how to define new z and x axis in case.qtl file by visualizing structure

2020-04-09 Thread Peter Blaha
But in that case you will NOT get the splitting given in that paper, but an approximate eg/t2g splitting. The a1g (d-z2) orbital in V2O3 does NOT point towards an oxygen, but in between 3 oxygens. Am 09.04.2020 um 03:48 schrieb Wasim Raja Mondal: Dear All,              Anybody have defined x