Re: [Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Peter Blaha
You wrote in an earlier email, that you put 3 4f electrons in the core and correspondingly reduced the number of valence electrons by 3. This would correspond to a Pr2+ state, which is not ver common. Although I've not studied that much these compounds, I'd put only 2 4f electrons into the

Re: [Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Laurence Marks
If the DOS, both occupied and unoccupied match, the Fermi energy must. Probably a difference in definition and/or smearing. Or you are losing electrons somehow. On Thu, Jan 30, 2020 at 12:55 PM Anup Shakya wrote: > Dear Prof. Blaha, > > I have plotted both the band structure and the DOS in

[Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Anup Shakya
Dear Prof. Blaha, I have plotted both the band structure and the DOS in Rydberg and they match very nicely with each other. Regarding the comparison with the reported DOS and band structure one needs to give some shift to the Fermi level so that it matches the reported results. Is it because of

Re: [Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Laurence Marks
Not directly related to the mismatch, I don't expect either using +U or putting the 4f into the core to be correct. For LnScO3 (PRM 2 (2018) 025001) -eece was significantly better, combined with +U to estimate the effect of a valence hole on the width (Slater-Janek, similar to LDA 1/2). Different

Re: [Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Peter Blaha
Of course there is a DOS in the papers you quoted. And they show where EF is. Plot both, the band structure and the DOS in Ry and put them next to each other. There must be some mistake, but I don't know where On 1/30/20 12:36 PM, Anup Shakya wrote: Dear Prof. Blaha, The DOS for

Re: [Wien] something about the density matrix

2020-01-30 Thread Peter Blaha
You can check the "update" page for a history and the main changes. I'm quite sure that the 2016 dmat in lapw2 was not correct in all cases (symmetries). On 1/30/20 1:54 PM, 姜若诗 wrote: I used version: WIEN2k_16.1, We don’t update the most recent version. Maybe it’s something cause it wrong.

Re: [Wien] something about the density matrix

2020-01-30 Thread 姜若诗
I used version: WIEN2k_16.1, We don’t update the most recent version. Maybe it’s something cause it wrong. I will use WIEN2k_19.1 to do the same thing, If I still have problem, I will let you know. Thank you very much! Regards, Jasmine. From: Peter Blaha Sent: Thursday, January 30, 2020 8:28

Re: [Wien] something about the density matrix

2020-01-30 Thread Peter Blaha
Are you using the most recent WIEN2k_19.1 version ?? (Previous versions could have had some bug). If yes, Please send me the struct file (to my private email) and I'll try to verify (and fix) the problem. Regards Peter Blaha On 1/30/20 1:16 PM, 姜若诗 wrote: Yes, everything is right, but the

Re: [Wien] something about the density matrix

2020-01-30 Thread 姜若诗
Yes, everything is right, but the answer is wrong. However, I try to use x lapwdm to create the new density matrix, the answer is different with the former density matrix which calculate by x lapw2. And the new trace of density matrix is agree with the partial charges sphere d orbital. ( equal

[Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Anup Shakya
Dear Prof. Blaha, The DOS for this material has not been reported so I could not compare it. In both cases without SOC and with SOC the DOS and band structures calculated in Wien2K by me match with each other. I have taken the correct Fermi energy from case.scf file and for SOC I have performed

Re: [Wien] something about the density matrix

2020-01-30 Thread Tran, Fabien
In principle they should agree. Is this block of the density matrix really corresponding to atom 1 and the corresponding spin? From: Wien on behalf of ??? Sent: Thursday, January 30, 2020 11:23 AM To: wien@zeus.theochem.tuwien.ac.at Subject: [Wien] something

[Wien] something about the density matrix

2020-01-30 Thread 姜若诗
Dear all, I have a question, below is the data of case.scf. Why the trace of the density matrix is not consistent with the :QTL001, partial charges sphere of the d orbital(L = 2)? Did something wrong? Looking forward to your reply. Thank you. Regards, Jasmine

Re: [Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

2020-01-30 Thread Peter Blaha
Does the DOS match (without shift ?) Did you put the correct EF into case.insp (from the calculations with/without SO) ? Did you runx spaghetti -so ?? On 1/30/20 8:28 AM, Anup Shakya wrote: Dear All,          I have been trying to perform calculations for PrBi by keeping the Pr 4f

Re: [Wien] "LAPW2: semicore band-ranges too large" error in LSDA+SO

2020-01-30 Thread Peter Blaha
Are you using the most recent WIEN2k version ? There were a couple of bugs fixed in symmetso and init_so. Anyway, with this magnetization direction your symmetry has been reduced and maybe even the number of non-equivalent atoms has been increased. I suggest that AFTER init_so