[Wien] strange behavior of QTL's

[Peter Blaha]

It is not particular strange for 4f systems with many atoms.
The 4f bands are very narrow and if the potential of an atom is shifted 
with respect to others by eg. 0.2 Ry, it can easily lead to full (or no) 
occupancy of the seven 4f bands.

Most likely, the mixer realized that he was pushing too much and has 
made a "BackTrack" step (check :mix).

Most likely, one would have seen this also in

:VCOUL034 ATOMNUMBER= 34 Gd      VCOUL-ZERO = -0.15646E+01
which jumps in this iteration.

Important is that it converges. A system with 34 (or more) atoms and one 
(or more?) 4f atoms is not easy to converge.

Peter Blaha

Am 13.10.2022 um 14:53 schrieb Lyudmila Dobysheva via Wien:
> Dear all,
> 
> I calculate systems with f-electrons, and have noticed some strange 
> convergence during the cycles (not once and not in one system).
> In the first, second, forth iterations the filling of the f-bands is 
> rather reasonable, but what happens in the 3,5,6 ites with the down charge?
> 
> 1ite
> :QTL034: 1.0233 2.8713 0.2151 6.9471.
> :QTL034: 1.0176 2.8442 0.1827 0.0900.
> 2ite
> :QTL034: 1.0242 2.8740 0.2214 6.9507.
> :QTL034: 1.0182 2.8467 0.1878 0.0990.
> 3ite
> :QTL034: 1.0359 2.9067 0.3241 6.9840.
> :QTL034: 1.0293 2.8831 0.2661 6.9510.
> 4ite
> :QTL034: 1.0251 2.8773 0.2304 6.9555.
> :QTL034: 1.0194 2.8503 0.1944 0.1155.
> 5ite
> :QTL034: 1.0320 2.8969 0.2838 6.9750.
> :QTL034: 1.0257 2.8716 0.2367 6.2775.
> 6ite
> :QTL034: 1.0320 2.8965 0.2829 6.9747.
> :QTL034: 1.0257 2.8710 0.2360 6.1812.
> 
> :DIS ( 1.5148058 for atom 34 spin 2) 0.4262988
> :DIS ( 1.4717688 for atom 34 spin 2) 0.4068688
> :DIS (10.3075097 for atom 34 spin 2) 0.3936483
> :DIS ( 1.4038974 for atom 34 spin 2) 0.3839130
> :DIS ( 8.9221003 for atom 34 spin 2) 0.3558997
> :DIS ( 8.7445558 for atom 34 spin 2) 0.3531774
> 
> This worsens the whole convergence, IMHO. Is this behaviour normal?
> Finally, the systems usually converge, with normal results, and there is 
> no such jumps closer to the end of the cycle. This strangeness is only 
> at first iterations, and does not affects the final results.
> 
> Best regards,
> Lyudmila Dobysheva
> ------------------
> http://ftiudm.ru/content/view/25/103/lang,english/
> Institute of Physics and Technology,
> Udmurt Federal Research Center, Ural Br. of Rus.Ac.Sci.
> 426000 Izhevsk Kirov str. 132
> Russia
> ---
> Tel. +7 (34I2)43-24-59 (office), +7 (9I2)OI9-795O (home)
> Skype: lyuka18 (office), lyuka17 (home)
> E-mail: lyuka17 at mail.ru (office), lyuka17 at gmail.com (home)
> _______________________________________________
> Wien mailing list
> Wien at 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

-- 
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300
Email: peter.blaha at tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at
-------------------------------------------------------------------------