[Wien] problem with SPAGHETTI

[katrusiat]

Dear Dr. Blaha,

Actually, the commands were all the time with the flag -p like x lapw1c -p -band, x spaghetti -p. I have calculated bands many times before for various compounds, so probably you are right and the problem lies in the specific features of the system that is being studied.

The system is really very big: the BSCCO cuprate, doped with an oxygen atom, accordingly a supercell is produced plus vacuum of appr. 15 A is added in c-direction. There are 77 inequivalent atoms, and the symmetry is 8 Cm:

CXZ LATTICE,NONEQUIV.ATOMS: 77 8 Cm
MODE OF CALC=RELA unit=bohr
 27.740304 58.366723 27.740304 90.000000 90.000000118.377323

The matrix was truncated from required 7.0 to 5.46 (accordingly, the warning in scf), but then I let it be like that because I did not need especial accuracy for energy etc. I was primarily interested in bandstructure. Also, the 14000x14000 matrix was already a lot to calculate.

The scf cycle was well converged. The NUME = 1000.

The strange thing is that in several .output1_* files (total - 24  the data are written to the end, while in most of the files they are cut.







-- 
Яндекс.Открытки. От всей души http://cards.yandex.ru/