Hi, all: I met some convergence issue when I made the charged state calculation. The situation is, when I made a charged state calculation, the SCF can’t achieve convergence. But in all neutral and most charged states, this issue won’t show up but only in certain charged states. For your convenience, the calculation I’m making is an isolated vacancy on diamond surface. I observed that the defect levels tend to be highly unlocalized which makes the related bands very dispersive. In this sense, the magnetic moment exhibits to be highly unstable:
:MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.53572 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.18060 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.10316 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.11329 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.11916 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = -0.04368 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.06194 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.22852 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.24276 :MMTOT: SPIN MAGNETIC MOMENT IN CELL = 0.20449 Also, By checking the :DIS label: :DIS : CHARGE DISTANCE ( 0.0591612 for atom 49 spin 1) 0.0126272 :DIS : CHARGE DISTANCE ( 0.1016792 for atom 49 spin 1) 0.0145533 :DIS : CHARGE DISTANCE ( 0.0768202 for atom 49 spin 1) 0.0186368 :DIS : CHARGE DISTANCE ( 0.0771968 for atom 49 spin 1) 0.0136267 :DIS : CHARGE DISTANCE ( 0.0825579 for atom 49 spin 1) 0.0120297 :DIS : CHARGE DISTANCE ( 0.0805263 for atom 49 spin 1) 0.0113403 :DIS : CHARGE DISTANCE ( 0.0750162 for atom 49 spin 1) 0.0110824 :DIS : CHARGE DISTANCE ( 0.0710646 for atom 49 spin 1) 0.0105623 :DIS : CHARGE DISTANCE ( 0.0486714 for atom 49 spin 1) 0.0087271 :DIS : CHARGE DISTANCE ( 0.0630542 for atom 49 spin 1) 0.0120210 the atom 49 (surface C on the edge of unit cell) seems to be the issue. I doubted it’s due to the k mesh I’m using is too coarse. Thus, I tried a finer k mesh (increase from 3x3x2 to 5x5x2) but it didn't make a difference still. Before using a finer k mesh, can anyone give a suggestion on this? If any other information about my calculation is needed, please let me know. thanks! Wenhao _______________________________________________ Wien mailing list [email protected] http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/[email protected]/index.html
