question regarding integrated DOS.

Message

applelinux · #1 Post by **applelinux** » Thu Oct 12, 2006 9:57 pm

Hi, I did some DOS calculations for Al and Cr. I found that the integrated DOS value for Al will remain at 10 when energy is above some values. As a result the DOS has a drop to zero above that energy. The same thing happened to Cr too with the integrated DOS remains at 20. I used PAW potential and follow the standard procedure in the manual. The calculations for Au and Pt look fine...Can anyone give some suggestions?

The following is part of the DOSCAR for Al calculation. Note the fermi energy determined from this calculation is 6.54707907. The integrated DOS has correct value, i.e. 3 at the fermi level.

20.431 0.1874E+00 0.9903E+01
20.491 0.1734E+00 0.9914E+01
20.551 0.1702E+00 0.9924E+01
20.611 0.1391E+00 0.9934E+01
20.671 0.1311E+00 0.9942E+01
20.731 0.1265E+00 0.9950E+01
20.792 0.1206E+00 0.9957E+01
20.852 0.1120E+00 0.9964E+01
20.912 0.1003E+00 0.9970E+01
20.972 0.8994E-01 0.9976E+01
21.032 0.9092E-01 0.9981E+01
21.092 0.8899E-01 0.9987E+01
21.152 0.6601E-01 0.9992E+01
21.212 0.4168E-01 0.9995E+01
21.273 0.1726E-01 0.9997E+01
21.333 0.9621E-02 0.9997E+01
21.393 0.8212E-02 0.9998E+01
21.453 0.6904E-02 0.9998E+01
21.513 0.5697E-02 0.9999E+01
21.573 0.4591E-02 0.9999E+01
21.633 0.3588E-02 0.9999E+01
21.693 0.2685E-02 0.1000E+02
21.754 0.0000E+00 0.1000E+02
21.814 0.0000E+00 0.1000E+02
21.874 0.0000E+00 0.1000E+02
21.934 0.0000E+00 0.1000E+02
21.994 0.0000E+00 0.1000E+02
22.054 0.0000E+00 0.1000E+02
22.114 0.0000E+00 0.1000E+02
22.174 0.0000E+00 0.1000E+02
22.234 0.0000E+00 0.1000E+02
22.295 0.0000E+00 0.1000E+02
22.355 0.0000E+00 0.1000E+02
22.415 0.0000E+00 0.1000E+02
22.475 0.0000E+00 0.1000E+02
22.535 0.0000E+00 0.1000E+02
22.595 0.0000E+00 0.1000E+02
22.655 0.0000E+00 0.1000E+02
22.715 0.0000E+00 0.1000E+02
22.776 0.0000E+00 0.1000E+02
22.836 0.0000E+00 0.1000E+02

#2 Post by **admin** » Thu Oct 19, 2006 12:19 pm

please check how large NBANDS was (in OUTCAR, unless set explicitely, it is calculated automatically by VASP). If the highest band is at about 21.693, there will be no more states to integrate, though EMAX may be set higher.
If you want to get the DOS up to higher energies, please explicitely increase NBANDS in INCAR.

applelinux · #3 Post by **applelinux** » Thu Oct 19, 2006 10:28 pm

thanks for reply...It is solved by increasing the NBANDS...