Dear all,
it seems to me that there is a small issue/bug in the way the .DOWN.LDOS and
.DOWN.RDOS files are printed during a TBTrans calculation.
Indeed, when performing a spin-polarized calculation, the .DOWN. files do not
contain the DOS for the spin down channel in the leads but the sum of the spin
UP and DOWN channels.
I don't know if this was done on purpose but it puzzled me for a while so I
decided to mention it on the mailing list ;o)
In order to obtain the correct UP and DOWN channels, lines 756 and 774 of the
tbtrans.F file have to be changed from
bulkdosL(iEn)=bulkdosL(iEn)-(dimag(zbulk)/Pi)*wkxy(ikxy)
bulkdosR(iEn)=bulkdosR(iEn)-(dimag(zbulk)/Pi)*wkxy(ikxy)
to
bulkdosL(iEn)=-(dimag(zbulk)/Pi)*wkxy(ikxy)
bulkdosR(iEn)=-(dimag(zbulk)/Pi)*wkxy(ikxy)
Best regards,
Simon
Dr. Simon M.-M. Dubois
Research Associate
University of Cambridge
Theory of Condensed Matter (TCM) Group
Cavendish Laboratory
19, J. J. Thomson Avenue
Cambridge CB3 0HE, U.K.
