The reason why USPP have been introduced is not for some form of masochism but because for certain elements of the first row (C-N-O-F) or for first row transition metals or rare earths NCPP would require very high cutoff. Most other elements (including Se,Te) can be treated with much lower cutoffs and do not need USPP construction... It is convenient therefore to use USPP for the hard elements while you can stick to NCPP when this is not needed. And you CAN mix different types.
And always test convergence of your calculations. stefano Gianluca Giovannetti wrote: > Dear Paolo, > > thank you for the reply. > Probably i`m missing something in your statement. :-) > Are you suggesting to use > > Fe.pbe-sp-van.UPF<http://www.quantum-espresso.org/pseudo/1.3/UPF/Fe.pbe-sp-van.UPF> > <http://www.quantum-espresso.org/pseudo/1.3/UPF/Te.pbe-rrkj.UPF> > Te.pbe-rrkj.UPF<http://www.quantum-espresso.org/pseudo/1.3/UPF/Te.pbe-rrkj.UPF> > Se.pbe-van.UPF<http://www.quantum-espresso.org/pseudo/1.3/UPF/Se.pbe-van.UPF> > > pseudopotential files evenif there are belonging to different types? > > cheers, > > Gianluca > > On Mon, Jun 21, 2010 at 9:59 AM, Paolo Giannozzi <giannozz at > democritos.it>wrote: > > >> On Monday 21 June 2010 08:04, Gianluca Giovannetti wrote: >> >> >>> i would appreciate much more to have Te Vanderbilt ultrasoft >>> pseudopotential. >>> >> why? the big advantage of ultrasoft PPs is to require a much >> smaller cutoff. In the case of Fe-Te-Se systems, Fe is by far >> the hardest element. The cutoff needed for ultrasoft Fe might >> well be sufficient for norm-conserving Te (and Se as well). >> >> P. >> -- >> Paolo Giannozzi, Democritos and Udine University >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum >
