[Wien] unoccupied DOS continuum states

[Jorissen Kevin]

	-----Oorspronkelijk bericht----- 
	Van: Peter Blaha [mailto:pblaha at zeus.theochem.tuwien.ac.at] 
	Verzonden: wo 10/22/2003 2:34 
	Aan: wien at zeus.theochem.tuwien.ac.at 
	CC: 
	Onderwerp: Re: [Wien] unoccupied DOS continuum states
	
	

	> we calculate the unoccupied DOS at high energies (up to 70 eV above the Fermi level)because we are interested in the excitation spectra. Our results are systematically too low at high energies compared to experiments and to calculations with a hydrogenic model.
	> Possible reasons for the discrepancy are :
	> * DFT is not suited for excited states.
	
	This is for sure one of the reasons. IT's a "fundamental" reason and only
	more evolved methods like GW or TD-DFT might help.
	
	> * the (L)APW+lo basis set is only good in a limited energy region.
	
	This could be a "technical" problem with our linearized methods. Try to
	add local orbitals at high energies (eg. 3Ry) and check if the spectrum
	is sensitive to it.
	
	
	                                      P.Blaha
	

	We have done that, of course, and yes, the spectra are sensitive to it.  Unfortunately we cannot add more than one extra linearization energy, and we cannot be sure that adding more lo's would not improve the situation even further.

	Kevin.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 4826 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20031023/cf19b85e/attachment.bin