Hi Marcos, Thank you, I looked through the SIESTA archive and found Item #110 from May 2003 where Javier Junquera responded to this very topic in more detail.
When I use fhi98pp for generating PP for use with ABINIT, it naively seems like I don't encounter this limitation. For instance in the case of Zr, I can use the neutral or ionic (+2) electronic configuration to generate the PP. I have confirmed this by looking at the *.dat file that are produced by fhi98pp program. They explicitly state the net charge for the reference electronic configuration use in generating the PP. The *.cpi files which contain the actual PP are also different. (I ran 'diff' on the files). Note however that there is still only one PP for each angular momentum channel. In both cases, there is only one pseudowave function generated for each angular momentum channel. It is generated for the lowest n with same l; in this example there is 4s, but no 5s pseudowave function generated by the program. The pseudowave functions generated between neutral and ionic configuration are not identical either. The 4d pseudowave function in the ionic configuration seems to be slightly contracted while the other pseudo wave function don't experience much of a change. Either one of these PP (constructed from the ionic or neutral) configuration yield nearly identical equilibrium geometries for FCC and HCP Zr. The bulk calculation obviously contained the 5s electrons. In summary, fhi98pp seems to be able to generate PP for multiple n same l. Or at least it some how takes into account the 5s orbital. Alberto Garcia wrote to me today stating that ATOM is setup for only one orbital per angular momentum channel. So, is this a limitation of ATOM or is this actually a limitation of the troullier-martins procedure? How is fhi98pp, taking into account the 5s electrons? I would appreciate your feedback and welcome the insight of others. Bests, On Thu, 2004-08-12 at 18:33, Marcos Verissimo Alves wrote:
Nichols, I believe a similar topic has already come up in the list some months ago. As far as i remember, there is no way of generating a psp the way you want to do with the implementation of the atom program as it comes with siesta. You will have to choose which s channel you want, the 4s or 5 s, and this is a limitation of the troullier-martins procedure. Recently, I remember having seen a paper by J. L. Martins in which this problem is overcome, but I don't know if it would require to change the atom program AND siesta as well. Best, Marcos On Thu, 12 Aug 2004 11:54:21 -0500, Nichols A. Romero wrote > Hi, > > I'm trying to generate a Zr pseudopotential (PP) with s and p semi-core > states using ATOM with a neutral atom configuration. > > The neutral atom configuration for Zr is > 1s^2 2s^2 2p^6 3s^2 3p^6 3d^10 | 4s^2 4p^6 4d^2 5s^2 > > I want to keep the 4s and 4p electron as semi-core states. I only > need one projector per angular momentum channel (i.e. use the same projector > for 4s and 5s). However, when I use the following input file with > ATOM, no PP is produced. I don't even get an error flag though. > > pg Zr > tm2 > n=Zr c=car > 0.0 0.0 0.0 0.0 > 6 4 > 4 0 2.00 0.00 > 4 1 6.00 0.00 > 4 2 2.00 0.00 > 5 0 2.00 0.00 > 1.35 1.45 1.55 1.35 > > If I remove the information for n=5 l=0 and change the number of valence > orbitals to 3, I end up with a PP generate for the ionic > configuration Zr+2. > > Does anyone have any ideas as to how I can do this for neutral Zr > configuration? > > Bests, > -- > Nichols A. Romero > > 4-133 ESB, MC 704 508 East Michigan Avenue > 1110 West Green Street Apartment 12 > Urbana, IL 61801-3080 Urbana, IL 61801-5176 > (217) 333-6108 (217) 417-5210 Marcos Verissimo Alves Departamento de Fisica Universidade Federal de São Carlos São Carlos, SP - Brasil *********************************** Marcos Verissimo Alves Physics Department Universidade Federal de Sao Carlos São Carlos, SP - Brazil
-- Nichols A. Romero 4-133 ESB, MC 704 508 East Michigan Avenue 1110 West Green Street Apartment 12 Urbana, IL 61801-3080 Urbana, IL 61801-5176 (217) 333-6108 (217) 417-5210
