Another way to get feff to run is to define the stoichiometry. I found that the following cards work. Change the SCF definition to the following SCF 3.46808 0 30 0.05 and change the potential definition to POTENTIALS * ipot z [ label l_scmt l_fms stoichiometry ] 0 26 Fe -1 -1 1 1 26 Fe -1 -1 1000 2 29 Cu -1 -1 1 This makes feff assume that there are many Fe atoms for each Cu atom, which was what I was assuming from your feff.inp. Cheers, Josh

Message: 1 Date: Thu, 21 Feb 2008 13:52:31 -0600 From: "Matt Newville" Subject: Re: [Ifeffit] problem with feff.inp To: "XAFS Analysis using Ifeffit" Message-ID: Content-Type: text/plain; charset=ISO-8859-1

It looks like the SCF calculation is having a hard time with the new potential. The input file you sent only has atoms out to 2.48 Ang (though perhaps you truncated it when posting???), but the

SCF 4 0 30 0.2 1

in your feff.inp means to make a self-consistent potential out to 4 Ang. Without any atoms out that far, I can imagine it would have problems. I think you'll probably need a larger cluster than a single shell for a reasonable XANES simulation.

Hope that helps,

--Matt