Dear Antonio, It appears that you've just experienced a common problem with hexagonal space groups that one often needs a shift of the origin to get the "correct" crystal structure. According to the atoms documentation (click on SHIFT in Webatoms) there should be a note about this when you try to run, but I didn't see it anywhere when I tried WEBATOMS today. Anyway the fix is to introduce a MS shift vector of (0 0 .25), as in the atoms.inp file I sent you. I think it might be useful if an appropriate shift were introduced by default into the atoms database for very common lattice structures like graphite, or otherwise to detail more precisely the coordinates of the shift that one has to introduce by hand into atoms.inp (which is less desirable). Maybe Bruce Ravel can consider these friendly suggestions in future releases of atoms. I hope you don't mind, but I've forwarded this msg to the ifeffit list. Cheers, John ---------- Forwarded message ---------- Date: Mon, 16 Jan 2006 09:20:39 +0100 From: Antonio Politano To: 'John J. Rehr' Subject: R: R: XANES calculations Yes, I understand about stoichiometry. But I don't have understood why using an atoms.inp file taken from the Atoms.inp archive I have six nearest neighbours of the central assorber at 1,42 A. In your feff.inp file under the ATOMS card there are only three nearest neighbours at 1,42 A (this is correct!!!). Thank you very much!!! Antonio Politano