Hello Dr.Mie, I have used vdW psps in Graphite and the agreement is very good, provided one uses optimized basis. I normally set the size of the first zeta orbital and leave the program to calculate the size of higher zeta orbitals (from the split.norm parameter). With the typical setting of DZP and energy shift as small as 35 meV did not give acceptable values of lattice parameter. I have tested with psps having NLCC too. Both work.
Thank you - Madan On Tue, Nov 9, 2010 at 6:52 PM, Mie Andersen <[email protected]> wrote: > Dear all, > > I have used the atom pseudo generator to create a relativistic vdW > pseudopotential (vwr) for Pt. When running a lattice constant optimization > calculation I get the following warning: > > Vna: Cut-off radius for the neutral-atom potential: 8.743299 > Vna: WARNING: Cut-off radius for the neutral-atom potential, rVna = > 8.74330 > Vna: WARNING: Cut-off radius for charge density = 4.91988 > Vna: WARNING: Check atom: Look for the sentence: > Vna: WARNING: LOCAL NEUTRAL-ATOM PSEUDOPOTENTIAL > Vna: WARNING: Increasing the tolerance parameter EPS > Vna: WARNING: might be a good idea > > Can anyone help me with the meaning? And where should I find this tolerance > parameter EPS? > > Furthermore I get an optimized lattice constant about 5% larger than the > experimental value. Could this be a problem of the pseudopotential, or is it a > problem of the vdW functional? > > Kind regards, > > Mie > -- ---------------------------------------- Madan Mithra .L.M Senior Research Fellow Dept.of Physics Indian Institute of Science Bangalore - 560 012 INDIA ......................................... Phone : 91-80-22932313 (Lab) FAX : (080) 360 2602 ----------------------------------------
