On Fri, 12 Jan 2007, Jess Kondor wrote: JK> Hello everyone,
jess, JK> Does anybody have a good Ba ultrasoft pseudopotential in LDA approximation? JK> I generated one from vanderbilt library, but I feel did something wrong - in case nobody responded yet and just as a caveat for everybody else. there is one thing that i noticed when re-creating pseudopotentials with the parameters from the vanderbilt library: many of these parameter sets are rather old and it appears that the polynomial fitting algorithm has been changed since, so that some fine-tuning is required for heavier elements. you can tell from looking at the output file from the pseudopotential generation, as it contains warnings about negative densities. adding constraints to the polynomial fitting to eliminate them or bring those as close to zero as possible improves in my experience those potentials massively. without they seem ok in minimal tests, but as soon as you start looking at more sensitive properties, the potentials start generating inconsistent results. JK> using it together with Ti and O from pwscf web-site on simple test I cannot JK> get convergence more than 1.d-10 (moreover, on the last scf steps warning JK> 'some eigenvalues do not converge' appears' - not a good sign according JK> manual). jess, if you're still interested, please let me know. it is rather straightforward to add a few constraints and redo the potential. i can do this for you, if you would be willing to run a few tests (eg. lattice constant, bulk modulus etc.,) also for a couple other functionals, so that - if tested ok - they could be added to the pseudopotential library... cheers, axel. JK> JK> Regards, JK> JK JK> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot.
