Pseudopotential choice: differences in adsorption energies
Posted: Thu Mar 24, 2011 8:55 pm
Hi,
There appear to be major differences between US-PW91 PPs and the non-ultrasoft ones when it comes to computing atomic adsorption energies on magnetic compounds (such as Co and Fe). I've seen that US PPs aren't great for these types of metals, however the difference between the two is huge: 20-100 kcal/mol depending on coverage (much larger difference at lower coverage). In addition, the adsorption energies computed using non-US PPs appear to be really high.
Or are these PPs not best in describing Co and Fe containing compounds? In which case, which PAW potentials should be used?
Care to comment?
There appear to be major differences between US-PW91 PPs and the non-ultrasoft ones when it comes to computing atomic adsorption energies on magnetic compounds (such as Co and Fe). I've seen that US PPs aren't great for these types of metals, however the difference between the two is huge: 20-100 kcal/mol depending on coverage (much larger difference at lower coverage). In addition, the adsorption energies computed using non-US PPs appear to be really high.
Or are these PPs not best in describing Co and Fe containing compounds? In which case, which PAW potentials should be used?
Care to comment?