NiGe PDOS

[Dhirendra]

Hi, I am testing the basis set on NiGe. I using the http://journals.aps.org/prb/pdf/10.1103/PhysRevB.75.085326 (Fig 3) to check for the PDOS. With DoubleZetaDoublePolarized basis I am not getting the PDOS onto Ge as resembling to the paper. The PDOS on Ni are resembling. But in the HGH Tier4 basis, even PDOS onto Ni (for s and p orbitals) are different from the paper.
1. Is it expected to see a significant difference in the PDOS with different basis sets?
2. Is it wise to compare the PDOS results of VAPS code(paper) and ATK?
3. The d-orbital in Ge is 3d or 4d?

Density mesh cutoff is set quite high at 425 Hartree. The Kmesh was 16x16x24, and for dos  calculations 25x25x25.

[Umberto Martinez]

Please attach your full input script for more detailed comments.
1)  yes and no. complete and good basis set should give similar results. did you optimized  the structure in both cases?
for Ge I would compare lower states, which are not reported in your reference. As you can see these states are very low and quite "noisy" around the Fermi level.
2) not sure if it wise it for sure fine to do that.
3) 3d.

[Dhirendra]

Sincere apologies Unberto, some how the Density Mesh Cutoff was not set to 425 Hartree. So I was still using the default 75 Hartree which is way too small for NiGe in HGHbasis. With 425 Hartree, PDOS on Ni are resembling to the reference. But PDOS on Ge are the same as in the image. I ll share the script and the results, need some time for that.