Many thanks! It was very helpful!

Best,
Alex

Hirshfeld and Becke methods are two ways of defining atomic spaces, they are not the ways of defining atomic orbitals in molecular environment. So they cannot be used in this context.

I suggest removing diffuse functions while using def2-TZVP instead when performing DOS analysis, which is adequate.

Best,

Tian

I'm now in Multiwfn...so please help.

I study the valence band of the Li6B4O9 cluster in the B3LYP/6-31+G* basis set using GAMESS (US). I'd like to get s- p- partial DOS of each group atoms in the valence band of cluster. My basis set includes diffuse functions.

I totally agree with the red note "It is very important to note that when diffuse functions are presented, the PDOS maps plotted based on Mulliken or SCPA methods will be meaningless!" in Multiwfn manual.

But I can't find commands to generate fragments to build PDOS (with conditions) using Hirshfeld or Becke methods as in the case of Mulliken:
-1 // Define fragments
1 // Redefine fragment 1
cond // Use conditions to select basis functions. You will be prompted to input three conditions
1-5 // The first condition is that the basis functions must belong to the heavy atoms 1~5
[Press ENTER button] // The second condition is the index range of basis functions. Press ENTER button directly means basis function index is arbitrary
X // The third condition is that the type of basis function should be PX
q // Save fragment 1

Is it possible to build PDOS using Hirshfeld or Becke methods?

Thanks in advance!
Alex