MO cube file index for ROKS calculation

[Xiaoming Wang]

Hello,

I am doing a triplet state calculation with SIC, OT, ROKS and MULTIP=3. The 
relaxed structure is correct (compared to literatures) and the spin 
component of each atom  also agrees well with literatures. However, when I 
plot the mo_cube files, I get lost. I have 2688 electrons in total with 
1345 of spin 1 and 1343 of spin 2, and I want to look at the orbitals of 
the electron and hole of the triplet state. So I should see the 
WFN_1344_2-1_0.cube for the hole and WFN_1345_1-1_0.cube for the electron. 
However, after visualizing the cube files, the WFN_1344_2-1_0.cube file is 
not what I expect for hole but the WFN_1345_1-1_0.cube is. Then, I get 
confused.  After searching the groups, I find that printing MO_CUBE with 
ADDED_MOS will give some misleading indexes. So how about the ROKS case?

Part of my inputs are as follows:
-----------
  &DFT
    MULTIPLICITY 3
    ROKS
    &SIC
     SIC_METHOD MAURI_SPZ
     SIC_SCALING_A 0.25
     SIC_SCALING_B 0
    &END SIC
    &OT
      ROTATION
      PRECONDITIONER FULL_KINETIC
      ENERGY_GAP 0.001
    &END OT
    &PRINT
     &MO_CUBES
      WRITE_CUBE T
      NHOMO 2
      NLUMO 2
     &END MO_CUBES
    &END PRINT
-----------------------

Best,
Xiaoming


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170928/93b82ab5/attachment.htm>