Example: LFDFT: g-tensor of Cu(H2O)6

Download Cu_LFDFT_gtensor.run

#!/bin/sh

# Application of the Ligand Field DFT approach for the calculation
# of the g-tensor. The g-tensor is only calculated for doublet states.
# This example calculates the g-tensor of the ground and excited doublet states
# of [Cu(H2O)6]^2+ with a Cu^2+ d^9 electron configuration.

# First an average of configuration calculation (AOC) is performed, where 9
# electrons are equally distributed over the 5 orbitals that have the most
# dominant Cu 3d character. Depending on the electron configuration this might
# be a non-trivial task. Symmetry NOSYM should be specified.


SCM_LFDFT="$AMSHOME/examples/adf/Cu_LFDFT_gtensor/LFDFT"
export SCM_LFDFT

$AMSBIN/ams <<eor
System
  Atoms
   Cu        0.000000    0.000000    0.000000
   O         0.000000    0.000000    2.325873
   O         0.000000    0.000000   -2.325873
   O        -2.004009    0.000000    0.000000
   O         2.004009    0.000000    0.000000
   O         0.000000   -2.004009    0.000000
   O         0.000000    2.004009    0.000000
   H         0.000000    0.773171   -2.924278
   H         0.000000   -0.773171   -2.924278
   H        -0.794410    2.570047    0.000000
   H         0.794410    2.570047    0.000000
   H        -2.570047    0.000000   -0.794410
   H        -2.570047    0.000000    0.794410
   H         0.794410   -2.570047    0.000000
   H        -0.794410   -2.570047    0.000000
   H         2.570047    0.000000   -0.794410
   H         2.570047    0.000000    0.794410
   H         0.000000   -0.773171    2.924278
   H         0.000000    0.773171    2.924278
  End
  Charge 2
End

task SinglePoint

Engine adf
  Symmetry NOSYM
  IrrepOccupations
    A 78 1.8 1.8 1.8 1.8 1.8
  End
  basis
     Core None
     Type TZP
  End
  XC
    GGA PBE
  End
EndEngine
eor

# When the AOC calculation is ready, you need to make sure that indeed the
# partially occupied orbitals are dominantly d orbitals. In the ADF output you
# can find the character of the MOs in the list of all MOs, ordered by energy,
# with the most significant SFO gross populations.

# Next the LFDFT calculation is performed including spin-orbit coupling (soc 1),
# which is needed for the g-tensor calculation. In this case there is 1
# shell, and the nlval for 3d is '3 2'. The MO indices should be the
# fractionally occupied levels of the AOC calculation (in this case 40 41 42 43 44).

$AMSBIN/lfdft << eor
 adffile ams.results/adf.rkf
 nshel 1
 nlval 3 2
 MOind 40 41 42 43 44
 soc 1.0
 DegeneracyThreshold 1.0E-5
eor