SAPT (DFT) m062x

SAPT
1

I am new to this topic and tools and trying to find out from forums and related topics. I have made this input file, which run perfectly.

memory 24 gb
set basis 6-311++G**
molecule {
0 1
Br -0.18869 -1.19358 0.00000
F -1.17551 -2.68523 0.00000

–
0 1
Cl -0.29956 2.65302 0.00000
Cl 1.36405 1.49954 0.00000
}
energy(‘sapt0’)

when I tried to do DFT-SAPT, and modified the input,

memory 24 gb
set basis 6-311++G**
molecule {
0 1
Br -0.18869 -1.19358 0.00000
F -1.17551 -2.68523 0.00000

–
0 1
Cl -0.29956 2.65302 0.00000
Cl 1.36405 1.49954 0.00000
}

set {
e_convergence 1e-8
d_convergence 1e-8
sapt_dft_grac_shift_a 0.136
sapt_dft_grac_shift_b 0.136
SAPT_DFT_FUNCTIONAL M06-2x
}
energy(‘sapt(dft)’)

It shows an error.

Fatal Error: Vx: RKS does not support rotated V builds for MGGA’s

I don’t want to change the basis nor the functional and want to extend this theoretical level for free radicals (unrestricted DFT).
Can anyone help me?

2

SAPT DFT for MGGA is not implemented, nor is it on the horizon. It is unclear if MGGA’s provide much benefit here and the theory is quite complex.

3

Coukd you perhaps clarify (or simply provide a paper citation) why do you belive mgga’s wouldn’t be beneficial in sapt?

4

Increasing the individual monomer theory level with a corresponding increase in complexity of the intermolecular expressions likely only yields performance increases up to a certain point. For an extreme example we could use MP2 for intermonomer expressions where each monomer’s wavefunction is expressed at FCI will likely not improve accuracy.

For SAPT(DFT) the density is fairly converged so Elst/Exch terms likely will not change. It could help in Induction some, but Disp would be similar to GGA as that expansion will not include MGGA terms. It is an interesting research question for sure, this is merely my guess.

1 Like