do_cosmo_smd false

[Carlos Nieto]

Hi,

Trying to getting familiarise with NWCHem and SMD solvation schemes, I compared the result of the calculations of the same water molecule invoking do_cosmo_smd true/false. In the attached image you got 3 examples, all of them with the same input directives except the presence/absence of the cosmo SMD block:

------------------------------------------------

start molecule

title water

geometry units angstroms print xyz autosym
O       -0.01600       -0.02270        0.00000
H       -0.32330        0.93860        0.00000
H        0.99270        0.00800        0.00000
end
 
basis
* library 6-31+G**
end
 
dft
xc m06-2x
disp vdw 3
fukui
print "Fukui information"
odft
end
 
task dft optimize

-----------------------------------------

The result are described as follows:

1) No solvent block: final energy in yellow arrow

2) Cosmos solvent block as follows, main directive false (final energy in red arrow):

--------------------

cosmo
do_cosmo_smd false
end

-------------------

3) Cosmos solvent block as follows, main directive true using water as solvent (finale nergy in green arrow):

-------------

cosmo
do_cosmo_smd true
solvent water
end

----------------

THE QUESTION is: how come invoking the cosmos directive as false NWCHem executes some type of calculation, even when no solvent directive is introduced?

What is the meaning of the so called electrostatic solvation energy in this case?

Hope anyone can clarify these... (if anyone wonders why, I´m designing Knime based workflows and for NWChem input generation and this was a situation it happened in the course of this design...)

Thank you very much!

[Edoardo Aprà]

>THE QUESTION is: how come invoking the cosmos directive as false NWChem executes some type of calculation, even when no solvent directive is introduced?

I am assuming you are referring to the following input block (correct me if I am wrong)

cosmo
 do_cosmo_smd false
end

This cosmo input section does enable the solvation by using the default  York-Karplus model, as described in https://nwchemgit.github.io/COSMO-Solvation-Model.html#cosmo-solvation-model.

Instead, using the input section

cosmo
 do_cosmo_smd true
end

the cavity–dispersion–solvent-structure (CDS) term is computed using the SMD model (more details in https://nwchemgit.github.io/SMD-Model)

[Carlos Nieto]

Thank you kindly!! But, for the YK model: what solvent parameters uses if no solvent is defined? water perhaps?

Carlos Nieto

[Edoardo Aprà]

You can have a look at the following section of the documentation for the default values used in the COSMO module of NWChem

https://nwchemgit.github.io/COSMO-Solvation-Model.html#cosmo-input-parameters