What you exactly want to know? The keyword for running PCM-SMD calculation or the value of the mentioned parameters of a particular solvent? The keyword is simple SCRF(SMD, solvent=XXXX). If you want to know the value of the parameter search "Minnesota Solvent Descriptor Database" probably available in Prof Truhlar's webpage. 

Hi Dr.Debasish Mandal,

Thank you for your prompt reply. I would like to know the keywords to input the descriptors in Gaussian 09 Software. For example, we type the keyword "eps" for dielectric constant. 

the keywords that I wanted to know are

Molar volume;

Thermal expansion coefficient;

absoulute temperature;

numeral density;

hydrogen bond basicity;

surface tension at interface;

carbon aromaticity;

electronegative halogenicity;

Thanks

 G09-SMD solvation model used the following seven solvent descriptors:

• index of refraction

• Abraham's hydrogen bond acidity

• Abraham's hydrogen bond basicity

• surface tension

• dielectric constant

• aromaticity

• electronegative halogenicity

and the corresponding keywords are

SolventName=

EpsInf=X

HBondAcidity=X

HBondBasicity=X

SurfaceTensionAtInterface=X

Eps=X

CarbonAromaticity=X

ElectronegativeHalogenicity=X

The value of these parameter you may get from "Minnesota Solvent Descriptor Database" 

Mr/Miss/Mr. Chin,

You can pay attention to this discussion:

https://www.researchgate.net/post/how_can_I_apply_Dimethylformamide_as_solvent_when_I_want_to_optimize_an_structure_using_Gaussian_09

Hi Dr.Debasish Mandal,

I have tried to use the keywords but the software said " syntax error in stoichiometry"

For more details, i have attached the input and output files. Thank you for your help 

Dear Dr.Bojidarka B. Ivanova,

I have read through the post and seems like there is no answer for that post. That's why I raised another post just to focus on this question. Thank you.

MrMr/Miss/Mr. Chin,

As I have understood you would like to run computations using SMD approach, meaning the Truhlar's universal solvation model based on solute electron density. If yes, then as I have written you should use 'solvent=generic' and to specify the parameters in the 'molecule specification section'. The input should contain:

# method/basis set scrf(SMD,Solvent=Generic,Read)

molecule specification section:

coordinates

stoichiometry=CxHxOz...

solventname=xxxxxxx

eps=xxxxx

epsinf=xxxxxx

…

as shown in the Mr. Mandal' posting and the discussion which I have provided (the link above).

Otherwise you could use the original SMD program package shown here: https://comp.chem.umn.edu/solvation/

Dear Dr.Bojidarka B. Ivanova,

Thank you for your prompt reply. I have successfully run one model using solvent=generic. However, I have encountered a syntax error " syntax error in stoichiometry" for the ionic liquid solvents. These are the files from the simulation. thank you very much.

Hi Dr.Bojidarka B. Ivanova and Dr.Debasish Mandal,,

I have solved the problem by replacing "stoichiometry=bmimBF4" with "stoichiometry=C8H15B1F4N2". Thanks for your help. However, I would like to know if "stoichiometry" will change the result of the calculation? For example, I replace stoichiometry=h2o for the above job, will it change the result? Thank you. 

Mr/Miss/Mrs. Chin,

Yes, it is changed, because of you should obtain different non-electrostatic and nuclear repulsion energy contributions.

In general, using non-standard inputs there is important to verify the compatibility of the input parameters to a model of solvent molecules, and when there is needed a new parameter should be introduce in the molecule specification section (point 4 for example, below).

You could pay attention to the following computationa of NH3+ in solution using SMD with different parameters and PCM, including the stoichiometric composition (points 1 and 2).

Point 3 constains data about PCM only.

In point 4 are shown same computations like those in points 1 and 2 but with inconsistent VMol and RSolv parameters. In such as cases Vmol = xxxx (new parameter) should be given in the 'molecule specification section'.

The non-electrostatic energy term could vary, in this case from 0.0099755434 to 0.0055812683 a.u. (or 6.25 - 3.51 kcal.mol-1).

So, the molecular design of precise computations of the energetics in quantitative terms needs more effort.

The input coordinates of the NH3+ are:

N 0 -0.348343 0.000000 0.246307

H 0 0.696655 0.000000 0.246307

H 0 -0.696671 0.000000 -0.738922

H 0 -0.696671 0.853241 0.738907

H 0 -0.696671 -0.853241 0.738907

For example: Computations of NH3+ by M062x/sto-3g

1. SMD, stoichiometry=C1H4O1

........................

--- end of non-standard input.

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

Polarizable Continuum Model (PCM)

=================================

Model : PCM.

Atomic radii : SMD-Coulomb.

Polarization charges : Total charges.

Charge compensation : None.

Solution method : Matrix inversion.

Cavity type : VdW (van der Waals Surface) (Alpha=1.000).

Cavity algorithm : GePol (No added spheres)

Default sphere list used, NSphG= 5.

Lebedev-Laikov grids with approx. 5.0 points / Ang**2.

Smoothing algorithm: Karplus/York (Gamma=1.0000).

Polarization charges: spherical gaussians, with

point-specific exponents (IZeta= 3).

Self-potential: point-specific (ISelfS= 7).

Self-field : sphere-specific E.n sum rule (ISelfD= 2).

Solvent : Generic,

Eps = 32.630000

Eps(infinity) = 1.758000

RSolv = 1.855000 Ang.

Molar volume = 40.700000 cm**3/mol

..........................

GePol: Fraction of low-weight points (