BSSE is usually estimated using the Counterpoise Method. Can I recommend that you take a look at David Sherril’s summary titled “Counterpoise Correction and Basis Set Superposition Error”. http://vergil.chemistry.gatech.edu/notes/cp.pdf

Thank you so much @Frederick Bennett. But still i am not getting which formula i have to used for BSSE determination.

Equation 3 in Sherrill's documents give you an estimate of the basis set superposition error. What software are you using for your calculation?

I am using Gaussian09 @ Frederick Bennett

Then you are in luck, Gaussian 09 automates the entire process with the “Counterpoise” keyword.

Thank you very much for your response @ Frederick Bennett

Dear Amarjyoti,

If your problem is still unresolved, please let me know so that I explain you this problem through a simple and explicit example.

Sincerely,

Saeed

Saeedreza Emamian

Can you share an example?

Thanks in advance

Renjith Thomas

Dear Renjith,

Suppose that you are going to calculate interaction energy (I.E.) for intermolecular hydrogen-bonded H3N...HOH complex.

Initially, you should optimize this system at a desirable computational level such as B3LYP-D3(BJ)/def2-TZVPP. Such optimization should be followed by a frequency calculation at the same level to ensure there is not any imaginary frequency. Now, you can save the optimize geometry and attempt to calculate I.E. Note that calculation of I.E. demands highly accurate computational levels. In this sense, while CCSD(T) calculations extrapolated toward CBS (complete basis set) are very common, you can also employ CCSD(T)/jul-cc-pVTZ level. In addition, while it is preferred to use aug-cc-pVTZ instead of jul-cc-pVTZ, very recently and in DOI: 10.1002/jcc.26068, we showed that jul-cc-pVTZ is as robust as aug-cc-pVTZ and can be used to obtain very accurate values of I.E. in a less expensive cost.

To reach this purpose, taking the optimized structure of complex, your Gaussian input should be as follows:

# ccsd(t)/jul-cc-pvtz counterpoise=2

Title Card Required

0 1,0 1,0 1

O -0.03836200 1.54613400 0.00000000 1

H 0.06356500 0.57566500 0.00000000 1

H 0.85354600 1.90026300 0.00000000 1

N -0.03836200 -1.37627100 0.00000000 2

H -1.03728700 -1.54652400 0.00000000 2

H 0.34780000 -1.83229000 0.81779500 2

H 0.34780000 -1.83229000 -0.81779500 2

Note that there are three 0 1 which indicate charge and total multiplicity of complex, monomer 1, and monomer 2, respectively. In addition, as you can see at the end of each line within molecular specification, atoms belonging to monomer 1 are characterized with number 1 while those of monomer 2 with number 2.

Now, you can run this job.

After completion and at the end of output file, you can find what you are looking for:

BSSE energy = 0.000638404757

complexation energy = -6.62 kcal/mole (raw)

complexation energy = -6.21 kcal/mole (corrected)

The first value is the BSSE energy in a.u. The second value (raw) is the uncorrected value of I.E (No CP-I.E.) corresponded to E(complex)-E(monomer 1)-E(monomer 2).

The third value (corrected) is the CP-corrected I.E. associated with E(complex)-E(monomer 1)-E(monomer 2)+BSSE.

Note that this method of I.E. calculation does not consider deformation energy (DE). DE, particularly for small and rigid monomers, is very small and can safely be ignored (I suggest you to carefully read the above-mentioned DOI). Meanwhile, such a complete report is provided since Gaussian 16. I think (I am not sure, please check it) Gaussian 09 is not able to provide CP-corrected I.E. and you have to do some manual calculations.

I hope this can help you.

Sincerely,

Saeed

Thank you so much Sir @Saeedreza Emamian . I have to take the coordinate after optimization and rerun with the keyword " counterpoise=2 " for BSSE.