For tautomers, conformers (rotamers, etc.), the principle is to calculate the Gibbs energy (not “free energy” please…) of the largest number of “stable” structures (no imaginary vibrational frequency), and deduce the composition (mole fractions…) from the equilibrium constants K = exp(-ΔG/RT). To reduce the problem, select those conformers which are supposed to contribute significantly to the mixture: first, select the most stable as reference (fix arbitrary ΔG = 0, not mandatory but practical), then select others with a higher Gibbs energy within a given limit (for example with ΔG < 11.4 kJ/mol the contribution to the mixture is > 1% at 298K). When the mole fractions of the mixture is calculated, you can estimate the “mean” thermodynamic values (ΔE, ΔH, ΔS…) of the mixture by a weighted average. As you can see, this procedure include some arbitrary choice, with a “chemical feeling”. I don’t know if Gaussian offers a simplified procedure. By the way, there are between 1000 and 2000 references delivered by Google Scholar, if you search with combinations of “acrylic acid”, rotational, conformers, "gas phase", "ab initio".

Hi Shivam,

The molecule spends most of its time in its optimized conformer at room temperature. So how about just optimizing the structure (at 0 K) and taking the corrected free energy of the optimized structure at room temperature? The thermal correction to Gibbs free energy is printed in the output file after optimization.

-Saeed

For tautomers, conformers (rotamers, etc.), the principle is to calculate the Gibbs energy (not “free energy” please…) of the largest number of “stable” structures (no imaginary vibrational frequency), and deduce the composition (mole fractions…) from the equilibrium constants K = exp(-ΔG/RT). To reduce the problem, select those conformers which are supposed to contribute significantly to the mixture: first, select the most stable as reference (fix arbitrary ΔG = 0, not mandatory but practical), then select others with a higher Gibbs energy within a given limit (for example with ΔG < 11.4 kJ/mol the contribution to the mixture is > 1% at 298K). When the mole fractions of the mixture is calculated, you can estimate the “mean” thermodynamic values (ΔE, ΔH, ΔS…) of the mixture by a weighted average. As you can see, this procedure include some arbitrary choice, with a “chemical feeling”. I don’t know if Gaussian offers a simplified procedure. By the way, there are between 1000 and 2000 references delivered by Google Scholar, if you search with combinations of “acrylic acid”, rotational, conformers, "gas phase", "ab initio".