Hello everyone!
I am studying a complexation reaction with a bidentate ligand the type:
Ligand + [Zn(H2O)4]2+ --> [Zn(Ligand(H2O)2]2+ + 2H3O+
The geometries of all reaction components were optimized in gas phase at DFT level followed by frequency calculation without the occurrence of imaginary frequencies. Then, for each component was obtained the data for Sum of Electronic and Thermal Free Energy in gas phase.
Then, the energy calculations of each component considering the implicit solvent effect (water and methanol) from the PCM and SMD approaches were performed. All calculations were performed using Gaussian 09 package.
I have read some articles on different thermodynamic property calculations involving various thermodynamic cycles. Knowing that, for example, for a reaction component A the energy correction ESolv.Effect is calculated as follows:
A (gas) --> A (Solv)
ESolv.Effect = E (PCM) - E (gas) or
ESolv.Effect = E (SMD) - E (gas)
Is there any consistency in calculating the deltaG of the reaction in solution considering the solvent effect (PCM or SMD) as below?
deltaG (sol) = SUM (Product) - SUM (Reagent)
deltaG (sol) = [E (PCM) - Sum of Electronic and Thermal Free Energy for [Zn(Ligand(H2O)2]2+ + E (PCM) - Sum of Electronic and Thermal Free Energy for 2H3O+ ] - [ E (PCM) - Sum of Electronic and Thermal Free Energy for Ligand + E (PCM) - Sum of Electronic and Thermal Free Energy [Zn(H2O)4]2+]
In this case, I would be just correcting the electron energy in the deltaG calculation, assuming that the geometries in the presence of the solvent correspond to the minimum potential energy surface for each gas phase optimized component.
I will be grateful if you can give your opinion on this.
De Souza, L. A.
Hi Leonardo, I'd just like to add that the 1 M correction is slightly more tricky than it seems from the previous answer. The correction of 1.89 kcal/mol is not general and this value is only valid for systems at 298.15K whose concentration is 1 M (i.e. for a system at 400 K the correction is higher than 1.89). Also, not sure whether your reaction is carried out in water, but if it is, the concentration correction for the water molecules released after the complexation reaction should be around 55.3 M since it's the concentration of water in the media.
You can use the GoodVibes program to calculate dG in solution. GoodVibes is a program that applies all the previous corrections for you (using the "-c" and '--media' options) and includes also other corrections such as quasiharmonic vibrational RRHO, single-point, multi-conformational and symmetry corrections:
https://github.com/bobbypaton/GoodVibes
This is a list of good articles about this topic:
- Plata and Singleton, J. Am. Chem. Soc. 2015, 137, 3811−3826
- Goddard and coworkers, J. Phys. Chem. B 2008, 112, 9709–9719
Hi Juan! Thanks for the suggestion of the articles and software. I already downloaded them.
No problem Leonardo, we just released a new version of GoodVibes this month with many new functions, please send us any questions or suggestions to improve the program!
Juan V. Alegre-Requena GoodVibes does not generate the same thermodynamic properties as the source file (when the same conditions are maintained). Hope they can fix the error.
Hi Jackson J. Alcázar,
We did extensive testing on that topic and we got the same standard state corrections for substrates and solvents compared to the previous authors I mentioned above. However, GoodVibes includes other kinds of corrections that might lead to different results when comparing thermodynamic values from other articles (especially quasi-harmonic corrections, since they started becoming popular a few years ago and many papers before 2017-2018 don't include these corrections). Would you send me the command line you're using with GoodVibes and the results you're observing? Also, there is an article about GoodVibes that includes more details on how the program calculates thermodata:
--> Luchini G, Alegre-Requena JV, Funes-Ardoiz I and Paton RS. GoodVibes: automated thermochemistry for heterogeneous computational chemistry data. F1000Research 2020, 9(Chem Inf Sci):291 (https://doi.org/10.12688/f1000research.22758.1)
Thanks for using the software!
Juan V. Alegre-Requena without correction-quasi harmonics does not generate the same results as gaussian. Although the zero-point energy is reproduced, its thermal corrections are different.
It is evident the different options that goodvibes offers. However, I think it would be good if the results were reproduced to build confidence as the Shermo program does.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Thermodynamics