As part of my MSc thesis I want to calculate the dynamic NMR spectra of a synthesized organic compound, the log file of the compound is attached to this question. I want to do this work using Gaussian 03.
In fact, the methyl groups attached to C11 (of the single ring stands over the xanthene base) have two different chemical shifts at 90°C, while my supervisor and I guess that at the higher temperatures they may reveal the same chemical shift - a chemical shift which is related to the coalescence temperature.
So, I do not know how to calculate this part of my MSc thesis and I would be grateful if someone give me some helpful comments to do this task.
Running NMR with gaussian, right? According to one gaussian tutorial text studied, you'll have to carry out a full optimisation on that molecule (as seen in your file) with DFT(6-31+g(d)) basis set since it is an organic molecule. Then, calculate the NMR with Hartree fock at the same level. Just suggesting.
Although I am not familiar with Gaussian I doubt whether you can calculate NMR "spectra" with this Software. You may calculate NMR Parameters like chemical shifts but these alone do not make a spectrum. If you molecule lacks symmetry (Point Group C1) you will find different shifts for every single atom in this calculated and fixed configuration. The observable spectrum is (provided the exchange ist fast enough) the average of the shifts of all contributing molecular configurations, scaled by their population. Maybe you will have to calculate the NMR Parameters of all relevant configurations and use these for construction of the dynamic spectra, but for calculation of the spectra themselves (the curves) you will probably need a software that takes into account time-dependent phenomena (lifetimes of specific configurations and nuclear relaxation rates), since dynamic NMR is related to reaction kinetics.
Gaussian will only calculate chemical shifts and J couplings. If I remember correctly (its been a while) the J-coupling accuracy is pretty low. As Wolfgang said, to get the actual spectra you will also need a program that can simulate the actual lineshape exchange rates and nuclear relaxation rates.
I haven't tried it, but it looks like WInDNMR can do this
http://www.chem.wisc.edu/areas/reich/plt/windnmr.htm
Dear Monsurat, Wolfgang, and Jeffrey. many thanks for the helpful comments.
- Badges
- Science topic