I am a gaussian beginner and i am not sure about what should be done for this job.
1- Should I draw a water molecule near to the site of proposed intermolecular H-bonding with the studied compound?
2- Should water molecule be loose or linked through H-bond to the compound ????
3- Should I freeze the compound, while allowing water molecule to rotate?? How ?
4- Should I do energy minimization or potential energy surface scan or both ???
Dear Muhammad,
The atoms of the water molecule should be drawn and about 3-4 Angstrom far from the atom in the compound that is expected to hydrogen bond with water.
No restriction is needed. The only problem you might face is once the angle X---O-H reaches 180 degrees the job is stopped. In order to solve this problem you freeze the angle to 179 degrees and rerun the optimization.
Hoping this will be helpful,
Rafik
Ok First of all download Avogadro. It is free access program. You can draw all your molecules there (water etc.) and you will see hydrogen bonds there. Why it is a little better then GaussView? Because you can see H-bonds;] Parameters for H-bond in the picture (cutt off radius 3A, angle 120o. When you draw your molecules you click "Extensions" then "Gaussian" and open window with parameters for calculations: In the picture you can see Calculation-e.g. Frequencies (this will optimize your structure and you will get IR spectra), Theory e.g. B3LYP it give you DFT calculations. Basis set e.g. 6-31G(d,p) - d,p -this is important enlargement for H-bonds. Next don't forget mark checkpoint if you want get IR spectra or Raman. Processors e.g. 4 you can parrarel your calculations for e.g. 12 processors.
For clarity now:
1.Should I draw a water molecule near to the site of proposed intermolecular H-bonding with the studied compound?
You can but if you choose Geometry Optimization your water molecules should come closer to each other, rotate to get minimum energy and in the end of calculations you will see all H-bonds. (in Avogadro)
2. Should water molecule be loose or linked through H-bond to the compound ????
No, You want calculate energy and structures for different molecules and get knowlage H-bonds appear or not. Once again I recommend Avogadro because in GaussView you will not see H-bonds unfortunately.
3- Should I freeze the compound, while allowing water molecule to rotate?? How ?
If you make your calculations you want to Optimize Structure to get ground state. In water your optimize ground state looks like two water molecules with O-O distance c.a. 2.8A with H-bond. If you want freeze atoms and get energy choose Single Point energy in Calculations (the simplest way). If you want freeze simply oxygen atoms you should make * F Freeze all Cartesian coordinates for oxygen
(read http://www.gaussian.com/g_tech/g_ur/k_opt.htm)
4- Should I do energy minimization or potential energy surface scan or both ???
Only energy optimization.
Mr. Kahattab,
And to all this mentioned by Mr. Karaman you ca conduct both in equilibrium and nonequilibrium solvation PCM
Hello Muhammed,
Without knowing fully what your job is, you might also consider doing your geometry optimisations and (if you're doing them) frequency calculations using the Gaussian counterpoise keyword.
I'm assuming here you are not adding a complete solvent shell, but just a single water molecule.
You should probably start with an H...O hydrogen bonding distance of about 1.9 Angstroms, roughly along the geometry you might expect the water molecule to adapt.
Past adding the counterpoise option, if you want it, there's nothing extra you need in adding water for hydrogen bonding (or, say, smaller dispersion bound molecules like rare gas atoms, though the counterpoise correction is important in such instances). Just put them in roughly the right place, at roughly the right separation.
Also, a tip I always pass on to people I teach Gaussian to. Whilst molecular building programs are good, they quite often don't allow for symmetry past C1, at least not without you explicitly requesting it (I don't speak for any particular program here, but generally this is true).
As frustrating as it might be at first, I would suggest, where practical, that you try to visualise the molecule you're building in 3D and construct your geometry without the aid of any builder.
I've always found this particularly helpful when you're trying to build molecules of higher symmetry, as the symmetry of the optimisation is frozen at the symmetry *implicit* in the start geometry.
I see that people have given various answers to your problem, based on various possible scenarios for your intended job, as it's not entirely clear what you want to do. Could tell us a little more detail about your job?
Thanks for ur detailed answer and apologies for my delayed response. I just need to figure out how many water molecules could bind to the studied compound. is it 1, 2 ,3, ....etc
should I calculate hydrogen bond energy via single point energies or should I use CBS-QB3 calculation? or should I use "Counterpoise" keyword? I got more confused after I read these answers :/
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments