Firefly and PC GAMESS-related discussion club

 
Learn how to ask questions correctly  
 
 

 

Alexey Nikitin
agilemolecule@gmail.com

I had a problem with the calculation of the torsion potential in the molecule FOOF.
When rotating the torsion angle two strange effects observed:
1.     F-O bond length is changed up to 0.2 A
2.     The energy barrier of rotation differs by almost 100 kcal / mol (!!!), depending on the method: HF, MP2, MP3, MP4. Moreover MP3 worse than MP2.

(I used the basis sets: 6-311+G(2d,p), cc-pVDZ and cc-pVQZ)

Could you be so kind to explain what the problem is and how to overcome it?
And more general question. I still considering the calculation of torsion angles in the cc-pVQZ MP2 as a very robust procedure. When it does not work?

[ This message was edited on Wed Oct 3 '12 at 12:25pm by the author ]