PC GAMESS/Firefly-related discussion club

Learn how to ask questions correctly

arvey s. oliveros
yevra_16@yahoo.com

I have been studying the ionization energy of phenylsilane derivatives through quantum chemical calculation using PCGAMESS. GABEDIT is the open graphic user interface that i used in generating the images of the orbitals so that i could see the HOMO/LUMO interaction and their overlapping in relation to stabilizing its ionized state (MULT=2 ICHARGE=1).

The geometries were optimized at HF/6-31G* and correlations were then accounted by running single point energy calculation. For open-shell, i have employed: ROB3LYP, UB3LYP, ROMP2 and UMP2. I have been worried about spin contamination for unrestricted calculation. Phenylsilane radical cation for instance has an S^2=0.86 (for UMP2/6-31G**) and S^2=0.763 (UB3LYP). Because of this, I decided to use ROMP2 and ROB3LYP as an alternative for open shell calculation and i read their outputs on GABEDIT. I compared the images of restricted and unrestricted calculations and they gave different images.

My question is:
1. Is it reliable to use ROB3LYP (The fact that it is restricted) to generate images of orbitals from which a qualitative explanation of stabilization can be drawn?

2. What could be the reason of having different orbital images generated from the restricted compared to the unrestricted open shell?

3. what is the best thing i could do for me to decide which among the methods give a qualitatively correct picture of the orbitals?

Thanks A lot! I know that my question is not really relevant to the PCGAMESS software but i am hoping that you could still help me on this problem... good day!