I mean how can I tell gaussian that there is only coulombic interaction between two molecules, for instance like in the Cu(MeCN)2 PF6 complexe?
good afternoon sr Mathieu Morency
the only thing you should do in gaussian is a calculation of energy, because if you make an optimization calculation the molecule will go to a gaseous state, I recommend that you draw the molecule with gaussview according to the crystal structure and execute the calculation of energy
Mathieu: Are you trying to calculate the salt as a solid? In that case you need to specify periodic boundary conditions in the calculation. This will keep the system from "evaporating".
On the other hand, if you are trying to calculate Cu(MeCN)2 PF6 as an ion pair, there is no reason to treat this any differently than a typical (but loosely-bound) molecule. In other words, you *can* do geometry optimization on an ion pair.
You mean by just separate the to ions by a reasonable distance so that gaussian condisiders no convalent bond between the atoms of each ions? The purpose of my main question is that I'm trying to find new starting material for a mechanism pathway. The reaction I'm investigating is spontaneous at room temperature so I'm seeking for a negative delta G. The reaction happens in liquid state. I already tried some starting materiel wich involve the copper coordinated to different ligands but I wanted to see the impact of starting with the hexafluorophophate salt. Nevertheless, I'm wondering if it is really usefull to do that since in liquid state the salt dissociates and the ions are solvated... Do you think it could be good to start with an ion alone? Thanks.
Liquids are hard to model. You could see what happens with the two ions alone. There is a crude solvent model called SCRF (self consistent reaction field) which is built into Gaussian, and a more sophisticated solvent model called COSMO-RS that requires a separate program.
- Badges
- Science topic