Dispersion force?
Currently, dispersion forces calculation is becoming hot and challenging topic in computational chemistry. Although many scientific papers have been published but none of them represent the clear definition and explanation about this phenomena. Dispersion force is a non-covalent bond intermolecule interaction. This kind of interaction is quite important and we can find in many areas, such as DNA stacking, interaction of graphite layers, molecular recognation behavior, and so on. This kind interaction is usually existed in the ring-type molecules which contain pi bonding system. It is weak long-range attractive force between molecule and exist even without charge or with permanent charge. It is known as a complicated quantum system.
In order to approach the complexcity of dispersion system, people tried first to use the simpler configuration. In aromatic pi-ring interaction, we can devide into three non-covalent bond that usually exsist: pi-cation, pi-hydrogen, and pi-pi type. Among these, pi-cation bond has the largest energy which is around 10-35 kcal/mol and the others two has lower at 2-4 kcal/mol. The pi-pi bond will be stronger in the bigger molecules as this bond was occured among electron in the pi system of two moieties.
Computation calculation
How to calculate it? It is rather tricky ways. Different from the common method, the calculation to find it slightly different from the others computational calculation. It probably there is no perfect method or function that addrest directly to this issue. Usually people will use point-to-point calculation, means that you calculate the optimized structure and corrected by counter-poise calculation. It is because when you calculate one moeity, you will use another function of the other moieties to calculate it. Counter-poise calculation will calculate the current moiety and give ‘zero’ for the other functions.
Simple HF calculation is not applicable for this case as the function has ‘too little’ describing this interaction. Common density functional theory (DFT) is also give a little un-corrected value. So, 2nd order Moller-Plesset (MP2) could be nice combined with moderat basis sets, eg. 6-31G(d), 6-31G*, 6-31+G*. MP2 is known to be overestimate the real value, means produce more negative value interm of correlation energy compared to couple-clusters method CCSD but it require less computational time. Just to give an imagination for how long to do the calculation, for dimers phyrene, using home PC-spect, I spent one week long only for the optimization!!!!
If you want to know detail about this, you can read the review paper about dispersion forces. Please klik this paper to download it.
