My favorite PAH

I have made several PAHs and discovered a few new ones. There are so many stories behind them, but one stands out by far as my favorite.

Tetrabenzo[a,cd,f,lm]perylene (1) was one of my new ones, but pride of discovery only scratches the surface.

First big clue that it was unique was its UV spectrum. Its isomer, tetrabenzo[a,cd,j,lm]perylene (2), is typical of many PAHs, a rising series of bands. Isomer (1) has this sort of pattern, but the highest wavelength band is much lower than the next below it.

When the fluorescence excitation spectra of the two isomers were collected, both showed the same pattern of increasing bands (and their emission spectra were mirror-image matches, too). “Curiouser and curiouser” So the transition from ground to excited state was normal, but something else happens in the UV.

An impurity could concievably cause a difference in the excitation spectra through some specific quenching mechanism, but an impurity would increase the UV intensity somewhere.

The solution comes through molecular modelling. Using around eight different modelling programs, the drawn structure of (2) always reduces to a structure where the bridging ring twists so that the two halves are out of plane to each other.

Modelling (1) sometimes gave this same type of structure, a twist of the bridging ring. But sometimes the drawn structure reduced to another 3-D conformation, one where the long axis of the substructure with four ortho-fused rings flexed to reduce the steric strain.

In all eight models, the twist versus flex structure of (2) were widely separated, 10 k cal/ mole to 15 kcal/ mole. In (1) the differences ranged from 0.5 kcal/ mole to 2 kcal/ mole. A very small difference.

So what happens? Two conformations exist in room-temperature solution, one slightly higher in energy and less abundant. UV light must convert the flex conformers to the twist ones before the molecule can be excited. A slight energy use in the highest wavelength band.

When I described these results in a hallway discussion at a meeting on polycyclics, the cluster of scientists – all well-known in spectroscopy and PAH circles – started brainstorming ways to prove this and ideas for experiments. These led to a collaboration using Shpol’skii spectroscopy and another using polarized light spectroscopy. Both found a mixture of two species in excitation and only one in emission.

One interesting idea that was never done because I ran out of the compound was an NMR study of temperature and solvent effects. Both ought to change the confortmer distributions in a controlable way. An elegant study waiting for someone to redo the synthesis (not hard since I could do it – other than making the 5-ring cyclic ketone starting material).