Excited State Delocalization in DNA

One persistent topic in photobiology has been the question of how many bases are involved in the absorption of a photon of UV light, or in other words how delocalized the absorbing states are. There have been estimates ranging from completely localized monomeric states to delocalization over the whole helix. The question is challenging to study experimentally because the delocalization of the wavefunction is a quantum phenomenon without a direct experimentally observable counterpart. It is difficult to study from a computational point of view because of the extended system size, environmental interactions, and structural disorder. Computational studies using exciton models could not include structural disorder well. Explicit quantum chemistry studies had troubles because of smaller QM regions.

That is why we thought there was need for another more extended study. We used a QM region of eight nucleobases, which could be treated by TDDFT thanks to the GPU based TeraChem code, we did extended sampling by molecular dynamics using the Amber code (also GPU based), and we connected the two by a QM/MM scheme. In total we computed 6000 excited states. To analyze these systematically, we used the TheoDORE code. The results are shown in a new article "Electronic delocalization, charge transfer and hypochromism in the UV absorption spectrum of polyadenine unravelled by multiscale computations and quantitative wavefunction analysis" that just appeared in Chemical Science.

The main results of the study are:

• Photon absorption occurs predominantly through a collective excitation of two neighboring nucleobases.
• Full charge transfer (CT) states are only present at higher energies but states with non-neglible CT admixture account for about 50% of the spectral intensity.
• The experimentally observed hypochromism occurs through perturbed monomer states rather than excitonic or CT interactions.