Quantum Dynamics of DNA Excited State Relaxation Driven by Base Stacking and Pairing Interactions

DNA exhibits remarkable photostability, largely attributed to ultrafast internal conversion pathways that dissipate absorbed UV photon energy through conical intersections (CIs). While isolated nucleobases undergo rapid deactivation on sub-picosecond timescales, their incorporation into stacked and paired DNA architectures introduces significant alterations to excited-state dynamics. Base stacking and Watson-Crick pairing give rise to delocalized excitonic and charge-transfer (CT) states that display extended lifetimes and complex relaxation behavior. In this work, we combine a linear vibronic coupling (LVC) Hamiltonian with an effective transformation to a structured system-bath representation, enabling explicit inclusion of key vibrational modes that drive nonadiabatic transitions. The hierarchical equations of motion (HEOM) method is employed to simulate population dynamics in systems ranging from adenine monomers to stacked heptamers, as well as A–T paired motifs and a combined (A–A)–T trimer. Our simulations show that increasing the number of stacked bases accelerates the initial decay rate (lifetimes decreasing from ∼300 fs in dimers to ∼230 fs in heptamers) but reduces the overall recovery to the ground state by up to ∼40%, indicating population trapping in long-lived excitonic states. Base-pairing effects are modeled via proton-coupled electron transfer (PCET) using a double Morse potential representation. For both A–T and C–G pairs, PCET was found to suppress direct ground-state recovery, further stabilizing excited-state populations. In the combined stacking–pairing configuration, PCET-mediated pairing reduced deactivation more effectively than stacking alone, especially in short oligomers. This vibrationally resolved framework captures the quantitative interplay between excitonic coupling, CIs, and PCET, providing mechanistic insight into how DNA structural motifs govern the fate and lifetime of photoexcited states.