QIPIRC

It has been recognized that the initial steps of natural photosynthesis harness the available light energy with almost unit efficiency (typical measured rates range from 95-99%). Despites decades of intense research, a clear understanding of the mechanisms behind this remarkably efficient transport process remain elusive [1]. Recently, experiments to probe the dynamics of delocalized exciton states in light-harvesting complexes and the Fenna-Metthew-Olson (FMO) complex provided direct evidence of the presence of quantum coherence between multiple chromophoric sites.  This result has led to the suggestion of identifying quantum coherence as a likely cause for the highly efficient energy transfer in these systems [2].  The situation, however, appears to be more complicated, as illustrated by the fact that pure quantum mechanics cannot explain the observed transfer rates while current observations can be accounted for if coherent transport were to be supported by the presence of a certain level of dephasing noise [3,4].



Fig.1: The FMO complex on the left is composed of 7 pigments that are loosely bound to form a complex. Excitons may enter the complex, e.g. site 6 (blue), and are then transported to site 3 (red) where energy is transferred to the reaction centre where chemical reactions are initiated. The r.h.s. depicts a simplified model of this complex where each pigment is represented as a single site and the interaction between sites is an excitation number preserving hopping term in a Hamiltonian. The exciton may be destroyed via spontaneous emission and the complex may suffer dephasing noise from a phonon bath (due to the many possible vibrational modes of the complex).

Fig. 2: Increasing the level of dephasing leads to an increased success probability for transport of energy through the FMO complex. The noise free setting (solid line) leads to a transfer probability of ~50% while in the presence of an optimal dephasing rate this may rise to ~98%

There are several processes that contribute to facilitate the dephasing assisted transport and these have been identified in [5]. The key observation is to realize that when an exciton enters the complex, it may explore different paths and hence experience constructive and destructive interference. Destructive interference closes off certain propagation channels and may in fact lead to population trapping due to cancellation of transition amplitudes. Dephasing noise inhibits this destructive interference and may as a result release trapped population (see fig 3a). Destructive interference depends on the existence of fixed phase relationships in the quantum state of the system and can therefore be affected by static disorder whereby different sites have different energies.  This asymmetry leads to a time evolution of relative phases and thus the conversion from destructive to constructive interference (see fig. 3b). The latter is in striking difference to the destructive effect that static disorder plays in the process of Anderson localization. Finally, dephasing resulting from energy level fluctuations will enhance the overlap between neighbouring energy level and hence facilitate transport. The latter scenario may already be understood at a classical level, as illustrated in fig 3c., where initially forbidden transitions become possible when energy gaps are reduced.

The discovery of dephasing assisted transport and the elucidation of the fundamental mechanisms that are leading to it opens up the road to better understanding the dynamics of photosynthetic complexes and may help us designing more efficient nano-scale devices for energy transport which could potentially lead to improved artificial light harvesting systems.