Charge-Transfer Excitons at Organic Semiconductor Surfaces and Interfaces

When a material of low dielectric constant is excited electronically from the absorption of a photon, the Coulomb attraction between the excited electron and the hole gives rise to an atomic H-like quasi-particle called an exciton. The bound electron-hole pair also forms across a material interface, such as the donor/acceptor interface in an organic heterojunction solar cell; the result is a charge-transfer (CT) exciton. On the basis of typical dielectric constants of organic semiconductors and the sizes of conjugated molecules, one can estimate that the binding energy of a CT exciton across a donor/acceptor interface is 1 order of magnitude greater than k(B)T at room temperature (k(B) is the Boltzmann constant and T is the temperature). How can the electron-hole pair escape this Coulomb trap in a successful photovoltaic device? To answer this question, we use a crystalline pentacene thin film as a model system and the ubiquitous image band on the surface as the electron acceptor. We observe, in time-resolved two-photon photoemission, a series of CT excitons with binding energies <= 0.5 eV below the image band minimum. These CT excitons are essential solutions to the atomic H-like Schrodinger equation with cylindrical symmetry. They are characterized by principal and angular momentum quantum numbers. The binding energy of the lowest lying CT exciton with 1s character is more than I order of magnitude higher than k(B)T at room temperature. The CT1s exciton is essentially the so-called exciplex and has a very low probability of dissociation. We conclude that hot CT exciton states must be involved in charge separation in organic heterojunction solar cells because (1) in comparison to CT1s, hot CT excitons are more weakly bound by the Coulomb potential and more easily dissociated, (2) density-of-states of these hot excitons increase with energy in the Coulomb potential, and (3) electronic coupling from a donor exciton to a hot CT exciton across the D/A interface can be higher than that to CT1s as expected from energy resonance arguments. We suggest a design principle in organic heterojunction solar cells: there must be strong electronic coupling between molecular excitons in the donor and hot CT excitons across the D/A interface.