The performance of efficient organic solar cells depends critically on the morphology of the active layer. Both, electron-hole separation and charge carrier transport to the electrodes benefit from the formation of suitable aggregates or crystallites. Using a blend of the polymer PM6 and the non-fullerene acceptor Y6 as example, I shall demonstrate how we can use relatively simple absorption and emission spectroscopy, in combination with a careful Franck-Condon analysis, to identify the formation of aggregates in films made for solar cells. The talk will be based on two papers whose abstracts I give below.

(1)

In organic solar cells, the resulting device efficiency depends strongly on the local morphology and intermolecular interactions of the blend film. Optical spectroscopy was used to identify the spectral signatures of interacting chromophores in blend films of the donor polymer PM6 with two state-of-the- art nonfullerene acceptors, Y6 and N4, which differ merely in the branching point of the side chain. From temperature-dependent absorption and luminescence spectroscopy in solution, it is inferred that both acceptor materials form two types of aggregates that differ in their interaction energy. Y6 forms an aggregate with a predominant J-type character in solution, while for N4 molecules the interaction is predominantly in a H-like manner in solution and freshly spin-cast film, yet the molecules reorient with respect to each other with time or thermal annealing to adopt a more J-type interaction. The different aggregation behavior of the acceptor materials is also reflected in theblend films and accounts for the different solar cell efficiencies reported with
the two blends.

(2)

In an endeavor to understand why the dissociation of charge-transfer (CT) states in a PM6:Y6 solar-cell is not a thermally activated process, measurements of energy-resolved impedance as well as of intrinsic photoconduction are employed. This study determines the density of states distributions of the pertinent HOMO and LUMO states and obtains a Coulomb binding energy (Eb,CT ) of ≈150 meV. This is 250 meV lower than the value expected for a pair of localized charges with 1 nm separation. The reason is that the hole is delocalized in the polymer and the electron is shared among Y6 molecules forming a J-like aggregate. There are two key reasons why this binding energy of the CT state is not reflected in the temperature dependence of the photocurrent of PM6:Y6-diode: i) The e–h dissociation in a disordered system is a multi-step process whose activation energy is principally different from the binding energy of the CT state and can be substantially less than Eb,CT , and ii) since dissociation of the CT state competes with its intrinsic decay, the dissociation yield saturates once the rate of dissociation grossly exceeds the rate of intrinsic decay. This study argues that these conditions are met in a PM6:Y6-solar cell.