The lab-scale power conversion efficiency of organic solar cells has recently breached the 20% mark thanks to the introduction of Y-series acceptors. Key to these advancements is the efficient exciton separation at low energy level offsets between acceptor exciton (S_1,A) and interfacial charge transfer state (CT) as well as barrierless separation of the interfacial CT states.[1],[2] While the origin of the efficient charge generation has been widely investigated, a mechanistic understanding of the free charge recombination kinetics and energy level offset related voltage losses remains elusive. For example, previous studies concluded that despite the low energy offset between S_1,A and CT, and the masking of the CT emission and absorption spectra by the S_1,A, the majority of free charge recombination proceeds through interfacial CT states.[3] Yet the CT state properties and non-radiative recombination mechanisms are difficult to assess from the optical spectra complicating the voltage loss analysis.

Here, I show transient optoelectronic (charge extraction and transient photovoltage) results that enable us to study the lifetime of free charges as a function of the charge carrier density and quasi-Fermi level splitting. Such an approach is particularly useful for low-offset Y-series blends where the CT properties are difficult to assess experimentally. In an interfacial CT state mediated recombination model, the lifetime is expected to crucially depend on the interfacial area[4] and energetics that govern the thermal equilibrium between CT states and free charges[5]. Contrary to that, I present evidence that the lifetime of free charges in polymer:Y-acceptor blends is independent of the interfacial area, charge carrier density or CT state energy. Independent of these factors, the free carrier lifetime is solely determined by the quasi-Fermi level splitting. I highlight the difference to the expected CT recombination model by comparison of different planar and bulk heterojunction systems,⁶ morphology dependent electrostatic energy level shifts,⁷ and variation of the interfacial energetics. To rationalise our experimental findings, an analytical model reproducing experimental trends will be presented. The model suggests that free charge recombination in high efficiency polymer:Y-acceptor blends is limited by the back hole transfer to the Y-acceptor and subsequent recombination through acceptor excitonic states – that we assign to dark triplet states due to the low emission yields.  Since I conclude that acceptor mediated – rather than CT state mediated – presents the bottleneck of free charge recombination in these polymer:Y-acceptor blends, addressing molecular properties of the Y-acceptor such as the (triplet) exciton energy and lifetime are key to reducing the free charge recombination related voltage losses in high performance, low-offset polymer:Y-series blends to realise their full potential.