The rate of photoinduced electron transfer (PET) (κPET), quantum yield of PET (QYPET), and charge extraction yield (EQE) are determined for a series of donor–acceptor (DA) organic photovoltaic systems, comprising low-band-gap polymer donors and the phenyl-C61-butyric acid methyl ester (PCBM) acceptor. The energetic alignment of these polymer donors relative to PCBM provides driving forces for PET (ΔG PET) in the range of 0.18–0.57 eV. Femtosecond transient absorption (TA) spectroscopy was used to assess the PET kinetics and QYPET, while time-resolved charge extraction (TRCE) measurements were employed to assess EQE. Near unity QYPET was observed in DA blend films with a ΔG PET of 0.57 and 0.30 eV, whereas no resolvable PET was observed with a ΔG PET of 0.18 eV. For the DA blends that exhibit PET, both κPET and QYPET appear independent of ΔG PET, with an average κPET of 420 fs for the 70% PCBM blends. An increase in nanosecond charge separation yield (TA) and EQE (TRCE) between DA systems was observed, which appears not to be due to the PET process but rather the subsequent recombination processes. DA systems should be designed to minimize ΔG PET, minimizing associated losses in device open-circuit potential; however, picosecond bimolecular recombination severely limits achievable charge extraction yields in these DA systems.

A variety of charge extraction (CE) techniques have been developed to measure charge density and recombination coefficients in bulk heterojuction (BHJ) solar cells. Charge recombination during charge extraction as a major limitation of this method has not been systematically quantified. Herein, we report CE measurements using a newly designed fast switch, which enables the application of a reverse bias to the solar cells facilitating charge extraction. With applied reverse bias, more than 40 % increase in the extracted charge was obtained in solar cells with thicker active layers or with fast recombination. The measured charge carrier lifetime increased by up to a factor of three at sufficiently high applied biases (up to 8 V), suggesting significant errors in CE measurements without applied bias. The increased extracted charges with increasing applied bias are attributed to a combination of three cases: i) slightly faster charge extraction due to the larger electric field; ii) increased charge extraction rate at high light intensities when the transients are space charge disturbed; iii) increased charge separated lifetime during charge extraction attributed to the spatial separation of the electron and hole density due to the applied electric field.

No abstract