Two prominent derivatives of the conjugated polymer PCPDTBT (C- and Si-bridged) blended with [60]PCBM exhibit clear differences in their charge separation efficiencies in organic solar cell devices. The discrepancy has often been attributed to improved bulk properties such as the charge carrier mobility and a more favorable phase separation. However, fundamental differences between both materials seem to exist regarding charge transfer and charge separation. Photoluminescence quenching experiments suggested long-lived charge transfer states at the heterojunction interface [1]. Here we show that the origin of the discrepancy is a strongly increased back-electron-transfer (BET) for the C-bridged derivate in comparison to the Si bridged one. The present knowledge of charge separation in polymer:fullerene blends is largely based on findings from optical spectroscopy. These techniques require that the CT states recombine and are thus rather insensitive to long-lived CT states. In contrast, EPR spectroscopy can directly probe the presence of CT states, separated polarons or triplet excitons in one experiment without the necessity that they recombine to yield a detectable signal [2]. Using time-resolved electron paramagnetic resonance (EPR) spectroscopy in conjunction with optical excitation we study charge separation in PCPDTBT blends with [60]PCBM. A direct comparison between the two PCPDTBT derivatives reveals a remarkable influence of the bridging atom (carbon vs. silicon) on the EPR spectra. While the EPR signatures of photogenerated positive polarons in C- and Si-bridged PCPDTBT are virtually identical, significant differences are observed with respect to the spin-relaxation behavior. The spin-lattice relaxation time of positive polarons in C-PCPDTBT at low temperature (T = 80 K) is found to be more than two orders or magnitude longer than in the Si-bridged polymer derivative. This surprisingly slow relaxation can be rationalized by polarons trapped in defect states that seem to be absent in blends comprising Si-PCPDTBT. Transient EPR signals attributed to charge transfer (CT) states and separated polarons are in general smaller in blends with C-PCPDTBT as compared to those with the silicon-bridged polymer. We propose that triplet exciton formation occurs via the excited CT state, thus diminishing the probability that the CT state forms free charge carriers in blends of C-PCPDTBT with PCBM. This hypothesis is confirmed by direct detection of strong triplet signatures in C-PCPDTBT:PCBM blends. The shape of the transient EPR spectra reveals that the triplet excitons are (in contrast to those formed in pristine polymer films) not generated by direct intersystem crossing, but are caused by back-electron-transfer from the interface into polymer domains. This BET triplet is not observed in blends containing the Si-bridged polymer, indicating efficient singlet exciton splitting and subsequent charge carrier separation at the Si-PCPDTBT/PCBM interface [3].
[1] Scharber, M. C., et al., Influence of the Bridging Atom on the Performance of a Low-Bandgap Bulk Heterojunction Solar Cell. Adv. Mater. 2010, 22, 367-370. [2] Behrends, J.; et.al., Direct Detection of Photoinduced Charge Transfer Complexes in Polymer Fullerene Blends. Phys. Rev. B 2012, 85, 125206. [3] Kraffert, F.; et.al.,Charge Separation in PCPDTBT:PCBM Blends from an EPR Perspective Journal of Physical Chemistry C,2014, 118,28482-28493.