Search

In this work, the harmful effect of triplet excitons on the lifetime of poly(p-phenylene vinylene) (PPV)-based organic light emitting diodes is demonstrated. A model system has been designed in which the triplet population in a PPV derivative is increased by blending the polymer with the triplet sensitizer platinum (II) octaethylporphyrine ketone (PtOEPK). The increase in the triplet concentration of the polymer affects both the initial decay and the long-term evolution of luminance and leads to a drastic acceleration of device degradation, reflected in a dramatically shortened lifetime.

We report the photovoltaic performance of a low-bandgap polymer:perylene diimide (PDI) photovoltaic blend and study the exciton to charge carrier conversion in the photoactive layer by Vis-NIR broadband transient absorption spectroscopy over a dynamic range from pico- to microseconds. Power conversion efficiencies of 1.2 % are obtained from the polymer:PDI blends with a maximum EQE of about 30 %, which is significantly below the performance of the same polymer with fullerene as acceptor indicating that severe loss processes exist that limit the photocurrent. From the evolution of the transient absorption spectra we conclude that the photovoltaic performance of the polymer:PDI blends is mainly limited by inefficient exciton harvesting and dissociation at the interface. However, once free charge carriers are generated in the blend they can be extracted as photocurrent as their recombination occurs on a timescale similar to the time typically needed for charge extraction from the photoactive layer. Hence, strategies to improve the efficiency of polymer:PDI blends should aim at increasing exciton harvesting at the heterojunction and the dissociation efficiency into free charges at the interface.

The dependence of the thin film morphology and excited-state dynamics for the low-bandgap donor-acceptor copolymer poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b']-dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (PCPDTBT) in pristine films and in blends (1:2) with [6,6]-phenyl-C(61)-butyric acid methyl ester (PCBM) on the use of the solvent additive 1,8-octanedithiol (ODT) is studied by solid-state nuclear magnetic resonance (NMR) spectroscopy and broadband visible and near-infrared pump-probe transient absorption spectroscopy (TAS) covering a spectral range from 500-2000 nm. The latter allows monitoring of the dynamics of excitons, bound interfacial charge-transfer (CT) states, and free charge carriers over a time range from femto- to microseconds. The broadband pump-probe experiments reveal that excitons are not only generated in the polymer but also in PCBM-rich domains. Depending on the morphology controlled by the use of solvent additives, polymer excitons undergo mainly ultrafast dissociation (100 fs) in blends prepared without ODT or diffusion-limited dissociation in samples prepared with ODT. Excitons generated in PCBM diffuse slowly to the interface in both samples and undergo dissociation on a time scale of several tens of picoseconds up to hundreds of picoseconds. In both samples a significant fraction of the excitons creates strongly bound interfacial CT states, which exhibit subnanosecond geminate recombination. The total internal quantum efficiency loss due to geminate recombination is estimated to be 50% in samples prepared without ODT and is found to be reduced to 30% with ODT, indicating that more free charges are generated in samples prepared with solvent additives. In samples prepared with ODT, the free charges exhibit clear intensity-dependent recombination dynamics, which can be modeled by Langevin-type recombination with a bimolecular recombination coefficient of 6.3 × 10(-11) cm(3) s(-1). In samples prepared without ODT, an additional nanosecond recombination of polaron pairs is observed in conjunction with an increased intensity-independent trap-assisted nongeminate recombination of charges. Furthermore, a comparison of the triplet-induced absorption spectra of PCPDTBT with the charge-induced absorption in PCPDTBT:PCBM blends reveals that triplets have a very similar excited-state absorption spectrum compared to the free charge carriers, however, in contrast have a distinct intensity-independent lifetime. Overall, our results suggest that whether free charges or strongly bound CT states are created upon dissociation of excitons at the PCPDTBT:PCBM interface is determined instantaneously upon exciton dissociation and that once formed strongly bound CT states rapidly recombine and thus are unlikely to dissociate into free charges. The observation of a significantly larger bimolecular recombination coefficient than previously determined for poly(3-hexylthiophen-2,5-diyl):PCBM (P3HT:PCBM) and PCDTBT:PCBM samples indicates that nongeminate recombination of free charges considerably competes with charge extraction in PCPDTBT:PCBM photovoltaic devices.

The precise mechanism and dynamics of charge generation and recombination in bulk heterojunction polymer:fullerene blend films typically used in organic photovoltaic devices have been intensively studied by many research groups, but nonetheless remain debated. In particular the role of interfacial charge-transfer (CT) states in the generation of free charge carriers, an important step for the understanding of device function, is still under active discussion. In this article we present direct optical probes of the exciton dynamics in pristine films of a prototypic polycarbazole-based photovoltaic donor polymer, namely poly[N-11''-henicosanyl-2,7-carbazole-alt-5,5-(4',7'-di-2-thienyl-2',1',3'-benzothiadiazole)] (PCDTBT), as well as the charge generation and recombination dynamics in as-cast and annealed photovoltaic blend films using methanofullerene (PC(61)BM) as electron acceptor. In contrast to earlier studies we use broadband (500-1100 nm) transient absorption spectroscopy including the previously unobserved but very important time range between 2 ns and 1 ms, which allows us not only to observe the entire charge carrier recombination dynamics but also to quantify the existing decay channels. We determine that ultrafast exciton dissociation occurs in blends and leads to two separate pools of products, namely Coulombically bound charge-transfer (CT) states and unbound (free) charge carriers. The recombination dynamics are analyzed within the framework of a previously reported model for poly(3-hexylthiophene):PCBM (Howard, I. A. J. Am. Chem. Soc. 2010, 132, 14866) based on concomitant geminate recombination of CT states and nongeminate recombination of free charge carriers. The results reveal that only ~11% of the initial photoexcitations generate interfacial CT states that recombine exclusively by fast nanosecond geminate recombination and thus do not contribute to the photocurrent, whereas ~89% of excitons create free charge carriers on an ultrafast time scale that then contribute to the extracted photocurrent. Despite the high yield of free charges the power conversion efficiency of devices remains moderate at about 3.0%. This is largely a consequence of the low fill factor of devices. We relate the low fill factor to significant energetic disorder present in the pristine polymer and in the polymer:fullerene blends. In the former we observed a significant spectral relaxation of exciton emission (fluorescence) and in the latter of the polaron-induced ground-state bleaching, implying that the density of states (DOS) for both excitons and charge carriers is significantly broadened by energetic disorder in pristine PCDTBT and in its blend with PCBM. This disorder leads to charge trapping in solar cells, which in turn causes higher carrier concentrations and more significant nongeminate recombination. The nongeminate recombination has a significant impact on the IV curves of devices, namely its competition with charge carrier extraction causes a stronger bias dependence of the photocurrent of devices, in turn leading to the poor device fill factor. In addition our results demonstrate the importance of ultrafast free carrier generation and suppression of interfacial CT-state formation and question the applicability of the often used Braun-Onsager model to describe the bias dependence of the photocurrent in polymer:fullerene organic photovoltaic devices.