The Impact of Driving Force on Electron Transfer Rates in Photovoltaic Donor–Acceptor Blends

The effect of acceptor energy level on electron transfer rate in blends of the polymer solar-cell material poly[4,8-bis[(2-ethylhexyl)oxybenzo1,2-b:4,5-b′dithiophene-2,6-diyl]3-fluoro-2-[(2-ethylhexyl)carbonylthieno3,4-bthiophenediyl]] (PTB7) is studied using time-resolved fluorescence. Fast electron transfer in less than 2 ps is observed for a driving force between 0.2 and 0.6 eV and the electron transfer is slower outside this range. This dependence is described by Marcus theory with a reorganization energy of ≈0.4 eV. Organic semiconductors have great potential for the development of large-scale, flexible, and semitransparent solar panels. The primary excitations in organic materials are strongly bound excitons therefore for efficient charge carrier generation it is necessary to use a heterojunction of two materials, one an electron donor and the other an electron acceptor. The free energy difference between the initial (exciton) and final (electron–hole pair) states is known as the “driving force” for electron transfer (ET). Photoinduced ET is a critical process for a wide range of biological, chemical, and physical systems, including natural1 and artificial photosynthesis,2 photocatalysis and excitonic photovoltaic devices.3 A key issue in solar cells is that to generate a high power conversion efficiency, it is desirable to have the smallest driving force necessary to generate free charges, as any excess will lead to increased thermalization losses and consequently a reduced open circuit voltage. A recent study has shown that the photocurrent generation efficiency at short circuit conditions is independent of the excess vibrational energy, suggesting that it is possible to develop efficient blends with a minimal driving force and consequently with almost no energy loss at the interface between donor and acceptor.4 Another recent study showed the existence of an optimal driving force for the highest relative efficiency of the mobile charge generation.5 Time-resolved spectroscopies have shown that ET between the donor and acceptor in bulk heterojunctions formed by blending donor and acceptor molecules, depends on blend morphology and occurs on timescales less than 100 fs to tens of picoseconds with longer times representing diffusion-assisted ET.3 In organic semiconductors, it is of great interest to separate ET from exciton diffusion to get better understanding of the influence of driving force on the critical process of charge generation. This will allow the optimization of the blend materials' electrical properties based on a complete understanding of the photophysics, potentially leading to a break-through in the power conversion efficiencies attainable in organic photovoltaic (OPV) solar cells. In this communication we address this issue by measuring ET rates from thermally relaxed excitons in a high photovoltaic efficiency conjugated polymer, to an assortment of acceptors with a range of electron affinities (EA). Very low loadings of acceptor are used to ensure that the acceptor sites are spread throughout the film so that the blend morphology has no influence on the rate of quenching. Comparably low concentrations of similar quenchers have been shown to remain intimately mixed in the blend3, 6, 7 and a plot demonstrating that acceptor 4 remains intimately mixed well above the concentration used of 0.005 nm−3 is shown in the Supporting Information. The donor polymer used was poly[4,8-bis[(2-ethylhexyl)oxybenzo1,2-b:4,5-b′dithiophene-2,6-diyl]3-fluoro-2-[(2-ethylhexyl)carbonylthieno3,4-bthiophenediyl]] (PTB7), a highly efficient photovoltaic material,8 chosen to be as relevant as possible to the study of OPV optimization. The influence of quenching by Förster resonance energy transfer (FRET) was minimized and the effect of exciton diffusion was determined independently and taken into account. Fast ET in > k-1 then kq ≈ k1 and quenching is diffusion limited. We make the observation that quenching kinetics in Figure 2 are very similar for acceptors 5, 6, and 7, which give the fastest quenching indicating that quenching is diffusion limited in these blends. We can fit quenching kinetics for these acceptors with the time dependent kq = k0t−h as shown by the solid line in Figure 2 with k0 = 9.1 nm3 ps−0.66 and h = 0.34. In conclusion, the presence of an optimum driving force for ET is shown by strong dependence of the fluorescence lifetime of the donor doped with a small concentration of the acceptor. Fitting the measured rates to a Marcus model allowed us to determine that the reorganization energy λ = 0.4 ± 0.2 eV for ET from PTB7 to a range of acceptors, suggesting that it is mainly the polymer which reorganizes upon ET. Our study shows that ET from thermally relaxed excitons can be very efficient for processes with the optimal driving force. Understanding that the reorganization energy of the polymer is important for charge separation is very important to the community, as being able to produce a polymer with a small reorganization energy will be extremely beneficial in the development of highly efficient solar cells. This is because it lowers the energetic driving force required to rapidly separate excitons and hence allows a greater proportion of the photon energy to be harvested. This could be one contributing factor to why the highest efficiency P3HT solar cells (power conversion efficiency 6.43%) experience voltage losses (Eg −VOC) of 1.4 eV21 whereas the highest efficiency PTB7 devices experience voltage losses of just 0.9 eV.8 Sample Preparation: PTB7 with a molecular weight of 92 000 Da and a polydispersity of 2.6 was purchased from 1-Material. Acceptors 2, 4, and 5 were purchased from Solenne. Synthesis and purification of acceptors 1, 3, 6, 7, 8, and 9 is described in the Supporting Information. Blend films with low concentration of acceptors were produced by spin-coating from a 13.6 mg mL−1 chlorobenzene solution of PTB7 containing ≈0.07 mg mL−1 of acceptor. The exact quantity of acceptor was chosen so that a film spun from this solution would give a number density of acceptors of 0.005 nm−3 assuming a PTB7 mass density of 1.12 g cm−3.22 The films were spin-coated from solutions onto fused silica substrates at 2000 rpm under a nitrogen atmosphere. Time-Resolved PL: It was measured in a nitrogen atmosphere with a synchroscan streak camera C6860 from Hamamatsu. The excitation for blends with low concentration of acceptors was with 100 fs pulses at 400 nm and 80 MHz repetition rate. Blends with high concentration of acceptors shown in Figure 3a were excited at 650 nm and 100 kHz. This work was funded by Engineering and Physical Sciences Research Council of the UK (grants EP/I00243X and EP/J009016) and the European Research Council of the European Union (grant 321305). I.D.W.S. also acknowledges support from a Royal Society Wolfson Research Merit Award. AJW acknowledges the Scottish Doctoral Training Centre in Condensed Matter Physics for financial support. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.