UC Santa Barbara

Organic Photovoltaics (OPVs) harvest energy directly from sunlight. They comprise a molecular or polymeric carbon-based donor:acceptor blend (bulk heterojunction or BHJ) in between two electrodes. OPVs underwent tremendous progress in the past few years, now reaching Power Conversion Efficiencies (PCEs) of over 19%. In contrast to their well-established inorganic counterparts, OPVs are lightweight, thin, and flexible, and they can be tuned on the molecular level, allowing the modification of color and transparency. These unique properties make OPVs ideal candidates for integrated energy harvesting solutions as their optical tunability and solution-processability are favorable for integration into buildings. However, to date, OPVs do not meet the requirements for widespread commercialization. I) High-performing systems suffer from reproducibility challenges due to batch-to-batch variations and II) a systematic in-depth understanding of the structure-property relationships is absent. Lastly, III) Semitransparent OPVs (ST-OPVs) still suffer from poor performances even though theoretical calculations predict high PCEs even for systems with 100% average visible transmittance (AVT). This work focuses on the understanding of these limitations through a multidimensional approach, including both experimental and simulation-based methods. After an introduction to OPVs in Chapter 1, various techniques used in this work are discussed in Chapter 2. Chapter 3 investigates residual catalysts traces as a possible cause for batch-to-batch variations, addressing I) and II). The systematic addition of Pd(PPh3)4 to PTB7-Th:IOTIC-4F devices and its effect on the morphology and the optoelectronic processes is studied. A drop in performance is observed that is due to altered material properties and different loss mechanisms, but the system showed robustness to 0.75% Pd(PPh3)4, an amount typically not exceeded after purification. Next, Chapter 4 presents a new approach to unravel the optoelectronic processes under short-circuit conditions, facilitating the study of II). We propose a new method to obtain the geminate prefactor Pg, the mobility-lifetime product µτ, and the extraction efficiency η, using only standard measurements and simulations. Our simple method also predicts the optimal device configuration and active layer thickness with greater accuracy than optical transfer matrix simulations, providing a fast and cost-effective alternative to the experimental device optimization. The following chapters focus on narrow-band gap ST-OPVs for integrated energy harvesting solutions, addressing III). Chapter 5 identifies changes in ST-OPVs that are concomitant with increased transparency. Reduced generation rates and altered generation rate profiles lead to a reduced open-circuit voltage (Voc) and changes in the recombination dynamics. We show that high-purity and low-trap-density active layers are crucial. Furthermore, transparent devices are sensitive to shunt-leakage, highlighting the need for high-quality active layers even more. The impact of surface recombination decreases with increased AVT and limitations due to high series resistance decrease, suggesting considering a wider range of transparent electrode materials. Chapter 6 focuses on the interfacial recombination in the narrow-band gap system PCE10:COTIC-4F. Our findings indicate that ZnO is the most suitable front electrode due to low interfacial recombination, efficient charge extraction, and favorable energy level alignment. Other electrodes are studied, including ZnO functionalized with PFN-Br, PEDOT:PSS and CPE-K. The present results show that interfacial recombination plays a significant role in narrow band gap OPV systems and future research in this direction will be necessary to overcome the PCE bottleneck arising from surface traps. Lastly, Chapter 7 introduces an experimental approach to obtain OPVs with highly transparent active layers, addressing II) and III). Devices with donor concentrations of 40% to 20% show high average visible transmittance (AVT) values of 64% to 77%. The AVT increases with lower donor concentration due to reduced visible range donor absorption and increased near-IR acceptor absorption. We investigate morphology, charge generation, charge recombination, and charge extraction, and propose further optimization of the interfaces and active layer morphology.