Free charge photogeneration in a single component high photovoltaic efficiency organic semiconductor

Organic photovoltaics (OPVs) have emerged as a potential solution for cost-effective and flexible solar energy generation. However, their commercialization has been hindered by challenges in achieving efficiencies comparable to inorganic counterparts. A key difference lies in the primary photoexcitation products: inorganic semiconductors like silicon produce both excitons (bound electron-hole pairs) and free charges, with their ratio dependent on the material's dielectric constant. In organic semiconductors, however, the low dielectric constant (typically 3-4) results in the predominant formation of strongly bound Frenkel excitons with high binding energies. For decades, this has dictated OPV design, relying on molecular heterojunctions between donor and acceptor materials to split these excitons into free charges for current generation. The traditional heterojunction approach involves creating an interpenetrating network of donor and acceptor materials, leading to complexities in device fabrication and optimization. This approach also introduces inherent voltage losses and instabilities associated with the interfaces between the donor and acceptor domains. While some field-assisted exciton dissociation and extrinsic charge formation in neat homopolymers and small molecules have been reported, substantial, or majority, free charge formation (rather than polaron pair formation that undergoes bimolecular recombination) has remained elusive until the recent advent of non-fullerene acceptors (FREAs). FREAs, and particularly the Y6 molecule and its derivatives, have spurred a significant rise in OPV efficiency. Unexpected observations with Y6 include apparent barrierless free charge generation in blends with PM6, an increase in charge generation efficiency with light intensity, and delocalized excitons or intra-moiety intermediate states with charge-transfer-like character. These observations have been explained by invoking charge-transfer (CT) states. The high refractive indices and hence high complex dielectric constants of FREAs, particularly Y6, are crucial. The Saha-Langmuir equation, commonly used for inorganic semiconductors, predicts a substantial intrinsic free charge fraction in Y6 based on its low exciton binding energy and high dielectric constant. This paper investigates this phenomena in detail, exploring the mechanisms involved and the implications for OPV device design.

The literature extensively documents the challenges and progress in organic photovoltaics. Early work focused on establishing the basic principles of exciton generation and dissociation in organic semiconductors, highlighting the need for efficient heterojunctions to overcome the low dielectric constants that favor exciton formation (Halls et al., 1995; Tang, 1986). Subsequent research aimed at optimizing the morphology and energy level alignment of donor-acceptor blends, aiming to enhance charge separation and reduce recombination losses (Benduhn et al., 2017; Karuthedath et al., 2020). The development of non-fullerene acceptors (FREAs) such as Y6 represented a significant advance, pushing efficiency records to unprecedented levels (Hou et al., 2018; Cui et al., 2020; Yuan et al., 2019; Lin et al., 2015; Yan et al., 2018; Wang & Zhan, 2021; Cheng et al., 2018). Studies have explored the role of charge transfer states in efficient Y6-based devices (Wang et al., 2020; Zhu et al., 2020; Zhang et al., 2020; Liu et al., 2020; Perdigón-Toro et al., 2020), while others have investigated the impact of material properties like dielectric constant and refractive index on device performance (Chen et al., 2014; Kerremans et al., 2020; Park et al., 2021). However, prior works have not decisively established the predominance of intrinsic free charge generation in neat organic semiconductors to the degree investigated here.

This research employed a multi-faceted approach combining experimental techniques and theoretical calculations. The experimental methodology involved several advanced spectroscopic techniques: 1. **Intensity-dependent Photoluminescence Quantum Efficiency (PLQE) Measurements:** These measurements, performed using a Q-switched frequency-doubled Nd:YAG laser, assessed the relationship between excitation density and PLQE to discern the nature of charge recombination (radiative bimolecular vs. monomolecular). Different Y6 film thicknesses were analyzed to investigate the impact of reabsorption. 2. **Ultrafast Transient Absorption (TA) Spectroscopy:** A home-built setup with an amplified Ti:sapphire laser and an optical parametric amplifier (TOPAS) was used. TA spectra and kinetics were measured at various excitation densities to track the evolution of singlet excitons and free charges/CT states. Blends with hole-accepting materials (PTB7-Th and poly-TPD) were analyzed for comparison. 3. **Time-Resolved Terahertz Spectroscopy:** This technique, utilizing a dual lock-in approach, probed the conductivity of neat Y6 films to reveal the presence of free charges. 4. **Ultrafast Transient Photoluminescence Spectroscopy:** Photoluminescence up-conversion measured the kinetics of singlet excitons. Samples were measured under vacuum or inert nitrogen to avoid degradation. 5. **Photovoltaic Device Fabrication and Characterization:** Single-component Y6 devices with different cathode structures (ITO/PEDOT:PSS/Y6/LiF/Al and ITO/PCP-Na/Y6/LiF/Al) were fabricated and tested. Devices using blends with PTB7-Th were also created and their External Quantum Efficiencies (EQEs) were measured. Intensity-dependent short-circuit current (Jsc) measurements were used to evaluate recombination processes. Theoretical calculations were performed using Density Functional Theory (DFT) with Gaussian 09 and the long-range corrected CAM-B3LYP functional. Electronic coupling between exciton and CT states was calculated. Density of states (DOS) calculations, using the VOTCA-CTP package, evaluated energetic landscapes for charges in a Y6 thin film based on molecular dynamics simulations. The simulations used the OPLS3e force field in Desmond, building a simulation cell of 8x8x6 unit cells from the Y6 crystal structure.

The key findings of this research demonstrate the unexpected prevalence of intrinsic free charge generation in the Y6 molecule and provide a new framework for understanding OPV operation. 1. **Intrinsic Free Charge Generation:** Intensity-dependent PLQE measurements showed a peak in PL efficiency with increasing excitation density, indicating bimolecular radiative recombination of free charges. This finding was supported by ultrafast transient absorption spectroscopy, which revealed the presence of free charge/CT states with distinct spectral signatures and kinetic behaviors compared to singlet excitons. Time-resolved terahertz spectroscopy provided further evidence for free charge formation. 2. **High Exciton Dissociation Probability:** Kinetic modeling of the experimental data revealed a high exciton dissociation probability (60-90%) at AM1.5 light intensity, with a significant free charge fraction (65-99%) under steady-state illumination. 3. **Bimolecular Recombination and Hole Traps:** Despite the high intrinsic charge generation, the efficiency of single-component Y6 devices was limited by rapid bimolecular recombination and hole traps. Intensity-dependent Jsc measurements confirmed significant bimolecular recombination at 1 sun intensity. 4. **Improved Efficiency with Donor Material:** The addition of small amounts of the donor polymer PTB7-Th to Y6 effectively slowed down charge recombination, improving device EQE significantly, despite having almost identical exciton splitting kinetics. This demonstrates that the major role of the donor polymer in state-of-the-art blends is to reduce charge recombination, not primarily exciton splitting. 5. **Quantum Chemical Insights:** Quantum-chemical calculations illuminated the underlying mechanism of free charge generation. Strong coupling between exciton and CT states, along with a distinct polarization pattern resulting in an inherent donor-acceptor system in the Y6 crystal structure, are found to favor free charge formation. A bimodal density of states for holes confirms this driving force for charge generation. This is unique to the crystal structure itself, not an interfacial effect. This implies a new pathway for optimizing OPV devices by focusing on crystal engineering to tune the energy levels.

This research fundamentally alters our understanding of organic photovoltaic mechanisms. The demonstration of substantial intrinsic free charge generation in neat Y6 challenges the long-held assumption that excitons are the predominant initial photoexcitation product in organic semiconductors. The findings suggest a revised four-step process, considering exciton generation, exciton diffusion, exciton-free charge interconversion, and charge transport and recombination. Exciton diffusion length becomes a complex function of exciton lifetime and interconversion with free charges. The high yield of free charges in Y6, coupled with rapid recombination, shifts the focus in device design towards minimizing recombination losses rather than solely maximizing exciton dissociation at heterojunctions. The observed improved performance of Y6 blends with small amounts of donor material may be explained by the suppression of recombination due to level bending at the interface, acting to pull holes toward the donor material and electrons away from it, reducing recombination before they reach electrodes.

This study provides compelling evidence for intrinsic free charge generation in the high-efficiency organic semiconductor Y6. This discovery challenges existing paradigms in OPV research, shifting focus from solely optimizing exciton splitting to minimizing charge recombination in device design. The high free charge yield combined with rapid recombination in single-component devices, and the effect of small quantities of donor material on slowing recombination, provides new pathways for achieving high-efficiency OPVs via crystal engineering and energy level tuning through packing structure and molecular orientation. Future research should investigate strategies to enhance the dielectric constant, optimize crystal packing to reduce CT-ground state coupling, and explore doped organic p-n junctions.

While this study provides strong evidence for intrinsic free charge generation in Y6, some limitations should be noted. The kinetic models used to analyze experimental data involve simplifications and assumptions. The exact contributions of various recombination pathways (e.g., bimolecular vs. trap-assisted recombination) may be challenging to fully quantify. Furthermore, the study primarily focuses on Y6, and the generalizability of these findings to other organic semiconductors needs to be further investigated. The role of disorder in the crystal structure and its influence on charge generation and recombination warrant further research. Finally, the device performance of single-component Y6 devices remains low despite high intrinsic charge generation.