Photocurrent generation in organic bulk heterojunction (BHJ) solar cells is most commonly understood as a process which predominantly involves photoexcitation of the lower ionization potential species (donor) followed by electron transfer to the higher electron affinity material (acceptor) [i.e., photoinduced electron transfer (PET), which we term Channel I]. A mirror process also occurs in which photocurrent is generated through photoexcitation of the acceptor followed by hole transfer to the nonexcited donor or photoinduced hole transfer (PHT), which we term Channel II. The role of Channel II photocurrent generation has often been neglected due to overlap of the individual absorption spectra of the donor and acceptor materials that are commonly used. More recently Channel II charge generation has been explored for several reasons. First, many of the new high-efficiency polymeric donors are used as the minority component in bulk heterojunction blends, and therefore, the acceptor absorption is a significant fraction of the total; second, nonfullerene acceptors have been prepared, which through careful design, allow for spectral separation from the donor material, facilitating fundamental studies on charge generation. In this article, we review the methodologies for investigating the two charge generation channels. We also discuss the factors that affect charge generation via Channel I and II pathways, including energy levels of the materials involved, exciton diffusion, and other considerations. Finally, we take a comprehensive look at the nonfullerene acceptor literature and discuss what information about Channel I and Channel II can be obtained from the experiments conducted and what other experiments could be undertaken to provide further information about the operational efficiencies of Channels I and II.

An optical-frequency dielectric constant of 4.6 leads to improved charge generation efficiency in an organic semiconductor homojunction photovoltaic device.

between the active layer and the electrodes should lead to an improved V oc . [ 11 ] The conundrum faced in tuning the V oc is how to modify the ionization potential of the interlayer on which the organohalide perovskite layer is deposited while maintaining the optimum fi lm quality. For example, it has previously been reported that the V oc of PVSCs formed by physical vapor deposition onto ultrathin organic semiconductor interlayers could be modifi ed by changing the ionization potential of the interlayer. [ 11 ] However, the problem observed was that not all the interlayers gave the same nor indeed optimum organohalide perovskite fi lm.With the focus on solution processing for large area PVSCs, solving the ionization potential-fi lm quality issue is important. Attaining the desired organohalide perovskite fi lm essentially boils down to matching the surface energies of the deposition surface (electrode interlayer) and the evolving fi lm after casting. Titanium dioxide (TiO 2 ) in the form of either a mesoporous scaffold or a dense thin fi lm has the ideal surface energy and the required insolubility for solution processed PVSCs. [ 22,23 ] On the other hand, poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) can be processed at low temperatures, and possesses orthogonal solubility to the organohalide perovskite precursors and a suitable surface energy for fi lm growth. Recently, Choi et al. reported a conjugated polyelectrolyte (CPE-K) as a hole transport layer. [ 24 ] However, their devices with PEDOT:PSS and CPE-K interlayers exhibited low V oc s (<1 V), limited by the ionization potentials of PEDOT:PSS (−5.1 eV) and CPE-K (−4.9 eV), respectively. [ 24,25 ] More recently, in a similar approach using conjugated polymers with poly(electrolyte) side chains, planar devices with V oc s of greater than 1 V have been achieved. [ 26,27 ] It is important to note that the maximum possible V oc for a nonexcitonic homojunction solar cell is ultimately defi ned by the difference in electrode work functions after the thermodynamic limit has been considered. In our experience, other organic semiconductors, including those previously used in physical vapor deposited PVSCs, are unsuitable for solution processing due to the mismatch in surface energies. That is, the casting solution of the organohalide perovskite precursors simply does not wet the work function modifying interlayers although we do note one report that claims suitable fi lms can be formed on poly-TPD if very high spin-coating speeds are used. [ 3 ] Therefore, it is critical to develop interlayers that can meet three requirements: i) have an appropriate ionization potential to match the valence band energy of the chosen organohalide perovskite; ii) provide a suitable surface energy to allow for high-quality polycrystalline fi lm formation using solution processing methods; and iii) not be soluble in the solvent used to deposit the organohalide perovskite precursors. To date, these requirements are yet to be achieved in an interlayer and are the ...

Replacing all protons on a polymer by deuterium has a dramatic impact on spin-dependent properties of the material in devices.