Non-fullerene acceptor

Schematic of energy states in organic solar cells

The mechanism of current induction in organic solar cells involves a charge transfer. After electromagnetic absorption and exciton formation in the electron donor polymer, the excited electron is moved towards the acceptor conduction band (LUMO) as a result of the lower energy value than the donor LUMO. This process is called a charge separation, and the corresponding energy value ${\displaystyle E_{CS}}$ satisfies ${\displaystyle E_{CS}=E_{LUMO}^{A}-E_{HOMO}^{D}}$ where CS denotes charge separation, A denotes the acceptor and D denotes the donor molecule. Along with the Coulombic potential that needs to be surpassed, the maximum energy obtained from the process is defined as the Charge Transfer energy, ${\displaystyle E_{CT}}$. The difference between the optical excitation energy (the optical band gap energy, ${\displaystyle E_{g}^{opt}}$) and the charge transfer energy is the driving force of the system.[8]

An advantage of NF-OSCs over current fullerene-based OSCs is that the SMAs used are relatively compatible with donors, as a result of their electronic affinity tunability. Their compatibility originates from their LUMO-energy value similarity. The driving force is minimized to solely Coulombic contributions (<0.3 eV) with negligible charge separation loss. This results in low potential spillage, ${\displaystyle V_{loss}}$, which depends explicitly on the value of the driving force, along with radiative and non-radiative losses during the current induction process. Thus, for NF-OSCs, ${\displaystyle E_{loss}=qV_{loss}}$, with q the electron's charge, is minimized, leading to a higher useful energy output. The result is a high open-circuit voltage ${\displaystyle V_{OC}}$ of the solar cell compared to fullerene counterparts, with reports of values as high as 1.1V.[9] However, the diminished charge separation energy cost negatively influences the tendency of excited electrons in the donor conduction band to transport to the acceptor LUMO as it is less preferred energetically. This gives rise to the fact that electrons induced in the current are more energetic, but fewer electrons are induced. This means that the short-circuit current density ${\displaystyle J_{SC}}$ and the fill factor (FF) are decreased. [9] In terms of the PCE, the higher open-circuit voltage is compensated by the lower short-circuit current density and fill factor. Researchers showed that ultrafast charge separation is possible with negligible driving force. In fact, the electrical external quantum efficiency ${\displaystyle EQE_{EL}}$ is highest for donor-acceptor blends with lowest driving force.[9]

One of the main advantages of the non-fullerene acceptors is their ability to be tuned and customized by chemical modification. This in contrary to fullerene acceptors. It also immediately creates a bottleneck because of the huge amount of possibilities there are which could be applied as an SMA. A wide variety of SMAs are tested to be a successful acceptor, but two classes of SMAs have proven to give the best results concerning Power Conversion Efficiency (PCE) and have made the greatest attribute to the recent development in NFA-OSCs.[7]

Rylene diimides

Rylene diimides are, as said, one of the two main subclasses which are a basis for acceptor-molecules in modern NFA-OSCs. Rylene diimides are industrial dyes and can be divided into, once again, two subclasses: Perylene Diimides (PDIs) and Naphthalene Diimides (NDIs). Rylene diimides consist of a planar rylene framework and numerous constructions can be made by attaching certain subgroups and by using more PDI molecules in one acceptor. The mono-PDI molecule is shown in the figure on the right.

The basic structure of perylene diimides.

Rylene diimides are considered good acceptors because of their favourable properties. Rylene diimides usually have high electron mobility values ${\displaystyle \mu _{e}}$ due to intermolecular π-stacking.[7] These values are comparable to ones of fullerene acceptors. Furthermore, Rylene diimides also have a high absorbance spectrum in the visible area, high thermal and oxidative stability and their electric affinities can be tuned to a great extent by adding side groups and 3D-structure which leads to a significant higher open-circuit voltage (${\displaystyle V_{oc}}$)[1] Challenges that must be faced by designing and improving Rylene diimides based OSCs are mainly concerned by synthesis of PDIs because the planar structure of the molecule makes that it tends to aggregate into a crystal structure. This greatly enhances the domain size, larger than the preferred 20 nm, in the bulk heterojunction which leads to a lower charge transport ability. Researchers have tried to reduce this aggregation by three structure adaptions, all focused on enhancing the mobility of Rylene diimides molecules. The first approach is to link two PDI molecules with a single carbon bond, to form a so-called twisted dimer. The second synthesis forms highly twisted 3D-structures of PDI molecules and the third approach forms a fused-ring structure.[1] For all three possible ways, an example molecule is shown in the figure below. These derivatives are examples of acceptor-molecules which were tested and assessed in OSCs for their performance and PCE. Future research will focus on developing better PDIs resulting in higher PCE values for the OSC.

Three PDI based SMAs constructed using three different synthesis to decrease the domain size in the BHJ

In current research, rylene diimides (for small band-gap energy donors) and FREAs (for large band-gap energy donors) have shown the most potential for becoming commercially viable solar cell materials for bulk heterojunction blend cells. Wide band gap donors are known to enhance voltage and diminish current density, but in combination with FREAs both values can be relatively high.

There are still a lot of improvements to be made before an NFA-OSC can be commercially profitable. First of all, the PCE should be increased to at least 15% since this is the minimal value for commercial application.[7] As said, PCEs already have exceeded 13% [1] so recent development is on the right track. PCEs can be increased by designing even better NFAs, for instance, on the level of electron mobility NFAs still can increase a lot compared to FAs ( ${\displaystyle 3.3\cdot 10^{-4}cm^{2}V^{-1}s^{-1}}$ for the best NFAs compared to ${\displaystyle 7.0\cdot 10^{-4}cm^{2}V^{-1}s^{-1}}$ for the best FAs).[1] Improvements can also be made in the following aspects: better donor matching, tandem constructions, BHJ morphology and domain purity of the donor and acceptor.

Besides these theoretical research aspects, implementation in a life size commercial solar cell also brings a lot of challenges, such as easy and sustainable device fabrication methods and long-term stability of the organic compounds. Studies also show that with upscaling, the PCE in general drops.[7] On all of these areas, NFA-OSCs show great potential but it will take a lot of research before a solid non-fullerene acceptor-organic solar cell can compete with inorganic solar cells.

• Rylene dye