Proceedings of nanoGe Fall Meeting19 (NFM19)
DOI: https://doi.org/10.29363/nanoge.nfm.2019.200
Publication date: 18th July 2019
The photovoltaic properties of an organic solar cell (OSC) crucially depend on the morphology of the active layer. An optimal bulk morphology for a high efficiency OSC can be achieved by using solvent additives combined with thermal annealing during device fabrication [1, 2]. Understanding the appearance of an optimal morphology and the mechanisms occurring during the fabrication process is therefore indispensable for further development of materials and OSCs. While the size of pure material domains should be within the range of the exciton diffusion length, the exciton dissociation decisively depends on the composition and extent of the donor-acceptor interface. The crystallinity of the different materials and the relative crystal orientation have a strong influence on the charge transport mechanisms. These three morphological parameters appear to have the most impact on the photovoltaic properties of an OSC.
We demonstrate the combination of conventional transmission electron microscopy (TEM) and analytical TEM (ATEM) allowing the simultaneous visualization of all three morphological parameters in non-fullerene acceptor (NFA) based blends. The interest in NFAs such as ITIC does not only arise from the increase in efficiency and stability of NFA based OSCs in the last years [3]. Furthermore, in some NFA blends highly efficient OSCs can be fabricated with negligible driving force for charge separation [4]. Our results reveal in detail the morphological parameters that contribute to an optimal PBDB-T:ITIC blend agreeing with atomistic simulations performed on this system [5, 6]. More significantly, our data implies that the presence of the polymer PBDB-T in this system forces the formation of particular ITIC crystals during thermal annealing. This process is even more enhanced by using solvent additives such as 1,8-diiodooctane (DIO) or chloronaphthalene (CN). These results explain the increase in efficiency for an PBDB-T:ITIC based OSC in relation to fabrication parameters.
An example of correlative conventional TEM and ATEM imaging is shown in Figure 1 for a 30 nm thin non-annealed PBDB-T:ITIC blend processed with DIO. For visualizing the material distribution of NFA based blends low-energy-loss ATEM and machine learning were applied (cf., details in [7, 8] for fullerene based blends). For PBDB-T and ITIC only minor spectral differences are found in the electron energy loss spectra (see Figure 1 a). However, appropriate data processing yields inverse material contrast due to minor spectral fingerprints (Figure 1 b and c). A material distribution map is computed after applying machine learning to a whole series of inelastic images. Additionally, crystalline areas visualized by conventional TEM on the same sample position can be correctly assigned to material phases. The spatial correlation of material composition with crystallinity yields a full view of the nanoscale morphology enabling correct interpretation of molecular arrangement over the whole sample area (Figure 1 d). An exact measurement of the lamellar spacing in both the acceptor (marked in red) and the donor (marked in blue) regions shows different crystal configurations typical for both materials (Figure 1 e and f).
We will further present the morphology of PBDB-T:ITIC layers processed with DIO and CN combined with thermal annealing. In comparison to the morphology shown in Figure 1 the annealed layers show different morphological parameters in agreement with atomistic simulations [6]. Our results imply the drastic influence of the molecular structure of both materials on the morphology formation during the fabrication process. This may help in designing new polymers and small molecules not only matching in their energetic properties, but also their molecular structure.
The authors acknowledge funding by the Ministerium für Wissenschaft, Forschung und Kunst Baden-Württemberg, through the HEiKA materials research centre FunTECH-3D (MWK, 33-753-30-20/3/3), and the Large Scale Data Facility (sds@hd).
This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No713750. Also, it has been carried out with the financial support of the Regional Council of Provence- Alpes-Côte d’Azur and with the financial support of the A*MIDEX (n° ANR- 11-IDEX-0001-02), funded by the Investissements d'Avenir project funded by the French Government, managed by the French National Research Agency (ANR).).