Quantifying organic solar cell morphology: a computational study of three-dimensional maps

Abstract

Establishing how fabrication conditions quantitatively affect the morphology of organic blends opens the possibility of rationally designing higher efficiency materials, yet such a relationship remains elusive. One of the major challenges stems from incomplete three-dimensional representations of morphology, which is due to the difficulties of performing accurate morphological measurements. Recently, three-dimensional measurements of mixed organic layers using electron tomography with high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) provided maps of morphology with high resolution and detail. Using a simple, yet powerful, computational tool kit, these complex 3D datasets are converted into a set of physically meaningful morphology descriptors. These descriptors provide means for converting these large, complicated data sets (∼5×10 7 voxels) into simple, descriptive parameters, enabling a quantitative comparison between morphologies fabricated under different conditions. A set of P3HT:endohedral fullerene bulk-heterojunctions, fabricated under conditions specifically chosen to yield a wide range of morphologies, are examined. The effects of processing conditions and electrode presence on interfacial area, domain size distribution, connectivity, and tortuosity of charge transport paths are herein determined directly from real-space data for the first time. Through this characterization, quantitative insights into the role of processing on morphology are provided, as well as a more complete picture of the consequences of a three-phase morphology. The analysis demonstrates a methodology which can enable a deeper understanding into morphology control.

Electrode Materials, Thermal Annealing Sequences, and Lateral/Vertical Phase Separation of Polymer Solar Cells from Multiscale Molecular Simulations

“…These results are consistent with our previous results related to three-phase, three-dimensional morphology visualized with electron tomography. 37 This observation is also consistent with previous experimental work, 19 where authors speculated that as long as contact area between fullerene and cathode is nonzero (even as low as 3%), it enables electron extraction. However, the quality of percolation pathways can be affected by the partially blocking layer exposed by vertical segregation profile.…”

“…This is consistent with our previous analysis of experimental data (electron tomography data), where the volume fraction of the interfacial region between polymer-rich and fullerene-rich phase was smaller than 1%. 37 Note also that the samples contain a relatively high fraction of beads that we classify as pure P. Except case 1, this fraction varies from 13% to 15% of total number of beads. The value is slightly higher for steps corresponding to annealing.…”

“…A similar double well shape of vertical profile was also reported experimentally. 37 Substrate-induced phase separation was hypothesized as a mechanism behind this vertical distribution. 15,46 In ref 17, the extent of the zone was found to be comparable to the characteristic wavelength, 80 nm for P3HT:PCBM.…”

Computationally connecting organic photovoltaic performance to atomistic arrangements and bulk morphology

“…Additionally, removing assumptions about the differences between inter-and intra-chain transport and carrier delocalization by instead using QCCs to determine the electronic couplings between small chromophores in-situ, the quantitative discrepancy between these investigations and experimental data appears to be reduced. Coupling charge transport studies with structural analysis of the electrical network [122,153,154] or crystal packing [80,123], can help predict the morphological features that provide the best device performance. Extending the scope of the KMC simulations beyond the bulk, pristine domain to molecular and electrical interfaces for different materials allows for the prediction of full device performance characteristics such as power conversion efficiencies and J-V curves.…”

Domain Compositions and Fullerene Aggregation Govern Charge Photogeneration in Polymer/Fullerene Solar Cells

“…The occurrence of mixed domains, however, suggests that crystalline-crystalline interfaces are minimal. [ 21 ] Furthermore, the thermodynamics of mixing polymers with small molecules implies that interfaces with pure, amorphous polymers do not exist (because pure amorphous polymer phases do not exist). It follows that interfaces of types 4-6 shown in Figure 1 must dominate and mixed phases are prevalent at donor/acceptor interfaces.…”