Proceedings of 13th Conference on Hybrid and Organic Photovoltaics (HOPV21)
Publication date: 11th May 2021
Halide perovskites have attracted tremendous attention due to their excellent opto-electronic properties. This has enabled perovskite PV single junction to reach record efficiencies of 25.5%[1]. The progress of perovskite in PV applications is the result of the improvement of the opto-electrical properties of the perovskite absorber, as well as of the quality of the interface between transport layers and the perovskite absorber. Specifically, the recent introduction of self-assembled monolayers (SAMs) as charge transport layers on ITO has led to highly efficient devices[2-3], as well as a recent tandem perovskite/Si record of 29.5%[4]. Thus, SAM-based transport layers are expected to be part of the technological roadmap of perovskite PV.
So far, limited investigation has been carried out on the surface coverage of SAMs on ITO. Surface coverage is important because non-covered areas can result in low open circuit voltage and electrical shunts, thus reducing the device efficiency. To date, it has been shown that the SAM quality depends on the ITO crystallinity and morphology, as well as its surface treatment prior to SAMs processing[5-6]. In this study, we investigate the surface coverage of SAM on ITO using MeO-2PACz ([2-(3,6-dimethoxy-9H-carbazol-9-yl)ethyl]phosphonic acid) employed in a p-i-n structure of ITO/SAM/CsFAMA/C60/BCP/Cu by transmission electron microscopy (TEM) in combination with conductive atomic force microscopy (c-AFM) mapping. We observe that processing SAM directly on ITO can result in an inhomogeneous layer, with areas having low molecular density in TEM images. Supporting the TEM images analysis, c-AFM map also reveals exposed ITO areas due to insufficient SAM coverage in ITO+SAM sample. We attribute this inhomogeneity to large device performance spread seen in ITO+SAM devices. On the other hand, when adopting an atomic layer deposited NiO[7-8] between ITO and MeO-2PACz, the homogeneity, and thus, the surface coverage of the SAM improve. The cross-sectional TEM images of NiO+SAM device point out to a homogeneous SAM layer on NiO in contrast to direct growth on ITO. The homogeneous distribution of SAM molecules on NiO and the low lateral conductivity of NiO lead to narrower device efficiency distribution with high shunt resistance on average reaching more than 20% efficiency with NiO+SAM device. Notably, NiO layers do not require pre-treatment prior to the SAM solution process, which offers an advantage over ITO - eliminating the discrepancy in ITO pre-treatment and properties. We trust that this finding can further promote the benefit of using phosphonic acid-based molecules as contact layers in PSCs.