Perovskite solar cells achieved major breakthroughs in the past years reaching a record power conversion efficiency (PCE) of 23.7 % on small lab-scale.[1] Nevertheless, efficiencies of perovskite photovoltaic devices fabricated by scalable printing techniques such as blade coating still lag behind compared to cell performances prepared by non-scalable coating methods like spin coating.[2] Here, we report on impacts of a surfactant in absorber layers on fully solution-processable perovskite solar cells in the inverted (p‑i‑n) architecture partly fabricated by blade coating in order to promote the field of upscaling and close this efficiency gap.

Blade coating of the perovskite active layer was performed on a heated plate with substrate temperatures ranging from 120 to 145 °C. Careful tuning of parameters such as blading temperature and wet film thickness lead to a controlled drying and crystallization of the thin film to the perovskite crystal structure.

For coating the active layer, a lead(II) acetate trihydrate (Pb(CH₃COO)₂·3H₂O)-based precursor solution is utilized. Coating uniform perovskite layers over large areas (3x3 cm²) was improved by blending a surfactant[3] into the precursor solvent, which suppresses detrimental fluid dynamics[4, 5] during blade coating and thus guaranteeing a more homogeneous morphology on micrometer scale as indicated by scanning electron microscopy (SEM) and time of flight secondary ion mass spectrometry (TOF-SIMS) top-view images. Interestingly, the surfactant mainly resides at the perovskite surface as explored by TOF-SIMS depth profiles. Hence, the amphiphilic nature of the surfactant results in dewetting of the subsequently deposited electron transport layer (ETL) so that there seems to be the need of proceeding the device architecture with vacuum deposition methods.[3, 6] However, we can overcome this issue and managed to apply the electron transport material rather by solution-based processes via a precisely adjusted argon plasma treatment of the perovskite active layer. Thereby, the short, low-energy plasma step breaks off the long, hydrophobic chains of fatty acid residues of the surfactant molecules as analyzed by TOF-SIMS depth profiles. In the case of low surfactant concentrations, a too long plasma treatment results in destroyed methylammonium groups at the top of the perovskite film which may be the reason for reduced solar cell performances. These assumptions were confirmed through a deeper insight of additional X-ray photoelectron spectroscopy (XPS) measurements. Thus, the surfactant concentration and duration of plasma treatment was closely correlated with the resulting short circuit current and open circuit voltage of solar cell devices, respectively. By means of the plasma treatment, solution-processed perovskite solar cells containing a blade coated absorber layer with PCEs up to 10.9 % on a 0.24 cm² active area were achieved which is comparable to results of their spin coated counterparts.