From Groundwork to Efficient Solar Cells: On the Importance of the Substrate Material in Co‐Evaporated Perovskite Solar Cells

Abstract

Adv. Funct. Mater. 2021, 31, 2104482 The originally published article contains an error in the calculation of the mean free path of methylammonium iodine (MAI) according to equation (2) in the main manuscript (wrong diameter for the MAI molecule), which also resulted in too low values for the mean free path of MAI in Figure 1b and a wrong explanation in the introductory part of the manuscript in section 2.1. The corrected Figure 1b is as follows: Relating to this change, a short paragraph starting on page 4 in the main manuscript, left column second-to-last line (“Due to the larger diameter of the MAI molecule compared to PbI2 […] resulting in a loss of the directed effusive flux and a transition to a more CVD-like evaporation behavior.”) is not completely correct. The mean free paths of MAI and PbI2 are rather similar after correction. Therefore, the conclusion that the loss in the effusive component of the MAI evaporation is due to the increased number of collisions between MAI molecules is unlikely. However, the clear indications that the MAI evaporation in the employed process is not fully effusive, but at least in part of CVD-like nature (e.g., the high background pressure and the conversion of a PbI2 film in presence of the background MAI atmosphere as discussed in the supporting information) remains valid, but must have another primary cause. As mentioned in the original manuscript and shown by others, the decomposition of MAI into more volatile compounds is expected to be one reason for the strong rise in background pressure as well as the CVD-like evaporation characteristic. Given the fact that a strong CVD-like evaporation characteristic of MAI is observed in our process, the description of the employed process as well as its challenges are still valid in the original manuscript. Most importantly, the influence of the substrate material (here polarity) on the film formation is not influenced by this initially wrong interpretation of our process since the substrate dependency is linked to the interaction of the substrate surface and the condensing molecules in this complex mixed process environment as shown throughout the manuscript. The authors apologize for any inconvenience this error may have caused and thank Tomas Leijtens for making us aware of it.