Blade‐Coated Solar Minimodules of Homogeneous Perovskite Films Achieved by an Air Knife Design and a Machine Learning‐Based Optimization. Issue 3 (1st October 2022)

• Record Type: Journal Article
• Authors: Ramírez, Edwin Alexander; Velásquez, Juan Pablo; Flórez, Alejandra; Montoya, Juan Felipe; Betancur, Rafael; Jaramillo, Franklin
• Is Part Of: Advanced engineering materials. Volume 25:Issue 3(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-10-01
• Subjects: air knife -- blade coating -- inverted perovskite -- machine learning optimization -- perovskite solar minimodules -- power conversion efficiency (PCE) distribution map; Materials -- Periodicals; 620.11
• DOI: 10.1002/adem.202200964
• Languages: English
• ISSNs: 1438-1656

Abstract

Abstract : Perovskite solar technology stands on three different pillars: efficiency, stability, and processability. Focusing on processability, the field demands the achievement of optoelectronic grade active layers with high uniformity fabricated by industry‐compatible methods. Herein, the design and implementation of an air knife on a blade coating deposition system, which allows controlling the evaporation rate of the solvents improving the crystallization and uniformity of perovskite films in a p–i–n device are reported. The effect of doctor blade operational parameters is studied using image analysis combined with a machine learning method to identify the most relevant processing variables leading to a uniform perovskite layer with optimal thickness. After implementing the air knife system and finding the best processing conditions, a special perovskite solar minimodule enabling the evaluation of single inner subcells is fabricated. These perovskite subcells reach an average efficiency of 10.1% and remarkably all the subcells deviate less than 20% from this value over a large‐area substrate. These results demonstrate the promising potential of this fabrication method for low cost and high deposition rate photovoltaic devices, which is on the path to mass production. Abstract : Advancing perovskite solar cells from lab to fab require considering a large amount of novel processing variables, which can be overwhelming. Herein, a machine learning‐based optimization enabling highly uniform perovskite layers by employing air knife‐assisted doctor blade processing is presented. Finally, spatially distributed small area devices are tested to validate a highly uniform photovoltaic performance.