RoboMapper—On-Chip Robotic Micro-Experimentation Reveals Quantitative Structure-Property Relations in Hybrid Perovskites

Solution-processed semiconductors, such as metal halide perovskites (MHPs), have attracted tremendous attention over the past decade, promising to revolutionize the fields of photovoltaics, photonics, and other printed electronic applications. However, given the large compositional and chemical space of modern solution-processed semiconductors, meeting multiple material requirements together with device performance, operational stability, material toxicity and cost, as well as sustainable manufacturing is currently a fundamental challenge that is nearly impossible to overcome using trial-and-error and heuristics approaches. We introduce the RoboMapper, a novel compact robotic laboratory which integrates materials on a chip. We demonstrate end-to-end miniaturized, automated workflow from ink formulation to high-throughput multi-modal characterization for efficient data collection. Compared with traditional manual workflows and existing full-scale serial automation, the RoboMapper is shown to be 5-10 times faster and reduces the material cost, toxic waste, and greenhouse gas emissions by more than an order of magnitude. A state-of-the-art case study on FA_1-yCs_yPb(I_1-xBr_x)₃proves the ability of our platform to rapidly establish the quantitative structure-property relationship (QSPR) in a large and complex compositional space and provide accurate and practical guidance for the screening and selection of compositions among a wide range of options with an ideal bandgap as well as improved photo-stability for different target applications. This platform can be universally applied to solution processable materials such as organic semiconductors, quantum dots, and nanoparticles to pave the way towards the fully autonomous experimentation of ink-based semiconductor materials, ink formulations and (opto)electronic devices co-design with the guidance of artificial intelligence (AI).