Accelerating materials research: Jiyun’s journey with SPINBOT automation

Symbolic picture for the article. The link opens the image in a large view.

Jiyun has used fully automated SPINBOT platform for engineering solution-processed Perovskite thin films, addressing the persistent challenge of simultaneously optimizing processing parameters. SPINBOT employs unsupervised processing on hundreds of substrates, showcasing exceptional experimental control and high sampling variability. Utilizing the Bayesian optimization algorithm, the SPINBOT iteratively explores a complex parameter space, enhancing the quality and reproducibility of thin films. Guided by machine learning, the SPINBOT accelerates the optimization of perovskite solar cells through photoluminescence characterization. The study achieves an optimal film, leading to a champion power conversion efficiency of 21.6% with sustained performance reproducibility. Unsealed devices exhibit 90% retention of initial efficiency after 1100 hours of continuous operation under specific conditions. The paper highlighted as a Front Cover of Advanced Energy Materials Issue 48, Vol 17 cover