A chemical autonomous robotic platform for the synthesis of nanoparticles

The development and optimization of nanomaterials are often hindered by labor-intensive manual trial-and-error methods, leading to inefficiencies and inconsistent outcomes. To overcome these challenges, we have developed an innovative, data-driven, automated experimental system that combines artificial intelligence (AI) decision-making models with automated experiments. Our system utilizes a Generative Pre-trained Transformer (GPT) model to retrieve experimental methods and parameters and incorporates a closed-loop optimization process centered on the A* algorithm. The system can synthesize a variety of nanomaterials such as Au, Ag, Cu2O, and PdCu with diverse types, morphologies, and sizes according to the requirements of researchers. The automated experimental system demonstrated excellent efficiency and repeatability for nanomaterials synthesis. Using the A* algorithm, we conducted a comprehensive search for synthesis parameters targeting multi-target Au nanorods (Au NRs) with localized surface plasmon resonance (LSPR) in the range of 600 to 900 nm across 735 experiments, as well as for Au nanospheres (Au NSs) and Ag nanocubes (Ag NCs) in 50 experiments. In the repetitive experiment, the deviations of the characteristic UV-vis peak and the corresponding full width at half maxima (FWHM) of the Au NRs spectra with identical synthesis parameters do not exceed 1.1 and 2.9 nm, respectively. Throughout this process, researchers only need to edit scripts and input parameters for the automatic experimental platform during the initial stages, significantly reducing the demand for human resources in specific experiments. Finally, a comparative analysis of the search capabilities of the A* algorithm, Optuna, and Olympus confirms that the A* algorithm exhibits superior search efficiency, requiring significantly fewer iterations.