Smart Manufacturing of Metal Halide Perovskite Nanocrystals Enabled by an Intelligent Multi-Robotic Platform

Metal Halide Perovskite (MHP) quantum dots (QDs) have shown great potential for energy and display applications, due to its unique optical characteristics that can be accurately tailored by adjusting their shape, size, and composition(1). The surface ligation, which depends on an acid-base equilibrium reaction, can not only enhance the colloidal stability but also mediate the optical characteristics of MHP QDs(2). Utilizing diverse carboxylic acids as surface capping ligands leads to different growth mechanisms and thus diverse QD morphologies. Besides the surface ligand influence, the optical properties of MHP QDs are also strongly impacted by the processing conditions, including the precursor concentrations during synthesis and anion exchange reactions that facilitate swift and precise control of nanocrystal composition, enabling band gap engineering and extending the emission range across the entire visible spectrum(3). As a result, the optical characteristics of MHP QDs are profoundly influenced by both the ligand structure (a discrete parameter) and the reaction conditions (continuous parameters). The expansive and multidimensional nature of this parameter space presents a substantial challenge for comprehensive and effective exploration of the manufacturing conditions for MHP QDs to achieve optimal optoelectronic properties.<br/>Conventional synthesis strategies for MHP QDs often rely on manual flask-based techniques that are time-intensive, material-intensive, and laborious. Traditional methods may struggle to elucidate the intricate interdependencies of reaction and processing parameters in colloidal QD synthesis—hindering the discovery of optimal formulations, limiting our understanding of underlying mechanisms, and inspiring the creation of more efficient investigative techniques(4).<br/>In this work, we have developed a multi-robot, self-driving lab (SDL) aimed at accelerating synthesis studies of room temperature-synthesized and halide exchange–modified MHP QDs. This SDL facilitates the systematic, automatic, and efficient investigation of how ligand structure and precursor concentrations impact the bandgaps, photon-conversion efficiency, and nanocrystal size uniformity of MHP QDs. Subsequently, we used the multi-robot SDL to autonomously map the Pareto front of the optical properties for various target bandgaps of MHP QDs.<br/>We addressed the challenges of traditional applied and fundamental MHP QD research by exploring the science and engineering of a multi-robot, intelligent experimentation platform. We developed a closed-loop MHP QD manufacturing strategy by combining this intelligent platform with Machine Learning modeling and Bayesian Optimization for experiment planning. The SDL expedited the mapping of the ligand structures and processing conditions to the optical properties of MHP QDs—enhancing our understanding of how ligand structure can impact the shape, morphology, and optical properties of MHP QDs. The knowledge generated by the SDL furthers our ability to conduct the on-demand synthesis of MHP QDs with optimized optical properties, paving the way for the advancement of future energy and display technologies .<br/><br/>Reference:<br/>1. Y. Bai, M. Hao, S. Ding, P. Chen, L. Wang, Surface Chemistry Engineering of Perovskite Quantum Dots: Strategies, Applications, and Perspectives. Adv Mater 34, e2105958 (2022).<br/>2. A. Pan et al., Insight into the Ligand-Mediated Synthesis of Colloidal CsPbBr3 Perovskite Nanocrystals: The Role of Organic Acid, Base, and Cesium Precursors. ACS Nano 10, 7943-7954 (2016).<br/>3. Q. A. Akkerman et al., Tuning the Optical Properties of Cesium Lead Halide Perovskite Nanocrystals by Anion Exchange Reactions. J Am Chem Soc 137, 10276-10281 (2015).<br/>4. M. Abolhasani, E. Kumacheva, The rise of self-driving labs in chemical and materials sciences. Nature Synthesis 2, 483-492 (2023).