Machine Learning-guided Realization Of Full-color High-quantum-yield Carbon Quantum Dots

This article discusses a novel multi-objective optimization strategy that integrates machine learning techniques with iterative experiments and characterization to guide the hydrothermal synthesis of carbon quantum dots (CQDs). The approach is able to learn from limited data and uncover hidden relationships between synthesis parameters and target properties, resulting in full-color fluorescent CQDs with high photoluminescence quantum yield (PLQY). The workflow consists of four key components: database construction, multi-objective optimization formulation, MOO recommendation, and experimental verification.