Artificial Intelligence and Quantum Dots: Design, Optimization, and Applications

Abstract

Artificial Intelligence (AI) has revolutionised the production of Quantum Dots (QDs), providing solutions to problems such as variability in the materials used to produce QDs, toxicity from elements such as Lead and Cadmium, and the short lifetime of QDs. AI and Machine Learning (ML) provide the opportunity to improve the reverse engineering of QDs and batch production of QDs, which has decreased the number of experimental trials needed to develop QDs, especially for mass production, by more than 90%. The types of algorithms used in this process include Random Forest, Support Vector Regression (SVR), Graph Neural Networks, and Generative Adversarial Networks (GANs). These algorithms aid in predicting attributes such as Bandgap and Quantum Yield and developing alternative non-toxic materials used to create QDs. Furthermore, AI can be integrated into automated synthesis platforms, such as Microfluidics, to allow for QDs to be produced continuously in real-time with uniform properties throughout the production cycle. AI also uses computer vision techniques for the measurement of QD size and defects to reduce production errors and increase manufacturing throughput. In the area of biomedicine, AI provides the potential to create QDs that facilitate drug delivery and imaging, thus providing safer, less toxic alternatives. Future advancements in QD technology may include concepts such as digital twins, reinforcement learning, and explainable AI.