P. Leu
University of Pittsburgh,
United States
Keywords: machine learning, advanced optoelectronic materials, solar modules, light-emitting diodes (LEDs)
Summary:
This talk explores the use of machine learning techniques to design advanced optoelectronic materials. We begin by examining the optimization of glass for broadband and broad-angle antireflection, emphasizing the role of multi-objective Bayesian optimization in accelerating material discovery. We demonstrate how this optimization framework integrates both experimental data and simulations to efficiently develop multi-functional materials. Next, we highlight the development of high-transparency glass with enhanced antireflection properties, minimal haze, and oil-repellency. We explore key functionalities such as optical switchability, self-cleaning, stain resistance, and anti-fogging, discussing their impact on material performance. By incorporating bio-inspired design principles and advanced computational simulations, we showcase how novel optimization techniques enhance optical performance, transparency, and durability in optoelectronic applications. The insights from this work have broad implications for improving the efficiency and longevity of solar modules, light-emitting diodes (LEDs), and other optoelectronic devices.