Improved QLEDs with Machine Learning
Machine learning identified solvent formulations that produce highly homogeneous quantum dot films The post Improved QLEDs with Machine Learning appeared first on Physics World .
Quantum dot LEDs, or QLEDs, are attractive for displays and lighting due to their bright, pure colors and potential for low-cost production. However, achieving high performance in QLEDs is difficult because the quantum dots must be evenly distributed in a dense film. Poor distribution can result in lower efficiency and quicker device degradation.
Researchers have employed machine learning to optimize the solvent conditions used to create uniform quantum dot films. By analyzing five key solvent parameters, several solvents were evaluated. Atomic force microscopy was then used to evaluate film uniformity, revealing surface morphology and roughness. Three machine learning models were developed to predict film uniformity based on solvent properties.
Support Vector Regression emerged as the most accurate model and was used to determine an optimal mixed-solvent formulation. X-ray scattering confirmed that the resulting films had more homogeneous quantum dot packing. Devices fabricated using this optimized solvent formulation demonstrated higher efficiency and longer lifetimes than those made with single solvents.
The study underscores the importance of film homogeneity in QLED performance and showcases how machine learning can be effectively utilized to optimize solution-processed optoelectronic devices.
Written by urgent.news from Physics World's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
