Carbon dots based white light-emitting diodes with adjustable correlated color temperature guided by machine learning.
Xiao Wang,Bo Wang,Hongshuai Wang,Tianyang Zhang,Huihui Qi,Zhenyu Wu,Yurong Ma,Hui Huang,Mingwang Shao,Yang Liu,Youyong Li,Zhenhui Kang
DOI: https://doi.org/10.1002/anie.202103086
2021-03-23
Angewandte Chemie
Abstract:Carbon dots (CDs) have attracted extensive attention in white light-emitting diodes (LEDs) research due to their inherent optical advantages, such as emission wavelength tunability, high quantum yield (QY) and low photobleaching properties. However, their wide emission spectrum and the electron transfer between CDs make the fabrication of their light-emitting devices very complicated and difficult. Here, we show the fabrication of the CDs with green and orange emissions from PTCDA (perylene-3,4,9,10-tetracarboxylic dianhydride). Using these CDs as emitters, the orange (or green) CDs LEDs were fabricated, which show electroluminescence (EL) spectra centered at 560 nm (or 498 nm) with an external quantum efficiency (EQE) of 1.98% (1.76%) adhering a luminescence of 626 cd/m2 (or 519 cd/m2). We successfully constructed the machine learning model to predict PL CCT value. With the model, we achieve white photoluminescence (PL) emission with adjustable correlated color temperature (CCT) from 3093 to 11018 K via combining blue, green and orange CDs. We obtain the warm white CDs LEDs with CCT of 3107, 4071 and 4548 K, and cold white CDs LEDs with CCT of 5632 (CIE coordinates of (0.33, 0.33), EQE: 1.18%, luminescence: 598 cd/m2) and 6034 K accurately.
Medicine