Carbon-Dot-Based White-Light-Emitting Diodes with Adjustable Correlated Color Temperature Guided by Machine Learning

Here, we show the fabrication of the carbon dots (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 m(-2) (or 519 cd m(-2)). The machine learning was successfully used to predict PL CCT value. With the model, the white photoluminescence (PL) emission with adjustable correlated color temperature (CCT) from 3093 to 11018 K via combining blue, green, and orange CDs was achieved. Then, we obtained 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 m(-2)) and 6034 K accurately.

Automated summary

In this study, carbon dots (CDs) with green and orange emissions were fabricated and used for LED fabrication. Machine learning was utilized to predict and control the correlated color temperature of the white light emission, resulting in warm white and cold white LEDs.