Organic Chemistry

Anti-Kasha System by Design: A New Gateway for Cell Differentiation Through Machine Learning

Junyi Gong Hong Kong University of Science and Technology


Kasha’s rule, which claimed that all emissions of excitons are from the lowest excited state and independent of excitation energy, makes the utility of high energy excitons difficult and severely hinder the widespread application of organic photoluminescent materials in real-world. For decades, scientists try to break it to unleash the power of high energy excitons but get minimal progress, no rational guiding principles, and few applications. So far, anti-Kasha’s rule is still a purely academic concept. In this contribution, we proposed a designing principle for pure organic anti-Kasha’s rule system and synthesized a series of compounds by following this designing rule which are all display evident S 2 emission in dilute solutions as proposed. Besides, we introduced a convolutional neural network as an assistant to apply the anti-Kasha’s rule luminogens to cell differentiations with high accuracy (~98.3%), which provide a new direction of applications of anti-Kasha system.

Version notes

First commit.


Thumbnail image of Anti.pdf
download asset Anti.pdf 0.90 MB [opens in a new tab]

Supplementary material

Thumbnail image of ESI.docx
download asset ESI.docx 1 MB [opens in a new tab]