De Novo Creation of a Naked-Eye-Detectable Fluorescent Molecule Based on Quantum-Chemical Computation and Machine Learning

26 March 2021, Version 1
This content is a preprint and has not undergone peer review at the time of posting.


Correlations between molecular properties and structures, such as those between the absorption wavelength and conjugate length, are beneficial for designing materials and controlling their properties. However, determining the molecular structures that correlate with the target molecular properties (such as molecular fluorescence) is not an easy task. In this study, we have used a de novo molecule generator (DNMG) coupled with quantum-chemical computation (QC) to develop new fluorescent molecules, which are garnering significant attention in various disciplines. With massive parallel computation (1024 cores, 5 days), DNMG has produced 3,643 candidate molecules within the density functional theory (DFT; one of QC) framework. Among the generated molecules, we have selected an unreported molecule and synthesized it for photoluminescence spectrum measurement. Our experimental verification demonstrated that DNMG can successfully create a new molecule which emits fluorescence detectable by the naked eye, as predicted by the DFT.


de novo molecule generation
density functional theory
fluorescent molecule
machine learning


Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.