Artificial intelligence in the discovery and design of molecular semiconductors : a systematic review

Tools

Zollner, Malin and Moshfeghi, Yashar and Nematiaram, Tahereh (2026) Artificial intelligence in the discovery and design of molecular semiconductors : a systematic review. Digital Discovery. ISSN 2635-098X (https://doi.org/10.1039/D5DD00552C)

[thumbnail of Zollner-etal-DD-2026-Artificial-intelligence-in-the-discovery-and-design-of-molecular-semiconductors]
Preview
 Text. Filename: Zollner-etal-DD-2026-Artificial-intelligence-in-the-discovery-and-design-of-molecular-semiconductors.pdf
Final Published Version
License: Creative Commons Attribution 3.0 logo
Download (2MB)| Preview

Abstract

Artificial intelligence (AI) is rapidly transforming the discovery and design of molecular semiconductors by linking chemical structure to electronic function with unprecedented speed and accuracy. These materials underpin flexible, lightweight, and sustainable optoelectronic technologies, yet their optimisation has been limited by the immense chemical search space and the cost of exhaustive experimentation and quantum-chemical calculations. This systematic review presents a comprehensive, PRISMA-guided analysis of 237 studies published between 2010 and 2025 that apply AI and machine learning to molecular semiconductor research. The literature is organised into four interconnected domains: electronic structure and spectroscopic properties, photoactive materials, emissive materials, and charge transport. Across these areas, AI models have achieved near quantum-level precision in predicting key electronic and optical properties, enabled the generative design of high-efficiency photoactive and emissive compounds, and accelerated multiscale simulations of charge mobility. The review identifies major trends toward hybrid, data-efficient, and physics-informed learning frameworks while highlighting persistent barriers related to data quality, benchmark inconsistency, and limited interpretability. By consolidating diverse methodologies and findings, this work establishes a unified perspective on how AI can drive reproducible, scalable, and autonomous discovery of molecular semiconductors for next-generation electronic and photonic technologies.

ORCID iDs

Zollner, Malin ORCID logoORCID: https://orcid.org/0009-0000-9662-0869, Moshfeghi, Yashar ORCID logoORCID: https://orcid.org/0000-0003-4186-1088 and Nematiaram, Tahereh ORCID logoORCID: https://orcid.org/0000-0002-0371-4047;
CORE (COnnecting REpositories)