# An artificial intelligence-driven synthesis planning platform (PhotoCat) for photocatalysis

## TL;DR

The paper introduces PhotoCatDB and PhotoCat, an AI platform that improves photocatalytic reaction prediction and accelerates sustainable chemistry discovery.

## Contribution

The novel contribution is the creation of a large annotated database and a high-accuracy AI platform for photocatalysis.

## Key findings

- PhotoCat achieves 82.6% accuracy in photocatalytic reaction prediction.
- The platform experimentally validated four novel photocatalytic reactions with yields up to 75.3%.

## Abstract

While photocatalysis has emerged as a transformative tool in modern synthesis, AI-assisted reaction prediction faces significant challenges due to data limitations. We present PhotoCatDB - a curated, open-source database containing 26.7 K photocatalytic reactions with detailed mechanistic annotations, including 9.2 K multicomponent transformations. Leveraging this resource alongside 100 million molecular data points, we developed PhotoCat, a Transformer-based platform that achieves unprecedented accuracy in photocatalytic reaction prediction (82.6%), retrosynthesis (77.1%), and condition recommendation (88.5%). The platform’s capabilities were experimentally validated through the discovery of four novel photocatalytic reactions with yields up to 75.3%. This integrated approach establishes a new paradigm for data-driven innovation in photocatalysis, bridging computational prediction with experimental validation to accelerate discovery in sustainable chemistry.

Photocatalysis holds promise for sustainable synthesis, but AI-driven reaction prediction is hindered by limited data. Here, the authors introduce PhotoCatDB, a comprehensive database of 26.7 K reactions, and develop PhotoCat, a transformer-based tool achieving high accuracy in photocatalytic reaction prediction, retrosynthesis, and condition recommendation, as well as experimentally validate the newly discovered photocatalytic reactions.