![Author ORCID: We display the ORCID iD icon alongside authors names on our website to acknowledge that the ORCiD has been authenticated when entered by the user. To view the users ORCiD record click the icon. [opens in a new tab]](https://chemrxiv.org/engage/assets/public/chemrxiv/images/logos/orcid.png)
Abstract
The right solvent is a crucial factor in achieving environmentally friendly, selective, and highly converted chemical reactions. Artificial intelligence-based tools, often lack the ability to reliably predict reaction conditions such as the appropriate solvent. Here, we present a comprehensive investigation into the efficacy of data-driven machine-learning models for solvent prediction for a broad spectrum of single-solvent organic reactions. Remarkably, our models achieve a Top-3 accuracy of 86.88%, showcasing outstanding performance in predicting solvents from underrepresented classes. An uncertainty analysis revealed that the models' misclassifications could be explained by the fact that the reaction can be run in multiple solvents. In the experimental validation, 8 out of 11 reactions succeeded with the predicted solvent. Our work addresses a key challenge in organic synthesis and demonstrates the practical application of machine learning models in predicting reaction solvents for more efficient and sustainable chemical synthesis.
