Chemical Reaction Engineering and Catalysis: AI/ML Workflows and Self-Driving Laboratories

TL;DR

This paper discusses how AI/ML workflows and autonomous laboratories can revolutionize catalyst design and chemical reaction engineering for sustainable industrial processes.

Contribution

It introduces a comprehensive framework integrating AI/ML, high-throughput experimentation, and self-driving labs to accelerate catalysis research and development.

Findings

AI/ML workflows enable faster catalyst discovery.

Self-driving laboratories automate experimental processes.

Data-driven approaches improve reaction engineering efficiency.

Abstract

Chemical reaction engineering is key to industrial might and sustainable chemistry. This will be enabled using smart, efficient catalysts or catalysis ecosystems. This is possible with advanced artificial intelligence and machine learning (AI/ML) workflows that need to be employed as agentic AI projects. The fundamentals of catalysis need to be emphasized. A strong focus on catalyst design, mechanistic studies, reaction engineering, and scale-up must use ML-driven workflows, along with high-throughput experimentation (HTE) and an autonomous, self-driving laboratory (SDL). Laboratory experience and data-driven approaches are valuable when working together to accelerate this development. Parametrize and create a virtuous circle for data-driven discovery across heterogeneous, homogeneous, and biocatalysts to enable utility in many chemical process industries as agentic AI tasks. This article builds the case for discovery science in catalysis and continuous improvement in chemical reaction engineering with this new ecosystem.