(116b) Leveraging Small Data for Better Chemistry

Even though over 95% of goods around us come from the business of chemistry, developing new materials, formulations, and chemical production processes is a time-intensive and expensive challenge. Complex optimization efforts are commonly required to enhance production performance while avoiding harsh reaction conditions, high costs, polluting effects, and other limitations.

Data-driven machine-learning (ML) offers the possibility to dramatically reduce the resources needed for reaction optimization through supervised-learning models that predict reaction performance, catalytic behavior, and optimal reaction conditions. However, most ML algorithms require large datasets (i.e. 100’s-1000’s of points) to provide accurate predictions, while experimental data collection is expensive in human and material resources. The challenge of building large datasets can be averted with Bayesian Optimization frameworks that guide experimental campaigns through informed and smart decisions. Predictions are thus based on carefully selected training data, which ultimately reduces the number of experiments required to achieve reliable understanding of variable effects and reaction behavior.

Addressing fundamental questions regarding the limitations of a modified Bayesian Optimization approach, we elucidate a pathway for the design and implementation of intelligent experimental campaigns using an organic electrocatalytic cross-coupling transformation as a model reaction. The approach has been further validated on catalytic biomass conversion, dehydrogenation processes, and mechanocatalytic transformations, among others. Working at the intersection of ML and chemistry, we created SuntheticsML, an online, user-friendly, ML platform that leverages very small data (starting with only 5 data points) to enable the development of new chemical processes, materials, and formulations up to 15 times faster
without the need of coding or ML skills.

The insights from this work innovate on the future of reaction engineering, material design, and process optimization while lowering the barrier to the implementation of new chemical technologies. SuntheticsML accelerates lengthy and complex traditional optimization campaigns, harnessing the power of ML and small data to fast-track innovation, sustainability, and digitalization in the chemical industry, further reducing development waste, emissions, and resource consumption by up to 95%