Data for the chemistry of the future
At the Max Planck Singapore Center for Data-driven Chemistry, researchers use automated experiments and artificial intelligence to generate reliable data for chemical reactions and new materials
Which substances react with one another, and under which conditions? Which catalysts accelerate reactions most effectively? What mechanisms and kinetic laws govern chemical reactions? And what role do solvents play? To answer these and similar questions with the help of artificial intelligence (AI) and machine learning (ML), chemistry – like many other fields of science – needs one thing above all: reliable data, generated in reproducible experiments, in sufficient quantity and precision. Today, however, chemical experiments are often difficult to reproduce. Descriptions in the scientific literature are frequently incomplete and do not document all relevant parameters. Negative results – such as reactions that did not proceed as planned – are rarely published.
The Max Planck Singapore Center for Data-driven Chemistry therefore aims to generate reliable chemical data through automation and standardization. The goal is to create “AI-ready” datasets that make it possible to use artificial intelligence to gain fundamental insights into chemical reactions and the development of new materials.
“The collaboration with our colleagues in Singapore is an important step toward the digitalization of chemistry,” says Peter Seeberger. “Only a shared global approach can drive this important revolution in chemistry – which is why we are excited to embark on this groundbreaking research together.”
Researchers from the Max Planck Institute of Colloids and Interfaces in Potsdam, Germany and the Max Planck Institute for Dynamics of Complex Technical Systems in Magdeburg, Germany are working with partners from Singapore at the Nanyang Technological University (NTU), the National University of Singapore, and the Agency for Science, Technology and Research. The Singaporean partners conduct world-class research and face challenges similar to those in Germany when it comes to the digitalization of chemistry. Together, they aim to develop globally accepted data standards, attract international talent, and translate their findings into applications in the chemical and pharmaceutical industries. The systematic exchange of complex chemical process data across institutional and national boundaries could become a blueprint for future cooperation between research organizations and industry.
The participating Max Planck Institutes contribute complementary strengths. The team led by Peter Seeberger at the Max Planck Institute of Colloids and Interfaces studies complex chemical reactions in synthetic chemistry and automates synthesis processes, for example for the production of pharmaceuticals and vaccines. At the Max Planck Institute for Dynamics of Complex Technical Systems, the department headed by Kai Sundmacher develops data-driven methods to predict material properties, analyze the structure and kinetics of chemical reaction networks, and design molecules, materials, and process configurations for sustainable chemical production systems.
“Our vision is to explore large molecular design spaces using data-driven methods and to search for novel molecules and material structures with optimal properties,” says Kai Sundmacher. “These could include highly efficient catalysts for fine chemicals or environmentally friendly solvents for plastic recycling.”