“This new method saves a huge amount of time,” stated Paco Laveille, who led the research. “With a conventional approach, our experiments would have taken years.”

AI algo and robotics accelerating the process

The Swiss Cat+ technology is a platform involving the integration of AI algorithms and automated laboratory infrastructure to accelerate the discovery and development of new chemical catalysts, particularly focusing on metal catalysts.

This platform enables researchers to efficiently identify promising catalyst compositions, synthesize them in an automated manner, and test their effectiveness, aiming to advance various chemical processes, including synthesizing substances like methanol from CO2.

A robotic arm was employed to transfer glass vials into a centrifuge. "It is part of a robot unit that produces catalysts completely autonomously according to the specifications of an AI mode," noted the research.

This process assisted in identifying such promising catalyst compositions and developing to efficiently and cost-effectively synthesize the energy source methanol from CO2.

Methanol is essential for a sustainable hydrocarbon economy

According to a statement by scientists, methanol is considered a key element for a sustainable hydrocarbon economy, as it can be used as a fuel and as a raw material for various organic compounds. 

Developing efficient catalysts for methanol production contributes to sustainability by providing an alternative to traditional fossil fuels.

Researchers can significantly reduce the time and resources required for traditional trial-and-error approaches by leveraging AI.

Furthermore, the researchers have developed an automated synthesis and test laboratory as part of their methodology. 

Addressing the challenges, Christophe Copéret, a professor at the Laboratory of Inorganic Chemistry at ETH Zurich and co-initiator of the Swiss Cat+ project, explained:

“The chemical space in which we’re searching for catalysts comprises around 1020 possibilities – that’s one hundred billion billion. So we’re literally looking for a needle in the chemical haystack.”

Therefore, researchers preselected the catalyst ingredients mainly limited to relatively cheaper metals, including iron, copper, and cobalt.

In addition to the main metals, the researchers also included three commonly used doping elements and potassium in their considerations. They restricted their selection of carrier materials to four standard metal oxides. 

This led to a staggering 20 million potential combinations when accounting for various mixing ratios.

This laboratory has robotic systems capable of autonomously synthesizing and testing numerous catalyst compositions. By automating these processes, the researchers can rapidly generate and evaluate a large number of catalysts quickly, accelerating the overall research process.

The combination of AI-driven predictive modeling and automated laboratory infrastructure enables the Swiss Cat+ technology platform to expedite the discovery of new and improved metal catalysts, paving the way for advancements in various chemical processes. 

“At present, knowledge about catalysts for fuel production is based predominantly on expertise from the oil industry,” Copéret stated. 

“When it comes to reactions for use in the sustainable energy industry, reliable data is still largely lacking.” Despite this, AI algorithms and human research intelligence require this data before conducting targeted searches within the expansive realm of chemical possibilities.

“And that’s precisely the kind of high-quality, reproducible data our AI-assisted robot laboratory now delivers. It’s certain to take catalyst research a long way forward,” Laveille added.