Automation is under-used in materials research, and could accelerate discovery in quantum materials. According to a new comment published this week in Nature Materials, flexible, robotics-driven automation enables the rapid creation of new automated experiments and empowers researchers to solve problems faster and with greater agility.
Flexible automation capitalizes on synergistic advances in robotics, rapid prototyping, computer vision, and open-source software ecosystems. Its major benefit is that it is customizable across a range of inorganic and organic materials and applications, reducing manual steps in even the most intricate experiments.
Materials discovery is foundational for new technologies, particularly in the clean energy sector. But while the need for clean energy technologies is urgent, the process of discovering and engineering these prospective materials can take decades using conventional approaches. According to Curtis Berlinguette, Professor in UBC’s Departments of Chemistry and Chemical & Biological Engineering, automation could realize more efficient technologies sooner, and at lower cost.
“Flexible automation makes robotics accessible to materials researchers,” said Berlinguette. “The diverse experimental workflows used in materials science often require highly customized automation that is not commercially available. ‘Flexible automation’ is changing this situation because of the recent emergence of safer, cheaper, and more user-friendly robotics.”
Berlinguette and colleagues have refined an automation process that originated with Ada, an artificially intelligent, self-driving laboratory able to explore formulations for a type of thin-film material common to advanced solar cells and consumer electronics. While it is inherently challenging to automate materials experiments, Ada is able to process around 50 consecutive samples without error, with other robotic platforms showing similar promise.
Canada currently leads the world in applying flexible automation to materials science, the result of an $8 million investment from Natural Resources Canada in Project Ada.
“This project has taught researchers around the world how to automate diverse workflows in the laboratory,” said Berlinguette. “Flexible automation also enables the materials science community to realize the full potential of machine learning algorithms. We used flexible automation to build the world’s first ‘self-driving lab’ for clean energy materials.”
With substantial investment from the Government of Canada in a national quantum strategy, and leadership in quantum materials and technology at several leading institutes including the Stewart Blusson Quantum Matter Institute (Blusson QMI) at The University of British Columbia, advances in flexible automation and self-driving laboratories will further cement Canada’s leadership role in advanced materials innovation.
Berlinguette and his team will establish their hub, the Flexible Automation Lab, in Blusson QMI in 2022. Their move to this newly renovated, larger space will bring them closer to other Blusson QMI researchers, including those working on thin-films and coatings under the Atomistic approach to emergent properties of disordered materials Grand Challenge.