Open Collections

Parlane, Fraser G. L.

University of British Columbia

Clean energy materials are most frequently used in electrochemical devices such as solar cells, fuel cells, and electrochemical reactors. These devices have exponentially larger materials spaces that must be optimized in comparison to a single component. This increased complexity is due to the intermolecular interactions that arise at the interfaces between materials in these devices. The performance of these devices is determined by this interfacial chemistry. Both the characterization and control of these interfaces is required, for a knowledge-based approach to device optimization. Techniques that characterize the interfaces of materials in fully functional devices are, therefore, attractive. In Part 1, I show that dye-sensitized solar cells can be used as a characterization tool to probe interfacial interactions. I am able to manipulate the strength of a halogen-bonding interfacial interaction by exchanging a single atom, and consequently control the performance of the electrochemical device. Most scientific disciplines use automation to accelerate discovery in such large experimental spaces. Laboratory automation has yet to be extensively used in materials science. The diverse experimental workflows and extreme experimental conditions used in materials science often require highly customized automation. Flexible automation is the use of robotics to create reconfigurable, automated experiments, which has recently been enabled by the emergence of safer, cheaper, and more user-friendly robotics. Self-driving laboratories are one such opportunity for materials researchers to use flexible automation to move towards autonomous, hypothesis-driven experimentation. Self-driving laboratories are robots controlled by decision-making algorithms that can autonomously plan, execute, and learn from materials science experiments. In Part 2, I introduce Ada: our self-driving laboratory built with flexible automation for automated material discovery and optimization. I show that Ada can accelerate rapid materials discovery and has been successfully deployed to three different materials projects.

Text

2022-09-21

Attribution-NonCommercial-ShareAlike 4.0 International

http://hdl.handle.net/2429/82734

Chemistry

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-sa/4.0/