Events‎ > ‎AI3SD Event List‎ > ‎

*** Postphoned *** 04/03/2019 - AI3SD Seminar: Closed-loop materials research for solid-state batteries - University of Southampton

posted 10 Jan 2020, 06:50 by Samantha Kanza   [ updated 27 Mar 2020, 03:25 ]
NB: This event has been postponed until further notice. 

When: Wednesday 4th March, 10:00 - 11:00 

Where: Building 29/1101, University of Southampton 

Speaker:
Taro Hitosugi (Professor, Tokyo Institute of Technology)

Bio: 
Education: 1999 Ph.D. Graduate School of Engineering, The University of Tokyo 
Professional Career: 2015 - present 2007 - 2015 2003 - 2007 1999 - 2003
Professor, Tokyo Institute of Technology Associate Professor, Tohoku University Assistant Professor, The University of Tokyo Sony Corporation
Interests: electrochemistry, solid-state ionics, surface/interface, thin films, heterointerfaces, oxides, hydrides, scanning probe microscopy, materials informatics, artificial intelligence (AI), Bayes optimization, robot

Abstract: 
Solid-state lithium batteries are promising candidates for the next-generation rechargeable batteries with high energy and power density. Such high-performance batteries require the discovery of new solid-electrode and solid-electrolyte materials. Thus, a high-throughput methodology for rapid selection and development of new inorganic materials becomes crucial.
To this end, the integration of knowledge, experience, and intuition of researchers using robotics and artificial intelligence (AI) can accelerate progress in materials research [1]. Strategies combining high-throughput synthesis with machine learning have already produced new small organic molecules and biomaterials at ever faster rates [2, 3, 4]. However, the application of these techniques to inorganic materials research is still at its infancy. Therefore, to drastically accelerate its development, the inclusion of AI and robotics into inorganic materials research is essential.
In this study, we demonstrate the autonomous synthesis of inorganic compounds using robotics and Bayesian optimization. This system fully automates sample transfer, thin film deposition, physical-property characterization, and growth condition optimization. Our apparatus is also equipped with a robot arm that can access each satellite chamber for growth and characterization. Based on the data obtained from the physical-property characterization, the Bayesian-optimization algorithm is used to predict the next growth condition (closed-loop process). In this talk, we showcase this closed-loop process in reducing the resistivity of Nb-doped TiO2, a negative electrode material [5]. We also show that this autonomous synthesis can be applied to a wide variety of functional inorganic materials including magnetic, optical, electronic, and ionic properties.

References
[1] R. F. Service, Science 366, 1295 (2019). [2] M. Peplow, Nature 512, 20 (2014).
[3] R. D. King et al., Science 324, 85 (2009). [4] J. M. Granda et al., Nature 559, 377 (2018). [5] T. Hitosugi et al., Phys. Status Solidi A 207, 1529 (2010).