Steven Kench

IMSE Lunchtime Seminar Series

Molecular Engineering for Next Generation Batteries

The challenge that will be discussed in this session is:

Machine learning methods for battery electrode characterisation and optimisation

Join us for this informal webinar with Steven Kench. There will be an opportunity for question and answer after the presentation. To join this webinar you must register in advance and you will be emailed the joining instructions for the webinar.


The development of next generation batteries for electric vehicle relies on the improvement of charge rate, capacity and cycle life behaviour. These metrics are linked to electrode electrochemical properties such as tortuosity, which in turn are often strongly influenced by microstructure. Thus, microstructural optimisation is a promising approach for performance gains. This talk will discuss how a machine learning method called generative adversarial networks can be used in this field. Of particular interest is the ability to synthesise 3D microstructural datasets using 2D micrographs to avoid resolution limitations associated with 3D tomographic imaging. Furthermore, the properties of these synthetic volumes, such as porosity and surface area, may be directly controlled. These capabilities allow the rapid exploration of possible microstructures and are hence a powerful optimisation tool in conjunction with electrochemical modelling techniques.

If you have any questions about accessibility requirements please email Leah Adamson (IMSE Events Officer) on

More Webinars in the Molecular Engineering in next Generation Batteries Webinar Series

Registration is now closed. Add event to calendar
See all events