The lithium-ion battery (LIB) technology has reached a very high energy density at beginning-of-life (BOL), but with the significant life-times (≈ 10–15 years) targeted for EVs and stationary applications, the prediction of a chosen end-of-life (EOL) is of uttermost importance – regardless of the exact LIB chemistry. The design of life-time predictors, appropriate health evaluators, and not the least support to strategies to mitigate the ageing, constitute a very challenging task in view of the multiple materials degradation mechanisms of both shelf-life and cycle-life. The electrolyte composition, the active materials’ chemistry, the binders and additives of the electrodes, the fabrication process and the operating conditions – all effect the ageing. As the degradation process in general starts very slowly, the collection of degradation data is a time-consuming and costly task, why accelerated ageing protocols are implemented, but always questioned for representativeness.

This PhD thesis aims at developing and demonstrating a machine learning (ML) based computational platform, ultimately able to predict the battery life-time as function of a wide spectrum of cell properties and operation conditions. In the first stage it will be limited to LIBs, but employing a wide spectrum of chemistries, for which extensive life-time databases already exist. In the second stage one of the most promising next generation battery technologies, the sodium-ion battery (SIB), will be targeted. The student concrete deliverables of this PhD will be a set of ML computer codes predicting life-time, the predicted life-time data itself, as well as a collaborative space in the cloud to share data, results and codes with all the academic and industry club members of the ALISTORE European Research Institute which is funding this project.     

The work will mainly be carried out at the Laboratoire de Réactivité et Chimie des Solides (LRCS) at Université de Picardie Jules Verne in Amiens, France, with regular 3 months stays at the Department of Physics at the Chalmers University of Technology in Göteborg, Sweden (every 9 months).

The candidate should have a background in computational science, applied mathematics, programming and knowledge in chemistry and materials science. Knowledge on artificial intelligence, machine learning, data mining and battery field will be a plus. Excellent English both written and spoken is mandatory. The candidate should be highly dynamic, autonomous, mobile and with great team spirit.

ALISTORE European Research Institute is a federative research structure coordinated by CNRS, funded through academic member contributions and the membership fees of several that joined its associated Industrial Club. Such funds allow developing hand in hand, integrated and collaborative research oriented towards medium-term transfer to industry. More information of ALISTORE ERI and its members here: http://www.alistore.eu/ 

If you are interested, please send your CV, motivation letter and the contact details of 4 referees to Prof. Alejandro A. Franco ([email protected]) and Prof. Patrik Johansson ([email protected]).