Electrochemical Noise Analysis applied to new generation Li-ion batteries
Résumé
This work aims to assess the possibility for Electrochemical Noise Analysis (ENA) to be used as a diagnosis method for new generation Li-ion cells (LNMO, LTO, TNO). Since it is only based on electrical fluctuations generated by the system, this method has the undisputed advantage of being non-invasive, operando and low cost. Plus, the implementation of this method into the current systems would not require any additional hardware, since cell voltage measurements is mandatory in a battery pack. The latter point is critical for economical and practical concerns.
Nevertheless, the very low level of voltage fluctuations (<μV) provided by the batteries renders a proper measurement challenging. That may explain why the literature is still poor on this application. In this work, an ultra low noise instrumentation has been purposely designed and realized to extract the voltage noise signature of four Li-ion coin cells with different electrode materials. The noise of the developed conditioning circuit has been characterized, and demonstrates the ability to perform measurements with an ultra-low background noise compatible with the targeted requirements for the ENA application. The estimation of the power spectral density of the signal allows a statistical description of the cells voltage fluctuations under different operating conditions. It is shown that the noise signature has a 1/f flicker nature, regardless the cell type. The noise level apparently depends on electrode materials (figure 1). It is also found out that the voltage noise intensity is proportional to the DC current load (figure 2). Finally, overcharging of a Li/LNMO cell seems to induce sharp transient in the voltage signal. This work demonstrates the high sensitivity of the ENA method to operating conditions, and thus its great potential for new generation Li-ion batteries diagnosis.
Domaines
Electronique
Origine : Fichiers produits par l'(les) auteur(s)