THE PROJECT

OVERALL OBJECTIVES

Our starting point are the encouraging results obtained by the RD-PDE morphochemical model, called DIB. We already studied from the theoretical and numerical points of view solutions that are Turing patterns, oscillating Turing-Hopf solutions and spiral waves. We proved in all cases that the DIB model for different choices of the physico-chemical parameters can describe experimental pattern of electrodeposits (battery electrodes) in qualitative and quantitative ways. In the BAT-MEN project we essentially identify two lines of development detailed in the following tasks.

From one hand, we would like to improve the numerical techniques used so far, looking for different approximation strategies both in space and time in order to improve the accuracy and reduce the computational load required by the simulations. This goal is also related to the construction of accurate and rich training sets for Deep Learning techniques for parameter identification in the RD-PDE model. 

The second line of development aims to both modify the original DIB model and introduce new mathematical PDE models to include more realistic physics for the study of battery electrode degradation and charge-discharge processes. Moreover, a possible extension on stationary and evolving surfaces will be considered to further account for more realistic metal growth. In all cases, new theoretical studies to describe pattern formation in the new settings and dedicated numerical methods for approximation and simulations are required.

Another major challenge in battery modeling is to understand and control the formation of dendrites at electrodes, that seriously reduce the charging capacity and then the cycling life. To account for these material localization processes with final objective of tuning parameters to reduce this peculiar metal growth, a new Phase-Field (PF) modelling approach will be introduced, that we will call PF-Den.

Another objective in this project concerns the study of charge-recharge profiles in batteries with metal anodes in new post-lithium batteries, e.g zinc-air devices. Towards this aim, we start by the new modelling approach proposed in a very recent publication in our collaboration with the Battery Materials Engineering Laboratory (BMEL) of Milano Politecnico (prof. Bozzini).

In all tasks, the numerical solutions of the studied models (Stoch-DIB and BS-DIB, PF-Den, MAC) will be compared with experiment in qualitative and quantitative way. In the last case, the PDE numerical simulations will be fitted to experimental data in order to infer model parameters, both for comparison with experimental images (of dendrites and structured/unstructured patterns) and experimental time series recorded from laboratory charge-discharge processes in different battery operating conditions.