Long-Term Aging Mechanisms in Li-Ion Batteries for Renewable Energy Storage

The development of renewable energies such as photovoltaic or wind energy is strongly conditioned by the intermittent nature of these energy sources, often making the production and consumption periods time-shifted. Therefore, the efficiency of energy storage is of first importance to manage the mismatch between energy production and consumption. Li-ion technology is a good candidate to fulfill this task [1], but this kind of application requires much longer battery lifetimes than other applications such as mobile electronic devices or electric vehicles. Indeed, batteries performances shall be preserved for up to 20 years, which is twice as much as the current commercial applications for EV/HEVs.

This work aims at describing the long time aging processes of a Li-ion battery. We focused on aging mechanisms occurring at interfaces between electrodes and electrolyte, which are a key parameter for the retention of battery performances over time. We studied Li-ion cells composed of a lamellar oxide NMC Li(Ni_xMn_yCo_1-x-y)O₂positive electrode and a graphite negative electrode. These cells were submitted to accelerated aging tests to simulate a long cycle or calendar life: high rate cycling tests at room temperature or storage at 60°C in charged state at 4.1V. In this experimental approach, aged electrodes were analyzed by X-ray Photoelectron Spectroscopy (XPS) and Electrochemical Impedance Spectroscopy (EIS) which provide complementary information about interfaces.

Post-mortem characterisation of aged cells highlighted the surface degradation of the NMC active material after intense cycling. From XPS data this degradation is revealed by a change of the layered oxide surface stoichiometry: (i) the Ni/Mn/Co ratio is different form the pristine positive material and (ii)the presence of these three metals is observed at the surface of the negative electrode (see Fig.1) with a larger amount than commonly observed [2].This demonstrates the enhanced dissolution process of the metals at the surface of the active material in the electrolyte. Moreover, XPS analyses also provided information about the composition of the passivation films at the surface of both electrodes considering the aging conditions.

Additionally, Electrochemical Impedance Spectroscopy (EIS) revealed that calendar ageing leads to the formation of a more resistive passivation film than cycling ageing at the surface of the positive electrodes. This means that the passivation layer at the surface of the positive electrodes is thicker (or denser) after a long storing period than after long cycling. EIS spectra of negative electrodes display distorted profiles as shown in Fig. 2 due to the samples porosity. Taking into account the porosity was necessary to establish correct equivalent electrical circuits modeling the electrode behavior. This leads to complex circuits using transmission lines based on the DeLevie model [3]. The evolution of samples porosity was studied in parallel with BET measurements in order to validate the associated models.

A new generation of prototype cells was developed afterwards. Results show that they have a better aging behavior. The modification of the positive electrode allowed a drastic capacity fading reduction upon cycling. Passivation layers have also been impacted by the operated modifications. O 1s XPS core peaks revealed that the cathode passivation film is thinner. Moreover, the active material degradation is lowered in these new cells.

[1] K.C. Divya, J. Østergraad, Electric Power System Research, 79(2009) 511-520

[2] S. Verdier, L. El Ouatani, R. Dedryvère, F. Bonhomme, P. Biensan, D. Gonbeau, J. Electrochem. Soc., 154(2007) A1088-A1099

[3] O.E. Barcia, E. D'Elia, I. Frateur, O.R. Mattos, N. Pebere, B. Tribollet, Electrochimica Acta, 47(2002) 2109-2116

Aknowledgement: This work was funded by ANR (Agence Nationale de la Recherche) and is part of the VISION project.