The Dynamic Change of the Pore Size Distribution in Porous Electrodes of Lithium-Oxygen Batteries during Discharging

The specific energy of lithium-oxygen (Li-O₂) battery is usually less than 10% of the theoretical specific energy due to the limitation of mass (particularly oxygen) transfer in the porous electrode. The insoluble Li₂O₂ generated/depleted during the battery discharging/charging significantly changes the pore size distribution, mass transfer properties of the porous electrode, and the specific energy of the battery. This study develops a two dimensional, transient, and non-isothermal computational fluid dynamics model to simulate the mass transfer and electrochemical reactions within the battery and quantitatively calculate the change of the pore size distribution at various locations of the porous electrode of the battery.

The pore size distribution of a home-made gas diffusional electrode (GDE) is measured by nitrogen adsorption and mercury intrusion tests and the distribution is fitted with a log normal distribution. The scale factor of and the shape factor of the fitted log normal distribution are -2.5 and 0.5, respectively. The corresponding mean value of the pore size is 93 nm. To make the model manageable, the infinite pore sizes are binned into 7 bins and each bin has a cumulative distribution function (CDF) to describe the probability of a pore in the electrode that has the size defined by the bin. The number of bins has also been increased to 10 and 15 to make sure that the results are independent from the number of bins.

The solid products from the oxygen reduction reaction (ORR), which are mainly Li₂O₂, deposits on the electrode surface and redistributes the pore size in the electrode. Since the Li₂O₂ is generated non-uniformly in the electrode, properties of the porous electrode such as the average pore size, effective surface area, permeability, and electrical conductive resistance change non-uniformly in the electrode. This model simulates the discharging of a Li-O₂ battery at 0.2 mA/cm² for 520 hours, which equivalents to the specific energy density of about 2300 mAh/g_carbon, and can capture the non-uniform property change in the porous electrode. The model results show that properties of pores close to the air side change the most due to fast reactions in these pores. The permeability of pores close to the air side decreases from 7.92×10^-15 m² to 4.46×10^-15 m² and the average pore size decreases from 93 nm to 61 nm after the discharging.