Lithium-ion (Li-ion) batteries play a vital role in our daily lives by powering essentials like smartphones, electric vehicles, and energy storage systems. One of the main factors that lead to accelerated aging of Li-ion batteries is the lithium plating effect. An improved understanding of lithium plating holds the potential to extend lifespan, enhance performance, and reduce cost of lithium-ion batteries. Lithium plating occurs while charging in low temperatures or with high currents, as lithium-ions accumulate on the anode.Lithium plating can be separated into two parts, a reversible and an irreversible part. During relaxation, when the battery sits idle, or while discharging, reversible plating is stripped from the anode. Typically, when plating is stripped, a voltage plateau can be observed. In contrast, irreversible plating consists of lithium deposits that cannot be stripped during discharge, therefore causing a loss of capacity.In this work, we present a novel non-destructive method for determining a cell's threshold plating current using only electrical measurements. The method is validated using commercial 3.2Ah Panasonic NCR18650B cells at varying low temperatures and charging currents. The designed method requires the use of two identical cells, a reference and a test cell. After completely discharging both batteries at room temperature to reach 0% depth of discharge (DOD), the same plating tests at low temperatures are conducted on both cells. Once the cells are fully charged with a constant current of 0.5C, the reference cell experiences relaxation for 10 minutes. Simultaneously, a sine wave current is applied to the test cell, leading to small charging and discharging cycles. During the positive half-wave of the sine, plating occurs and during the negative half-wave reversible plating is stripped again. As plating only occurs, once the threshold plating current is exceeded, the sine amplitude must be greater than this threshold. It is important to note that, the charging current consists of intercalation current and plating current.Intercalation is the insertion of Li-ion into the layered structure of graphite during charging. As charging progresses, the surface concentration of intercalated lithium increases, but simultaneously, the number of vacancies for lithium intercalation at the graphite surface decreases. When the anode potential falls below 0V, thermodynamically allowing lithium plating, the intercalation current decreases and the plating current increases. With continued charging, the plating current continues to increase, while the intercalation current gradually diminishes.We assume that only the part of the sine exceeding this threshold is responsible for the lithium plating, while the portion below this threshold contributes to intercalation. In contrast, the negative half-wave is completely responsible for stripping. After 10 minutes, both cells are discharged with a low current while the remaining reversible plating is stripped. From the voltage and current measurements during discharge, we can compute how much reversible plating was stripped during the sine phase and during the discharge. To determine the plating current, the plated charge amount dQ on the positive sine half-wave is calculated by the difference between the stripped charge amount on the sine negative half-wave and the difference between the stripped charge amount of the reference cell and the test cell (c.f. attached figure). With the known plated charge amount dQ, the sine is integrated to find the threshold plating current.To validate the precision of this method, we subject an additional cell to charging at the same low temperature as the other cells. The charging involved varying currents from 0.04C to 0.4C in 10 steps, with a 1.5-hour interval between each step for the cell to cool down. To determine the threshold plating current, a differential voltage analysis (DVA) was conducted at each step, thus identifying the lowest current where a voltage plateau occurred, as the threshold plating current. Further refinement involves smaller steps to achieve a more accurate plating current. We assume this result is the true threshold plating current. Comparing with our novel method, in the worst case we achieve an error of less than 100mA. To ensure the reliability and reproducibility of this method across varying cells, additional plating tests, involving various cell chemistry types, are planned. Figure 1
Read full abstract