Is Transition Metal Dissolution a Major Degradation Process in Commercial Li-Ion Cells? Application of Impedance with Extended Sub-mHz Low-Frequency Range

Abstract

Transition metal dissolution is a well-known degradation process in Li-ion batteries correlated with capacity fade1. Transition metals from a lithium transition metal oxide cathode can dissolve which causes structural changes in the crystal structure leading to deterioration of the material. Additionally, dissolved transition metals can migrate to the anode and reduce on the surface, impacting the surface activity of the anode composite and increasing its impedance.It is generally known that electrochemical performance of a Li-ion insertion battery electrode is primarily defined by the active material chemistry, further it is strongly impacted by the material and electrode-level engineering (particle size and aggregation, electrode porosity and tortuosity, electrolyte amount and properties, conductive additive and binder choice and their distribution in electrode composite). Furthermore, impedance response of an insertion electrode exhibits a strong dependence upon lithiation level (x). The specific dependence of impedance of an electrode upon x should allow to help in separating the two individual contributions of the two electrodes in a full cathode-anode battery cell. Thus, not surprisingly, Electrochemical Impedance Spectroscopy (EIS) is a common electrochemical technique of choice for degradation studies of Li-ion batteries2. Typically, EIS spectra of batteries are measured in frequency spans from MHz (laboratory cells) or kHz (commercial cells) down to mHz ranges. However, impedance measurements down to lower minimal frequencies (below 1 mHz) are often avoided from the practical reasons due to the inevitably longer measurement times being required.In the present work cylindrical commercial 18650-size high-energy NMC-Graphite(+Si) cells were electrochemically (galvanostatically) cycled at various conditions. In the constant temperature regime (10°C, 23°C, 35°C or 45°C) the tested cells were systematically cycled within various depth of discharge (DoD) ranges to observe the effect of the voltage range on the evolution (decline) of cell capacity. EIS responses were measured at different selected states of charge (SoC), with the aim to help to resolve the individual contributions of cathode and anode. State of health (SoH) of a battery was determined by current integrated capacity measurement and was periodically monitored after specific number of cycles and correlated with the impedance growth. EIS measurements down to conventional 10 mHz as well as to very low frequencies (0.1 mHz) were performed and compared to determine what additional information and insights for developing of a physics-based battery cell model and degradation modes can be obtained by including the extended low-frequency range (down to 0.1 mHz). We systematically investigated evolution (increase) of impedance of commercial Li-ion cells including the extended low-frequency range.After completion of the cycling test, post-mortem analysis of the cell components was performed on selected cells. Cells were disassembled, electrodes carefully removed, properly washed and SEM-EDX and ICP-MS analysis was performed on the cathode and anode samples. The amount of transition metals in both cathode and anode was determined.We thoroughly confirmed that the impedance of the commercial (full cells) in general increases during the decrease of capacity at all the tested conditions. Importantly, EIS spectra measured down to very low frequencies show that in addition to the (expected) increase of the real component (Re(Z)) simultaneously also the imaginary component (Im(Z)) significantly increases. We provide novel physics-based explanation of the underlying phenomena. Electrochemical characterization of the half cells prepared from the harvested electrodes in general showed a decrease in capacity and increase in voltage hysteresis for both cathode and anode – the actual changes showing strong dependence upon conditions during cycling. EDX and ICP-MS analysis revealed that the composition of the cathode active material remained practically unchanged and that the amount of the transition metals (TMs) deposited on the anode did not increase greatly with the cycling – the total amount of detected TMs remaining well below 300 ppm (relative to the amount of TMs in pristine cathode). Gilbert, J. A., Shkrob, I. A. & Abraham, D. P. Transition Metal Dissolution, Ion Migration, Electrocatalytic Reduction and Capacity Loss in Lithium-Ion Full Cells. J. Electrochem. Soc. 164, A389–A399 (2017).Meddings, N. et al. Application of electrochemical impedance spectroscopy to commercial Li-ion cells: A review. J. Power Sources 480, 228742 (2020). Figure 1. a) Typical examples of EIS response of 811-NMC-Graphite(+Si) full cells measured at SoC = 0.5 (3.705 V) down to 1 mHz (dark colors) and down to 0.1 mHz (lighter dolors) for pristine cell (black/grey), cell cycled at 23°C (blue), and cell cycled at 45°C (red). The corresponding SoH values were: 1, 0.39, and 0.39, respectively. b) Comparison of capacity of the half-cells prepared with the cathodes and anodes harvested from the initial commercial cells (panel a). Figure 1