ABSTRACT
Increasing the operation voltage of electrochemical energy storage devices is a viable measure to realize higher specific energies and energy densities. A sufficient oxidative stability of electrolytes is the predominant requirement for successful high voltage applicability. The common method to investigate oxidative stability of LIB electrolytes is related to determination of the electrochemical stability window (ESW), on e.g. Pt or LiMn2O4 electrodes. However, the transferability of the obtained results to practical systems is questionable for several reasons. In this work, we evaluated the validity of the potentiodynamic based ESW method by comparing the obtained data with the results of galvanostatic based techniques, applied on commercial positive electrodes. We demonstrated that the oxidative stabilities, determined by the two techniques, are in good accordance with each other. However, the investigation of electrolytes being incompatible to Li metal, renders conventional ESW measurements useless when metallic Li is used as counter - and reference electrode in the ESW setup. For this reason, we introduced an alternative setup based on Li4Ti5O12 full cells. On the example of a butyronitrile-based electrolyte, we finally demonstrated that this electrolyte is not only reductively but also oxidatively less stable than common LiPF6/organic carbonate based electrolytes.
ABSTRACT
The 1st cycle Coulombic efficiency (CE) of LiNi1/3Co1/3Mn1/3O2 (NCM) at 4.6 V vs. Li/Li(+) has been extensively investigated in NCM/Li half cells. It could be proven that the major part of the observed overall specific capacity loss (in total 36.3 mA h g(-1)) is reversible and induced by kinetic limitations, namely an impeded lithiation reaction during discharge. A measure facilitating the lithiation reaction, i.e. a constant potential (CP) step at the discharge cut-off potential, results in an increase in specific discharge capacity of 22.1 mA h g(-1). This capacity increase during the CP step could be proven as a relithiation process by Li(+) content determination in NCM via an ICP-OES measurement. In addition, a specific capacity loss of approx. 4.2 mA h g(-1) could be determined as an intrinsic reaction to the NCM cathode material at room temperature (RT). In total, less than 10.0 mA h g(-1) (=28% of the overall capacity loss) can be attributed to irreversible reactions, mainly to irreversible structural changes of NCM. Thus, the impact of parasitic reactions, such as oxidative electrolyte decomposition, on the irreversible capacity is negligible and could also be proven by on-line MS. As a consequence, the determination of the amount of extracted Li(+) ("Li(+) extraction ratio") so far has been incorrect and must be calculated by the charge capacity (=delithiation amount) divided by the theoretical capacity. In a NCM/graphite full cell the relithiation amount during the constant voltage (CV) step is smaller than in the half cell, due to irreversible Li(+) loss at graphite.