Importance of Fluorine in High Voltage Electrolytes for LNMO||SiGr Cell Chemistry

Abstract

Lithium nickel manganese oxide (LNMO) and silicon/graphite (SiGr) are promising active materials for high voltage lithium ion batteries attributed to the high operating potential versus Li|Li⁺ of LNMO and the high specific discharge capacity of silicon. However, this cell chemistry exhibits rapid capacity fading, primarily attributed to electrolyte decomposition at the high operating voltage of 4.9 V. Here, a fluorinated electrolyte containing lithium hexafluorophosphate as conducting salt, as well as fluoroethylene carbonate and methyl (2,2,2-trifluoroethyl) carbonate as electrolyte solvents is introduced. The influence of the selected solvents on the interphase formation and galvanostatic cycling performance is analyzed using complementary electrochemical, spectroscopic, and safety-related techniques. The presence of fluorinated solvents enables a high oxidative stability of an electrolyte up to 5.0 V versus Li|Li⁺ and effective interphase formation. In comparison to cells with non-fluorinated electrolytes, the galvanostatic cycling performance demonstrates a considerable improvement, leading to a doubling of the achievable cycle life. Roll-over failure observed in the electrolyte with non-fluorinated solvents could be effectively suppressed for over 300 cycles and the resulting electrolyte formulation with fluorinated solvents is non-flammable. Additionally, by fine-tuning the electrolyte formulation, the extent of acetylcholinesterase inhibition, an indication of substance toxicity of the aged electrolyte could be reduced.

Keywords: high voltage electrolyte; lithium ion battery; lithium nickel manganese oxide electrode; non‐flammable electrolyte; silicon graphite electrode.

Figure 1

Mean specific discharge capacity versus cycle number a) and state of health versus cycle number profiles b) of LNMO||SiGr (20 wt.% Si) cells with four considered electrolyte formulations.

Figure 2

Electrochemical charge–discharge profile of LNMO||SiGr (20 wt.% Si) cells with non‐fluorinated a) and fluorinated electrolyte c) versus specific discharge capacity. The charge and discharge endpoints of LNMO||SiGr (20 wt.% Si) cells with non‐fluorinated b) and fluorinated electrolyte d) are plotted using the cumulative capacity as a function of cycle number.

Figure 3

Differential voltage versus capacity curves during the first discharge in a three‐electrode LNMO||SiGr (20 wt.% Si) cell with non‐fluorinated electrolyte.

Figure 4

Differential voltage versus discharge capacity curves of LNMO||SiGr (20 wt.% Si) cells with non‐fluorinated (a)and fluorinated solvents (b).

Figure 5

Cell polarization (∆V) as a function of cycle number of LNMO||SiGr cells with electrolytes containing non‐fluorinated and fluorinated solvents.

Figure 6

Nyquist plots of reassembled symmetric LNMO||LNMO a,b) and SiGr||SiGr c,d) cells containing non‐fluorinated and fluorinated electrolytes after 3 and 50 cycles.

Figure 7

Fitted core F 1s spectra for harvested SiGr electrodes with non‐fluorinated a,b) and fluorinated electrolyte c,d) after 3 and 50 cycles.

Figure 8

Fitted core C 1s spectra for harvested SiGr electrodes with non‐fluorinated a,b) and fluorinated electrolyte c,d) after 3 and 50 cycles.

Figure 9

SHINER spectra recorded from the surface of a SiGr electrode in the presence of the 1 m LiPF₆ in EC/EMC (3/7) a) and 1 m LiPF₆ in FEC/FEMC (3/7) b) electrolyte. Spectra were recorded prior to galvanostatic cycling and after the second cycle.

Figure 10

SET values  of four considered electrolyte formulations.

Figure 11

Specific discharge capacity a) and state of health b) of LNMO||SiGr (20 wt.% Si) cells as a function of the cycle number with different ratios of FEC and FEMC.

Figure 12

AChE inhibition [T/C, %] of 200 ppm solvent mixture or electrolyte after different numbers of cycles a) and AChE inhibition [T/C, %] of 2000 ppm of the solvent mixtures as applied in the electrolytes as well as the individual solvent/ACN mixtures b).Considered electrolytes: 1 m LiPF₆ in EC/EMC (3/7), 1 m LiPF₆ in FEC/FEMC (3/7), and 1 m LiPF₆ in FEC/FEMC (5/5). Paraoxon‐methyl (25 µm) was used as a positive control.