Page 179 - Read Online
P. 179

Page 4 of 14             Wang et al. Energy Mater 2023;3:300040  https://dx.doi.org/10.20517/energymater.2023.28

               polyvinylidene difluoride at a weight ratio of 80:10:10. The pastes were cast onto Celgard 2400 separators to
                                                                                   2
               form electrodes. The mass loading of graphite was controlled at ≈30 mg/cm , and the mass loading of
                                     2
               NCM811 was ≈18 mg/cm . Three layers of glass fiber (MA-EN-SE-01, Canrd) were used as the separators.
               Li||NCM811 cells were cycled from 2.5 to 5.0 V, and Li||graphite cells were cycled from 2.5 to 0 V.

               Characterizations
               After 100 times cycling, the cells were disassembled in a glove box, and all the electrodes were washed with
               the corresponding solvent (DME for TTME-d and DME-d systems, DEC for carbonate system) three times
               before characterizations. The samples were transferred for characterizations under an Ar atmosphere
               (without air contact). X-ray photoelectron spectroscopy (XPS) with a 200 μm X-ray gun source was
               performed on an Escalab Xi+ (Thermo Avantage), and the C 1s signal of 284.8 eV was set as an internal
               standard to correct the binding energies. The morphologies of graphite and cathode electrodes were
               characterized with a field emission scanning electron microscope (SEM) (TESCAM MIRA3). Aberration-
               corrected high-angle annular dark field-scanning transmission electron microscopy (HAADF-STEM)
               images were taken on Titan Cubed Themis G2 300. X-ray diffraction (XRD) patterns were characterized
               with Rigaku SmartLab equipped using a Cu Kα radiation source in the 2θ range from 0 to 90°, with a speed
               of 10°/min.


               RESULTS AND DISCUSSION
               Through the DFT calculation, the highest occupied molecular orbital (HOMO) energy value and the lowest
               unoccupied molecular orbital (LUMO) energy value of the different solvents and additives were
               investigated. As shown in Figure 1, the LUMO energy level of TTME (1.491 eV) is slightly lower than that
               of DME (2.316 eV) but still higher than that of general carbonate solvents and additives, indicating its
               strong reduction stability on the anode side. In addition, due to the high fluorine (F) substitution of the
               TTME molecule, its HOMO energy level (-7.758 eV) is obviously lower than those of traditional ether
               molecules and even lower than that of some carbonate solvents and additive molecules (DEC: -7.2779 eV,
               VC: -6.999 eV). This shows that the substitution of hydrogen (H ) by F can improve the oxidation stability
                                                                      2
               of ether molecules.

               For DME, although its reduction stability is the best among these solvents and additives, it would easily co-
               intercalate with graphite during charge/discharge, resulting in the stripping of graphite and the
               decomposition of the electrolyte [12-14] . This problem can be clearly seen from the CV curves in Figure 2A. As
               the discharge progresses, the CV curve of the Li||graphite half-cell using the ether electrolyte begins to show
               peaks when the voltage drops to ≈1.2 V, and as the discharge process continues, the peaks become larger,
               and the shape of the curves is irregular. During the delithiation process, the internal reactions of the cell
               with the ether electrolyte are also extremely unstable. As the voltage increases, small peaks not for the
               charge-discharge platform continue to appear in the CV curve. In the subsequent four cycles, such irregular
               small peaks are always observed, and none of them overlaps, indicating that various side reactions
               constantly occur. This shows that DME cannot form a stable SEI film on graphite when it is solely used, and
               the co-intercalation process with graphite leads to the collapse of the graphite structure and the
               decomposition of the electrolyte.

               The electrolyte developed in this work contains two ethers, DME and TTME, and the addition of TTME as
               the co-solvent prevents the occurrence of the above problem. Considering the entire electrolyte system, the
               addition of TTME weakens the interaction between DME solvent and Li  ions, enhancing the desolvation
                                                                             +
                                                                                               [19]
               process of Li  ions, which would be beneficial for inhibiting the co-intercalation with graphite . As can be
                          +
               seen from the CV curves of the Li||graphite half-cell using the fluorinated ether electrolyte in Figure 2B,
   174   175   176   177   178   179   180   181   182   183   184