正文阅读顺序审阅

Lithium ion batteries (LIBs) have been used in electric vehicles because of their high energy density, safety, and low price; however, these qualities need to be further enhanced to extend the maximum distance that electric vehicles can travel before needing to be recharged [1 -3]. Many novel and advanced types of cathode materials have been explored and had their potentials gauged to meet the requirements of a high cell energy density [4]. Lithium-rich layered oxide (LRM), xLi2MnO3 ⋅ (1-x)LiMO2 (M = Ni, Co, Mn) has attracted considerable attention as the most promising alternative for LIBs, due to its high reversible discharge capacity (over 250 mAh g 1 ) and high range of operating voltage (3.0 -4.8 V) [5 -8]. However, its application is limited by the poor cycling and rate performance of LRM, which leads to the destruction of the cathode material structure and an unstable

electrode/electrolyte interface [9].

To overcome these problems, several approaches have been explored in the past decade, including doping and coating [10 -13]. Recently, a simple and economical strategy has been developed: the formation of a protective interphase on the cathode through the preferential oxidation of additives ( < 5 wt%) added to the electrolyte. These additives have a higher highest occupied molecular orbital than solvents. This ensures that the electrolyte decomposition products do not react with the protective cathode interphase, resulting in the maintenance of interface stability. A variety of electrolyte additives with different molecular structures have been proposed. Tris(trimethylsilyl)phosphate [14,15] has been used as a film-forming additive on lithium-rich cathode materials to improve the electrochemical performance. Tris(trimethylsilyl) borate forms a modified solid electrolyte interphase (CEI) on the cathode, which reduces the capacity fading of a Li[Li0.2Mn0.54Ni0.13Co0.13]

O2/Li cell from 81% to 26% after 200 cycles [16]. Lan et al. reported [17] that the electrode capacity retention in an electrolyte without any additive was 40%, which is much lower than the 72% observed in an electrolyte with bis(trimethylsilyl)carbodiimide. The capacity retention of Li1.2Mn0.54Ni0.13Co0.13O2 cathodes was found to increase in an electrolyte with 1 wt% diphenyl disulfide [18]. After 150 cycles, the capacity fading of Li1.2Mn0.54Ni0.13Co0.13O2 was found to decrease from 27.4% in an electrolyte without any additive to 8.8% an electrolyte with 1% 1,3,6-hexanetricarbonitrile [19]. The cycling stability of Li1.2Mn0.525Ni0.175Co0.1O2 cathodes was improved by using 5 wt% di-(2, 2,2 trifluoroethyl) carbonate in the electrolyte, and as a result, the capacity fading decreased from 30% to 11% after 50 cycles [20]. Cha et al. [21] investigated Li-rich/graphite full cells with 1 wt% lithium difluoro (oxalate)borate (LiDFOB) in 1.3 M LiPF6 containing an EC:EMC:DMC electrolyte at a volume ratio of 3:4:3. LiDFOB optimizes the surface composition of both the graphite anode and Li-rich cathode. A stable CEI film was produced on the Li[Li0.2Mn0.56Ni0.16Co0.08]O2 electrode with the presence of the decomposition products of tri(hexafluoro-iso-propyl) phosphate (HFiP). The capacity loss for a Li[Li0.2Mn0.56Ni0.16Co0.08]O2 electrode in an electrolyte with 1% HFiP after 130 cycles was shown to be reduced from 35.5% to 26.7% [22]. Triphenyl phosphite (TPPi) was found to preferentially oxidize over the solvent at an oxidation potential of 4.2 V (vs. Li/Li + ), and the TPPi-derived layer can effectively inhibit electrolyte decomposition [23]. Tu et al. reported that TEP reacted with active oxygen and formed a protective layer that suppressed electrolyte decomposition and the structural destruction of the LRM [24]. Triethyl borate and tripropyl borate have been reported as electrolyte additives that improve the self-discharge property of LRM cathodes [25].

Some studies have reported that the improvement of the electrochemical behavior of NCM622-based cells and LNMO-based cells are a result of silicon-based electrolyte additives [26 -28]. In this paper, 1, 3-divinyltetramethyldisiloxane (DTMS) was used as an additive to enhance the electrochemical properties of a Li-rich layered oxide cathode (LRM). Results showed that 2 vol% DTMS effectivity decreased capacity loss of LRM/Li cell from 57.2% to 25%. The possible mechanism for this was investigated further though electrochemical and physical analyses.

The LRM cathode was composed of 80 wt% LRM (LRM300, Ningbo Li-rich Battery Material Technology Co., Ltd., China), 10 wt% polyvinylidene fluoride (PVDF) binder, and 10 wt% acetylene black in N -methyl-2-pyrrolidone. The slurry was coated on Al foil, and then vacuum dried at 100 ◦ C for 12 h. The CR2025-cell was fabricated in an Arfilled glove box with a Li sheet as the anode and Celgard 2325 as the separator. The base (BE) electrolyte contained 1 M LiPF6 and included EC/EMC/DMC (1:1:1, in weight). DTMS (Aladdin Co., Ltd., China) was then added into the BE electrolyte. Electrolytes with 2 vol% DTMS were used (Supplementary Material. 1).

A total of 1000 ppm of hydrofluoric acid (HF) was then added into the BE and BE + 2 vol%DTMS electrolytes to prepare HF-containing electrolytes using a 40 wt% HF aqueous solution [25]. All electrolytes were prepared in an Ar-filled glove box.

The charge -discharge performance of the cells was evaluated using a computer-controlled test system (CT2001A, China). The LRM/Li cells were pre-cycled in the following schedule: 0.1C (1C = 300 mA g 1 ) for three cycles and 0.2C for three cycles at 25 ◦ C between 2.8 and 4.8 V. The LRM/Li cells were then operated for 100 cycles at 0.5C to analyze the cycling performance. The rate capability of the cells was evaluated at a 0.5-1-2-3-5C charge/discharge current. The electrochemical

/Li) of EC, DMC, EMC, and DTMS.

impedance spectroscopy (EIS) was employed in the frequency range of 10 5 to 0.1 Hz with an amplitude of 5 mV, using a frequency response analyzer (VSP, Bio-logic).

To understand the influence of DTMS on the cycling performance of LRM/Li cells, scanning electron microscopy (SEM; S-4800, Hitachi), transmission electron microscopy (TEM; JEM-2010, JELO), XRD analysis (XRD; Bruker D8 ADVANCE) using Cu Ka radiation, X-ray photoelectron spectroscopy (XPS; ESCALAB 250Xi), and 19 F nuclear magnetic resonance ( 19 F NMR; AVANCE III 400 MHz) analyses were carried out. The cycled electrodes (LRM and Li) were disassembled, rinsed with dimethyl carbonate solvent, and dried under vacuum.

The Gaussian 09 software package was used for theoretical calculations. The equilibrium structures were determined using the B3LYP/6311G(d,p) level [28]. The oxidation potential (Eox) was obtained as follows[28]:

Eox (Li + /Li) = [G(M + ) - G(M)]/F - 1.4 V

The as-calculated oxidation potential and Li + binding affinity (Eb) are often used to forecast the oxidation tendency of an additive on the cathode surface [29,30]. The film-forming electrolyte additives have a lower oxidation potential than the electrolyte solvents. Therefore, the additives dominate the composition of the CEI film on the cathode. Furthermore, a low value of Li + Eb is suitable for the enrichment of the additive on the cathode surface [29]. The calculated Eox values of the solvents and DTMS are listed in Table 1. As per the theoretical calculation results, DTMS (5.58 V) has a lower Eox than EC (7.11 V), DMC (7.00 V), and EMC (6.99 V), indicating that DTMS has a lower oxidation stability than the electrolyte solvents. The optimized structures and solvent-Li + and DTMS-Li + Eb values are seen in Fig. 1. The Eb value of DTMS is lower than those of solvents. This indicates that the interaction between DTMS and Li + is weaker, so DTMS can easily accumulate at the cathode surface. DTMS may thus form a protective film at the positive electrode, as per the theoretical calculation results.

To study the interaction of DTMS with HF/F , the cycling performance of the coin cell in the BE + 1000 ppm HF electrolyte and the 2 vol % DTMS-containing + 1000 ppm HF electrolyte was analyzed (Fig. 6a). A significant capacity loss was observed after 40 cycles in the case of the

BE + 1000 ppm HF electrolyte: from 246.4 to 60.3 mAh g 1 . However, the discharge capacity in the 2 vol% DTMS-containing + 1000 ppm HF electrolyte changed from 240.3 to 190.6 mAh g 1 . DTMS can effectively suppress the damage to the cathode by HF and improve battery cycle stability. Fig. 6b and c shows the F 1s XPS spectra of the LRM electrodes in the BE + 1000 ppm HF and 2 vol% DTMS + 1000 ppm HF electrolyte. The LiF and Me -F peak intensities are much higher in the BE + 1000 ppm HF electrolyte than in the 2 vol% DTMS-containing + 1000 ppm HF electrolyte. The addition of DMTS could inhibit the fluoride formation.

NMR spectra are used for analyzing the composition changes after storage for 24 h in the electrolyte upon the addition of HF [32] (Fig. 7). A pair of peaks at 75 ppm can be found for all electrolytes, which are assigned to PF6 [25]. Unlike the BE (Fig. 7a) and DTMS-containing electrolytes (Fig. 7b), the BE + 1000 ppm HF electrolyte shows three additional peaks (Fig. 7c). The pair of weak peaks at 85 ppm is

Fig. 7. 19 F NMR spectra of BE and DTMS-containing electrolytes before (a and b) and after (c and d) 1000 ppm HF was added.

assigned to PO2F2 formed from the partial hydrolysis of LiPF6 [25], and the peak at 190 ppm corresponds to HF. However, the HF peak intensity for the 2 vol% DTMS-containing + 1000 ppm HF electrolyte is lower than that for the BE + 1000 ppm HF electrolyte (Fig. 7c and d). In addition, a

new small peak at 162 ppm corresponding to the BE + 2 vol%DTMS containing 1000 ppm of the HF electrolyte is shown in Fig. 7d, which can be attributed to the reaction between DTMS and HF/F .Supplementary Material 3 presents the 19 F NMR spectra of the DTMS-containing

electrolytes collected from the LRM/Li cells after evaluating their cycling performance. In them, the peak at 162 ppm corresponding to the products of the reaction of DTMS with HF/F disappears. Combined with the Si 2p XPS results, this demonstrates that DTMS is involved in film formation on the cathode.

The possible DTMS reaction on the interface of the cathode is shown in Fig. 8. We hypothesize that the Si -O bond can react with HF, Lewis acids, and EC [28,33].

In conclusion, the cycling performance of the LRM cathode can be greatly improved by adding DTMS. The capacity loss decreased from 57.2% to 25% at 0.5C after 100 cycles. Experimental results show that DTMS is preferentially oxidized on the LRM cathode surface to form a more stable CEI layer with Si -O and Si -F bonds, and that it also reacts with the HF/F from the electrolyte, meaning that it is effective for suppressing metal dissolution and enhancing the stability between the electrolyte and cathode.