Zhiqun Wanga,
Shaokang Tiana,
Shangda Lia,
Lei Li*ab,
Yimei Yinb and
Zifeng Ma
b
aSchool of Chemistry and Chemical Engineering, Shanghai Key Lab of Electrical Insulation and Thermal Aging, Shanghai Jiao Tong University, Shanghai 200240, China. E-mail: lilei0323@sjtu.edu.cn
bShanghai Electrochemical Energy Devices Research Center, School of Chemistry and Chemical Engineering, Shanghai Jiaotong University, Shanghai 200240, China
First published on 31st May 2018
Lithium sulfonate-grafted poly(vinylidenefluoride-hexafluoro propylene) P(VDF-HFP) ionomers are synthesized through covalent attachment of taurine and used as binder for the LiFePO4 cathode of lithium-ion batteries(LIBs). The incorporation of the ionomer binders will add ionic conducting channels inside the electrodes, and prevent electrolyte depletion during rapid charge–discharge processes. It leads to an improved performance of LIBs using the ionomer binders including cycling stability and rate capability compared to that of LIBs using non-ionic binders (PVDF and PVDF-HFP). Therefore, the lithium sulfonate-grafted P(VDF-HFP) ionomers offer a new route to develop high-power LIBs.
The binders with ionic conductivity are of great interest, due to the fact that they will enhance lithium ion conductivity with high rate capability of LIBs. Recent studies have shown that using lithiated ionomers as binders will provide a path between electroactive materials and electrolytes, which reduces the polarization resistance and increases the ionic conductivity within the electrodes. Li et al. used polyacrylic acid (PAA) as binder to lower the polarization resistance of the LiFePO4 cathode.40 By the addition of lithiated perfluorinated sulfonic ionomers such as Nafion41,42 and Dow43 to the lithium cathode, cycling stability at large current density of LIBs was improved. Shi et al. synthesized lithiated poly(perfluoroalkyl sulfonyl)imde (PFSILi) ionene and blended it with PVDF to make it as ionic binder for the LiFePO4/C cathode.44 The cathode with this ionic composite binder exhibited a higher rate capacity compared with the PVDF non-ionic binder. Chui et al. adopted lithiated perfluorosulfonate ionomer as the binder for LiMn2O4 cathodes.41 Wei et al. prepared the LiFePO4/C cathode with sulfonated polyether ether ketone with pendant lithiated fluorinated sulfonic groups (SPEEK-FSA-Li) as binder to reduce the Li+ concentration polarization and electrolyte depletion during rapid charge–discharge processes.45
Since poly(vinylidenefluoride-hexafluoro propylene) (P(VDF-HFP)) polymer has higher amorphousity and lower glass transition temperature (Tg), Hu et al. reported that the LiFePO4 cathode using P(VDF-HFP) as binder showed better electrochemical performance including cycling stability and rate capability compared to the LiFePO4 cathode using PVDF as binder.46 However, P(VDF-HFP) polymer still does not have intrinsic ionic functionality, so that it is difficult to improve rate performance of LIBs.
Herein, we synthesized lithium sulfonate-grafted P(VDF-HFP) ionomers with different content of Li+ and used them as binders for the LiFePO4 cathode of LIBs. It is well known that amine is easily grafted onto fluoropolymer main chains as functional groups owing to the strong polarity of C–F bonds.47,48 In our experiments, taurine, one of sulfur-containing amino acids, was directly grafted onto P(VDF-HFP) polymer chains through one reaction in solution (see Scheme 1). The content of Li+ of the lithium sulfonate-grafted P(VDF-HFP) ionomers was controlled by various taurine feed contents. Electrochemical performance of these ionomers as binders for the LiFePO4 cathodes of LIBs were investigated by electrochemical impedance spectroscopy, charge–discharge testing, cycling voltammetry and 180° peel testing. The performance of cathodes including cycling stability, rate capability and adhesion strength were improved compared to the cathodes using P(VDF-HFP) as binders. To the best of our knowledge, it is the first time to report that the lithium sulfonate-grafted P(VDF-HFP) ionomers were used as binders in LIBs.
Fig. 1a showed FTIR spectra of the P(VDF-HFP) polymer and the lithium sulfonate-grafted P(VDF-HFP) ionomers. It can be found that there was a new characteristic peak at 1635 cm−1 attributed to the symmetric stretching vibration of CN in the lithium sulfonate-grafted ionomers compared to the P(VDF-HFP) polymer.49 In addition, the characteristic peak at 1040 cm−1 belonging to the symmetric stretching vibration of O
S
O can also be clearly found. It indicated that tauric lithium was successfully grafted onto P(VDF-HFP) polymer chains. The peaks at 1178 and 1400 cm−1 were the stretching vibration of CF2 and CH2, respectively.50 It can be found that the intensity of both peaks of CF2 and CH2 reduced after grafting the tauric lithium.
To further verify this grafted reaction, the lithium sulfonated grafted P(VDF-HFP) was infiltrated in 1 M HCl solution for 72 h to substitute Li+ by H+. Then, the obtained sulfonic acid-grafted P(VDF-HFP) polymer was tested by 1H NMR. Fig. 1b shows 1H NMR spectra of P(VDF-HFP) polymer and the sulfonic acid-grafted P(VDF-HFP) with 15 wt% and 20 wt% contents of taurine. The new resonance at 7.7 ppm can be assigned to hydroxyl proton of sulfonic acid group in the sulfonic acid-grafted P(VDF-HFP) compared to the P(VDF-HFP) polymer.51 It can also be found that the intensity of this new resonance at 7.7 ppm increased with the increasing of the content of taurine. Both FTIR and 1H NMR results demonstrated that tauric lithium was successfully grafted to P(VDF-HFP) polymer in our experiments.
The content of Li+ (i.e. Li+ exchange capacity) in the lithium sulfonate-grafted P(VDF-HFP) ionomers was determined quantitatively by titration method. Table 1 showed the theoretical Li+ exchange capacity and the measured Li+ exchange capacity. Theoretical Li+ exchange capacity of 5 wt%, 15 wt% and 20 wt% taurine feed content were 3.91 × 10−4, 1.17 × 10−3 and 1.56 × 10−3 mmol g−1 respectively. Measured Li+ exchange capacity of 5 wt%, 15 wt% and 20 wt% taurine feed content were 3.81 × 10−4, 1.05 × 10−3 and 1.25 × 10−3 mmol g−1 respectively. The Li+ exchange capacity was increased from 0.38 mmol g−1 from the ionomer with 5 wt% taurine feed content to 1.25 mmol g−1 with 20 wt% taurine feed content.
The electrochemical stability of the lithium-sulfonate-grafted P(VDF-HFP) binders in the LiFePO4 cathode was tested by cyclic voltammetry at room temperature. Fig. 2 showed the CV profiles of the LiFePO4 cathodes with the P(VDF-HFP) and the grafted-P(VDF-HFP) binders at a scanning rate of 0.2 mV s−1 and room temperature. It can be found that there were a pair of oxidation and reduction peaks corresponding to Fe2+/Fe3+ redox couple for all the LiFePO4 electrodes. Compared to the electrode with P(VDF-HFP) binder, all the electrodes with the grafted-P(VDF-HFP) binders showed lower cathodic potential and higher anodic potential. Furthermore, the electrodes with the grafted-P(VDF-HFP) binders showed a smaller voltage difference between the oxidation and reduction peak potential than that of the electrode with P(VDF-HFP) binder. The voltage differences ranked from P(VDF-HFP) (0.95 V) > grafted-P(VDF-HFP)-5 (0.88 V) > grafted-P(VDF-HFP)-15 (0.83 V) > grafted-P(VDF-HFP)-20 (0.65 V), which indicated that the electrodes with the grafted-P(VDF-HFP) binders will reduce the electrochemical polarization. In addition, the electrode with the grafted-P(VDF-HFP)-20 binder had the lowest electrochemical polarization among all the electrodes.
![]() | ||
Fig. 2 Cyclic voltammograms of the LiFePO4 electrodes with P(VDF-HFP) and lithium sulfonate-grafted P(VDF-HFP) binders at the scanning rate of 0.2 mV s−1 and room temperature. |
The cycling performance of the LiFePO4 electrodes with the P(VDF-HFP) and grafted-P(VDF-HFP) binders were shown in Fig. 3. The cells were cycled at 1C under constant current conditions at room temperature. Fig. 3a showed the initial charge and discharge curves of the LiFePO4 electrodes with different binders. It was clear that all the electrodes with the grafted-P(VDF-HFP) binders showed a higher discharge plateau potential and a lower charge plateau potential than that of the electrodes with the P(VDF-HFP) binder. The similar results have also been reported in other ionomers as binders for LIBs.13–15 As shown in Fig. 3b, it was obvious that the cycling performance of the electrodes with the grafted-P(VDF-HFP) binders were better than that of the electrode with the P(VDF-HFP) binder. The capacity fading of the electrode with the P(VDF-HFP) binder was severe, and the discharge specific capacity declined from 145.7 to 99.0 mA h g−1 (capacity retention: 67.9%) after 50th cycles. For the electrodes with the grafted-P(VDF-HFP) binders after 50th cycles, the capacity decreased to 118.9 mA h g−1 (capacity retention: 79.8%), 132.8 mA h g−1 (capacity retention: 89.0%), and 140 mA h g−1 (capacity retention: 92.0%) for grafted-P(VDF-HFP)-5 binder, grafted-P(VDF-HFP)-15 binder and grafted-P(VDF-HFP)-20 binder, respectively.
Fig. 4 showed the rate performance of the electrodes with different binders at various discharge currents ranging from 0.5 C to 4 C. As expected, the discharge specific capacity of all the electrodes gradually decreased with the increasing discharge current density (i.e. rate). Compared to the electrode with P(VDF-HFP) binder, however, all the electrodes with the grafted-P(VDF-HFP) binders showed lower decrease in capacity, especially at higher rates. For example, at 4C as shown in Fig. 4, the discharge specific capacity of the electrodes with the grafted binders (grafted-P(VDF-HFP)-5 binder: 36.5 mA h g−1, grafted-P(VDF-HFP)-15: 78.8 mA h g−1, grafted-P(VDF-HFP)-20: 95.9 mA h g−1) were higher than that of the electrode with P(VDF-HFP) binder (24.7 mA h g−1). It indicated that the electrodes with the grafted-P(VDF-HFP) binders showed higher rate capability than that of electrode with the P(VDF-HFP) binder. For the grafted binders, the rate performance of the electrodes increases with the increasing content of Li+ (i.e. Li+ exchange capacity) of the lithium sulfonate-grafted P(VDF-HFP) ionomers. And the electrode with the grafted-P(VDF-HFP)-20 binder exhibited the best rate performance among all the electrodes. Due to P(VDF-HFP) binder does not have intrinsic ionic functionality, it will lead to salt concentration polarization and/or salt depletion within the electrodes during high rate charging and discharging, and finally result in the poorer rate capability of the electrodes.43 However, the lithium sulfonate-grafted P(VDF-HFP) ionomer can accommodate Li+ ions, the loss of Li+ ions in the electrolyte due to consumption via electrode reactions can be compensated (see Scheme 3).43 Therefore, the rate performance of the electrode with the ionomers will be enhanced.
![]() | ||
Fig. 4 Rate performance of the LiFePO4 electrodes with P(VDF-HFP) and lithium sulfonate-grafted P(VDF-HFP) binders. |
![]() | ||
Scheme 3 Illustration of a possible mechanism for Li-ion transport in the electrodes with P(VDF-HFP) and lithium sulfonate-grafted P(VDF-HFP) binders. |
Fig. 5 showed the cycling performance of the LiFePO4 electrodes at 60 °C and 1C rate. Normally, LiPF6-based electrolytes at higher temperature (>50 °C) will result in thermal decomposition of LiPF6 to form LiF, PF5 and HF, which will lead to not only the decomposition of the electrolyte, but also increase the resistance of LIBs and finally bring about fading of the capacity. In addition, the iron element of LiFePO4 will tend towards dissolving in the electrolyte due to the presence of HF. These issues caused that both LIBs with P(VDF-HFP) and grafted-P(VDF-HFP) binders showed poor thermal stability at 60 °C. However, the electrode with the grafted-P(VDF-HFP)-20 binder exhibited better performance compared to the electrode with P(VDF-HFP) binder (as shown in Fig. 5). In addition, XPS measurements was carried out to verify the stability of grafted-P(VDF-HFP) binders after the cycling test at higher temperature. As shown in Fig. 6a, N1s spectra of the LiFePO4 electrode with lithium sulfonate-grafted P(VDF-HFP) binder before cycling test can be fitted into three peaks at 398.2, 399.5, 401.0 eV, corresponding to pyridine-like, pyrrole-like and quaternary-like nitrogen, respectively.52 Compared to the electrode before cycling test, we can not find any change for the electrode after cycling test (see Fig. 6). These results indicated that the grafted-P(VDF-HFP) binder was stable during the cycling test, even at higher temperature. The reason maybe the CN double bond in the grafted-P(VDF-HFP) polymer was highly reactive towards nucleophiles due to the polarization by both CF3 and CF(CF3) perfluoroalkyl groups.
![]() | ||
Fig. 5 Cycling performance of the LiFePO4 electrodes with P(VDF-HFP) and lithium sulfonate-grafted P(VDF-HFP) binders at 60 °C and 1C rate. |
![]() | ||
Fig. 6 XPS of N1s of the LiFePO4 electrodes with lithium sulfonate-grafted P(VDF-HFP) binder before cycling test (a) and (b) after cycling test (32nd cycle) at 60 °C and 1C rate. |
Fig. 7a showed the EIS curves of the LiFePO4 electrodes with the P(VDF-HFP) and grafted-P(VDF-HFP) binders. An intercept at Zre-axis in the high frequency corresponds to the ohmic resistance (Rs). The semicircle in the middle frequency range was attributed to the charge-transfer reaction resistance (Rct) in the cathode-electrolyte interface. The straight line in the lower frequency region represented the Warburg impedance (W), which was associated with Li+ diffusion in the LiFePO4/C cathode. A simplified equivalent circuit was constructed to analyse the impedance spectra in Fig. 7a by ZSimpWin V 3.1 program. The lithium ion diffusion coefficient (DLi) can be calculated from the formula as following:53
![]() | (1) |
Z′ = Re + Rf + Rct + σwW−1/2 | (2) |
![]() | ||
Fig. 7 (a) EIS results of the LiFePO4 electrodes with P(VDF-HFP) and lithium sulfonate-grafted P(VDF-HFP) binders at frequencies from 0.01 Hz to 100 kHz. (b) Plots of Z′ − ω−1/2 circuit. |
Binders | Rs (Ω) | Rct (Ω) | DLi (cm s−1) |
---|---|---|---|
P(VDF-HFP) | 22.52 | 365.1 | 1.83 × 10−14 |
Grafted-P(VDF-HFP)-5 | 9.663 | 342.0 | 6.05 × 10−14 |
Grafted-P(VDF-HFP)-15 | 10.69 | 329.1 | 8.11 × 10−14 |
Grafted-P(VDF-HFP)-20 | 4.43 | 285.0 | 1.14 × 10−13 |
The adhesion ability between current collector and active material was a vital factor to choose suitable binder. In our experiments, 180 peeling test was employed to evaluate the adhesion strength between the coating electrode and the Al current. Since the electrode with the grafted-P(VDF-HFP)-20 binder showed the best electrochemical performance among all the electrodes based on the above-mentioned results, we further measured the adhesion ability of this electrode. The peeling strength of the electrodes with the P(VDF-HFP), grafted-P(VDF-HFP)-20 binders and PVDF are 0.96 ± 0.01 N, 1.22 ± 0.01 N and 0.4 N ± 0.01 N, respectively. The result indicated that the adhesion of the electrode with the grafted-P(VDF-HFP)-20 binder was stronger than that of the electrode with the P(VDF-HFP) binder and PVDF.
This journal is © The Royal Society of Chemistry 2018 |