- Email: [email protected]
graphite cells
Electrochimica Acta 221 (2016) 107–114
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Lithium difluorophosphate as an additive to improve the low temperature performance of LiNi0.5Co0.2Mn0.3O2/graphite cells Bowen Yanga , Hong Zhanga,* , Le Yub , WeiZhen Fanb , Donghai Huangb a
School of Mechanical and Electrical Engineering, Shannxi University of Science & Technology, Xi’an 710021, PR China Guangzhou Tinci Materials Technology Co. Ltd.,Guangzhou Key Laboratory of New Functional Materials for Power Lithium-ion Battery, Guangzhou 510760, PR China b
A R T I C L E I N F O
Article history: Received 30 July 2016 Received in revised form 2 October 2016 Accepted 6 October 2016 Available online 7 October 2016 Keywords: Lithium difluorophosphate electrolyte additive low temperature LiNi0.5Co0.2Mn0.3O2/graphite cells
A B S T R A C T
Lithium difluorophosphate (LiPO2F2) was used as an electrolyte additive to promote the low temperature performance of LiNi0.5Co0.2Mn0.3O2/graphite cells. The impact of LiPO2F2 on the solid electrolyte interface (SEI) film-formation on electrodes was demonstrated by various electrochemical methods and microscopy techniques, such as transmission electron microscopy (TEM), scanning electron microscope (SEM) as well as X-ray photoelectron spectroscopy (XPS), in the pouch cells and half cells. The results showed that the cells containing 1% LiPO2F2 performed 71.9% (20 C) and 57.93% (30 C) of initial capacity, while the cells without LiPO2F2 discharged only 49.41% and 9.6% of initial capacity under the same condition. In addition, the enhancement of cyclic performance at 0 C was attributed to a conductive and stable SEI film formed on the graphite by the sacrifice of LiPO2F2, which led to a low impedance and richer content of LiF and Li2CO3 in SEI components, as depicted in XPS. ã 2016 Elsevier Ltd. All rights reserved.
1. Introduction Owing to the characteristics of high energy density, memoryless effects and being environment friendly, lithium ion batteries (LIBs) have been widely used as an essential device for digital communication, electric vehicles and so on [1–3]. Unfortunately, LIBs are often subjected to great capacity loss at low temperature, which restricts their application in cold climates [1]. Although the substantial failure mechanism is still in dispute, the main reasons which lead to the poor low temperature performance have been attributed to the following points; (i) increased viscosity and poor Li+ conductivity, (ii) limitation of the SEI film-formation on graphite and the sluggish lithium kinetics over the SEI, (iii) the reduced diffusion of lithium ion in graphite, (iv) high polarization giving a rise to charge-transfer resistance on electrolyte-electrode interface [4,5]. In previous researches [6,7], many attempts were made to overcome the above barriers via optimizing solvent system or film-formation additives. Due to their high ionic conductivity and low viscosity, esters like ethyl propionate (EP), thyl acetate (EA) and methyl acetate (MA), are used as the cosolvent to enhance the cells performance at low temperature. However, the presence of linear carboxylic ester would affect long
* Corresponding author. Tel/fax: +86-29-86132696. E-mail address: [email protected] (H. Zhang). http://dx.doi.org/10.1016/j.electacta.2016.10.037 0013-4686/ã 2016 Elsevier Ltd. All rights reserved.
life cycle performance at an elevated temperature. From a comprehensive perspective, it is not a great option to use ester as the co-solvent. Due to its low melting point (48.8 C), propylene carbonate (PC) could expand the liquid temperature range of electrolyte [8]. But it is well known that PC, when used as a major solvent, will combine with lithium-ion co-inserting into graphite, which causes severe graphite exfoliation [9]. Whereas an appropriate concentration of PC in the electrolyte will have the slight negative influence on graphite, and more importantly, prevent ethylene carbonate (EC) from crystallizing out at low temperature. Changing electrolyte system is a feasible method to solve the problem of high viscosity and poor conductivity while it could not improve the sluggish diffusivity of lithium ion within the graphite. Adopting film-formation additive is beneficial to the intercalation of Li+ into graphite. A majority of low temperature additives are fluorinated compound, since the reduction of fluorochemicals is prior to that of EC and modifies the surface morphology and SEI composition. Fluoroethylene carbonate (FEC) is most commonly used additive for low temperature [6,8]. In addition, some lithium salts like lithium tetrafluoroborate (LiBF4) [10] and lithium bis(oxalato)borate (LiBOB) [11] also could promote the low temperature performance when used as the main salt or additive. In this paper, we reported the lithium difluorophosphate (LiPO2F2) was used as an electrolyte additive to improve the low
108
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
temperature performance of LIBs. According to the work by Ko-Eun Kim et al. [12], the high rate performance was strongly enhanced by the combination of LiPO2F2 and vinylene carbonate (VC). This work also revealed that LiPO2F2 could modify the SEI film formation on graphite, which led to a high ionic conductivity and stable SEI layer. Owing to the characteristics of LiPO2F2, we expect that LiPO2F2 can improve the low temperature performance of LiNi0.5Co0.2Mn0.3O2/graphite cells. Therefore, various electrochemical methods and spectroscopic techniques were employed to demonstrate the effect of LiPO2F2 on low temperature performance and illustrate the reaction mechanism of LiPO2F2 at both electrodes. 2. Experiment 2.1. Sample preparation The following chemicals are all purchased from Guangzhou Tinci Materials Technology Co., Ltd., China. A low-temperature type of electrolyte which was made by dissolving a 1 M LiPF6 in ethylene carbonate (EC), ethylmethyl carbonate (EMC) and propylene carbonate (PC) with a 4:7:1 weight ratio was regarded as a reference electrolyte. Adding 1 wt.% LiPO2F2 (LG Co. Ltd, Korean) to the reference electrolyte was supposed to be the researched one. A cathode slury was fabricated by mixing 94.5 wt.%LiNi0.5Co0.2Mn0.3O2 (Shenzhen Tianjiao Tech. Co,. Ltd.), 1.5 wt.% polyvinylidene fluoride (PVDF) and 4.0 wt.% conductive carbon (Super-P). The anode electrode contained graphite (Shenzhen BTR Co,. Ltd.), Carboxymethyl Cellulose (CMC), Super-P and Polymerized Styrene Butadiene Rubber (SBR) with a 95.3:1.4:1.5:1.8 weight ratio. Both electrodes were dried at 120 C for 10 hours under vacuum conditions before the cells assembled. The LiNi0.5Co0.2Mn0.3O2/ graphite cells were used to understand the impact of LiPO2F2 on low temperature performance. Moreover, in order to know the effect of LiPO2F2 on cathode and anode electrode, CR-2016 button cells made of LiNi0.5Co0.2Mn0.3O2/Li and graphite/Li, respectively, were used. 2.2. Electrochemical Measurements Cyclic voltammograms (CV) curves were conducted on graphite/Li button cells with the scan rate of 0.05 mV s1 over a range of 0.01–3 V. The cells were measured by an electrochemical work station (CHI660, Chenhua, China). The charge-discharge cyclic performance at 0 C and 20 C was determined with LiNi0.5Co0.2Mn0.3O2/graphite cells through a battery test instrument (CT3008W, Neware, China). A temperature controlling box (GDW100L, SUOYATE, China) was used to maintain the environment temperature at diverse low temperature. Before the normal battery tests, the LiNi0.5Co0.2Mn0.3O2/graphite cells were charged to 3.85 V
with a constant current (0.1 C) and then cycled 3 times with different rates of 0.1 C, 0.2 C and 0.5 C. After that, a lowtemperature cycling test was followed by 100 cycles at 0.5 C rate from 2.75 V to 4.20 V. For the sake of the low temperature discharge ability, cells were cycled 3 times with a rate of 1 C at room-temperature before putting them in temperature controlling box with different temperatures of 0 C,10 C,20 C and 30 C for 4 hours. And then the cells were discharged at 0.5 C rate to 2.75 V. Indeed, DC resistance (DCR) was also examined at the above temperatures. Before DCR measurement, pouch cells were charged to 100% SOC and then put them into the temperature controlling box at different temperatures for 4 hours and the last second voltage was recorded as U1. The cells were discharged at a specific rate of 0.5 C for 10 seconds using the final second potential as U2. Afterward, the DCR value was calculated by using the formula: DCR = (U2 U1)/I. Electrochemical impedance spectroscopy (EIS) measurements were conducted with frequency response analyzer (FRA, Solartron 1455A, Solartron, England), CHI660 for LiNi0.5Co0.2Mn0.3O2/graphite cells and half cells. For full cells, the EIS of 100th cycles at 0 C was measured with 100% SOC (approximately 4.2 V) over a frequency range of 100 kHz to 20 mHz at an amplitude of 10 mV. For half cells, in order to form a stable SEI film and investigate the impact of LiPO2F2 on electrodes interface resistance, 3 cycles at a rate of 0.05 C at room temperature need to be taken prior to EIS measurement at 0 C. The EIS test of LiNi0.5Co0.2Mn0.3O2/Li and graphite/Li half cells were performed at 4.2 V and 0.01 V, respectively. 2.3. Physical Characterization To further figure out the characteristic of electrode surface, the cycled cells were disassembled in an argon-filed glovebox and the collected electrodes were rinsed with pure DMC for 3 times to remove the residual lithium salt and the electrolyte on electrodes surface. Then the samples were dried in glovebox for 12 hours and taken into an argon-filed bottle before the analysis. The surface morphology of the electrodes was observed by scanning electron microscopy (SEM, ZEISS Sigma500, Carl Zeiss, Germany) and transmission electron microscopy (TEM, JEM-2100HR, JEOL, Japan). The surface element composition of the electrodes were analyzed by X-ray photoelectron spectroscopy (XPS, AZIS SUPRA, Kratos, British) using the Al Ka line as the X-ray source. 3. Results 3.1. Electrochemical behavior of the electrolytes on graphite anode In order to understand the impact of LiPO2F2 on graphite, Li/ graphite half cell was used to investigate the electrochemical behavior of the electrolyte which contains 1% LiPO2F2. As shown in
Fig. 1. Cyclic voltammograms of graphite/Li half cells in 1 M LiPF6 dissolved in EC: EMC: PC (4:7:1, wt %) with different contents of LiPO2F2: (a) without LiPO2F2; (b) with 1% LiPO2F2 with a scan rate of 0.05 mV s1.
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
109
Fig. 2. Discharge curves of LiNi0.5Co0.2Mn0.3O2/graphite cells with a rate of 0.5C discharged to 2.75 V at various temperatures: (a) without LiPO2F2; (b) with 1% LiPO2F2; (c) capacity retention of the cells discharged at various low temperatures.
Fig. 1a, the poor repeatability of the three CV curves suggested that side reactions had taken place without LiPO2F2 during the chargedischarge progress and thus led to the increase of irreversible capacity. However, the CV curve (Fig. 1b) of the cell containing 1% LiPO2F2 presented that the second cycle was perfectly coincided with the third cycle, indicating the addition of LiPO2F2 stabilized the SEI film. 3.2. Electrochemical properties of LiPO2F2 in pouch cells Fig. 2 compared the discharge performance at various low temperatures of the cells with and without LiPO2F2. In addition, the specific value of capacity retention was shown in Fig. 2c. The cells without LiPO2F2 presented merely 49.4% (20 C) of capacity retention comparing with at 25 C and delivered little discharge capacity when temperature declined further, as depicted in Fig. 2a. However, with additional 1% LiPO2F2, the capacity retention at 20 C and 30 C were boosted to 71.9% and 57.9%, respectively (Fig. 2b). Moreover, the operating voltage showed the same trend
with the capacity retention as the temperature gradually went down. Owing to the rapidly increased resistance and low electrochemical reaction of cells, high polarization was the key reason to the operating voltage loss [13]. Consequently, the cells without LiPO2F2 showed poor capacity retention at low temperatures. This indicated the addition of LiPO2F2 could decrease the polarization and discharge ability was enhanced at subzero temperatures. As shown in Fig. 3, the impact of LiPO2F2 on the low temperature cyclic performance of LiNi0.5Co0.2Mn0.3O2/graphite cells was examined at 0 C and 20 C. The challenge of cycling at low temperature was well recognized with the slow lithium kinetics on graphite, which probably led to the loss of active lithium and rapid charge capacity fade. As depicted in Fig. 3b, d, the coulombic efficiency of the cells cycling at 0 C and 20 Chad strongly supported the above view of the charge capacity fade leading to the poor low temperature performance of LIBs. As a result, the poor capacity retention of the pouch cells without LiPO2F2 at 0 C had been observed in Fig. 3a. Owing to high
Fig. 3. Cyclic performance and coulombic efficiency of LiNi0.5Co0.2Mn0.3O2/graphite cells with a rate of 0.5 C cycled at a potential range of 2.75–4.20 V at various low temperatures: (a) cyclic performance at 0 C; (b) coulombic efficiency at 0 C; (c) cyclic performance at 20 C; (d) coulombic efficiency at 20 C.
110
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
stability, and the method below was employed to explain the difference of capacity retention at low temperature. 3.3. Electrochemical impedance
Fig. 4. The DCR values ofLiNi0.5Co0.2Mn0.3O2/graphite cells with different concentrations LiPO2F2 at various low temperatures.
Fig. 5. Electrochemical impedance spectra of the LiNi0.5Mn0.2Co0.3O2/graphite cells with and without LiPO2F2 charged to 4.2 V at 1st cycle and 100th cycle at 0 C.
polarization, cells without LiPO2F2 displayed lower initial capacity (123.5 mAh g1) than that (132.1 mAh g1) of the cells containing 1% LiPO2F2. In addition, cells without LiPO2F2 experienced a fast capacity retention fading of 20.1% after 100 cycles at 0 C meanwhile the cells with 1% LiPO2F2 maintained a high retention of 96.7% at the same cycle number. Additionally, a sharp contrast was depicted in Fig. 3c when LiNi0.5Co0.2Mn0.3O2/graphite cells were cycled at 20 C. The cells with LiPO2F2 delivered 91.7 mAh g1 after 100 cycles and reached high capacity retention of 91%. However, the cells without LiPO2F2 experienced a discharge capacity drop from 60.1 to 9.6 mAh g1. The results suggested that a little amount of LiPO2F2 significantly improved the cycling
A new measurement of DCR was introduced to illustrate the different performance between the cells with and without LiPO2F2 at low temperature. As we know, DCR was used to characterize the discharge ability at a specific condition. As shown in Fig. 4, a growing gap of DCR values was observed between the cells with or without LiPO2F2. In particular, the DCR value of cells containing LiPO2F2 was 0.882 V while that of cells without LiPO2F2 sharply increased to 2.192 V at 20 C. This supported the hypothesis that the decrease of DCR value was considered to be the sacrifice of LiPO2F2, which contributed to better lithium kinetics on electrodes surface. To investigate the influence of LiPO2F2 towards interface impedance, electrochemical impedance spectroscopy (EIS) was employed to measure the impedance in pouch cells and half cells, respectively. As other literatures reported [14], a typical EIS curve composed of three parts: (i) high-frequency semicircle, (ii) middlefrequency semicircle, (iii) low-frequency straight line. The above three regions represented SEI resistance (RSEI), charge-transfer resistance (RCT) and Warburg resistance, respectively. Fig. 5 showed the EIS of LiNi0.5Co0.2Mn0.3O2/graphite cells with and without LiPO2F2 at 1st cycle and 100th cycle respectively, at 0 C. The addition of LiPO2F2 could significantly suppress the increased SEI and charge transfer resistance. By the contrast of EIS of 1st and 100th, a slight decrease of SEI resistance could be seen from the cell containing 1% LiPO2F2, which meant an ionic conductivity film was formed on electrode interface. Meanwhile it also restricted the sharp increase of charge transfer resistance compared to the EIS results of cell without LiPO2F2. In summary, the high conductive SEI film formed by LiPO2F2 facilitated the Li+ migration through electrodes interface and weakened the impact of the rapid increased Rct which was due to the non-conductive substance deposited on electrodes surface. Moreover, the half cells could simulate the state of intercalation/deintercalation lithium of electrodes and investigate the real reason for such a low resistance. So the EIS of graphite/Li and LiNi0.5Co0.2Mn0.3O2/Li half cells was also conducted at 0 C. Fig. 6a showed slight difference between the LiNi0.5Co0.2Mn0.3O2/Li cells with and without LiPO2F2. However, the EIS result of graphite/Li cells containing LiPO2F2 exhibited an obvious decrease of impedance value in both high and middle frequency regions (Fig. 6b) than that of cells without LiPO2F2, which suggested that the electrolyte containing 1% LiPO2F2 modified the SEI film on graphite and improved Li+ migration
Fig. 6. Electrochemical impedance spectra of the half cells at 0 C: (a) LiNi0.5Co0.2Mn0.3O2/Li cell at 4.2 V; (b) graphite/Li cell at 0.01 V. The half cells were cycled 3 times at room temperature before EIS measurement.
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
and diffusion though the surface. Consequently, the low impedance of pouch cells was attributed to the influence of LiPO2F2 on graphite, making a better lithium kinetics film at low temperature, and thus improved the low temperature performance of the LiNi0.5Co0.2Mn0.3O2/graphite cells. Meanwhile the result also strongly supported the view of the poor lithium kinetics on graphite restricting the application of LIBs in cold environment. 3.4. The effect of LiPO2F2 on electrodes surface morphology For the sake of the effect of LiPO2F2 on electrodes, transmission electron microscope (TEM) was used before and after 100 cycles at 0 C, and the results were shown in Fig. 7. Compared to the fresh
111
cathode, a smooth surface and a clear outline were observed in the cycled cathode of the cells with or without LiPO2F2 (Figs. 7a–c). Consistent with the EIS result of LiNi0.5Co0.2Mn0.3O2/Li cell, the TEM photos of cycled cathode also suggested that LiPO2F2 had little impact on cathode. However, a huge difference could be found on graphite anode, as depicted in Fig. 7d–f. Fig. 7e displayed the TEM image of cycled anode without LiPO2F2, whose graphite surface had been destroyed into numerous irregular particles and SEI almost vanished without trace relating to the rapid capacity fading of cyclic performance at 0 C. On the contrary, Fig. 7f showed a smooth and thin layer film on graphite surface which indicated the anode with LiPO2F2 formed a conductive film, facilitating the intercalation of Li+. This suggested that a high ionic conductive film
Fig. 7. TEM images taken from the surface of electrode: (a) a fresh cathode; (b) a cycled cathode without LiPO2F2; (c) a cycled cathode with LiPO2F2; (d) a fresh anode; (e) a cycled anode without LiPO2F2; (f) a cycled anode with LiPO2F2. The cycled electrodes were taken from pouch cells after 100 cycles at 0 C.
112
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
ensured the normal intercalation and deintercalation of Li+ through the graphite, which contributed to a better cyclic performance at low temperature. The morphology of the electrodes surface before and after 100cycles at 0 C was also observed with scanning electron microscopy (SEM) and the results were displayed in Fig. 8. Both the fresh and cycled cathode showed the similar morphology, which proved the view that LiPO2F2 exerted no side-effect on
cathode, once again (Fig. 8a–c). The graphs of graphite surface morphology were presented in Fig. 8d–f. As shown in Fig. 8e, some dendritic structure substance had been found on the surface of the cycled graphite without LiPO2F2. It was likely to be the lithium oxidations originated from the lithium dendrites or lithium precipitation exposed to the air before the SEM analysis. And this inactive lithium was attributed to the increased irreversible capacity. However, compared to the fresh graphite (Fig. 8d), a
Fig. 8. SEM images taken from the surface of electrode: (a) a fresh cathode; (b) a cycled cathode withoutLiPO2F2; (c) a cycled cathode with LiPO2F2; (d) a fresh anode; (e) a cycled anode without LiPO2F2; (f) a cycled anode with LiPO2F2. The cycled electrodes were taken from pouch cells after 100 cycles at 0 C.
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
thin SEI layer was observed in the surface of cycled graphite containing LiPO2F2 (Fig. 8e). The graphite was covered by a dense layer and some big particles adhered to the active materials. From the SEM results, the Li-ion deposition on graphite surface was effectively reduced through a layer film formed by LiPO2F2, and thus expanded the operating temperature range of LiNi0.5Co0.2Mn0.3O2/graphite cells. 3.5. X-ray photoelectron spectroscopy (XPS) analyses of graphite anode To further reveal the effect of LiPO2F2 on graphite, X-ray photoelectron spectroscopy (XPS) was used to analyze the chemical composition of graphite surface. The analysis and different elemental concentrations of the graphite surface were shown in Fig. 9 and Table 1. Compared with fresh graphite, an obvious decrease of C concentration could be found while the concentration of O, F and P were increased. It was assumed that the graphite surface was covered by the electrolyte decomposition products. Particularly, the concentration of F and P were higher on the graphite cycled with 1% LiPO2F2 than those on the graphite without LiPO2F2, hence, it could be concluded that the decomposition of LiPO2F2 played an important role in the SEI film formation. The C 1s spectrum of the fresh graphite consisted of three peaks. The first one at 284.3 eV was associated with graphite [15]. The next one at 285.0 eV was attributed to SBR binder [16]. The C-O
113
Table 1 Different element concentrations on fresh graphite and graphite form cycled cells with and without 1% LiPO2F2 at 0 C. Sample
C1s (%)
O1s (%)
F1s (%)
P2p (%)
Fresh Without LiPO2F2 With 1% LiPO2F2
87.14 41.61 43.95
12.86 50.51 43.06
4.53 8.81
3.35 4.18
bond appeared at 286.2 eV was due to the CMC binder. After 100 cycles at 0 C, a new characteristic peak of C¼O, at 289.3 eV, was observed in C1s spectra for graphite both cycled with and without LiPO2F2. C¼O bonds were probably the electrolyte decomposition products of lithium alky carbonates and polycarbonates. In addition, given the disappeared peak of SBR binder we assumed that the graphite had been covered by a SEI film or some organic compounds attached on the electrode surface. The O 1s spectrum of the fresh graphite showed tow main peaks which belonged to the oxygen atoms of the CMC binder. For the graphite cycled without LiPO2F2, there were also two main peaks, the weaken intensity of C O (533.8 eV) and especially high intensity peak of C¼O (532.0 eV) could be found [17]. In other word, the chief components of the cycled graphite without LiPO2F2 were the lithium alky carbonates or polycarbonates due to the decomposition of the solvents and salts. What’s more, a new peak at 531.5 eV, the characteristic peak of Li2CO3, could be found in the cycled graphite containing LiPO2F2 and it had not been observed in
Fig. 9. C 1s, O 1s, F 1s and P 2p XPS spectra of fresh graphite (top) and cycled graphite without LiPO2F2 (middle) and cycled graphite with 1% LiPO2F2 (bottom) at 0 C after 100 cycles.
114
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
the graphite without LiPO2F2, which suggested that the cycled graphite containing LiPO2F2 comprised of more inorganic components. The presence of Li2CO3 was closely related to a better battery performance owing to its less soluble property in the solvents, and thus effectively passivated electrode preventing electrolyte from further decomposing [18]. More importantly, the SEI film containing Li2CO3 had a better Li-ion conductivity making the Li+ intercalation much easier [19]. It was easy to see the difference of cycled graphite with and without LiPO2F2 in the F 1s spectrum. Both cycled anode consisted of three peaks. The peaks, at 684.3 eV and 685.9 eV, corresponded to the characteristics of LiF and LixPOyFz respectively .The third peak at 686.9 eV was attributed to LiPxFy, which was likely to be the residual LiPF6 on graphite although the electrodes was wished with DMC before XPS analysis [17]. The concentration of LiF in cycled anode with LiPO2F2 was much higher than that without LiPO2F2, which suggested that the proportion of LiF in SEI film were boosted on account of the decomposition of LiPO2F2 at low temperature cycling. Unlike the room-temperature condition, an increased content of LiF, in a reasonable range, not only reduced the SEI film resistance, but also improved its low temperature performance [6]. Comparing to the P 2p spectrum, the peak at 131.0 eV, characteristic of LixPOyFz, had the same trend of that with F 1s spectrum. The decomposition of LiPO2F2, especially P O bounds, made the stronger intensity of LixPOyFz [20]. These results all suggested that a conductive, protective SEI film formed on graphite anode and the improvement of low temperature performance was attributed to the addition of LiPO2F2. In addition, the XPS data of the work made by Ko-Eun Kim et al. showed a different result from this paper [12]. We found a lower Li2CO3 content than that of the reference. It came to conclusion diametrically opposed to our work. The main reasons affecting the content of Li2CO3 in SEI film were attributed to the graphite type, solvent system and hermetic-seals [21]. And the different solvent system makes greatest impact on the Li2CO3 content. From the previous work [22], an unpaired nucleophilic anion (CO32 or LiCO3) interacts with Li+ to form Li2CO3 in SEI film at low PC concentration. The solvent system using PC as co-solvent, other than DMC in Ko-Eun Kim et al. work, is the key factors of high Li2CO3 content. More importantly, the extra Li+ provided by LiPO2F2 also makes contribution to the high content of Li2CO3. Secondly, higher P O phosphate compounds in SEI film was consistent with the conclusion in this work which the peak of LixPOyFz contained P O phosphate compounds and showed high peak intensity than that without additive. 4. Conclusions LiPO2F2 was investigated as an electrolyte additive in LiNi0.5Co0.2Mn0.3O2/graphite cells. The CV curves suggested that LiPO2F2 could stabilize the SEI film formed on graphite anode, and significantly decreased the charge-transfer resistance as depicted in EIS data. In addition, the XPS results had demonstrated that such a low impedance was attributed to the richer content of LiF and the presence of Li2CO3 in SEI film. Moreover, TEM and SEM also intuitively demonstrated that LiPO2F2 had a tremendous impact on graphite while it exerted no side effect on cathode. Therefore, the addition of 1 wt% LiPO2F2 could significantly enhance the discharge characteristics of full cells at various low temperatures and the capacity retention had been increased from 79.9% to 96.7% at 0.5C rate after 100 cycles at 0 C. These results all powerfully supported that the addition of LiPO2F2 improved the stability and conductivity
of the SEI film on graphite, which is helpful for the intercalation of Li+ into graphite and the better lithium kinetics on graphite surface, leading to an improved low temperature and long cyclic performance in LiNi0.5Co0.2Mn0.3O2/graphite cells. Acknowledgments This work was financially supported by the science and technology projects of Guangdong Province (No.2014B010123002) and the science and technology projects of Guangzhou (2014Y2-00012). References [1] C.Y. Wang, G.S. Zhang, S.H. Ge, Terrence Xu, Y. Ji, X.G. Yang, Y.J. Leng, Lithiumion battery structure that self-heats at low temperatures, Nature 528 (2016) 515. [2] Y. Ji, Y.C. Zhang, C.Y. Wang, Li-Ion Cell Operation at Low Temperatures, J. Electrochem. Soc. 160 (2013) A636. [3] Y. Ji, C.Y. Wang, Heating strategies for Li-ion batteries operated from subzero temperatures, Electrochim. Acta 107 (2013) 664. [4] S.S. Zhang, K. Xu, T.R. Jow, Low temperature performance of graphite electrode in Li-ion cells, Electrochim. Acta 48 (2002) 241. [5] S.S. Zhang, K. Xu, T.R. Jow, The low temperature performance of Li-ion batteries, J. Power Sources 115 (2003) 137. [6] B.X. Liu, B. Li, S.Y. Guan, Effect of Fluoroethylene Carbonate Additive on Low Temperature Performance of Li-Ion Batteries, Electrochem. Solid-State Lett. 15 (2012) A77. [7] M.C. Smart, B.V. Ratnakumar, K.B. Chin, L.D. Whitcanack, Lithium-Ion Electrolytes Containing Ester Cosolvents for Improved Low Temperature Performance, J. Electrochem. Soc. 15 (2010) A1361. [8] L.X. Liao, P.J. Zuo, Y.L. Ma, Y.X. An, G.P. Yin, Y.Z. Gao, Effects of fluoroethylene carbonate on low temperature performance of mesocarbon microbeads anode, Electrochim. Acta 74 (2012) 260. [9] M.Y. Nie, Daniel P. Abraham, Daniel M. Seo, Y.J. Chen, Arijit Bose, Brett L. Lucht, Role of Solution Structure in Solid Electrolyte Interphase Formation on Graphite with LiPF6 in Propylene Carbonate, J. Phys. Chem. 117 (2013) 25381. [10] S.S. Zhang, K. Xu, T.R. Jow, Low-temperature performance of Li-ion cells with a LiBF4-based electrolyte, J. Solid State Electrochem. 7 (2003) 147. [11] Marshall C. Smart, Brett L. Lucht, Swapnil Dalavi, Frederick C. Krause, Bugga V. Ratnakumar, The Effect of Additives upon the Performance of MCMB/LiNixCo1xO2 Li-Ion Cells Containing Methyl Butyrate-Based Wide Operating Temperature Range Electrolytes, J. Electrochem. Soc. 159 (2012) A739. [12] K.E. Kim, J.Y. Jang, Inbok Park, M.H. Woo, M.H. Jeong, W.C. Shin, M. Ue, N.S. Choi, A combination of lithium difluorophosphate and vinylene carbonate as reducible additives to improve cycling performance of graphite electrodes at high rates, Electrochim. Acta 61 (2015) 121. [13] S.S. Zhang, K. Xu, T.R. Jow, A new approach toward improved low temperature performance of Li-ion battery, Electrochem. Commun. 4 (2002) 928. [14] J.G. Zhu, Z.C. Sun, X.Z. Wei, H.F. Dai, Studies on the medium-frequency impedance arc for Lithium-ion batteries considering various alternating current amplitudes, J. Appl. Electrochem. 46 (2016) 157. [15] R. Dedryvère, S. Leroy, H. Martinez, F. Blanchard, D. Lemordant, D. Gonbeau, XPS Valence Characterization of Lithium Salts as a Tool to Study Electrode/ Electrolyte Interfaces of Li-Ion Batteries, J. Phys. Chem. B 110 (2006) 12986. [16] L. El Ouatani, R. Dedryvère, C. Siret, P. Biensan, D. Gonbeau, Effect of Vinylene Carbonate Additive in Li-Ion Batteries: Comparison of LiCoO2/C, LiFePO4/C, and LiCoO2/Li4Ti5O12 Systems, J. Electrochem. Soc. 156 (2009) A468. [17] M.Q. Xu, L.S. Hao, Y.L. Liu, W.S. Li, L.D. Xing, B. Li, Experimental and Theoretical Investigations of Dimethylacetamide (DMAc) as Electrolyte Stabilizing Additive for Lithium Ion Batteries, J. Phys. Chem. 115 (2011) 6085. [18] Pallavi Verma, Pascal Maire, Petr Novák, A review of the features and analyses of the solid electrolyte interphase in Li-ion batteries, Electrochim. Acta 55 (2010) 6332. [19] Y.-K. Choi, K.-i. Chung, W.-S. Kim, Y.-E. Sung, S.-M. Park, Suppressive effect of Li2CO3 on initial irreversibility at carbon anode in Li-ion batteries, J. Power Sources 104 (2002) 13. [20] G.C. Yan, X.H. Li, Z.X. Wang, H.J. Guo, C. Wang, Tris (trimethylsilyl) phosphate: A film-formation additive for high voltage cathode material in lithium-ion batteries, J. Power Sources 248 (2014) 1306. [21] Kristina Edstrom, Marie Herstedt, Daniel P. Abraham, A new look at the solid electrolyte interphase on graphite anodes in Li-ion batteries, J. Power Sources 153 (2006) 380. [22] Doron Aurbach, Mikhail D. Levi, Elena Levi, Alexander Schechter, Failure and Stabilization Mechanisms of Graphite Electrodes, J. Phys. Chem. B 101 (1997) 2195.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Lithium difluorophosphate as an additive to improve the low temperature performance of LiNi0.5Co0.2Mn0.3O2/graphite cells Bowen Yanga , Hong Zhanga,* , Le Yub , WeiZhen Fanb , Donghai Huangb a
School of Mechanical and Electrical Engineering, Shannxi University of Science & Technology, Xi’an 710021, PR China Guangzhou Tinci Materials Technology Co. Ltd.,Guangzhou Key Laboratory of New Functional Materials for Power Lithium-ion Battery, Guangzhou 510760, PR China b
A R T I C L E I N F O
Article history: Received 30 July 2016 Received in revised form 2 October 2016 Accepted 6 October 2016 Available online 7 October 2016 Keywords: Lithium difluorophosphate electrolyte additive low temperature LiNi0.5Co0.2Mn0.3O2/graphite cells
A B S T R A C T
Lithium difluorophosphate (LiPO2F2) was used as an electrolyte additive to promote the low temperature performance of LiNi0.5Co0.2Mn0.3O2/graphite cells. The impact of LiPO2F2 on the solid electrolyte interface (SEI) film-formation on electrodes was demonstrated by various electrochemical methods and microscopy techniques, such as transmission electron microscopy (TEM), scanning electron microscope (SEM) as well as X-ray photoelectron spectroscopy (XPS), in the pouch cells and half cells. The results showed that the cells containing 1% LiPO2F2 performed 71.9% (20 C) and 57.93% (30 C) of initial capacity, while the cells without LiPO2F2 discharged only 49.41% and 9.6% of initial capacity under the same condition. In addition, the enhancement of cyclic performance at 0 C was attributed to a conductive and stable SEI film formed on the graphite by the sacrifice of LiPO2F2, which led to a low impedance and richer content of LiF and Li2CO3 in SEI components, as depicted in XPS. ã 2016 Elsevier Ltd. All rights reserved.
1. Introduction Owing to the characteristics of high energy density, memoryless effects and being environment friendly, lithium ion batteries (LIBs) have been widely used as an essential device for digital communication, electric vehicles and so on [1–3]. Unfortunately, LIBs are often subjected to great capacity loss at low temperature, which restricts their application in cold climates [1]. Although the substantial failure mechanism is still in dispute, the main reasons which lead to the poor low temperature performance have been attributed to the following points; (i) increased viscosity and poor Li+ conductivity, (ii) limitation of the SEI film-formation on graphite and the sluggish lithium kinetics over the SEI, (iii) the reduced diffusion of lithium ion in graphite, (iv) high polarization giving a rise to charge-transfer resistance on electrolyte-electrode interface [4,5]. In previous researches [6,7], many attempts were made to overcome the above barriers via optimizing solvent system or film-formation additives. Due to their high ionic conductivity and low viscosity, esters like ethyl propionate (EP), thyl acetate (EA) and methyl acetate (MA), are used as the cosolvent to enhance the cells performance at low temperature. However, the presence of linear carboxylic ester would affect long
* Corresponding author. Tel/fax: +86-29-86132696. E-mail address: [email protected] (H. Zhang). http://dx.doi.org/10.1016/j.electacta.2016.10.037 0013-4686/ã 2016 Elsevier Ltd. All rights reserved.
life cycle performance at an elevated temperature. From a comprehensive perspective, it is not a great option to use ester as the co-solvent. Due to its low melting point (48.8 C), propylene carbonate (PC) could expand the liquid temperature range of electrolyte [8]. But it is well known that PC, when used as a major solvent, will combine with lithium-ion co-inserting into graphite, which causes severe graphite exfoliation [9]. Whereas an appropriate concentration of PC in the electrolyte will have the slight negative influence on graphite, and more importantly, prevent ethylene carbonate (EC) from crystallizing out at low temperature. Changing electrolyte system is a feasible method to solve the problem of high viscosity and poor conductivity while it could not improve the sluggish diffusivity of lithium ion within the graphite. Adopting film-formation additive is beneficial to the intercalation of Li+ into graphite. A majority of low temperature additives are fluorinated compound, since the reduction of fluorochemicals is prior to that of EC and modifies the surface morphology and SEI composition. Fluoroethylene carbonate (FEC) is most commonly used additive for low temperature [6,8]. In addition, some lithium salts like lithium tetrafluoroborate (LiBF4) [10] and lithium bis(oxalato)borate (LiBOB) [11] also could promote the low temperature performance when used as the main salt or additive. In this paper, we reported the lithium difluorophosphate (LiPO2F2) was used as an electrolyte additive to improve the low
108
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
temperature performance of LIBs. According to the work by Ko-Eun Kim et al. [12], the high rate performance was strongly enhanced by the combination of LiPO2F2 and vinylene carbonate (VC). This work also revealed that LiPO2F2 could modify the SEI film formation on graphite, which led to a high ionic conductivity and stable SEI layer. Owing to the characteristics of LiPO2F2, we expect that LiPO2F2 can improve the low temperature performance of LiNi0.5Co0.2Mn0.3O2/graphite cells. Therefore, various electrochemical methods and spectroscopic techniques were employed to demonstrate the effect of LiPO2F2 on low temperature performance and illustrate the reaction mechanism of LiPO2F2 at both electrodes. 2. Experiment 2.1. Sample preparation The following chemicals are all purchased from Guangzhou Tinci Materials Technology Co., Ltd., China. A low-temperature type of electrolyte which was made by dissolving a 1 M LiPF6 in ethylene carbonate (EC), ethylmethyl carbonate (EMC) and propylene carbonate (PC) with a 4:7:1 weight ratio was regarded as a reference electrolyte. Adding 1 wt.% LiPO2F2 (LG Co. Ltd, Korean) to the reference electrolyte was supposed to be the researched one. A cathode slury was fabricated by mixing 94.5 wt.%LiNi0.5Co0.2Mn0.3O2 (Shenzhen Tianjiao Tech. Co,. Ltd.), 1.5 wt.% polyvinylidene fluoride (PVDF) and 4.0 wt.% conductive carbon (Super-P). The anode electrode contained graphite (Shenzhen BTR Co,. Ltd.), Carboxymethyl Cellulose (CMC), Super-P and Polymerized Styrene Butadiene Rubber (SBR) with a 95.3:1.4:1.5:1.8 weight ratio. Both electrodes were dried at 120 C for 10 hours under vacuum conditions before the cells assembled. The LiNi0.5Co0.2Mn0.3O2/ graphite cells were used to understand the impact of LiPO2F2 on low temperature performance. Moreover, in order to know the effect of LiPO2F2 on cathode and anode electrode, CR-2016 button cells made of LiNi0.5Co0.2Mn0.3O2/Li and graphite/Li, respectively, were used. 2.2. Electrochemical Measurements Cyclic voltammograms (CV) curves were conducted on graphite/Li button cells with the scan rate of 0.05 mV s1 over a range of 0.01–3 V. The cells were measured by an electrochemical work station (CHI660, Chenhua, China). The charge-discharge cyclic performance at 0 C and 20 C was determined with LiNi0.5Co0.2Mn0.3O2/graphite cells through a battery test instrument (CT3008W, Neware, China). A temperature controlling box (GDW100L, SUOYATE, China) was used to maintain the environment temperature at diverse low temperature. Before the normal battery tests, the LiNi0.5Co0.2Mn0.3O2/graphite cells were charged to 3.85 V
with a constant current (0.1 C) and then cycled 3 times with different rates of 0.1 C, 0.2 C and 0.5 C. After that, a lowtemperature cycling test was followed by 100 cycles at 0.5 C rate from 2.75 V to 4.20 V. For the sake of the low temperature discharge ability, cells were cycled 3 times with a rate of 1 C at room-temperature before putting them in temperature controlling box with different temperatures of 0 C,10 C,20 C and 30 C for 4 hours. And then the cells were discharged at 0.5 C rate to 2.75 V. Indeed, DC resistance (DCR) was also examined at the above temperatures. Before DCR measurement, pouch cells were charged to 100% SOC and then put them into the temperature controlling box at different temperatures for 4 hours and the last second voltage was recorded as U1. The cells were discharged at a specific rate of 0.5 C for 10 seconds using the final second potential as U2. Afterward, the DCR value was calculated by using the formula: DCR = (U2 U1)/I. Electrochemical impedance spectroscopy (EIS) measurements were conducted with frequency response analyzer (FRA, Solartron 1455A, Solartron, England), CHI660 for LiNi0.5Co0.2Mn0.3O2/graphite cells and half cells. For full cells, the EIS of 100th cycles at 0 C was measured with 100% SOC (approximately 4.2 V) over a frequency range of 100 kHz to 20 mHz at an amplitude of 10 mV. For half cells, in order to form a stable SEI film and investigate the impact of LiPO2F2 on electrodes interface resistance, 3 cycles at a rate of 0.05 C at room temperature need to be taken prior to EIS measurement at 0 C. The EIS test of LiNi0.5Co0.2Mn0.3O2/Li and graphite/Li half cells were performed at 4.2 V and 0.01 V, respectively. 2.3. Physical Characterization To further figure out the characteristic of electrode surface, the cycled cells were disassembled in an argon-filed glovebox and the collected electrodes were rinsed with pure DMC for 3 times to remove the residual lithium salt and the electrolyte on electrodes surface. Then the samples were dried in glovebox for 12 hours and taken into an argon-filed bottle before the analysis. The surface morphology of the electrodes was observed by scanning electron microscopy (SEM, ZEISS Sigma500, Carl Zeiss, Germany) and transmission electron microscopy (TEM, JEM-2100HR, JEOL, Japan). The surface element composition of the electrodes were analyzed by X-ray photoelectron spectroscopy (XPS, AZIS SUPRA, Kratos, British) using the Al Ka line as the X-ray source. 3. Results 3.1. Electrochemical behavior of the electrolytes on graphite anode In order to understand the impact of LiPO2F2 on graphite, Li/ graphite half cell was used to investigate the electrochemical behavior of the electrolyte which contains 1% LiPO2F2. As shown in
Fig. 1. Cyclic voltammograms of graphite/Li half cells in 1 M LiPF6 dissolved in EC: EMC: PC (4:7:1, wt %) with different contents of LiPO2F2: (a) without LiPO2F2; (b) with 1% LiPO2F2 with a scan rate of 0.05 mV s1.
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
109
Fig. 2. Discharge curves of LiNi0.5Co0.2Mn0.3O2/graphite cells with a rate of 0.5C discharged to 2.75 V at various temperatures: (a) without LiPO2F2; (b) with 1% LiPO2F2; (c) capacity retention of the cells discharged at various low temperatures.
Fig. 1a, the poor repeatability of the three CV curves suggested that side reactions had taken place without LiPO2F2 during the chargedischarge progress and thus led to the increase of irreversible capacity. However, the CV curve (Fig. 1b) of the cell containing 1% LiPO2F2 presented that the second cycle was perfectly coincided with the third cycle, indicating the addition of LiPO2F2 stabilized the SEI film. 3.2. Electrochemical properties of LiPO2F2 in pouch cells Fig. 2 compared the discharge performance at various low temperatures of the cells with and without LiPO2F2. In addition, the specific value of capacity retention was shown in Fig. 2c. The cells without LiPO2F2 presented merely 49.4% (20 C) of capacity retention comparing with at 25 C and delivered little discharge capacity when temperature declined further, as depicted in Fig. 2a. However, with additional 1% LiPO2F2, the capacity retention at 20 C and 30 C were boosted to 71.9% and 57.9%, respectively (Fig. 2b). Moreover, the operating voltage showed the same trend
with the capacity retention as the temperature gradually went down. Owing to the rapidly increased resistance and low electrochemical reaction of cells, high polarization was the key reason to the operating voltage loss [13]. Consequently, the cells without LiPO2F2 showed poor capacity retention at low temperatures. This indicated the addition of LiPO2F2 could decrease the polarization and discharge ability was enhanced at subzero temperatures. As shown in Fig. 3, the impact of LiPO2F2 on the low temperature cyclic performance of LiNi0.5Co0.2Mn0.3O2/graphite cells was examined at 0 C and 20 C. The challenge of cycling at low temperature was well recognized with the slow lithium kinetics on graphite, which probably led to the loss of active lithium and rapid charge capacity fade. As depicted in Fig. 3b, d, the coulombic efficiency of the cells cycling at 0 C and 20 Chad strongly supported the above view of the charge capacity fade leading to the poor low temperature performance of LIBs. As a result, the poor capacity retention of the pouch cells without LiPO2F2 at 0 C had been observed in Fig. 3a. Owing to high
Fig. 3. Cyclic performance and coulombic efficiency of LiNi0.5Co0.2Mn0.3O2/graphite cells with a rate of 0.5 C cycled at a potential range of 2.75–4.20 V at various low temperatures: (a) cyclic performance at 0 C; (b) coulombic efficiency at 0 C; (c) cyclic performance at 20 C; (d) coulombic efficiency at 20 C.
110
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
stability, and the method below was employed to explain the difference of capacity retention at low temperature. 3.3. Electrochemical impedance
Fig. 4. The DCR values ofLiNi0.5Co0.2Mn0.3O2/graphite cells with different concentrations LiPO2F2 at various low temperatures.
Fig. 5. Electrochemical impedance spectra of the LiNi0.5Mn0.2Co0.3O2/graphite cells with and without LiPO2F2 charged to 4.2 V at 1st cycle and 100th cycle at 0 C.
polarization, cells without LiPO2F2 displayed lower initial capacity (123.5 mAh g1) than that (132.1 mAh g1) of the cells containing 1% LiPO2F2. In addition, cells without LiPO2F2 experienced a fast capacity retention fading of 20.1% after 100 cycles at 0 C meanwhile the cells with 1% LiPO2F2 maintained a high retention of 96.7% at the same cycle number. Additionally, a sharp contrast was depicted in Fig. 3c when LiNi0.5Co0.2Mn0.3O2/graphite cells were cycled at 20 C. The cells with LiPO2F2 delivered 91.7 mAh g1 after 100 cycles and reached high capacity retention of 91%. However, the cells without LiPO2F2 experienced a discharge capacity drop from 60.1 to 9.6 mAh g1. The results suggested that a little amount of LiPO2F2 significantly improved the cycling
A new measurement of DCR was introduced to illustrate the different performance between the cells with and without LiPO2F2 at low temperature. As we know, DCR was used to characterize the discharge ability at a specific condition. As shown in Fig. 4, a growing gap of DCR values was observed between the cells with or without LiPO2F2. In particular, the DCR value of cells containing LiPO2F2 was 0.882 V while that of cells without LiPO2F2 sharply increased to 2.192 V at 20 C. This supported the hypothesis that the decrease of DCR value was considered to be the sacrifice of LiPO2F2, which contributed to better lithium kinetics on electrodes surface. To investigate the influence of LiPO2F2 towards interface impedance, electrochemical impedance spectroscopy (EIS) was employed to measure the impedance in pouch cells and half cells, respectively. As other literatures reported [14], a typical EIS curve composed of three parts: (i) high-frequency semicircle, (ii) middlefrequency semicircle, (iii) low-frequency straight line. The above three regions represented SEI resistance (RSEI), charge-transfer resistance (RCT) and Warburg resistance, respectively. Fig. 5 showed the EIS of LiNi0.5Co0.2Mn0.3O2/graphite cells with and without LiPO2F2 at 1st cycle and 100th cycle respectively, at 0 C. The addition of LiPO2F2 could significantly suppress the increased SEI and charge transfer resistance. By the contrast of EIS of 1st and 100th, a slight decrease of SEI resistance could be seen from the cell containing 1% LiPO2F2, which meant an ionic conductivity film was formed on electrode interface. Meanwhile it also restricted the sharp increase of charge transfer resistance compared to the EIS results of cell without LiPO2F2. In summary, the high conductive SEI film formed by LiPO2F2 facilitated the Li+ migration through electrodes interface and weakened the impact of the rapid increased Rct which was due to the non-conductive substance deposited on electrodes surface. Moreover, the half cells could simulate the state of intercalation/deintercalation lithium of electrodes and investigate the real reason for such a low resistance. So the EIS of graphite/Li and LiNi0.5Co0.2Mn0.3O2/Li half cells was also conducted at 0 C. Fig. 6a showed slight difference between the LiNi0.5Co0.2Mn0.3O2/Li cells with and without LiPO2F2. However, the EIS result of graphite/Li cells containing LiPO2F2 exhibited an obvious decrease of impedance value in both high and middle frequency regions (Fig. 6b) than that of cells without LiPO2F2, which suggested that the electrolyte containing 1% LiPO2F2 modified the SEI film on graphite and improved Li+ migration
Fig. 6. Electrochemical impedance spectra of the half cells at 0 C: (a) LiNi0.5Co0.2Mn0.3O2/Li cell at 4.2 V; (b) graphite/Li cell at 0.01 V. The half cells were cycled 3 times at room temperature before EIS measurement.
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
and diffusion though the surface. Consequently, the low impedance of pouch cells was attributed to the influence of LiPO2F2 on graphite, making a better lithium kinetics film at low temperature, and thus improved the low temperature performance of the LiNi0.5Co0.2Mn0.3O2/graphite cells. Meanwhile the result also strongly supported the view of the poor lithium kinetics on graphite restricting the application of LIBs in cold environment. 3.4. The effect of LiPO2F2 on electrodes surface morphology For the sake of the effect of LiPO2F2 on electrodes, transmission electron microscope (TEM) was used before and after 100 cycles at 0 C, and the results were shown in Fig. 7. Compared to the fresh
111
cathode, a smooth surface and a clear outline were observed in the cycled cathode of the cells with or without LiPO2F2 (Figs. 7a–c). Consistent with the EIS result of LiNi0.5Co0.2Mn0.3O2/Li cell, the TEM photos of cycled cathode also suggested that LiPO2F2 had little impact on cathode. However, a huge difference could be found on graphite anode, as depicted in Fig. 7d–f. Fig. 7e displayed the TEM image of cycled anode without LiPO2F2, whose graphite surface had been destroyed into numerous irregular particles and SEI almost vanished without trace relating to the rapid capacity fading of cyclic performance at 0 C. On the contrary, Fig. 7f showed a smooth and thin layer film on graphite surface which indicated the anode with LiPO2F2 formed a conductive film, facilitating the intercalation of Li+. This suggested that a high ionic conductive film
Fig. 7. TEM images taken from the surface of electrode: (a) a fresh cathode; (b) a cycled cathode without LiPO2F2; (c) a cycled cathode with LiPO2F2; (d) a fresh anode; (e) a cycled anode without LiPO2F2; (f) a cycled anode with LiPO2F2. The cycled electrodes were taken from pouch cells after 100 cycles at 0 C.
112
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
ensured the normal intercalation and deintercalation of Li+ through the graphite, which contributed to a better cyclic performance at low temperature. The morphology of the electrodes surface before and after 100cycles at 0 C was also observed with scanning electron microscopy (SEM) and the results were displayed in Fig. 8. Both the fresh and cycled cathode showed the similar morphology, which proved the view that LiPO2F2 exerted no side-effect on
cathode, once again (Fig. 8a–c). The graphs of graphite surface morphology were presented in Fig. 8d–f. As shown in Fig. 8e, some dendritic structure substance had been found on the surface of the cycled graphite without LiPO2F2. It was likely to be the lithium oxidations originated from the lithium dendrites or lithium precipitation exposed to the air before the SEM analysis. And this inactive lithium was attributed to the increased irreversible capacity. However, compared to the fresh graphite (Fig. 8d), a
Fig. 8. SEM images taken from the surface of electrode: (a) a fresh cathode; (b) a cycled cathode withoutLiPO2F2; (c) a cycled cathode with LiPO2F2; (d) a fresh anode; (e) a cycled anode without LiPO2F2; (f) a cycled anode with LiPO2F2. The cycled electrodes were taken from pouch cells after 100 cycles at 0 C.
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
thin SEI layer was observed in the surface of cycled graphite containing LiPO2F2 (Fig. 8e). The graphite was covered by a dense layer and some big particles adhered to the active materials. From the SEM results, the Li-ion deposition on graphite surface was effectively reduced through a layer film formed by LiPO2F2, and thus expanded the operating temperature range of LiNi0.5Co0.2Mn0.3O2/graphite cells. 3.5. X-ray photoelectron spectroscopy (XPS) analyses of graphite anode To further reveal the effect of LiPO2F2 on graphite, X-ray photoelectron spectroscopy (XPS) was used to analyze the chemical composition of graphite surface. The analysis and different elemental concentrations of the graphite surface were shown in Fig. 9 and Table 1. Compared with fresh graphite, an obvious decrease of C concentration could be found while the concentration of O, F and P were increased. It was assumed that the graphite surface was covered by the electrolyte decomposition products. Particularly, the concentration of F and P were higher on the graphite cycled with 1% LiPO2F2 than those on the graphite without LiPO2F2, hence, it could be concluded that the decomposition of LiPO2F2 played an important role in the SEI film formation. The C 1s spectrum of the fresh graphite consisted of three peaks. The first one at 284.3 eV was associated with graphite [15]. The next one at 285.0 eV was attributed to SBR binder [16]. The C-O
113
Table 1 Different element concentrations on fresh graphite and graphite form cycled cells with and without 1% LiPO2F2 at 0 C. Sample
C1s (%)
O1s (%)
F1s (%)
P2p (%)
Fresh Without LiPO2F2 With 1% LiPO2F2
87.14 41.61 43.95
12.86 50.51 43.06
4.53 8.81
3.35 4.18
bond appeared at 286.2 eV was due to the CMC binder. After 100 cycles at 0 C, a new characteristic peak of C¼O, at 289.3 eV, was observed in C1s spectra for graphite both cycled with and without LiPO2F2. C¼O bonds were probably the electrolyte decomposition products of lithium alky carbonates and polycarbonates. In addition, given the disappeared peak of SBR binder we assumed that the graphite had been covered by a SEI film or some organic compounds attached on the electrode surface. The O 1s spectrum of the fresh graphite showed tow main peaks which belonged to the oxygen atoms of the CMC binder. For the graphite cycled without LiPO2F2, there were also two main peaks, the weaken intensity of C O (533.8 eV) and especially high intensity peak of C¼O (532.0 eV) could be found [17]. In other word, the chief components of the cycled graphite without LiPO2F2 were the lithium alky carbonates or polycarbonates due to the decomposition of the solvents and salts. What’s more, a new peak at 531.5 eV, the characteristic peak of Li2CO3, could be found in the cycled graphite containing LiPO2F2 and it had not been observed in
Fig. 9. C 1s, O 1s, F 1s and P 2p XPS spectra of fresh graphite (top) and cycled graphite without LiPO2F2 (middle) and cycled graphite with 1% LiPO2F2 (bottom) at 0 C after 100 cycles.
114
B. Yang et al. / Electrochimica Acta 221 (2016) 107–114
the graphite without LiPO2F2, which suggested that the cycled graphite containing LiPO2F2 comprised of more inorganic components. The presence of Li2CO3 was closely related to a better battery performance owing to its less soluble property in the solvents, and thus effectively passivated electrode preventing electrolyte from further decomposing [18]. More importantly, the SEI film containing Li2CO3 had a better Li-ion conductivity making the Li+ intercalation much easier [19]. It was easy to see the difference of cycled graphite with and without LiPO2F2 in the F 1s spectrum. Both cycled anode consisted of three peaks. The peaks, at 684.3 eV and 685.9 eV, corresponded to the characteristics of LiF and LixPOyFz respectively .The third peak at 686.9 eV was attributed to LiPxFy, which was likely to be the residual LiPF6 on graphite although the electrodes was wished with DMC before XPS analysis [17]. The concentration of LiF in cycled anode with LiPO2F2 was much higher than that without LiPO2F2, which suggested that the proportion of LiF in SEI film were boosted on account of the decomposition of LiPO2F2 at low temperature cycling. Unlike the room-temperature condition, an increased content of LiF, in a reasonable range, not only reduced the SEI film resistance, but also improved its low temperature performance [6]. Comparing to the P 2p spectrum, the peak at 131.0 eV, characteristic of LixPOyFz, had the same trend of that with F 1s spectrum. The decomposition of LiPO2F2, especially P O bounds, made the stronger intensity of LixPOyFz [20]. These results all suggested that a conductive, protective SEI film formed on graphite anode and the improvement of low temperature performance was attributed to the addition of LiPO2F2. In addition, the XPS data of the work made by Ko-Eun Kim et al. showed a different result from this paper [12]. We found a lower Li2CO3 content than that of the reference. It came to conclusion diametrically opposed to our work. The main reasons affecting the content of Li2CO3 in SEI film were attributed to the graphite type, solvent system and hermetic-seals [21]. And the different solvent system makes greatest impact on the Li2CO3 content. From the previous work [22], an unpaired nucleophilic anion (CO32 or LiCO3) interacts with Li+ to form Li2CO3 in SEI film at low PC concentration. The solvent system using PC as co-solvent, other than DMC in Ko-Eun Kim et al. work, is the key factors of high Li2CO3 content. More importantly, the extra Li+ provided by LiPO2F2 also makes contribution to the high content of Li2CO3. Secondly, higher P O phosphate compounds in SEI film was consistent with the conclusion in this work which the peak of LixPOyFz contained P O phosphate compounds and showed high peak intensity than that without additive. 4. Conclusions LiPO2F2 was investigated as an electrolyte additive in LiNi0.5Co0.2Mn0.3O2/graphite cells. The CV curves suggested that LiPO2F2 could stabilize the SEI film formed on graphite anode, and significantly decreased the charge-transfer resistance as depicted in EIS data. In addition, the XPS results had demonstrated that such a low impedance was attributed to the richer content of LiF and the presence of Li2CO3 in SEI film. Moreover, TEM and SEM also intuitively demonstrated that LiPO2F2 had a tremendous impact on graphite while it exerted no side effect on cathode. Therefore, the addition of 1 wt% LiPO2F2 could significantly enhance the discharge characteristics of full cells at various low temperatures and the capacity retention had been increased from 79.9% to 96.7% at 0.5C rate after 100 cycles at 0 C. These results all powerfully supported that the addition of LiPO2F2 improved the stability and conductivity
of the SEI film on graphite, which is helpful for the intercalation of Li+ into graphite and the better lithium kinetics on graphite surface, leading to an improved low temperature and long cyclic performance in LiNi0.5Co0.2Mn0.3O2/graphite cells. Acknowledgments This work was financially supported by the science and technology projects of Guangdong Province (No.2014B010123002) and the science and technology projects of Guangzhou (2014Y2-00012). References [1] C.Y. Wang, G.S. Zhang, S.H. Ge, Terrence Xu, Y. Ji, X.G. Yang, Y.J. Leng, Lithiumion battery structure that self-heats at low temperatures, Nature 528 (2016) 515. [2] Y. Ji, Y.C. Zhang, C.Y. Wang, Li-Ion Cell Operation at Low Temperatures, J. Electrochem. Soc. 160 (2013) A636. [3] Y. Ji, C.Y. Wang, Heating strategies for Li-ion batteries operated from subzero temperatures, Electrochim. Acta 107 (2013) 664. [4] S.S. Zhang, K. Xu, T.R. Jow, Low temperature performance of graphite electrode in Li-ion cells, Electrochim. Acta 48 (2002) 241. [5] S.S. Zhang, K. Xu, T.R. Jow, The low temperature performance of Li-ion batteries, J. Power Sources 115 (2003) 137. [6] B.X. Liu, B. Li, S.Y. Guan, Effect of Fluoroethylene Carbonate Additive on Low Temperature Performance of Li-Ion Batteries, Electrochem. Solid-State Lett. 15 (2012) A77. [7] M.C. Smart, B.V. Ratnakumar, K.B. Chin, L.D. Whitcanack, Lithium-Ion Electrolytes Containing Ester Cosolvents for Improved Low Temperature Performance, J. Electrochem. Soc. 15 (2010) A1361. [8] L.X. Liao, P.J. Zuo, Y.L. Ma, Y.X. An, G.P. Yin, Y.Z. Gao, Effects of fluoroethylene carbonate on low temperature performance of mesocarbon microbeads anode, Electrochim. Acta 74 (2012) 260. [9] M.Y. Nie, Daniel P. Abraham, Daniel M. Seo, Y.J. Chen, Arijit Bose, Brett L. Lucht, Role of Solution Structure in Solid Electrolyte Interphase Formation on Graphite with LiPF6 in Propylene Carbonate, J. Phys. Chem. 117 (2013) 25381. [10] S.S. Zhang, K. Xu, T.R. Jow, Low-temperature performance of Li-ion cells with a LiBF4-based electrolyte, J. Solid State Electrochem. 7 (2003) 147. [11] Marshall C. Smart, Brett L. Lucht, Swapnil Dalavi, Frederick C. Krause, Bugga V. Ratnakumar, The Effect of Additives upon the Performance of MCMB/LiNixCo1xO2 Li-Ion Cells Containing Methyl Butyrate-Based Wide Operating Temperature Range Electrolytes, J. Electrochem. Soc. 159 (2012) A739. [12] K.E. Kim, J.Y. Jang, Inbok Park, M.H. Woo, M.H. Jeong, W.C. Shin, M. Ue, N.S. Choi, A combination of lithium difluorophosphate and vinylene carbonate as reducible additives to improve cycling performance of graphite electrodes at high rates, Electrochim. Acta 61 (2015) 121. [13] S.S. Zhang, K. Xu, T.R. Jow, A new approach toward improved low temperature performance of Li-ion battery, Electrochem. Commun. 4 (2002) 928. [14] J.G. Zhu, Z.C. Sun, X.Z. Wei, H.F. Dai, Studies on the medium-frequency impedance arc for Lithium-ion batteries considering various alternating current amplitudes, J. Appl. Electrochem. 46 (2016) 157. [15] R. Dedryvère, S. Leroy, H. Martinez, F. Blanchard, D. Lemordant, D. Gonbeau, XPS Valence Characterization of Lithium Salts as a Tool to Study Electrode/ Electrolyte Interfaces of Li-Ion Batteries, J. Phys. Chem. B 110 (2006) 12986. [16] L. El Ouatani, R. Dedryvère, C. Siret, P. Biensan, D. Gonbeau, Effect of Vinylene Carbonate Additive in Li-Ion Batteries: Comparison of LiCoO2/C, LiFePO4/C, and LiCoO2/Li4Ti5O12 Systems, J. Electrochem. Soc. 156 (2009) A468. [17] M.Q. Xu, L.S. Hao, Y.L. Liu, W.S. Li, L.D. Xing, B. Li, Experimental and Theoretical Investigations of Dimethylacetamide (DMAc) as Electrolyte Stabilizing Additive for Lithium Ion Batteries, J. Phys. Chem. 115 (2011) 6085. [18] Pallavi Verma, Pascal Maire, Petr Novák, A review of the features and analyses of the solid electrolyte interphase in Li-ion batteries, Electrochim. Acta 55 (2010) 6332. [19] Y.-K. Choi, K.-i. Chung, W.-S. Kim, Y.-E. Sung, S.-M. Park, Suppressive effect of Li2CO3 on initial irreversibility at carbon anode in Li-ion batteries, J. Power Sources 104 (2002) 13. [20] G.C. Yan, X.H. Li, Z.X. Wang, H.J. Guo, C. Wang, Tris (trimethylsilyl) phosphate: A film-formation additive for high voltage cathode material in lithium-ion batteries, J. Power Sources 248 (2014) 1306. [21] Kristina Edstrom, Marie Herstedt, Daniel P. Abraham, A new look at the solid electrolyte interphase on graphite anodes in Li-ion batteries, J. Power Sources 153 (2006) 380. [22] Doron Aurbach, Mikhail D. Levi, Elena Levi, Alexander Schechter, Failure and Stabilization Mechanisms of Graphite Electrodes, J. Phys. Chem. B 101 (1997) 2195.