Article Cite This: J. Am. Chem. Soc. 2018, 140, 17515−17521
pubs.acs.org/JACS
Cationic Surfactant-Based Electrolyte Additives for Uniform Lithium Deposition via Lithiophobic Repulsion Mechanisms Hongliu Dai,†,‡ Kai Xi,∥,‡ Xin Liu,† Chao Lai,*,† and Shanqing Zhang⊥ †
School of Chemistry and Materials Science, Jiangsu Normal University, Xuzhou, Jiangsu 221116, China Department of Materials Science and Metallurgy, University of Cambridge, Cambridge CB3 0FS, U.K. ⊥ Center for Clean Environment and Energy, School of Environment and Science, Griffith University, Brisbane, Queensland 4222, Australia J. Am. Chem. Soc. 2018.140:17515-17521. Downloaded from pubs.acs.org by UNIV OF SOUTH DAKOTA on 12/19/18. For personal use only.
∥
S Supporting Information *
ABSTRACT: Lithium metal is among the most promising anode materials for high-energy batteries due to its high theoretical capacity and lowest electrochemical potential. However, dendrite formation is a major challenge, which can result in fire and explosion of the batteries. Herein, we report on hexadecyl trimethylammonium chloride (CTAC) as an electrolyte additive that can suppress the growth of lithium dendrites by lithiophobic repulsion mechanisms. During the lithium plating process, cationic surfactant molecules can aggregate around protuberances via electrostatic attraction, forming a nonpolar lithiophobic protective outer layer, which drives the deposition of lithium ions to adjacent regions to produce dendrite-free uniform Li deposits. Thus, an excellent cycle of 300 h at 1.0 mA cm−2 and rate performance up to 4 mA cm−2 are available safely in symmetric Li|Li cells. In particular, significantly enhanced cycle and rate performance were achieved when the electrolyte with CTAC additives was used in lithium−sulfur and Li|LiNi0.5Co0.2Mn0.3O2 full cells. The effects of carbon chains, anions of surfactant, and electrostatic repulsion on the deposition of lithium anodes are reported. This work advances research in inhibiting Li dendrite growth with a new electrolyte additive based on cationic surfactants.
■
INTRODUCTION Lithium (Li) metal has been designated as one of the most promising anode materials for next-generation high-energy batteries due to its ultrahigh theoretical capacity (3860 mAh g−1) and low electrochemical potential (−3.04 V vs the standard hydrogen electrode).1−3 The advantages of higher energy densities can be attained if lithium is paired with sulfur or oxygen cathodes to form a full cell.4,5 However, commercial applications of pure Li-based batteries have been seriously hindered due to safety considerations and lifetime concerns caused by a combination of factors such as dendritic Li formation, large volume changes, and instability of the solid electrolyte interphase (SEI).1−3,6 During repeated discharge− charge processes, the growth of lithium dendrites can penetrate the separator, leading to internal short circuiting, thermal runaways, and even catastrophic cell failure with explosion. In addition, fresh Li dendrites can increase the side reactions between lithium and organic electrolytes, causing the continuous consumption of both. Such consumption can lead to drying of the electrolyte, corrosion of the Li anode, and even the generation of electrochemically inert Li, thus resulting in low Coulombic efficiency, fast capacity decay, and large polarization.1−3,6 Over the past four decades, various strategies such as employing solid-state electrolytes, optimizing electrolyte © 2018 American Chemical Society
composition, constructing an artificial SEI or polymer coating layer, and designing three-dimension Li-hosting materials have been developed in order to suppress Li dendrite growth.6−23 Among these different routes, three-dimensional Li-hosting materials show obvious advantages in achieving long cycle life and excellent rate performance due to its unique structural features,1−3 while in terms of the requirements for practical applications, modification of the electrolyte using additives is of particular interest.6,11 The general role of most additives is help to construct a stable SEI film in order to suppress Li dendrite formation. Nevertheless, sufficient passivation between the Li anode and electrolyte is rarely achieved due to the gradual consumption of additives that occurs during prolonged cycling.6,11 Recently, Zhang et al. proposed a novel Li protection strategy, called a “self-healing electrostatic shield mechanism”.17,18 During the Li plating process, specific cations assemble around the nucleation points of Li dendrites to form a positively charged region, and such electrostatic forces drive any further decomposition of Li+ to adjacent regions, in favor of producing a smooth lithium anode without dendrites. Ideally, plating additives are “reused” during cycling, and thus a prolonged cycle life can be obtained in comparison to other Received: August 20, 2018 Published: November 29, 2018 17515
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
Article
Journal of the American Chemical Society
outward lithiophobic layer due to the repulsive forces, and further Li decomposition is diverted to adjacent regions, allowing for the formation of a smoother Li surface (enlarged figure in Figure 1). In addition, the internal positively charged layer of polar groups can also effectively suppress the growth of Li dendrites via the electrostatic repulsive forces. By using this novel additive, Li+ distribution on the Li anode surface can be uniformly deposited, generating dendrite-free Li deposition and thus resulting in stable cycling performance.
electrolyte additives. However, these ions will also deposit on the surface of a lithium anode during cycling at high current densities, and complex procedures are still needed to obtain the additives, such as CsPF6.1,17,18 Then, an ionic liquid, 1dodecyl-1-methylpyrrolidinium bis(fluorosulfonyl)imide (Pyr1(12)-FSI), is employed as the plating additives for the same mechanisms.19 In combination with the synergistic effect of the FSI− anion and lithiophobicity of aliphatic chains, very promising electrochemical results were obtained. However, it should be noted that the cycle life is still not high enough (below 140 cycles at 0.5 mA cm−2). In addition, the high cost of such an additive cannot be neglected, and the adding amount is also relatively high according to the high molecular weight of Pyr1(12)-FSI, for which the molar ratio between ionic liquid and solvent is 1:12. Hence, developing highefficiency and low-cost electrolyte additives is still urgently desirable for the safe applications of metallic Li-based batteries. Since Li dendrite growth originates mainly from the spatial inhomogeneity of Li ion distribution during plating, one strategy is to more precisely control an even distribution of Li ions on the electrode surface.1−3 Inspired by the aforementioned works, herein, we proposed an effective but low-cost additive, namely, hexadecyl trimethylammonium chloride (CTAC), for anode protection, and systematically verify the lithiophobic repulsion mechanism for rendering a Li-based battery with long-term cycle life. As presented in Figure 1, in
■
EXPERIMENTAL SECTION
Characterizations. The morphological changes of the lithium foil after cycling in different electrolytes were characterized via scanning electron microscopy (SEM) measurements (Hitachi SU8010). Realtime observations of lithium dendrite growth were investigated by optical microscopy (Nikon SMZ1270), for which the lithium anode was sealed in a transparent optical cell (Figure S1). The assembly of the optical cell was conducted at room temperature in an argon-filled glovebox. The contact angle of the electrolyte on lithium metal was tested by an automatic contact angle measuring instrument (JC2000D3M). Electrolyte viscosity was measured using a TA DHR2 rheometer. Electrochemical Measurements. To investigate the electrochemical performance of the CTAC-containing electrolyte, symmetrical Li|Li cells were assembled. The working electrode and counter electrode were both Li foil. The electrolyte was lithium hexafluorophosphate (LiPF6, 1.0 M) dissolved in a mixture of propylene carbonate (PC), ethyl carbonate (EC), and diethyl carbonate (DEC) in a volume ratio of 1:4:5 with or without CTAC. The concentration of CTAC additives in the electrolyte was 1.2 mM, and the added CTAC is sufficient to cover the entire surface of the Li anode. For symmetric cell testing, 30 μL of electrolyte was used for long-term plating/stripping galvanostatic testing, at current densities of 1.0, 2.0, and 4.0 mA cm−2, respectively. For testing of the Li−S battery, a pPAN@S composite was used as a cathode, which was prepared according to previous reports.26,27 For the fabrication of electrodes, pyrolyzed polyacrylonitrile/sulfur (pPAN@S) composite or LiNi0.5Co0.2Mn0.3O2, carbon black, and binder were mixed with a weight ratio of 7:2:1. The loading mass of the sulfur electrodes is about 1.2−1.4 mg cm−2. The mass loading of the LiNi0.5Co0.2Mn0.3O2 electrode is about 1.2 mg cm−2. For the bare electrode, it was prepared by compressing a mixture of the acetylene black and binder (polytetrafluoroethylene, PTFE), in a weight ratio of 80:20. Li foil (China Energy Lithium Co., Ltd., >99.99%) was used as a counter electrode. The thickness of the Li foil is around 600 μm with a diameter of 14 mm. Galvanostatic measurements were performed on the battery testing system (LAND, CT2001A) at current densities of 400, 800, 1600, and 3200 mA g−1, within the voltage range of 1.0−2.8 V (vs Li+/Li). Electrochemical impedance spectra (EIS) were studied using an electrochemical workstation (Solartron 1287) with an amplitude of 5 mV and a frequency range between 10 mHz and 100 kHz. Cyclic voltammetry (CV) experiments were conducted using a CHI 600D, at a scan rate of 0.1 mV s−1. Differential capacitance− potential curves were obtained by impedance methods (Solartron 1287). All measurements were conducted using a frequency of 1 kHz and a signal of 5 mV amplitude, within a voltage range of −0.6−0.6 V, at a scan rate of 10 mV s−1. The capacitance can be calculated by the following equation:
Figure 1. Schematic illustration of the effect of CTAC addition on the Li decomposition process.
the initial plating process, Li+ is reduced at the anode and begins to unavoidably form protuberant tips as it deposits onto the electrode surface. It is well established that electric charges tend to accumulate over these tips, which in turn attracts further Li ions to deposit in the same area, eventually leading to dendrite formation in conventional organic electrolytes.1 In contrast, after the introduction of CTAC additives, more CTA+ ions will be adsorbed around the protuberant tips via electrostatic interactions with outward facing nonpolar groups. Meanwhile, the surface free energy of the protuberant tips can be expected to increase as compared to the smooth electrode surface, and thus a greater concentration of CTAC molecules will be adsorbed at the solid−liquid interface of protuberances as compared to a smooth surface.24,25 Accordingly, driven by the electrostatic interactions and surface free energy, large-scale adsorption of CTAC molecules in protuberant regions will be favored, thus ensuring formation of a cross-linked lithiophobic protective layer (Figure 1). As the plating process progresses, polarized solvent Li+ is unable to transport through this
C = (2πf Zim)−1 where C is the capacitance, Zim is the imaginary component of the impedance, and f is the frequency of the ac perturbation.
■
RESULTS AND DISCUSSION In Situ Optical Microscopy Study of Lithium Deposition. The Li deposition process was recorded first via in situ optical microscopy (Figure S1) in order to illustrate 17516
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
Article
Journal of the American Chemical Society
the entire Li foil can all be observed. With the addition of CTAC, a dense and smooth deposition of Li can be observed even after 300 cycles. Based on these findings, it can be concluded that Li dendrite growth can be entirely suppressed via the addition of CTAC to produce a smooth and stable interface, allowing for excellent cycling performance. Electrochemical Performance of Li|Li Symmetric Cells. Symmetric cells were assembled in order to evaluate the electrochemical cycle stability and rate performance of the Li anodes, with the results presented in Figure 4. As shown in
the effect of CTAC additives. As presented in Figure 2a, without CTAC additives, the generation of protuberances
Figure 2. In situ optical microscopy observations of the Li deposition process: (a) in untreated electrolyte; (b) in electrolyte with CTAC additives; (c) continuous deposition of the same Li foil in an alternative electrolyte. The thickness of the Li foil is about 600 μm.
happens after plating for 2 min at a current density of 4.0 mA cm−2, and highly dendritic and mossy Li growth are observed after 6 min. This formation will continuously consume both metallic Li and electrolyte and, therefore, result in low Coulombic efficiency. In contrast, with the inclusion of CTAC additives, the Li surface remains smooth and free of dendrites, even at high current densities (Figure 2b). To further illustrate the role of CTAC additives, the same sample of Li foil was first deposited for 0.5 min at a current density of 4.0 mA cm−2 in untreated electrolyte, then for another 4.0 min in an electrolyte containing CTAC additives, and finally for another 4.0 min, again in untreated electrolyte. As presented in Figure 2c, the continuous growth of Li dendrites is completely suppressed when the electrolyte contains CTAC additives. Characterization of the Surface of a Lithium Anode. The resulting smooth Li surface can be clearly illustrated by SEM images. Figure 3 shows SEM images of the Li anode after 20 and 300 cycles, at a current density of 1.0 mA cm−2 with a fixed capacity of 0.5 mAh cm−2 in symmetric Li|Li cells. Without CTAC additives, fracture of the lithium anode is shown to occur after 20 cycles. After a further 300 cycles, massive dendritic, mossy Li deposits, dead Li, and cracking of
Figure 4. Cycling performance comparisons of symmetric cells fabricated without or with CTAC additives, at current densities of (a) 1.0 mA cm−2 with a fixed capacity of 0.5 mAh cm−2, (b) 2.0 mA cm−2 with a fixed capacity of 1 mAh cm−2, and (c) 4.0 mA cm−2 with a fixed capacity of 2 mAh cm−2. Enlarged figures on the right show detailed voltage profiles with cycle time indicated. (d) Rate performance of symmetric cells measured at current densities of 1.0, 2.0, and 4.0 mA cm−2 for 0.5 h in both the Li-stripping and Li-plating processes of each cycle. (e) EIS of the Li anode before cycling and after first cycle at a current density of 1.0 mA cm−2 with fixed capacity of 0.5 mAh cm−2.
Figure 3. SEM images of Li anodes plated with or without CTAC additives after (a) 20 and (b) 300 cycles at a current density of 1.0 mA cm−2 and a fixed capacity of 0.5 mAh cm−2 in symmetric Li|Li cells.
Figure 4a−c, symmetric Li|Li cells with CTAC additives are stable after cycling for 300 h at 1.0 mA cm−2 with a fixed capacity of 0.5 mAh cm−2 and for 100 h at 2.0 mA cm−2 with a fixed capacity of 1 mAh cm−2 and 4.0 mA cm−2 with a fixed capacity of 2 mAh cm−2, respectively. On the other hand, without CTAC additives, significant fluctuations in voltage, followed by an abrupt drop-off during cycling, can be observed. As the number of cycles is increased, the magnitude of polarization that occurs in symmetric Li|Li cells in untreated electrolyte is much higher than that observed in an electrolyte containing CTAC additives. This elevated level of polarization is a result of dendritic, mossy, and dead Li that is formed after repeated plating/stripping (Figure 3). A detailed comparison of the voltage profiles is also shown in the enlarged figure in Figure 4a−c. At a current density of 1.0 mA cm−2, the Li anode with CTAC additives gives an overpotential of ∼68.0 mV, 17517
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
Article
Journal of the American Chemical Society which increases slightly to 79.0 mV at a current density of 2.0 mA cm−2 and then to 182 mV at a current density of 4.0 mA cm−2. However, the Li anode plated without CTAC additives displays a much higher overpotential, even resulting in a short circuit. Meanwhile, the stability of bare Li anodes, either with or without CTAC additives, is further evaluated and compared by measuring the cycle curves under abusing current densities of 1.0, 2.0, and 4.0 mA cm−2 after 0.5 h in both the Li stripping and plating processes, corresponding to a Li plating/stripping capacity of 0.5, 1, and 2 mAh cm−2, respectively, as illustrated in Figure 4d. The voltage hysteresis for the Li anode with CTAC additives is well retained, and a low overpotential from ∼73.0 mV at 1.0 mA cm−2 to ∼116.0 mV at 4.0 mA cm−2 can be obtained, well consistent with the above results, which exhibits significant advantages over bare Li foil counterparts. In addition, EIS measurements were conducted to illustrate the kinetic behavior of the Li anodes, with the resulting Nyquist plots obtained from different electrolytes given in Figure 4e. Before cycling, the charge transfer impedance associated with the semicircular arc in the high frequency range of the untreated electrolyte (111.2 Ω) is slightly lower than that with the CTAC electrolyte (116.7 Ω).28,29 In order to explain this phenomenon, contacting angle and viscosity testing were conducted. As presented in Figure S2, the introduction of CTAC additives can greatly decrease the contact angle and increase the wettability between the electrolyte and lithium anode. This acts to reduce the viscosity of the electrolyte and enhance its conductivity.30,31 Both influences can decrease the interfacial charge transfer resistance, which is contrary to our experimental results. Thus, the increasing of charge transfer resistance can mainly be attributed to the adsorption of surfactant molecules on the surface of the solid electrode that act as a physical barrier to hinder Li+ transport. After the first cycle, the SEI resistance and charge transfer impedance for devices containing CTAC additives is much smaller than that produced without CTAC additives, as presented in Figure S3, and this is due to the creation of a stable solid−liquid interface driven by CTAC. To further reveal the lithiophobic repulsion mechanisms that occur, the effects of anions, concentration of CTAC, and the length of carbon chains on Li deposition are also investigated. As presented in Figure 5a, hexadecyl trimethyl ammonium bromide (CTAB) is used as an additive to verify the effect of chloride ions, demonstrating a similar cycling performance to that of CTAC additives in a Li|Li symmetric cell. This indicates that the type of anions has little effect on the anode protection, while the length of the carbon chain is decisive. This can be confirmed by comparisons of electrochemical tests using dodecyl trimethylammonium chloride (DTAC) and stearyl trimethylammonium chloride (STAC) additives, in Figure 5b and c. The cell using CTAC additives shows much higher cycling stability relative to that of DTAC, and its voltage profiles are in better agreement compared to those using STAC additives, suggesting that CTAC is sufficiently effective to drive uniform Li deposition. In order to verify the effect of electrostatic repulsion, 2-chloroethyl trimethylammonium chloride (CCC), which contains only a short and polar chloroethyl group, is also investigated as an electrolyte additive. As presented in Figure 5d, after about 40 cycles, significant increasing of the overpotential can be observed, and then a short circuit occurs. For CTAC, the cations with three methyl groups may not be able to produce a dense positive “shielding” on the protuberances like that of Cs+ ion due to the
Figure 5. Cycling performance comparisons of symmetrical cells using electrolytes with (a) CTAB additives, (b) DTAC additive, (c) STAC additives, and (d) CCC additives under a current density of 2.0 mA cm−2 with a fixed capacity of 1.0 mAh cm−2.
steric hindrance.17 These results suggest that lithiophobic repulsion of long aliphatic chains plays key roles for anode protection in common CTAC additives. Moreover, Figure S4 illustrates the effect of concentration on the cycling performance of Li|Li symmetric cells. Compared to that of 1.2 mM CTAC, poor cycling performance can be observed, suggesting that the growth of Li dendrites is not adequately restricted, regardless of the higher or lower concentration of CTAC additives. It follows that lower concentrations (0.5 mM) cannot generate a dense lithiophobic protective layer on the surface of a protuberance, yet contrary to our expectations, it is not any more effective to suppress the formation of Li dendrites using an electrolyte with high concentration of CTAC additives (5.0 mM). This may result from the nature of the surfactant, as massive surfactant molecules will adsorb on the entire surface of Li foil, and there are no obvious differences between the protuberant and smooth regions at higher concentration. In addition, CTAC molecules will aggregate to form micelles in electrolyte at high concentration,32,33 and resultantly, the properties of the electrolyte also change. As shown in Figure S5, when the concentration of CTAC is increased to 5.0 mM, the contacting angle will greatly increase as compared to that at low concentrations (1.2 mM), and the viscosity is even higher than that for untreated electrolyte, all of which will act to deteriorate the electrochemical performance of the Li anodes. We also attempted to investigate the effect of anionic surfactants, such as sodium dodecyl sulfate, on Li anodes, but unfortunately, these additives were insoluble in carbonate solvent. Electrochemical Performance of Lithium−Sulfur and Li|LiNi0.5Co0.2Mn0.3O2 Full Cells. The advantages of using CTAC additives were further verified in lithium−sulfur (Li−S) full batteries, using pPAN@S composite as the cathode. As demonstrated in Figure 6a, the values of discharge capacity 17518
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
Article
Journal of the American Chemical Society
untreated electrolyte barely discharge at all and charge at a high current density of 3200 mA g−1 with huge fluctuations of Coulombic efficiency, even during the second cycle (Figure S6b). In addition, the discharge−charge profiles are also well retained at different current densities in the CTAC-containing electrolyte, indicating superior kinetic behavior compared to that with untreated electrolyte (Figure 6d and Figure S7b). The excellent reversibility of Li−S batteries in CTACcontaining electrolyte can also be illustrated via CV measurements. As presented in Figure S8, after the initial activation, the CV curves show substantial overlap, which is consistent with the cycling performance results. To summarize, it is evident that the electrochemical performance of Li−S batteries using CTAC-containing electrolyte, including the utilization of active materials, cycle performance, rate performance, and Coulombic efficiency, shows dominant advantages over that without CTAC additives. Long-term cycling performance at a current density of 1600 mA g−1 is illustrated in Figure 6e. Li−S batteries can demonstrate a long cycle life at such high current densities for up to 500 cycles using CTAC-containing electrolyte, while a rapid decay of discharge capacity can be observed after just 200 cycles with untreated electrolyte. After disassembling the battery, the electrolyte dries out (Figure S9) due to the growth of Li dendrites and the continuous consumption of electrolyte during cycle. This results in battery failure after around 300 cycles in untreated electrolyte. To further reveal the modifying mechanisms of CTAC electrolyte additives in Li−S batteries, EIS and SEM measurements were also conducted. Figure S10 illustrates the EIS results of Li anodes in Li−S batteries using electrolytes with or without CTAC additives. Before cycling, the charge transfer resistance obtained using untreated electrolytes is significantly lower than that from electrolytes containing CTAC additives (Figure S10a). This suggests that adding CTAC increases the charge transfer resistance before cycling to a certain extent regardless of the cathode and anode material. While similar to that in symmetric Li|Li cells, after 20 cycles, the SEI layer resistance (36.4 Ω) for the Li anode in electrolyte with CTAC electrolyte is significantly lower than that in untreated electrolyte (85.6 Ω). Similar results are also obtained for pPAN@S cathodes (Figure S11). Figure S12 is SEM images of a Li anode from a Li−S battery, after 500 cycles, at a current density of 1600 mA g−1. The top surface of Li foil still exhibits a relatively smooth surface without obvious and pronounced Li dendrites. In contrast, fractured and mossy Li can be readily observed using an untreated electrolyte. Both the above results can help to explain the enhanced electrochemical performance of Li−S batteries by using an electrolyte with CTAC additives, even when the added amount of electrolyte is relatively small. To further verify the applicability of CTAC additives, Li| LiNi0.5Co0.2Mn0.3O2 full cells were also assembled, and the results are given in Figure 7. As presented in Figure 7a, although the initial charge−discharge capacity in blank electrolyte is slightly higher than that in electrolyte with CTAC additives, the full cells in electrolyte with CTAC additives show much better rate performance than that in blank electrolyte after the first cycle. With increasing the rate to 5C, a high discharge capacity of 102.1 mAh g−1 still can be obtained, as well as well-retained voltage profiles shown in Figure 7b. The long cycling performance at the rate of 2C is given in Figure 7c. It is obvious that the full cell in electrolyte with CTAC additives presents an ultrastable cycling perform-
Figure 6. Cycling performance of pPAN@S composite cathodes tested at various current densities with either (a) 30 or (c) 15 μL of electrolyte. Discharge−charge profiles of pPAN@S composite cathodes tested at various current densities with either (b) 30 or (d) 15 μL of electrolyte. (e) Long-term cycling behavior of the pPAN@S composite cathode, tested at a current density of 1600 mA g−1, with or without CTAC additives.
remain relatively stable at 1442.3 (10th), 1400.4 (20th), 1352.8 (30th), and 1256.2 (40th) mAh g−1 at increasing current densities of 400, 800, 1600, and 3200 mA g−1, respectively, and the Coulombic efficiency is also retained above 99.1% after the initial cycle. In particular, when the current density returns to 1600, 800, and 400 mA g−1, the discharge capacity is shown to recover well; that is, after 100 cycles, a peak discharge capacity of 1397.2 mAh g−1 can still be obtained at the current density of 400 mA g−1, indicating improved reversibility of sulfur cathodes in electrolyte with CTAC additives.34−37 In contrast, composite cathodes can show a higher initial discharge capacity in untreated electrolyte, but this drops quickly to 198.3 mAh g−1 at a current density of 3200 mA g−1. Although the discharge capacity can be recovered as the current density is lowered, the composite cathodes demonstrate poor cycle performance with fluctuating Coulombic efficiency (Figure S6a). Figure 6b illustrates the discharge−charge profiles of the pPAN@S cathodes in electrolyte with CTAC additives, and it can be observed that the potential plateau is well retained even at the high current density of 3200 mA g−1, while for untreated electrolyte, the significant potential gap between charge/discharge plateaus can be clearly observed with increasing of current density (Figure S7a). For Li−S batteries, an excess of electrolyte is often desired in order to enhance the discharge and cycle performance, but this also results in a decrease in energy density.38−40 Accordingly, we reduced the amount of electrolyte to 15 μL and evaluated the electrochemical behavior of pPAN@S composite cathodes in electrolytes containing CTAC additives. As shown in Figure 6c, a slight decrease of discharge capacity can be observed, but high stable cycle performance and Coulombic efficiency are still well retained. However, the Li−S batteries prepared with 17519
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
Article
Journal of the American Chemical Society
reducing the consumption of electrolyte during cycling can be verified effectively in Li−S batteries, as batteries can present a long cycle life even by adding a relatively small amount of electrolyte (Figure 6e). (v) After the reconstruction of smooth Li surfaces, CTAC molecules can rearrange on the surface of the Li anode as in the initial step. In addition, they can also reversibly break off into the electrolyte, due to the electrostatic repulsion forces that occur during the stripping process, thus ensuring long-term cycle life.
■
CONCLUSION In summary, we have proposed an additive that promotes a lithiophobic repulsion mechanism in order to suppress the growth of Li dendrites, and CTAC was selected for such design to achieve uniform Li deposition during the electrochemical process. During the plating process, CTAC additives create a lithiophobic protective layer on the surface of a protuberance, blocking Li decomposition in these regions and contributing to enhanced electrochemical performance. In symmetric Li|Li cells, the Li anode demonstrated high cycle stability for 300 h at a current density of 1 mA cm−2 and excellent rate performance up to 4 mA cm−2. When applied to Li−S batteries, it demonstrates stable cycling and high rate performance. At a current density of 3200 mA g−1, assembled Li−S batteries can deliver a peak discharge capacity of 1256.2 mAh g−1. In particular, even when the amount of electrolyte is reduced to 15 μL, it can still present a high reversible capacity (962.8 mAh g−1 at 3200 mA g−1) and long cycling life (up to 500 cycles). Meanwhile, enhanced cycle and rate performance also can be obtained in the Li|LiNi0.5Co0.2Mn0.3O2 full cells using electrolyte with CTAC additives. Our work not only illustrates an efficiency mechanism for suppressing dendritic lithium growth but also introduces effective additives in the form of cation surfactants for the production of safer and longer life Li-based batteries.
Figure 7. (a) Rate performance of Li|LiNi0.5Co0.2Mn0.3O2 full cells with or without CTAC additives. (b) Voltage profiles of Li| LiNi0.5Co0.2Mn0.3O2 full cell in electrolyte with CTAC additives. (c) Cycling performance of Li| LiNi0.5Co0.2Mn0.3O2 full cells with or without CTAC additives. 1C = 270.0 mAh g−1.
ance up to 300 cycles with a high discharge capacity of 127.2 mAh g−1, while fast capacity decay can be observed in blank electrolyte just after 100 cycles. Based on all the above results, it can be concluded that CTAC is a highly efficient electrolyte additive to inhibit the growth of Li dendrites and can be easily applied in practical metal Li-based batteries. Also, it should be noted that the electrochemical stability and reversible adsorption−desorption of CTAC additives on the surface of lithium anodes are key for such protective mechanisms, and this can be illustrated via electrochemical measurements of bare electrodes in Figure S13, SEM-EDS analysis in Figure S14, and differential capacitance−potential curves in Figure S15, respectively. As presented in Figure S13, the bare electrode in different electrolytes shows a similar cycle and discharge performance, indicating the electrochemical stability of CTAC during the discharge−charge process in full cells. In addition, there is basically no elemental nitrogen that is detected on the Li surface shown in Figure S14, indicating that the CTA+ does not participate in the reaction process of the SEI film. For differential capacitance−potential curves, it is apparent that the differential capacitance obtained using CTAC additives is much lower than that from untreated electrolytes in negative polarization, as the surfactant ions replace the Li+ to produce a thicker ionic layer.41−43 For positive polarization, a capacitance peak occurs at 0.28 V, which is conventionally attributed to the desorption of adsorbed additives from the lithium anode.41−43 Accordingly, the detailed role of CTAC during the Li plating process can be classified as follows: (i) CTAC is adsorbed as individual ions to form a first layer on the surface of the Li foil through electrostatic forces; (ii) when protuberances are formed during the plating process, adsorption increases dramatically around the tip regions, and “hemimicelles” can form through lithiophobic interactions between the hydrocarbon chains in the polar electrolyte; (iii) during further plating, solvated Li+ is unable to transport through the lithiophobic regions, and the positively charged shield of inter polar groups can also act as a second obstacle to suppress the reduction reaction of Li+ on the protuberant tips; (iv) the dense lithiophobic protective layer can block the contact between the protuberances and the polar solvent, reducing the side reactions between solvent and fresh Li. The role of
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.8b08963.
■
Instruments of the optical microscopy, viscosity, and contact angle of modified electrolytes, fitting results of the EIS, cycle performance of Li−Li cells with different concentrations of additives, charge−discharge curves and Coulombic efficiency of Li−S battery in blank electrolyte, CV curves, EIS analysis, SEM images of lithium anode after cycling, cycle curves of bare electrode, differential capacitance curves (PDF)
AUTHOR INFORMATION
Corresponding Author
*
[email protected] ORCID
Kai Xi: 0000-0003-0508-7910 Chao Lai: 0000-0002-6021-6343 Shanqing Zhang: 0000-0001-5192-1844 Author Contributions ‡
H. Dai and K. Xi contributed equally.
Notes
The authors declare no competing financial interest. 17520
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
Article
Journal of the American Chemical Society
■
(31) Ding, M. S.; Cresce, A.; Xu, K. J. Phys. Chem. C 2017, 121, 2149. (32) Yan, J.; Wang, D.; Bu, F.; Yang, F. F. J. Solution Chem. 2010, 39, 1501. (33) Sepulveda, L.; Cortés, J. J. Phys. Chem. 1985, 89, 5322. (34) Zhang, Z.; Kong, L.-L.; Liu, S.; Li, G.-R.; Gao, X.-P. Adv. Energy Mater. 2017, 7, 1602543. (35) Zhou, T. H.; Lv, W.; Li, J.; Zhou, G. M.; Zhao, Y.; Fan, S. X.; Liu, B. L.; Li, B. H.; Kang, F. Y.; Yang, Q.-H. Energy Environ. Sci. 2017, 10, 1694. (36) Xing, L.-B.; Xi, K.; Li, Q. Y.; Su, Z.; Lai, C.; Zhao, X. S.; Kumar, R. V. J. Power Sources 2016, 303, 22. (37) Hu, G. J.; Sun, Z. H.; Shi, C.; Fang, R. P.; Chen, J. P.; Hou, X.; Liu, C.; Cheng, H.-M.; Li, F. Adv. Mater. 2017, 29, 1603835. (38) Hagen, M.; Fanz, P.; Tübke, J. J. Power Sources 2014, 264, 30. (39) Fang, R. P.; Zhao, S. Y.; Sun, Z. H.; Wang, D.-W.; Cheng, H.M.; Li, F. Adv. Mater. 2017, 29, 1606823. (40) Cleaver, T.; Kovacik, P.; Marinescu, M.; Zhang, T.; Offer, G. J. Electrochem. Soc. 2018, 165, A6029. (41) Costa, R.; Pereira, C. M.; Silva, F. RSC Adv. 2013, 3, 11697. (42) Lecœur, J.; Rousset, S. J. Electroanal. Chem. 2002, 519, 18. (43) Lockett, V.; Sedev, R.; Ralston, J. J. Phys. Chem. C 2008, 112, 7486.
ACKNOWLEDGMENTS This work is financially supported by the Key Research and Development Program of Xuzhou (KC17004) and National Natural Science Foundation of China (Nos. 51871113 and 51572116). We thank Prof. R. Vasant Kumar (University of Cambridge) for his contribution to the language revision.
■
REFERENCES
(1) Cheng, X.-B.; Zhang, R.; Zhao, C.-Z.; Zhang, Q. Chem. Rev. 2017, 117, 10403. (2) Lin, D. C.; Liu, Y. Y.; Cui, Y. Nat. Nanotechnol. 2017, 12, 194. (3) Guo, Y. P.; Li, H. Q.; Zhai, T. Y. Adv. Mater. 2017, 29, 1700007. (4) Noorden, R. V. Nature 2014, 507, 26. (5) Carbone, L. S.; Greenbaum, G.; Hassoun, J. Sustain. Energy Fuels 2017, 1, 228. (6) Shen, X.; Liu, H.; Cheng, X.-B.; Yan, C.; Huang, J.-Q. Energy Storage Materials 2018, 12, 161. (7) Duan, H.; Yin, Y.-X.; Shi, Y.; Wang, P.-F.; Zhang, X.-D.; Yang, C.-P.; Shi, J.-L.; Wen, R.; Guo, Y.-G.; Wan, L.-J. J. Am. Chem. Soc. 2018, 140, 82. (8) Liu, K.; Pei, A.; Lee, H. R.; Kong, B.; Liu, N.; Lin, D.; Liu, Y. Y.; Liu, C.; Hsu, P.-C.; Bao, Z.; Cui, Y. J. Am. Chem. Soc. 2017, 139, 4815. (9) Liu, Y. Y.; Lin, D. C.; Yuen, P. Y.; Liu, K.; Xie, J.; Dauskardt, R. H.; Cui, Y. Adv. Mater. 2017, 29, 1605531. (10) Liu, Y. Y.; Cui, Y. Joule 2017, 1, 643. (11) Zhang, H.; Eshetu, G. G.; Judez, X. C.; Li, L.; RodriguezMartínez, M.; Armand, M. Angew. Chem., Int. Ed. 2018, 57, 15002. (12) Yang, C.-P.; Yin, Y.-X.; Zhang, S.-F.; Li, N.-W.; Guo, Y.-G. Nat. Commun. 2015, 6, 8058. (13) Zhang, R.; Chen, X.; Shen, X.; Zhang, X.-Q.; Chen, X.-R.; Cheng, X.-B.; Yan, C.; Zhao, C.-Z.; Zhang, Q. Joule 2018, 2, 764. (14) Yang, C.; Yao, Y. G.; He, S. M.; Xie, H.; Hitz, E.; Hu, L. B. Adv. Mater. 2017, 29, 1702714. (15) Liu, L.; Yin, Y.-X.; Li, J.-Y.; Wang, S.-H.; Guo, Y.-G.; Wan, L.-J. Adv. Mater. 2018, 30, 1706216. (16) Yan, K.; Sun, B.; Munroe, P.; Wang, G. X. Energy Storage Materials 2018, 11, 127. (17) Ding, F.; Xu, W.; Graff, G. L.; Zhang, J.; Sushko, M. L.; Chen, X.; Shao, Y. Y.; Engelhard, M. H.; Nie, Z.; Xiao, J.; Liu, X. J.; Sushko, P. V.; Liu, J.; Zhang, J.-G. J. Am. Chem. Soc. 2013, 135, 4450. (18) Ding, F.; Xu, W.; Chen, X.; Zhang, J.; Shao, Y. Y.; Engelhard, M. H.; Zhang, Y. H.; Blake, T. A.; Graff, G. L.; Liu, X. J.; Zhang, J.-G. J. Phys. Chem. C 2014, 118, 4043. (19) Yoo, D.-J.; Kim, K. J.; Choi, J. W. Adv. Energy Mater. 2018, 8, 1702744. (20) Zhao, J.; Liao, L.; Shi, F.; Lei, T.; Chen, G.; Pei, A.; Sun, J.; Yan, K.; Zhou, G.; Xie, J.; Liu, C.; Li, Y.; Liang, Z.; Bao, Z.; Cui, Y. J. Am. Chem. Soc. 2017, 139, 11550. (21) Xue, P.; Liu, S.; Shi, X.; Sun, C.; Lai, C.; Zhou, Y.; Sui, D.; Chen, Y.; Lian, J. Adv. Mater. 2018, 30, 1804165. (22) Liu, W.; Lin, D.; Pei, A.; Cui, Y. J. Am. Chem. Soc. 2016, 138, 15443. (23) Lopez, J.; Pei, A.; Oh, J. Y.; Wang, G.-J. N.; Cui, Y.; Bao, Z. J. Am. Chem. Soc. 2018, 140, 11735. (24) Zhu, B.-Y.; Gu, T. Adv. Colloid Interface Sci. 1991, 37, 1. (25) Somusundurun, P.; Shrolri, S.; Huung, L. Pure Appl. Chem. 1998, 70, 621. (26) Wang, J. L.; Yang, J.; Xie, J. Y.; Xu, N. X.; Li, Y. Electrochem. Commun. 2002, 4, 499. (27) Wang, J. L.; Yang, J.; Xie, J. Y.; Xu, N. X. Adv. Mater. 2002, 14, 963. (28) Ogihara, N.; Itou, Y.; Sasaki, T.; Takeuchi, Y. J. Phys. Chem. C 2015, 119, 4612. (29) Petibon, R.; Aiken, C. P.; Sinha, N. N.; Burns, J. C.; Ye, H.; VanElzen, C. M.; Jain, G.; Trussler, S.; Dahn, J. R. J. Electrochem. Soc. 2013, 160, A117. (30) Hwang, J.-Y.; Kim, H. M.; Sun, Y.-K. J. Electrochem. Soc. 2018, 165, A5006. 17521
DOI: 10.1021/jacs.8b08963 J. Am. Chem. Soc. 2018, 140, 17515−17521
