Suppression of Lithium Dendrite Formation by ... - ACS Publications

Research Article www.acsami.org

Suppression of Lithium Dendrite Formation by Using LAGP-PEO (LiTFSI) Composite Solid Electrolyte and Lithium Metal Anode Modified by PEO (LiTFSI) in All-Solid-State Lithium Batteries Chunhua Wang,† Yifu Yang,*,† Xingjiang Liu,*,‡ Hai Zhong,‡ Han Xu,‡ Zhibin Xu,‡ Huixia Shao,† and Fei Ding‡ †

College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072, P.R. China National Key Lab of Power Sources, Tianjin Institute of Power Sources, Tianjin 300384, P.R. China

‡

S Supporting Information *

ABSTRACT: The formation of lithium dendrites is suppressed using a Li1.5Al0.5Ge1.5(PO4)3−poly(ethylene oxide) (LAGP-PEO) composite solid electrolyte and a PEO (lithium bis(trifluoromethane)sulfonimide) [PEO (LiTFSI)]-modified lithium metal anode in all-solid-state lithium batteries. The effects on the anode performance based on the PEO content in the composite solid electrolyte and the molecular weight of PEO used to modify the Li anode are studied. The structure, surface morphology, and stability of the composite solid electrolyte are examined by X-ray diffraction spectroscopy, scanning electron microscopy, and electrochemical tests. Results show that the presence of a PEO-500000(LiTFSI) film on a Li anode results in good mechanical properties and satisfactory interface contact features. The film can also prevent Li from reacting with LAGP. Furthermore, the formation of lithium dendrites can be effectively inhibited as the composite solid electrolyte is combined with the PEO film on the Li anode. The ratio of PEO in the composite solid electrolyte can be reduced to a low level of 1 wt %. PEO remains stable even at a high potential of 5.12 V (vs Li/Li+). The assembled Li-PEO (LiTFSI)/LAGP-PEO/LiMn0.8Fe0.2PO4 all-solid-state cell can deliver an initial discharge capacity of 160.8 mAh g−1 and exhibit good cycling stability and rate performance at 50 °C. KEYWORDS: lithium metal anode, lithium dendrites, PEO (LiTFSI), composite solid electrolyte, all-solid-state lithium battery

■

INTRODUCTION Lithium (Li) metal is considered as the most desired anode for next-generation rechargeable batteries because of its high theoretical specific capacity of 3860 mAh g−1 and lowest electrode potential (−3.040 V vs standard hydrogen electrode). 1−4 However, Li dendrite growth and low Coulombic efficiency during iterative Li deposition/stripping cycles are two major hurdles to the practical application of Limetal-based energy-storage systems.5,6 In particular, undesired mossy Li generation or Li dendrite growth can expedite the capacity fade of the battery and induce many serious security issues (e.g., fire, burning, and explosion) during repeated Li deposition/stripping cycles when using liquid electrolytes.7 One promising solution to these issues is the design of all-solidstate batteries using solid electrolytes in place of liquid ones. Solid electrolytes exhibit higher melting point and a nonflammable property. Over the past decades, substantial research has been conducted on solid-state electrolytes, such as NASICON-type Li+ ion conductors with the general formula LiM2(PO4)3 (M = Sr, Ti, Ge, Zr, Sn, etc.). Among the discovered solid-state electrolytes, Li1.5Al0.5Ge1.5(PO4)3 (LAGP) conductive micro© 2017 American Chemical Society

crystalline glass is regarded as a desirable candidate for inorganic solid electrolyte materials. The material exhibits high Li+ ion conductivity, excellent chemical stability, broad applicable temperature range, and low cost.8,9 Li+ ion conductive microcrystalline glass is fabricated by crystallizing its basic glass at high temperature such that the deformation easily occurs during annealing.10,11 Unfortunately, many obvious flaws must be improved to realize the practical application of LAGP in all-solid-state Li batteries. First, a dense and uniform electrolyte membrane is difficult to produce directly using LAGP raw material. Second, the solid LAGP material shows poor interface contact performance with electrodes. Finally, LAGP is unstable during contact with the Li metal.12 When poly(ethylene oxide) (PEO) is mixed with Li salt, such as Li[N(SO2CF3)2] (LiTFSI), the PEO can form a polymer Li+ ion conductor.13−15 PEO (LiTFSI) electrolyte shows excellent electrochemical stability. Furthermore, PEO (LiTFSI) electroReceived: January 8, 2017 Accepted: March 23, 2017 Published: March 23, 2017 13694

DOI: 10.1021/acsami.7b00336 ACS Appl. Mater. Interfaces 2017, 9, 13694−13702

Research Article

ACS Applied Materials & Interfaces

Figure 1. All-solid-state Li-PEO (LiTFSI)/LAGP-PEO (LiTFSI)/LiMFP cells.

lyte is also flexible and can produce a thin film.16 Therefore, this polymer Li+ ion conductor can combine with LAGP to form an inorganic/organic composite solid electrolyte. Thus, the filmforming performance and interface contact of LAGP with electrodes can be improved.17,18 However, the PEO content is exceedingly high (20 wt %) in the reported composite solid electrolyte,18 such that the cutoff voltage of charging needs to be set lower than the decomposition voltage of PEO (4.3 V).19 In another aspect, the cutoff voltage of charging for the majority of cathode materials equals to or exceeds 4.3 V. Therefore, the high PEO content in the composite solid electrolyte has limited its application. To broaden the stable electrochemical window of the composite solid electrolyte, the PEO content in the composite solid electrolyte has been successfully reduced to 1 wt % in our laboratory. However, by simply reducing the PEO content of the composite solid electrolyte, the composite solid electrolyte cannot guarantee a good contact between the electrolyte and Li anode surface. Accordingly, we have attempted to coat a thin layer of PEO (LiTFSI) polymer Li+ ion conductor on Li anode. The coating layer of PEO (LiTFSI) can form a bridge connecting the LAGP solid electrolyte with the Li anode but preventing the direct contact of LAGP from Li metal. In this case, the side reaction between LAGP and Li anode can be avoided.12 On the basis of these works, we proposed in the present paper a simple method of suppressing Li dendrites by using PEO (LiTFSI)-modified Li anode/composite solid electrolyte interface in all-solid-state Li batteries. The effect of the PEO molecular weight used to modify Li surface upon Li deposition was studied.

■

97:3, and 95:5. On the basis of the PEO content in the mixture, the samples were designated as LAGP-PEO1, LAGP-PEO3, and LAGPPEO5, respectively. Upon achieving satisfactory homogenization of the mixture, the solution was cast onto a flat Teflon plate with a doctor blade, and the solvent was slowly evaporated in an argon-filled glovebox. Residual solvent was completely removed by heating the electrolyte at 70 °C. Afterward, the composite solid electrolyte was cut into the required shape and size. Electrode Preparation and Cell Assembly. Home-prepared LiMn0.8Fe0.2PO4 (LiMFP) was used as the active material for the positive electrode as described in the literature.22 Thin-film positive electrodes were prepared by casting onto an aluminum current collector a THF slurry of the LiMFP active material (70 wt %) mixed with carbon (Super P, MMM Carbon) conductive additive (10 wt %) and PEO-500000 (LiTFSI)/THF solution (20 wt %). PEO (LiTFSI) was used as an ionic conductor as well as a polymer binder in the positive electrode. The electrode was vacuum-dried at 70 °C for 24 h. The mass loading of the cathode material is 2.16 mg cm−2. The Li negative electrode consisted of a Li foil pressed on a stainless-steel current collector and coated with a thin layer of PEO (LiTFSI) film on the surface. A certain amount of PEO (LiTFSI)/ THF solution was cast on Li negative electrodes with a doctor blade, followed by solvent evaporation at 70 °C. In the preparation of the surface film, the thickness can be well controlled at about 100 nm by accurate controlling of the volume of the PEO (LiTFSI)/THF solution with certain concentrations on unit area of Li electrode. Finally, all-solid-state Li-PEO (LiTFSI)/LAGP-PEO (LiTFSI)/LiMFP cells were assembled to a 2430 coin-type cell in an argon-filled glovebox. The schematic illustration of the process is depicted in Figure 1. Characterization of the Composite Solid Electrolyte. X-ray diffraction (XRD) patterns of the solid electrolyte materials were obtained using an X-ray diffractometer (RigakuM2400) with Cu Kα radiation. The morphologies of the electrolyte materials were examined by scanning electron microscopy (SEM, S-4800, Hitachi, Japan). The mechanical properties of the PEO (LiTFSI) polymer membranes (length × width: 6 × 15 mm) were tested using a universal testing machine at room temperature. Electrochemical Tests. A symmetrical stainless-steel/electrolyte/ stainless-steel cell was assembled to test the ionic conductivity by using a two-electrode AC impedance method. The measurements were carried out using a PARSTAT 2273 impedance analyzer in a frequency range from 10 Hz to 1 MHz with an amplitude of 10 mV within a temperature range of 30 to 90 °C. Linear sweep voltammetry (LSV) experiments were performed using a three-electrode cell in which a stainless-steel plate served as the working electrode and metallic Li as the counter and reference electrodes. The experiments were conducted to investigate the electrochemical stability of the composite solid electrolytes. The potential was scanned at a rate of 0.1 mV s−1 and 50 °C. LSV experiments were also performed to explore the effect of PEO (LiTFSI) films modified in the interface of LAGP-PEO1 and stainless steel on the Li+ reduction potential at 50 °C. In this test, metallic Li was used as counter and reference electrodes, and the potential of the working electrode was scanned at a rate of 5.0 mV s−1. Meanwhile, cyclic voltammetry (CV) was performed using a threeelectrode cell, where LiMFP acted as the working electrode and the Li metal as the counter and reference electrodes. The electrochemical performance of the Li-PEO (LiTFSI)/LAGP-PEO (LiTFSI)/LiMFP cells were tested using battery-testing equipment (LAND CT2001A,

EXPERIMENTAL SECTION

Preparation of the PEO (LiTFSI) Solution. PEO with different molecular weights (Mn = 6000, 100000, 500000, and 900000, hereafter denoted as PEO-6000, PEO-100000, PEO-500000, and PEO-900000, respectively) and LiTFSI were purchased from J&K Scientific Ltd. The salt was vacuum-dried at 100 °C for 24 h before use. PEO and LiTFSI were dissolved in tetrahydrofuran (THF) at a molar ratio of EO:Li = 8:1,20 and the solution was stirred at 70 °C for 12 h in a sealed container before use. Preparation of the Composite Solid Electrolyte. Stoichiometric amounts of lithium carbonate, aluminum oxide, germanium oxide, and ammonium dihydrogen phosphate were employed as raw materials to prepare LAGP through a high-temperature solid-statefusion method.21 Raw materials were mixed by planetary ball milling at 450 rpm for 12 h. The materials were then heated in an alumina crucible at 700 °C for 2 h to release the gaseous products that evolved from the raw materials, including carbon dioxide, ammonia, and vapor. Subsequently, the temperature was elevated to 1450 °C and kept for 2 h to produce the melt. The melt was then quenched on a stainlesssteel mold and yielded a transparent glass pellet. The pellet was annealed at 950 °C for 12 h to ensure the formation of glass ceramic LAGP. The glass ceramic was ground into powder and sieved with 300-mesh sieves. The LAGP powder and PEO-500000(LiTFSI)/THF solution were mixed uniformly in separate solid proportions of 99:1, 13695


Research Article


Figure 2. XRD patterns of electrolytes. (a) PEO (LiTFSI) polymer with different PEO molecular weights. (b) LAGP-PEO (LiTFSI) composite solid electrolytes with different PEO contents.

Table 1. Peak fwhm of Samples at 14.7° and 25.3° samples

PEO-6000(LiTFSI)

PEO-100000(LiTFSI)

PEO-500000(LiTFSI)

PEO-900000(LiTFSI)

LAGP-PEO1

LAGP-PEO3

LAGP-PEO5

fwhm (deg)

0.081

0.173

0.210

0.274

0.161

0.173

0.183

China) within a voltage range of 2.5 to 4.5 V. All cells were laid aside for 2 h at the required temperature before measurement.

■

RESULTS AND DISCUSSION Characteristics of the Composite Solid Electrolyte. The XRD patterns of PEO (LiTFSI) with different PEO molecular weights and those of LAGP-PEO (LiTFSI) with different PEO contents are presented in Figure 2. Increased molecular weight of PEO caused the diffraction peak position of PEO (LiTFSI) to shift slightly toward the lower angle direction, and peak intensity was reduced (Figure 2a). These findings indicate that the degree of crystallinity of PEO (LiTFSI) depends on the PEO molecular weight. The correspondence of the full width at half-maximum (fwhm) of the diffraction peak at 14.7° with different PEO molecular weights is summarized in Table 1. The results in Figure 2a also show that the crystallinity of PEO (LiTFSI) decreases with increased PEO molecular weight. Figure 2b presents the XRD patterns of LAGP-PEO (LiTFSI) with different PEO-500000 contents. The position of all diffraction peaks of LAGP-PEO (LiTFSI) does not move with increased PEO content, but peak strength decreases with augmented PEO content. The fwhm of the diffraction peak at 25.3° increases with elevated PEO content (Table 1). In addition, the peaks of PEO (LiTFSI) and LiTFSI are not observed in the patterns of all composite LAGP-PEO (LiTFSI) electrolytes. It is proposed23−27 that this phenomenon is due to the fact that LAGP, as the inorganic filler, can weaken the crystalline formation of PEO (LiTFSI) and that PEO (LiTFSI) with amorphous phase is the desired one. Thus, it can be speculated that the structure of the LAGPPEO (LiTFSI) composite electrolyte is composed of LAGP particles with a PEO-500000(LiTFSI) amorphous layer in between. As is well-known, the conductivity of PEO (LiTFSI) is very poor,13−15 so the PEO (LiTFSI) content in the composite solid electrolyte should be controlled to be as low as possible. The SEM images of composite electrolytes with different PEO500000 contents are presented in Figure 3. It can be seen from Figure 3a that a pure LAGP slice consists of glass and ceramic phases. Figure 3b reveals that a dense and uniform electrolyte membrane was obtained with a PEO-500000(LiTFSI) content

Figure 3. SEM images of the LAGP slice and the composite solid electrolytes: (a) LAGP slice, (b) LAGP-PEO1, (c) LAGP-PEO3, (d) LAGP-PEO5.

of 1 wt % with some holes fully filled by PEO-500000(LiTFSI). It can be speculated that the amorphous PEO-500000(LiTFSI) formed a transition phase as a bridge between the LAGP particles. A PEO-500000(LiTFSI) thin layer was coated onto the surface of LAGP particles with increased PEO-500000(LiTFSI) content to 3 wt % (Figure 3c). With further increased PEO-500000(LiTFSI) content to 5 wt %, the LAGP particles were dispersed into the PEO-500000(LiTFSI) material phase (Figure 3d). This result means that LAGP particles were separated by PEO-500000(LiTFSI). Given that the conductivity of PEO (LiTFSI) was much lower than that of LAGP (Figure 4a), the optimal content of PEO-500000(LiTFSI) was preliminarily considered to be 1 wt %. However, this value must be verified by electrochemical tests. The relationships of the ionic conductivity of LAGP, PEO (LiTFSI), and the LAGP-PEO (LiTFSI) composite solid electrolytes with temperature are shown in Figure 4a. Phase transition between the crystalline and amorphous phases of pure PEO (LiTFSI) membrane was observed between 30 and 13696


Research Article


direction of Li+ transport. The Li+ ion conduction in the composite solid electrolyte flows along the channel with minimum migration activation energy. Generally, given that the ionic conductivity of the LAGP crystal phase was much greater than that of the PEO (LiTFSI) amorphous phase, the Li+ ion conduction of the composite solid electrolyte mainly depended on the fast ionic conduction of LAGP particles.18 Particularly, ion transport likely prefers the crossing of LAGP particles and the portion with the lowest thickness of the PEO (LiTFSI) layer in between LAGP particles. These conditions provide the lowest ion transport resistance. Therefore, LAGP-PEO1(LiTFSI) understandably holds the highest ion conductivity among the composite solid electrolytes. Performance of the Composite Solid Electrolyte. The electrochemical stability of the LAGP-PEO1(LiTFSI) composite solid electrolytes was evaluated by LSV measurements at 50 °C, and the results are shown in Figure 5. In the cathodic scan (Figure 5a), a reductive current occurred around 0 V (vs Li/ Li+), which corresponded to the reductive deposition of Li+ ion onto the electrode.18 During anodic scan (Figure 5b), an oxidative current began to increase at around 5.12 V (vs Li/ Li+), which was associated with the oxidative decomposition of PEO-500000. 18,28 The decomposition potential of the composite solid electrolyte was higher than that of pure PEO (Figure S1), indicating that LAGP can improve the electrochemical stability of the composite solid electrolyte.18 Both LAGP and PEO are polar molecules. Thus, a dipole−dipole interaction exists between the two molecules. The dipole interaction can change the electron transition energy level of PEO to elevate the oxidation decomposition potential of PEO. In the composite electrolyte the majority of the component is LAGP, and this can significantly affect the behavior of PEO. Furthermore, LAGP also functioned as a lithium ionic conductor, without degradation. This can make the electrolyte be more stable.29−31 Given these results, it is expected that the composite solid electrolyte involves an electrochemical operation window broader than that of the PEO-based polymer electrolyte for Li batteries. To further confirm the effectiveness of LAGP on suppressing Li dendrite formation, Li/LiMFP cells with two different electrolytes were compared with charge and discharge cycling at 0.1 C current and 50 °C. The discharge capacity of the cell with the PEO-500000(LiTFSI) electrolyte significantly decayed during cycling (Figure 6). Sun and co-authors pointed out that the good stability was seen at the PEO (LiTFSI) polymer electrolyte/cathode interface.32 It showed that rapid attenu-

Figure 4. (a) Relationship of the ionic conductivity of the solid electrolytes with temperature. (b) Li+ ion transport mechanism in the composite solid electrolyte with different contents of PEO (LiTFSI): (b-1) LAGP-PEO1, (b-2) LAGP-PEO3, (b-3) LAGP-PEO5.

70 °C, and this transition did not follow the classical Arrhenius equation. Furthermore, the ionic conductivity of PEO (LiTFSI) decreased with increased PEO molecular weight. However, the ionic conductivities of PEO (LiTFSI) with different molecular weights were similar at 50 °C. With increased temperature, a linear relationship was obtained between the conductivity and temperature for the LAGP ceramic plate and the LAGP-PEO (LiTFSI) composite electrolyte membranes. This relationship followed the classical Arrhenius equation. In addition, the ionic conductivities of LAGP ceramic plate and pure PEO (LiTFSI) membrane were the highest and lowest, respectively; the difference between the two values was about 2 orders of magnitude. For the LAGP-PEO (LiTFSI) composite electrolyte membrane, the ionic conductivity decreased with increased PEO-500000 content. To vividly describe the influencing mechanism of PEO (LiTFSI) addition to the conductivity of composite solid electrolytes, a schematic illustration of the conduction path of the Li+ ion in composite solid electrolytes with different PEO (LiTFSI) contents is drawn in Figure 4b. The arrow shows the

Figure 5. LSV curves of the composite solid electrolyte at 50 °C: (a) cathodic scan, (b) anodic scan (scan rate: 0.02 mV s−1). 13697


Research Article


film was about 100 nm, such that the resistance of PEO (LiTFSI) film can be ignored. These observations indicate that the contact performance of the PEO with the electrode decreased with increased PEO molecular weight. The reduction potential of the working electrode modified with PEO6000(LiTFSI) was the highest among all electrodes modified with other PEO polymers. This phenomenon may result from the partial melting of PEO-6000(LiTFSI) at 50 °C. The reduction potentials of Li+ on the working electrode modified with PEO-100000(LiTFSI) and PEO-500000(LiTFSI) were contiguous, so that their interfacial contact resistances showed only slight differences. However, both reduction potentials were lower than those of electrodes modified with PEO-6000(LiTFSI). The reduction potential on the working electrode modified with PEO-900000(LiTFSI) was the lowest among all reduction potentials obtained. The rate capabilities of solid-state Li-PEO (LiTFSI)/LAGPPEO1/LiMFP cells within the voltage range 2.5−4.5 V at 50 °C are shown in Figure 7b. The Li anodes in the test cells were modified by PEO (LiTFSI) polymer membranes with different PEO molecular weights. It can be seen that the discharge capacity of all cells at 0.1 C rate in the first cycle was almost the same, although the PEO molecular weight used to modify the Li anode varies among different cells. The reduction potentials of the electrodes modified with PEO-100000(LiTFSI) and PEO-500000(LiTFSI) were similar, such that their rate performances are basically the same (Figure 7a). However, with increased discharge rate, the discharge capacity and the voltage platform of the discharge decreased with increased PEO molecular weights modified on the Li anode. Furthermore, with increased charge rate, the voltage platform of the charge increased with increased PEO molecular weights modified on the Li anode. In addition, the resistance of the PEO (LiTFSI) film can be ignored. Therefore, this result can be explained by the worsening change of interface contact performance of PEO (LiTFSI) film with increased PEO molecular weight. This trend led to deteriorating ion-shuttle performance. The cyclic performance of all-solid-state Li-PEO (LiTFSI)/ LAGP-PEO1/LiMFP cells within the voltage range 2.5−4.5 V at 0.2 C and 50 °C is presented in Figure 7c. Although the cell with a Li anode modified with PEO-6000(LiTFSI) exhibited a higher specific capacity and better rate performance than those of the others, the decay of the former’s discharge capacity, along with the increase in cycle number, was faster than those of the other cells. This result may be explained by the good interface, poor mechanical strength, and partial melting of PEO-6000(LiTFSI) at 50 °C. As a result, the deposition of Li+ ions onto the surface of the Li anode squeezes the PEO and causes it to deform and spread to surrounding areas in the charge process. However, in the discharge process, the partially deformed PEO can no longer be recovered. This result further generates voids in the interface between the solid electrolyte and Li anode, resulting in poor contact and eventually leading to capacity fade. With increased PEO molecular weight to 100000, the discharge capacity of the test cell is basically the same as that of the Li anode modified with PEO-6000(LiTFSI). However, the decay of the discharge specific capacity is slow. This effect can be attributed to the superior mechanical strength of PEO-100000(LiTFSI) over that of PEO-6000(LiTFSI) for the Li anode. The cell with a Li anode modified with PEO-500000(LiTFSI) exhibited the best cycling performance, which is mainly attributed to its optimized mechanical strength and improved interface contact. For the cell with the

Figure 6. Discharge capacities of the Li/LiMFP cells assembled with different solid electrolytes as a function of cycle number (0.1 C, cutoff voltage: 2.5−4.5 V, 50 °C).

ation of the capacity was seen regardless of the cathode. In addition, some studies have also proved that the PEO is not able to inhibit the formation of Li dendrites.32−35 Therefore, Li dendrites can be formed on the surface of Li metal and kept growing in the cycling process, resulting in rapid capacity degradation. In other words, only PEO-500000(LiTFSI) film cannot inhibit Li dendrite formation. In contrast, the cell with LAGP-PEO1 electrolyte maintained a very stable discharge capacity for 50 cycles. Thus, Li dendrite penetration across the electrolyte phase was effectively hindered. Effect of PEO (LiTFSI) Modification of the Li Anode on Li+ Reduction. To gain deep insight into the function of PEO (LiTFSI) modification on Li anode performance, the mechanical properties of PEO (LiTFSI) polymer membranes with different PEO molecular weights were investigated, and the results are listed in Table 2. The PEO-6000(LiTFSI) Table 2. Mechanical Properties of PEO (LiTFSI) Polymer Membranes with Different PEO Molecular Weights Mr (PEO) tensile strength (MPa) elongation at break (%)

6000

100000

500000

900000

0.11 50

0.32 150

0.81 185

1.7 45

membrane clearly possesses the worst mechanical property. As expected, the tensile strength of PEO (LiTFSI) polymer membranes increases with increased PEO molecular weight. However, the elongation at break of PEO (LiTFSI) polymer membranes demonstrated a phenomenon of initial increase and then decrease with increased PEO molecular weight. The PEO500000(LiTFSI) membrane exhibited the highest elongation at break among all PEO membranes with different PEO molecular weights. To evaluate the effects of the PEO (LiTFSI) modification of the anode on the Li+ reduction reaction, a set of tests was designed with stainless-steel as the working electrode. The surface was modified with PEO (LiTFSI) of different molecular weights, and the results are shown in Figure 7a. At the same current density as for Li+ reduction, the reduction potential of the PEO-modified stainless-steel electrode decreased with increased PEO molecular weight. In another aspect, the ionic conductivities of PEO (LiTFSI) with different molecular weights were similar at 50 °C. The thickness of PEO (LiTFSI) 13698


Research Article


Figure 7. (a) LSV curves of the stainless-steel electrode modified with PEO (LiTFSI) with different molecular weight at 50 °C (scan range: −0.5− 2.0 V). (b) Rate performance of Li/LAGP-PEO1/LiMFP cells with different PEO molecular weights for anode modification: (b-1) PEO6000(LiTFSI), (b-2) PEO-100000(LiTFSI), (b-3) PEO-500000(LiTFSI), (b-4) PEO-900000(LiTFSI). (c) Cycle performance of Li-PEO (LiTFSI)/LAGP-PEO1/LiMFP cells (cutoff voltage: 2.5−4.5 V, 50 °C).

Figure 8. SEM image of the fresh Li (a) and the images of Li anodes modified with PEO (LiTFSI) polymer membranes with different PEO molecular weights after 200 cycles (b−e): (b) PEO-6000(LiTFSI), (c) PEO-100000(LiTFSI), (d) PEO-500000(LiTFSI), (e) PEO900000(LiTFSI). (f) SEM image of the Li anode with PEO-500000(LiTFSI) as electrolyte after 50 cycles.

Li anode modified with PEO-900000(LiTFSI), the cycling performance was inferior to those of the others. Although it presents good mechanical properties, the interface contact was very poor. After 120 cycles, the discharge capacity sharply dropped. Hence, it can be conjectured that Li dendrites formed, and that they have pierced the PEO film during the charge and discharge cycles. Overall, PEO-500000(LiTFSI) film as the transition layer prevented Li from reacting with LAGP and formed a satisfactory connection between the Li anode and LAGP-PEO composite electrolyte. We then analyzed the behavior of the PEO-modified Li anode with different PEO molecular weights, as well as the differences in Li/LiMFP cells with LAGP-PEO (LiTFSI) and PEO-500000(LiTFSI) electrolytes. To achieve this goal, Li metal anodes were disassembled from Li/LiMFP cells after cycling within the voltage range 2.5−4.5 V. These anodes were

then washed for characterization. Figure 8 shows the SEM images of the Li metal anode under different conditions. Although the composite electrolyte was used, a large number of Li dendrites were retained on the anode surface when PEO6000 was used to modify the Li anode (Figure 8b). Consequently, the cycle performance was attenuated (Figure 7c). With increased PEO molecular weight to 100000 (Figure 8c), a few dead Li were still observed, which resulted in a slow attenuation of the cycle performance. Furthermore, Li dendrites piercing the PEO film were distinctly observed on the Li anode modified with PEO-900000(LiTFSI) (Figure 8e). This effect may be explained by the poor interface contact, which resulted in the loss of the activity of some Li sites. Moreover, Li dendrites were formed in the long charge/discharge process. Interestingly, almost no dead Li was noted on the Li anode surface modified with PEO-500000(LiTFSI). This observation 13699


Research Article


Figure 9. (a) Charge/discharge curves. (b) Rate performance of Li-PEO-500000(LiTFSI)/LAGP-PEO1/LiMFP cell (cutoff voltage: 2.5−4.5 V, 50 °C). (c) CV curve of LiMFP at a scan rate of 0.1 mV s−1 in 2.5−4.75 V at 50 °C.

correspond well with the charge−discharge plateaus shown in Figure 9a. Therefore, the satisfactory cyclic performance of the Li-PEO-500000(LiTFSI)/LAGP-PEO1/LiMFP cell was mainly supported by the excellent composite solid electrolyte and the modification of PEO-500000(LiTFSI) on Li anode.

clearly confirms that the battery with the Li anode modified with PEO-500000(LiTFSI) possessed good mechanical properties and interface contact during the charge/discharge process. In comparison, the image of the Li anode with PEO500000(LiTFSI) as electrolyte is shown in Figure 8f. A large number of Li dendrites are present in the figure, and these dendrites induced the attenuation of cycle performance. The findings suggest that simply using PEO-500000(LiTFSI) as electrolyte cannot inhibit the Li dendrite formation. However, PEO-500000(LiTFSI) film as the transition layer can suppress Li dendrite formation in combination with the LAGP-PEO1 composite solid electrolyte. Electrochemical Performances of LiMFP in All-SolidState Batteries. The cycling performance of the all-solid-state Li-PEO-500000(LiTFSI)/LAGP-PEO1/LiMFP cell within the voltage range 2.5−4.5 V at 0.2 C rate and 50 °C is displayed in Figure 9a. The cell presents two charge voltage plateaus at around 3.6 and 4.23 V, corresponding to the oxidation of Mn2+/Mn3+ and Fe2+/Fe3+, respectively.36 The cell delivered an initial discharge capacity of 161.7 mAh g−1 with a Coulombic efficiency of 92.4%. After 10 cycles, the Coulombic efficiency was increased to above 99%. The excellent cycling performance should be attributed to the good mechanical property and interface contact of the PEO-500000(LiTFSI), as well as the satisfactory characteristics of the LAGP-PEO-500000(LiTFSI) composite solid electrolyte. In addition, the cell had a very large polarization at room temperature, as shown in Figure S2. Furthermore, the cell also had an excellent cycling performance at room temperature (Figure S3). The rate discharge performance of the Li-PEO-500000(LiTFSI)/LAGP-PEO1/LiMFP cell at 50 °C is shown in Figure 9b. The cell was activated by 10 cycles at low current density before cycling, and all the charging in the rate discharge test was under 0.1 C. The discharge capacity at 1 C was still as high as 115 mAh g−1. After the rated current cycle test, the capacity at 0.1 C was recovered quickly. To verify the origin of the plateaus in Figure 8a, a CV test was conducted for the LiMFP working electrode between 3.0 and 4.75 V. The other conditions include a scan rate of 0.1 mV s−1 and Li modified with PEO-500000(LiTFSI) as counter and reference electrodes. The result is presented in Figure 9c. The CV response shows that the electrochemical Li+ extraction/ insertion process of LiMFP occurs reversibly. Two pairs of redox peaks were observed. The oxidation peaks at 3.6 and 4.23 V are attributed to the oxidation of Fe2+ to Fe3+ and Mn2+ to Mn3+, respectively. Meanwhile, the reduction peaks at 3.5 and 3.96 V are ascribed to the reduction of Fe3+ to Fe2+37 and Mn3+ to Mn2+,38 respectively. The anodic and cathodic peaks

■

CONCLUSIONS In this work, the effects of LAGP-PEO (LiTFSI) composite solid electrolyte and PEO modification of the Li anode on Li anode performance were investigated. Different molecular weights of PEO and the composition of the solid electrolytes were studied in detail. It is found that the PEO-500000(LiTFSI) film modified on the Li anode possessed good mechanical properties and interface contact between the Li anode and solid electrolyte. The composite solid electrolyte of LAGP-PEO-500000(LiTFSI) can provide satisfactory performance although the PEO content was as low as 1 wt %. The combination of the use of LAGP-PEO (LiTFSI) composite solid electrolyte and the modification of Li anode with PEO500000(LiTFSI) can effectively prevent the Li anode from dendrite growth. As such, the materials provide a satisfactory cycling performance of Li/LiMFP battery. The all-solid-state Li-PEO-500000(LiTFSI)/LAGP-PEO1/LiMFP cell delivers a high initial discharge capacity of 160.8 mAh g−1 and exhibits good cycling and rate performance at 50 °C. In addition, the Li+ ion conductive mechanism of the composite electrolyte with LAGP as the main content was proposed. It is believed that the problem of Li dendrites can be effectively solved through surface modification of the Li anode and the use of LAGP-PEO1 composite solid electrolyte for all-solid-state Li batteries.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b00336. Additional details including LSV curves and the electrochemical performance at room temperature (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Yifu Yang: 0000-0002-4949-4502 Notes

The authors declare no competing financial interest. 13700


Research Article


■

Based on a Poly(perfluoroalkylsulfonyl)-imide Ionene Polymer. J. Mater. Chem. A 2014, 2 (38), 15952−15957. (20) Nookala, M.; Kumar, B.; Rodrigues, S. Ionic Conductivity and Ambient Temperature Li Electrode Reaction in Composite Polymer Electrolytes Containing Nanosize Alumina. J. Power Sources 2002, 111 (1), 165−172. (21) Xu, X.; Wen, Z.; Wu, X.; Yang, X.; Gu, Z. Lithium Ionconducting Glass-ceramics of Li1.5Al0.5Ge1.5(PO4)3‑xLi2O (x = 0.0− 0.20) with Good Electrical and Electrochemical Properties. J. Am. Ceram. Soc. 2007, 90 (9), 2802−2806. (22) Xu, H.; Zong, J.; Ding, F.; Lu, Z.-W.; Li, W.; Liu, X.-J. Effects of Fe2+ Ion Doping on LiMnPO4 Nanomaterial for Lithium Ion Batteries. RSC Adv. 2016, 6 (32), 27164−27169. (23) Capiglia, C.; Mustarelli, P.; Quartarone, E.; Tomasi, C.; Magistris, A. Effects of Nanoscale SiO2 on the Thermal and Transport Properties of Solvent-free, Poly(ethylene oxide) (PEO)-based Polymer Electrolytes. Solid State Ionics 1999, 118 (1−2), 73−79. (24) Wieczorek, W.; Such, K.; Wyciślik, H.; Płocharski, J. Modifications of Crystalline Structure of PEO Polymer Electrolytes with Ceramic Additives. Solid State Ionics 1989, 36 (3), 255−257. (25) Capuano, F.; Croce, F.; Scrosati, B. Composite Polymer Electrolytes. J. Electrochem. Soc. 1991, 138 (7), 1918−1922. (26) Wieczorek, W.; Siekierski, M. A Description of the Temperature Dependence of the Conductivity for Composite Polymeric Electrolytes by Effective Medium Theory. J. Appl. Phys. 1994, 76 (4), 2220− 2226. (27) Shubha, N.; Prasanth, R.; Hng, H. H.; Srinivasan, M. Study on Effect of Poly (ethylene oxide) Addition and In-situ Porosity Generation on Poly (vinylidene fluoride)-glass Ceramic Composite Membranes for Lithium Polymer Batteries. J. Power Sources 2014, 267, 48−57. (28) Feng, J. K.; Lu, L.; Lai, M. O. Lithium Storage Capability of Lithium Ion Conductor Li1.5Al0.5Ge1.5(PO4)3. J. Alloys Compd. 2010, 501 (2), 255−258. (29) Seki, S.; Kobayashi, Y.; Miyashiro, H.; Mita, Y.; Iwahori, T. Fabrication of High-voltage, High-capacity All-solid-state Lithium Polymer Secondary Batteries by Application of the Polymer Electrolyte/Inorganic Electrolyte Composite Concept. Chem. Mater. 2005, 17 (8), 2041−2045. (30) Zaghib, K.; Armand, M.; Gauthier, M. Electrochemistry of Anodes in Solid-state Li-ion Polymer Batteries. J. Electrochem. Soc. 1998, 145 (9), 3135−3140. (31) Zeng, X. X.; Yin, Y. X.; Li, N. W.; Du, W. C.; Guo, Y. G.; Wan, L. J. Reshaping Lithium Plating/Stripping Behavior via Bifunctional Polymer Electrolyte for Room-Temperature Solid Li Metal Batteries. J. Am. Chem. Soc. 2016, 138 (49), 15825−15828. (32) Sun, B.; Xu, C.; Mindemark, J.; Gustafsson, T.; Edstrom, K.; Brandell, D. At the Polymer Electrolyte Interfaces: The Role of the Polymer Host in Interphase Layer Formation in Li-batteries. J. Mater. Chem. A 2015, 3 (26), 13994−14000. (33) Liu, S.; Imanishi, N.; Zhang, T.; Hirano, A.; Takeda, Y.; Yamamoto, O.; Yang, J. Lithium Dendrite Formation in Li/ Poly(ethylene oxide)-Lithium Bis(trifluoromethanesulfonyl)imide and N-Methyl-N-propylpiperidinium Bis(trifluoromethanesulfonyl)imide/Li Cells. J. Electrochem. Soc. 2010, 157 (10), A1092−A1098. (34) Liu, S.; Imanishi, N.; Zhang, T.; Hirano, A.; Takeda, Y.; Yamamoto, O.; Yang, J. Effect of Nano-silica Filler in Polymer Electrolyte on Li Dendrite Formation in Li/Poly(ethylene oxide)Li(CF3SO2)2N/Li. J. Power Sources 2010, 195 (19), 6847−6853. (35) Brissot, C.; Rosso, M.; Chazalviel, J. N.; Baudry, P.; Lascaud, S. In Situ Study of Dendritic Growth in Lithium/PEO-salt/Lithium Cells. Electrochim. Acta 1998, 43 (10−11), 1569−1574. (36) Sun, Y.-K.; Oh, S.-M.; Park, H.-K.; Scrosati, B. Micrometersized, Nanoporous, High-volumetric-capacity LiMn0.85Fe0.15PO4 Cathode Material for Rechargeable Lithium-ion Batteries. Adv. Mater. 2011, 23 (43), 5050−5054. (37) Oh, S.-M.; Myung, S.-T.; Choi, Y. S.; Oh, K. H.; Sun, Y.-K. Coprecipitation Synthesis of Micro-sized Spherical LiMn0.5Fe0.5PO4

ACKNOWLEDGMENTS This research was supported by funding from the National Natural Science Foundation of China (no. 21233004).

■

REFERENCES

(1) Kim, H.; Jeong, G.; Kim, Y.-U.; Kim, J.-H.; Park, C.-M.; Sohn, H.J. Metallic Anodes for Next Generation Secondary Batteries. Chem. Soc. Rev. 2013, 42 (23), 9011−9034. (2) Xu, W.; Wang, J.; Ding, F.; Chen, X.; Nasybulin, E.; Zhang, Y.; Zhang, J.-G. Lithium Metal Anodes for Rechargeable Batteries. Energy Environ. Sci. 2014, 7 (2), 513−537. (3) Zhamu, A.; Chen, G.; Liu, C.; Neff, D.; Fang, Q.; Yu, Z.; Xiong, W.; Wang, Y.; Wang, X.; Jang, B. Z. Reviving Rechargeable Lithium Metal Batteries: Enabling Next-generation High-energy and Highpower Cells. Energy Environ. Sci. 2012, 5 (2), 5701−5707. (4) Kozen, A. C.; Lin, C. F.; Pearse, A. J.; Schroeder, M. A.; Han, X. G.; Hu, L. B.; Lee, S. B.; Rubloff, G. W.; Noked, M. Next-Generation Lithium Metal Anode Engineering via Atomic Layer Deposition. ACS Nano 2015, 9 (6), 5884−5892. (5) Sun, F.; Zielke, L.; Markoetter, H.; Hilger, A.; Zhou, D.; Moroni, R.; Zengerle, R.; Thiele, S.; Banhart, J.; Manke, I. Morphological Evolution of Electrochemically Plated/Stripped Lithium Microstructures Investigated by Synchrotron X-ray Phase Contrast Tomography. ACS Nano 2016, 10 (8), 7990−7997. (6) Tsai, C. L.; Roddatis, V.; Chandran, C. V.; Ma, Q. L.; Uhlenbruck, S.; Bram, M.; Heitjans, P.; Guillon, O. Li7La3Zr2O12 Interface Modification for Li Dendrite Prevention. ACS Appl. Mater. Interfaces 2016, 8 (16), 10617−10626. (7) Gachot, G.; Grugeon, S.; Armand, M.; Pilard, S.; Guenot, P.; Tarascon, J.-M.; Laruelle, S. Deciphering the Multi-step Degradation Mechanisms of Carbonate-based Electrolyte in Li Batteries. J. Power Sources 2008, 178 (1), 409−421. (8) Thokchom, J. S.; Kumar, B. The Effects of Crystallization Parameters on the Ionic Conductivity of a Lithium Aluminum Germanium Phosphate Glass−ceramic. J. Power Sources 2010, 195 (9), 2870−2876. (9) Adachi, G.-y.; Imanaka, N.; Aono, H. Fast Li+ Conducting Ceramic Electrolytes. Adv. Mater. 1996, 8 (2), 127−135. (10) Fu, J. Fast Li+ Ion Conducting Glass-ceramics in the System Li2O-Al2O3-GeO2-P2O5. Solid State Ionics 1997, 104 (3−4), 191−194. (11) Xu, X.; Wen, Z.; Gu, Z.; Xu, X.; Lin, Z. Lithium Ion Conductive Glass Ceramics in the System Li1.4Al0.4(Ge1‑xTix)1.6(PO4)3 (x = 0−1.0). Solid State Ionics 2004, 171 (3−4), 207−213. (12) Kotobuki, M.; Hoshina, K.; Kanamura, K. Electrochemical Properties of Thin TiO2 Electrode on Li1+xAlxGe2‑x(PO4)3 Solid Electrolyte. Solid State Ionics 2011, 198 (1), 22−25. (13) Choi, B.-K. Optical Microscopy Study on the Crystallization in PEO-salt Polymer Electrolytes. Solid State Ionics 2004, 168 (1−2), 123−129. (14) Quartarone, E.; Mustarelli, P.; Magistris, A. PEO-based Composite Polymer Electrolytes. Solid State Ionics 1998, 110 (1−2), 1−14. (15) Armand, M. Polymer Solid Electrolytes-an Overview. Solid State Ionics 1983, 9, 745−754. (16) Zheng, Q.; Ma, L.; Khurana, R.; Archer, L. A.; Coates, G. W. Structure-property Study of Cross-linked Hydrocarbon/Poly(ethylene oxide) Electrolytes with Superior Conductivity and Dendrite Resistance. Chem. Sci. 2016, 7 (11), 6832−6838. (17) Stoeva, Z.; Martin-Litas, I.; Staunton, E.; Andreev, Y. G.; Bruce, P. G. Ionic Conductivity in the Crystalline Polymer Electrolytes PEO6:LiXF6, X = P, As, Sb. J. Am. Chem. Soc. 2003, 125 (15), 4619− 4626. (18) Jung, Y.-C.; Lee, S.-M.; Choi, J.-H.; Jang, S. S.; Kim, D.-W. All Solid-state Lithium Batteries Assembled with Hybrid Solid Electrolytes. J. Electrochem. Soc. 2015, 162 (4), A704−A710. (19) Shi, Q.; Xue, L.; Qin, D.; Du, B.; Wang, J.; Chen, L. Single Ion Solid-state Composite Electrolytes with High Electrochemical Stability 13701


Research Article

ACS Applied Materials & Interfaces Cathode Material for Lithium Batteries. J. Mater. Chem. 2011, 21 (48), 19368−19374. (38) Saravanan, K.; Ramar, V.; Balaya, P.; Vittal, J. J. Li(MnxFe1‑x)PO4/C (x = 0.5, 0.75 and 1) Nanoplates for Lithium Storage Application. J. Mater. Chem. 2011, 21 (38), 14925−14935.

13702


Suppression of Lithium Dendrite Formation by ... - ACS Publications

Recommend Documents